United States
Environmental Protection
Agency
Health Effects Research
Laboratory
Research Triangle Park NC 27711
EPA-600/9-81-001
January 1981
Research and Development
Proceedings of the
Research Planning
Workshop on Health
Effects of Oxidants
-------
EPA 600/9-81-001
January 1981
PROCEEDINGS OP THE
RESEARCH PLANNING WORKSHOP ON HEALTH EFFECTS OF OXIDANTS
Raleigh, North Carolina
January 28-30, 1980
Edited by
Northrop Services, Inc.
Environmental Sciences
P.O. Box 12313
Research Triangle Park, NC 27709
Project Officer
Jim Smith
Research Advisory and Special Studies Office
Health Effects Research Lab
Research Triangle Park, NC 27711
HEALTH EFFECTS RESEARCH LABORATORY
OFFICE OF RESEARCH AND DEVELOPMENT
U.S. ENVIRONMENTAL PROTECTION AGENCY
RESEARCH TRIANGLE PARK, NORTH CAROLINA 27711
-------
DISCLAIMER
This report has been reviewed by the Health Effects Research Laboratory,
U.S. Environmental Protection Agency, and approved for publication. Mention
of trade names or commercial products does not constitute endorsement or
recommendation for use.
ii
-------
ABSTRACT
Presentations at the Research Planning Workshop on Health Effects of
Oxidants (Raleigh, North Carolina, January 28-30, 1980) are documented. The
participants include scientists, administrators, and regulatory personnel from
the following agencies and institutions: U.S. Environmental Protection
Agency, Brookhaven National Laboratory, Oak Ridge National Laboratory,
Lawrence Berkeley Laboratory, Inhalation Toxicology Research Institute, EPA
Science Advisory Board, California Air Resources Board, University of
California - Los Angeles, University of Southern California, University of
California - Davis, University of California - Santa Barbara, and University
of Rochester.
The focus of the entire volume is the EPA-funded research that is planned
or in progress under Theme 1 ("Health Effects of Criteria and Non-Criteria
Pollutants from Fossil Fuel Combustion") of the Energy Interagency Health and
Ecological Effects Program. The relevance of component projects to EPA
regulatory activities is a frequent topic of informal discussion.
Chapter topics fall into five general categories: regulatory activities
and concerns, general overviews of research programs, specific animal studies,
specific human studies, and biomathematical modeling. Regulatory chapters
iii
-------
review the legislative framework within which EPA operates and suggest lines
of research relevant to regulatory needs. General program overviews are
presented for each of the agencies and institutions represented.
Approximately half of the total 40 chapters describe animal
experimentation. Most of the reported studies involve inhalation exposures to
ozone, nitrogen dioxide, or sulfur dioxide followed by examination of
biochemical, morphometric, or functional parameters. In vitro methods receive
some attention.
With respect to human studies, both epidemiologic and experimental
projects are reported. The experimental protocols involve inhalation
exposures to ozone or nitrogen dioxide followed by measurements of pulmonary
function. Chapters on biomathematical modeling discuss projections of human
absolute dosage and of the health effects of energy-related air pollution.
The final chapter is a transcript of the Panel Discussion which concluded
the workshop. The issues raised involve formal review of EPA-funded research
for scientific merit and relevance to regulatory activities.
iv
-------
CONTENTS
ABSTRACT iii
ABBREVIATIONS xiii
1. INTRODUCTION 1
Robert E. Lee
2. PROGRAM PLANNING FOR ENERGY HEALTH EFFECTS FROM CRITERIA
AND NON-CRITERIA POLLUTANTS 3
William Frietsch, III
Energy Interagency Health and Ecological Effects Program ... 3
Health Effects Program Area 7
Concluding Remarks 14
Reference - 15
Workshop Commentary 15
3. ROLE OF THE SCIENCE ADVISORY BOARD IN RESEARCH PLANNING
AND REVIEW 16
James L. Whittenberger
Introduction 16
Committee Makeup and Orientation 17
Steps in the Review Process 18
Recommendations 19
Concluding Remarks 19
Reference ..... 20
4. OVERVIEW OF REGULATORY AND COMPLIANCE REQUIREMENTS 21
Michael H. Jones
Introduction and Overview 21
Regulatory Options Under the Clean Air Act 24
Development of National Ambient Air Quality Standards .... 24
Issues in Development of the Ozone Standard 27
Current Research Needs 29
References • 30
-------
5. CRITERIA DOCUMENT REVIEW AND REVISION 31
Beverly E. Tilton
Introduction 31
The Framework 32
The Procedures 34
The Content 40
Current Research Needs 43
References 46
Workshop Commentary 46
6. TOXICOLOGICAL RESEARCH STRATEGIES IN RELATION
TO EPA REGULATORY NEEDS 47
Edward L. Alpen
Introduction 47
Comments on EPA Recommendations • 48
Reference ..... 49
Workshop Commentary 49
7. CORRELATING EPIDEMIOLOGIC, CLINICAL, AND BIOLOGICAL
RESEARCH ON OXIDANTS 53
David L. Coffin
Introduction 53
Previous Research .. 55
Promising Avenues for Future Research .... 57
Concluding Remarks 59
8. OVERVIEW OF CURRENT AND PLANNED RESEARCH
AT BROOKHAVEN NATIONAL LABORATORY 60
Robert T. Drew
Life Sciences Research at Brookhaven National Laboratory ... 60
Brookhaven Inhalation Toxicology Facility .... 61
9. INCORPORATING OXIDANTS IN ASSESSMENTS
OF ENERGY-RELATED HEALTH EFFECTS 70
Samuel C. Morris
The Nature of a Health Effects Assessment 70
Special Challenges of Energy-Related Assessments 72
Incorporating the Consideration of Oxidants 74
Research Needs • 76
Acknowledgment 77
References • 77
vx
-------
10. INTERACTIONS BETWEEN HYPERTENSION
AND OXIDANT AIR POLLUTANTS 78
Robert T. Drew, Daniel L. Costa,
Sonja Haber, and Junichi Iwai
Introduction ..... 78
Methods 79
Results and Discussion 81
Workshop Commentary 88
11. THE EFFECTS OF OZONE ON RAT ERYTHROCYTES
AFTER EXPOSURE IN VIVO 91
Karen M. Schaich and Donald C. Borg
Introduction 91
Initial Studies 96
Effect of Dietary Tocopherol 101
Discussion 106
Summary and Overview of Current Research 118
References 119
Workshop Commentary 121
12. PATHOGENESIS OF CHRONIC LUNG DISEASE:
THE ROLE OF TOXICOLOGICAL INTERACTION 126
Hanspeter Witschi
Introduction 126
Background 127
Methods and Results 128
Discussion 129
Continuing Studies 131
References 132
Workshop Commentary 133
13. OVERVIEW OF CURRENT AND PLANNED RESEARCH
BY THE RESPIRATORY UNIT, OAK RIDGE NATIONAL LABORATORY
PART 1. RAT TRACHEAL TRANSPLANT SYSTEM 135
Ann C. Marchok
Introduction and Background ..... 135
Controlled Delivery of the Test Agent(s) 137
Increased Specification of End Points 139
References 143
Workshop Commentary 144
vii
-------
14. OVERVIEW OF CURRENT AND PLANNED RESEARCH
BY THE RESPIRATORY UNIT, OAK RIDGE NATIONAL LABORATORY
PART 2. TRACHEAL WASHING SYSTEM
AND OXIDANT INHALATION EXPERIMENTS 145
Walden E. Dalbey
Introduction 145
Irritant Gases as Cofactors
in Nitrosamine-Induced Tumorigenesis 14*>
Tracheal Washing System 148
Effects of Inhaled Nitrogen Dioxide
on Pulmonic Lesions in Rats •
15. EARLY STAGES OF RESPIRATORY TRACT CANCER: A REVIEW 153
Carol A. Heckman, C. C. Scott/
A. C. Olson, and F. Snyder
Introduction 153
The Ether Lipid Marker 156
The Cell Shape Marker 160
Anomalies in Respiratory Tract Carcinogenesis 164
References 167
Workshop Commentary 169
16. CARDIOVASCULAR EFFECTS OF OZONE AND CADMIUM INHALATION
IN THE RAT 171
Nathanial W. Revis, T. Major, and Walden E. Dalbey
Introduction 171
Background 172
Methods and Results 174
Recommendations for Further Research ... 178
Workshop Commentary 179
17. EFFECTS OF NITROGEN DIOXIDE AND 3-METHYLFURAN INHALATION
ON THE SMALL AIRWAYS IN THE MOUSE 180
Wanda M. Haschek
Introduction 180
Methods 182
Results 183
Discussion 184
References 186
Workshop Commentary 187
viii
-------
18. OVERVIEW OF RESEARCH AT LAWRENCE BERKELEY LABORATORY 189
Edward L. Alpen
Introduction 189
Cocarcinogenic Effects of Nitrogen Dioxide
and Sulfur Dioxide in the" Mouse 189
Effects of Ozone on Serum Lipoprotein Concentrations
in the Guinea Pig *.. 194
19. OVERVIEW OF RESEARCH
AT THE UNIVERSITY OF CALIFORNIA - DAVIS 198
Marvin Goldman
Introduction 198
Approach to Research 199
Current Efforts 199
20. PULMONARY EFFECTS OF OZONE IN THE RAT AND MONKEY 201
Donald L. Dungworth
Introduction 201
Choice of Animal Models 201
Exposure Regimens 202
Studies of Adaptation
During Chronic Low-Level Exposure . 203
Other Studies 205
Comparison of Experimental and Epidemiologic
Data for Ozone and Sulfur Oxides 206
References 206
Workshop Commentary .. 207
t
21. BIOLOGICAL EFFECTS OF FLY ASH FROM COAL COMBUSTION 209
Otto Raabe
Introduction 209
Sample Collection 210
Sample Characterization • 210
Biological Test Systems ..•. 211
Exposure Techniques 211
Preliminary Results 213
Current Studies 214
Workshop Commentary 214
ix
-------
22. OVERVIEW OF RESEARCH
AT THE INHALATION TOXICOLOGY RESEARCH INSTITUTE .......... 216
Joe L. Mauderly
9 1 6
Introduction ........................ »•
Operation and Funding ....................
Facilities and Staff ..................... 217
Research Capabilities .................... 218
Current Projects
23. CELLULAR (IN VIVO) AND BIOCHEMICAL CHANGES
FOLLOWING INHALATION OF ACID SULFATES:
RAPID SCREENING TESTS TO DETERMINE
THE PULMONARY RESPONSE TO COMBINED POLLUTANT EXPOSURES ...... 222
Rogene F. .Henderson
Problem ........................... 222
Approach ......................... •• 223
Planned Research ....................... 225
Acknowledgments ...... ................. 226
References .......................... 226
Workshop Commentary .. ..... . ............. 227
24. RESPIRATORY TOXICOLOGY OF NITROGEN OXIDES ............. 229
John A. Pickrell
Problem ........................... 229
Approach ... ........................ 231
Planned Research ....... .... ............ 235
Acknowledgments ....... ..... ........... 238
References .......................... 239
Workshop Commentary .... ..... . ........... 239
25. ACUTE-CHRONIC LOSS OF LUNG FUNCTION
FOLLOWING INHALATION OF ACID SULFATES ............... 243
Steven A. Silbaugh
Problem ........................... 243
Approach ........................... 243
Planned Research ..... ........ <.... ..... 246
Acknowledgments ............. . ........ . 250
Reference .......................... 250
Workshop Commentary ......... . ........... 250
-------
26. LUNG CLEARANCE MECHANISMS
FOLLOWING INHALATION OF ACID SULFATES 256
Ronald K. Wolff
Problem 256
Approach 257
Planned Research ..... 263
Acknowledgments 264
References 264
Workshop Commentary 265
27. OVERVIEW OF CURRENT AND PLANNED RESEARCH
BY THE INHALATION TOXICOLOGY BRANCH 266
Donald E. Gardner
Introduction 266
Staff, Facilities, and Budget 266
Goals for Criteria Pollutant Research 268
Pollutants Studied 270
Model Systems in Use 270
Model Systems Under Development ... 272
Studies to Be Replicated 274
Collaboration with Other Investigators 275
References 276
28. SOME SPECIFIC STUDIES PLANNED
BY THE INHALATION TOXICOLOGY BRANCH 277
Judith A. Graham
Introduction 277
Chronic Nitrogen Dioxide Exposure
Using Mouse Infectivity Model 277
Pentobarbital-Induced Sleeping Time
in the Mouse 279
Sensitivity to Bronchoconstricting Agents 282
Studies to Identify Susceptible Populations 284
Workshop Commentary • 285
xi
-------
29. BIOMATHEMATICAL MODELING OF OXIDANT TOXICITY ........... 289
Fred J. Miller
Introduction ......................... 289
Nasal Pharyngeal Removal ................... 290
Gas Transport ........................
Modeling Process ........... • ........... 291
Modeling Absolute Dosage: Some Preliminary Results ..... 292
Quantitative Biochemical Data on the Mucous Layer:
A Major Research Need ...................
References ............ • ............. 297
Workshop Commentary ..................... 298
30. OVERVIEW OF CURRENT AND PLANNED RESEARCH
BY THE HUMAN STUDIES DIVISION ................... 3°1
Robert E. Lee
Introduction ................ . ........ 301
Epidemiologic Program .......... . ......... 302
Clinical Program ............. . ......... 303
Workshop Commentary ..................... 303
31. HUMAN PULMONARY ADAPTATION TO OZONE ................ 304
Edward D. Haak, Jr.
Introduction ......................... 304
Protocol ........................... 305
Results and Discussion .................... 306
Concluding Remarks ...................... 312
Workshop Commentary ..................... 313
32. OZONE-INDUCED HYPERREACTIVITY AS MEASURED
BY HISTAMINE CHALLENGE IN NORMAL HEALTHY SUBJECTS ......... 314
Milan J. Hazucha
Introduction .................. .. ..... 314
Protocol ..................... . ..... 315
Results ........................... 316
Discussion ...... . ....... . ........... 322
References .......................... 324
Workshop Commentary ..................... 325
xii
-------
33. RESPONSE OF NORMALS AND ASTHMATICS
TO LOW-LEVEL NITROGEN DIOXIDE 328
Joel F. Ginsberg
Introduction 328
Protocol 329
Preliminary Results 331
Preliminary Conclusions ... 338
References 340
Workshop Commentary ............... 340
34. HEALTH EFFECTS OF AIR POLLUTANTS
IN THE TEXAS GULF COAST AREA 341
Robert S. Chapman
Introduction ..... 341
Background 342
Planning Study 343
Feasibility Assessment ...... 347
Tentative Protocol 350
References 353
Workshop Commentary 353
35. OVERVIEW OF RESEARCH AND REGULATORY ACTIVITIES
OF THE CALIFORNIA AIR RESOURCES BOARD 358
John R. Holmes
Introduction ......... 358
Background 358
Research Program 359
California Air Pollution Standards 364
Workshop Commentary • 367
36. LUNG INJURY AND DEPLETION IN THE MOUSE
FOLLOWING EXPOSURE TO NITROGEN DIOXIDE 368
Russell P. Sherwin
Introduction '• 368
Background • 368
Methods and Results 370
Discussion 376
References 377
Addendum * 377
xiii
-------
37. PULMONARY AND PSYCHOPHYSIOLOGICAL EFFECTS OF OZONE 378
Steven M. Horvath
Introduction 378
Pulmonary Effects of Ozone
in Combination with Exercise 378
Pulmonary Habituation to Ozone
in Combination with Exercise 380
Psychophysiological Studies 381
References • 381
Workshop Commentary 382
38. EPIDEMIOLOGIC STUDIES OF OXIDANT
HEALTH EFFECTS IN THE LOS ANGELES AREA 383
Roger Detels
Introduction 383
Study Sites 384
Pulmonary Function Testing 385
Preliminary Results • 387
Discussion 389
Workshop Commentary .... 390
39. PROPOSED SCIENTIFIC PROGRAM
OF THE COOPERATIVE STUDY OF OXIDANT
HEALTH EFFECTS IN THE LOS ANGELES AREA 395
Stanley V- Dawson
Introduction 395
Epidemiologic Studies 395
The Integrated Program 398
Improvement of Population Exposure Estimates . 400
Human Laboratory Exposures 400
Animal Laboratory Studies 403
Overall Aspects 405
Workshop Commentary 406
40. PANEL DISCUSSION 411
MAILING ADDRESSES: AUTHORS AND DISCUSSANTS 428
xiv
-------
ABBREVIATIONS
Ab's
AChE
APH
ARB
BHT
BP
CASAC
CEQ
COMT
DEE
DEN
DMBA
DNA
DOE
2 , 3-DPG
ECAO
FEV1
FIV
FRC
FVC
FWS
FY
GPx
GRase
6SH
G6PD
Hb
Hct
HRP
ITB
IVPL
LALN
LDH
MAA
MAO
MDH
MetHb
MMAD
antibodies
acetylcholinesterase
acetylphenyl hydrazine
Air Resources Board
butylated hydroxytoluene
ben zopyrene
Clean Air Scientific Advisory Committee
Council on Environmental Quality
catechol O methyltransferase
Department of Energy and Environment
diethylnitrosamine
dimethylbenzanthracene
deoxyribonucleic acid
Department of Energy
2,3-diphosphoglycerate
Environmental Criteria and Assessement Office
forced expiratory volume
forced inspiratory volume
functional residual capacity
forced vital capacity
Fish and Wildlife Service
fiscal year
glutathione peroxidase
glutathione reductase
reduced glutathione
glucose-6-phosphate dehydrogenase
hemoglobin
hematocrit
horseradish peroxidase
Inhalation Toxicology Branch
isolated ventilated perfused lung
lung-associated lymph nodes
median lethal concentration
lactate dehydrogenase
macroaggregated albumin
monoamine oxidase
malate dehydrogenase
methemoglobin
mass median aerometric diameter
xv
-------
MMEF — maximal midexpiratory flow rate
MW — maximal voluntary ventilation
NAAQS — National Ambient Air Quality Standards
NAS — National Academy of Science
NASA — National Aeronautics and Space Administration
NBS — National Bureau of Standards
NIEHS -- National Institute of Environmental Health Sciences
NIH — National Institutes of Health
NIOSH — National Institute of Occupational Safety and Health
NMU — N-methyl-N-nitrosourea
NOAA — National Oceanic and Atmospheric Administration
NOEL — "no observed effects" level
NPSH — nonprotein sulfhydryl
OAQPS — Office of Air Quality Planning and Standards
OMB — Office of Management and Budget
ORD — Office of Research and Development
ORNL — Oak Ridge National Laboratory
PCH — polycyclic hydrocarbons
pdf — probability density function
PFT — pulmonary function test
PK — pyruvate kinase
PMN — polymorphonuclear
RAW — airway resistance
RBC's — red blood cells
RTP — Research Triangle Park
SAB — Science Advisory Board
S.D. — standard deviation
SOD — superoxide dismutase
SRAW -- specific airway response
TBA — thiobarbituric acid reactive products
TGV — thoracic gas volume
TIC — trypsin inhibitory capacity
TLC — thin layer chromatography
TVA — Tennessee Valley Authority
UCD — University of California - Davis
UCI — University of California - Irvine
UCLA — University of California - Los Angeles
USDA — U.S. Department of Agriculture
USGS — U.S. Geological Survey
UV — ultraviolet
VC — vital capacity
3MF -- 3-methylfuran
6PGD — 6-phosphoglycerate dehydrogenase
xvi
-------
1. INTRODUCTION
Robert E. Lee
Health Effects Research Laboratory, MD-51
Office of Research and Development
U.S. Environmental Protection Agency
Research Triangle Park, NC 27711
On January 28-30, 1980, the Health Effects Research Laboratory (Office of
Research and Development, U.S. Environmental Protection Agency, Research
Triangle Park, North Carolina) sponsored a Research Planning Workshop on
Health Effects of Oxidants. The meeting was held in Raleigh, North Carolina
and was attended by representatives of several governmental and private
agencies and institutions.
The workshop focused on the research projects that are planned or in
progress under Theme 1 ("Health Effects of Criteria and Non-Criteria
Pollutants from Fossil Fuel Combustion") of the Energy Interagency Health and
Ecological Effects Program. Over the three days, attendees heard
approximately 30 presentations on these efforts. Also presented were the
views of EPA personnel who will use the results of such research in the
regulatory process. The purpose of the present volume is to document each
presentation and the often lively discussion that followed.
-------
1. INTRODUCTION I»ee
The subsequent chapters underscore the fact that research is a very human
activity. In addition to the essential ingredients of money, facilities, and
instrumentation, it is still the individual researcher's creativity,
dedication, and often luck that may mean the difference between successful,
reproducible results and failure. The individual researcher will always be
the key to successful research. Accordingly, the basic workshop goals were to
encourage the exchange of knowledge, understanding, and interpretation of
oxidants health effects research and, most importantly, to provide a forum for
individual creativity, ideas, and perspective on the additional research
needed to develop the data base that is essential for protecting the public
health. To that end, the workshop was a notable success.
-------
2. PROGRAM PLANNING FOR ENERGY HEALTH EFFECTS
FROM CRITERIA AND NON-CRITERIA POLLUTANTS
William Frietsch, III
Energy Effects Division, RD682
Office of Research and Development
U.S. Environmental Protection Agency
Washington, DC 20460
ENERGY INTERAGENCY HEALTH AND ECOLOGICAL EFFECTS PROGRAM
Background; The First Five Years
The Energy Interagency Health and Ecological Effects Program involves
coordination and funding of energy-related research by EPA and 10 other
federal agencies. The program was created in 1973 after the Arab oil embargo
emphasized the impending acceleration of our country's energy development and
the need to address resultant environmental and health concerns. Creation of
the program was driven by what is commonly known as the "King-Muir Report"
written by Don King of the State Department and Warren Muir of the Council on
Environmental Quality (CEQ). With the assistance of Dr. Steven Gage and
personnel from about 20 federal agencies, King and Muir assembled an initial
five-year energy program that involved EPA and 10 other agencies. The Office
of Management and Budget (OMB) and CEQ approved the program for a five-year
period to begin with fiscal year (FY) 1975 and end with FY 1979.
-------
2. ENERGY HEALTH EFFECTS PROGRAM PLANNING Frietsch
Table 2-1 indicates the basic program that remained in place for the
first five years. The same agencies continue to participate. An important
distinction is that research at the National Laboratories, unlike research at
the listed agencies, is not funded on a "pass through" basis. ("Pass through"
refers to the procedure by which funds appropriated to EPA are "passed
through" to other agencies.) In the case of the National Laboratories,
existing projects were transferred to EPA.
Recent Planning; The Second Five Years
About 18 months ago, OMB and CEQ approved the continuation of the program
for a second five years. At that time, those of us involved in program
management decided to evaluate what program changes, if any, should be
implemented. Two major criteria for program development in the second five
years were identified: relevancy to agency programmatic needs and a client
relationship. The first of these refers not only to EPA needs but also to the
needs of the other participating pass through agencies. The second criterion
stems in part from frequent criticism that EPA research lacks a "client" or
"customer." We in EPA have became very cognizant of this criticism and have
found that, in order to survive the budget cycles, we need to have a client
who wants the research, who works with us throughout development and conduct
of the research, and who receives the output of the research.
With respect to the sources of research requirements, our first goal was
to respond to national energy needs and problems. We negotiated with other
-------
2. ENERGY HEALTH EFFECTS PROGRAM PLANNING Frietsch
TABLE 2-1. ENERGY INTERAGENCY HEALTH AND ECOLOGICAL EFFECTS PROGRAM:
PARTICIPANTS AND MAJOR PROGRAM AREAS FOR THE FIRST FIVE YEARS
Program Area
Air Transport Measurement
Ecological and and Health
Agency* Effects Transformation Instrumentation Effects
EPA X
National
Laboratories X
DOE X
FWS X
NBS
NOAA X X
TVA X X
NIOSH
NASA
USGS
USDA X
NIEHS
X
X
X
X
X
X
X
X
X
X
Definitions of agency abbreviations: DOE, Department of Energy; FWS, Fish
and Wildlife Service; NBS, National Bureau of Standards; NOAA, National
Oceanic and Atmospheric Administration; TVA, Tennessee Valley Authority;
NIOSH, National Institute of Occupational Safety and Health; NASA, National
Aeronautics and Space Administration; USGS, U.S. Geological Survey; USDA,
U.S. Department of Agriculture; NIEHS, National Institute of Environmental
Health Sciences.
-------
2. ENERGY HEALTH EFFECTS PROGRAM PLANNING Frietsch
agencies to ensure sufficient emphasis in such areas as acid rain, synthetic
fuels, air transport, visibility, and complex terrain.
Of course, consideration of EPA programmatic needs was also extremely
important in the planning effort. About three years ago, Dr. Gage set up four
pilot Research Committees (Mobile Sources, Drinking Water, Pesticides, and
Gaseous and Inhalable Particulates). In these Committees he brought together
staff members of EPA program offices and researchers working for the Office of
Research and Development (ORD). The objective was to jointly plan the ORD
program, to assure its responsiveness to program office needs.
This Research Committee system has worked extremely well: there are 12
Committees for FY 1980. With very minor exceptions, the whole ORD program is
strongly influenced by the Committees. From the researchers' point of view,
the Committees provide the advantage of close coordination with the program
offices (the "clients"). Researchers have a forum in which to express their
own particular concerns (e.g., the need for longer-term research). On the
other side of the fence, the advantage to the regulatory offices is clear:
I
they have a strong voice in deciding what research will be performed.
As a third source of research requirements, we considered the needs of
the other participating agencies. The whole character of the interagency
program is based on negotiation of research projects that have high priority
in terms of mutual concern. For example, DOE is extremely concerned about
synthetic fuels; so is EPA. Our challenge in negotiating such programs is to
-------
2. ENERGY HEALTH EFFECTS PROGRAM PLANNING Frietsch
identify the particular projects which are most relevant to both agencies.
The system forces and fosters a good working relationship between EPA and the
other participating agencies.
HEALTH EFFECTS PROGRAM AREA
Table 2-2 outlines the Health Effects program area for FY 1980. This
research effort is divided into a total of six "Themes"; funding totals ~$18
million. Theme 1, Health Effects from Criteria and Non-Criteria Pollutants
from Fossil Fuel Combustion, has so far been the major focus of our FY 1981
planning exercise (discussed below).
Theme 3, Fossil Fuel Leachate Hazards, refers to the leaching of
materials from fossil fuel combustion into waters and soils. The main EPA
clients for this work are the regulatory personnel concerned with drinking
water.
Theme 4, Toxicological Studies of Specific Energy-Related Agents,
consists of research on cadmium performed at the Comparative Animals Research
Laboratory at Oak Ridge National Laboratory. Originally, this Theme involved
four or five compounds, but we subsequently discovered that most of the other
agents are covered by ORD base program activities.
Theme 5, Relative Risk Assessments for Specific Energy Systems, seeks to
compare the known risks of conventional combustion with those of the newer
-------
TABLE 2-2. HEALTH EFFECTS PROGRAM AREA:
THEMES, PARTICIPANTS, AND FUNDING LEVELS FOR FISCAL YEAR 1980
Theme
Funding (thousands of dollars)8
HEKL-RTP HERL-CIN DOE.,
DOE NIEHS NIOSH
Total
by Theme
M
8
*
00
1. Health Effects of Criteria
and Non-Criteria Pollutants
from Fossil Fuel Combustion
2. Development and Validation
of Bioassay Screens and
Predictor Test Protocols
3. Fossil Fuel Leachate Hazards
4. Toxicological Studies of
Specific Energy-Related
Agents (Cadmium)
5. Relative Risk Assessments
for Specific Energy Systems
(Fluidized Bed Combustion
Versus Conventional Combustion)
6. Advanced Fossil
Fuel Cycle Hazards
2050
500
600
2368
4065
1150
2150
775
400
1217
500 1912
300
5568
7215
600
775
2312
1517
W
*1
I
CO
V
&
8
?a
>
3
a
Total by Agency:
2450
1100
7208
1217
3800
2212
17987
aDefinitions of agency abbreviations: HERL-RTP, Health Effects Research Laboratory (EPA-ORD), Research Triangle
Park, NC; HERL-CIN, Health Effects Research Laboratory (EPA-ORD), Cincinnati, OH; DOEX, Department of Energy
Projects transferred to EPA; DOE, Department of Energy; NIEHS, National Institute of Environmental Health
Sciences; NIOSH, National Institute of Occupational Safety and Health.
-------
2. ENERGY HEALTH EFFECTS PROGRAM PLANNING Frietsch
fluidized bed combustion. The National Institute of Occupational Safety and
Health has a major role in this Theme. There are ~14 epidemiologic projects
focusing on workplace environments.
Planning for 1980; Phase I
One of the first steps in our planning exercise for FY 1980 was to
discuss with the EPA program offices their energy-related research
requirements. Walter Barber of the Office of Air Quality Planning and
Standards (OAQPS) indicated a primary need for additional health effects
studies to support review and revision of National Ambient Air Quality
Standards for both Criteria and Non-Criteria Pollutants. Barber also stressed
the need for additional epidemiologic and clinical studies of human health.
His reason for this request was that he finds human data much more convincing
to Congress and to the public interest groups. These groups are not easily
convinced of the relevance of animal tojtficological data in establishing or
revising standards. Barber was careful to point out that he recognizes the
importance of animal data, and only advocates more of a balance between human
and animal studies.
For the Energy Health Effects program, at least, Barber's was a valid
criticism: at that time we had no epidemiology and a fairly low level of
clinical work. We responded by dedicating almost half of the 1980 Energy
Health Effects program to epidemiology and clinical work. We still have a
large animal toxicology program (as does the ORD base program). The key
-------
2. ENERGY HEALTH EFFECTS PROGRAM PLANNING Frietsch
question of what constitutes an "appropriate balance" between human and animal
work is a difficult one for which there may be no definitive answer.
Following these discussions with OAQPS, the Energy Health Effects program
was structured into the six Themes shown in Table 2-2. Theme 1 consists
partly of animal inhalation toxicology projects inherited from the National
Laboratories via an OMB mandate that $14 million worth of existing projects in
the health/ecological effects area be transferred to EPA. The intent of the
transfer was to broaden EPA capabilities in this area by allowing us to draw
upon the tremendous talent and expertise residing in the National
Laboratories. At the same time, ~$14 million worth of control technology work
was transferred from EPA to DOE. OMB attached several ground rules to the
transfer: Only projects involving conventional combustion would be
transferred to EPA; no human work would be affected. Also, no project changes
would take place in FY 1979 and 1980. The two-year moratorium on project
changes gave those of us in EPA a welcome opportunity to deal with our own
heavy workload and to become better acquainted with the National Laboratories
investigators.
With respect to any changes for FY 1981, we prefer to keep project
resources within the particular National Laboratory in which they reside, and
we are absolutely committed to maintaining the total resources within the
National Laboratory system. Thus, we might shift a particular project from
Brookhaven to Oak Ridge, for example, but we will not take those project
resources out of the National Laboratory system.
10
-------
2. ENERGY HEALTH EFFECTS PROGRAM PLANNING Frietsch
Also included in Theme 1 are some clinical projects carried over from the
first five years, as well as some new epidemiologic work*
Planning for 1981; Phase II
About six months ago we embarked upon Phase II of planning for the Energy
Health Effects program. As it would have been impossible to perform an
in-depth evaluation of the entire $18-million program, our strategy was to
focus on that part of the program which responds to OAQPS needs. With the
assistance of Richard Dowd we assembled a small Science Advisory Board (SAB)
subcommittee to assist us in this effort. Dr. James L. Whittenberger was
named Chairman of the subcommittee (see Chapter 3 of this volume) .
Our first public meeting was held in Washington on November 13-14, 1979.
This was an "information meeting": no particular advice was requested or
received. Rather, the intent was to explain our general planning philosophy
and procedure. The meeting agenda was split into three major areas. First,
OAQPS staff described their research requirements in as much detail as
possible. Secondly, ORD staff presented the ORD base program. In this, our
goal was to promote coordination and minimum overlap between the Energy Health
Effects program and the base program (which has been in existence for -15
years). The final presentation was on the Energy Health Effects program
itself. After outlining the program as it stood at that time, we asked the
SAB subcommittee to work with us over the next few meetings to identify
11
-------
2. ENERGY HEALTH EFFECTS PROGRAM PLANNING Frietsch
changes needed to make the program more responsive to OAQPS needs and to
ensure better coordination with the ORD base program.
Subsequent to this meeting, we developed what is called the "straw man"
document (Miller et al. 1979). The straw man document proposes changes in the
Energy Health Effects program to enhance its responsiveness to regulatory
needs and improve coordination with the ORD base program.
A second public meeting was held in Washington on December 18-19, 1979.
The entire two days were devoted to discussions of the straw man document.
The SAB subcommittee concurred with most of the recommendations contained in
that document and added recommendations of its own.
In January 1980, Dr. Charles Nauman, our health expert in the Energy
Effects Division, and this author made a quick "round robin" trip to the
National Laboratories that would be affected by the proposed changes. We
discussed the impact of the changes and received their feedback. In several
cases we discovered misinterpretations, on our part, of their work; thus, the
discussions were of considerable benefit.
In our visits to the National Laboratories, we repeatedly encountered
concern about the sometimes short-range nature of EPA programs. With regard
to Theme 1, it is significant that the Clean Air Act requires standard
revisions and revaluations on a five-year recurring basis. Walter Barber of
OAQPS has stressed the importance that Theme 1 programs continue for five-year
12
-------
2. ENERGY HEALTH EFFECTS PROGRAM PLANNING Frietsch
periods, to permit acquisition of the best information to make regulatory
decisions. And, indeed, we have endeavored to structure the Theme 1 projects
on five-year recurring bases.
Aside from the needs of EPA (OAQPS), our Phase II planning exercise has
also considered the needs of the other agency that participates in Theme 1:
the National Institute for Environmental Health Sciences (NIEHS). NIEHS has
expressed concern that the Theme 1 research program gives too much attention
to EPA regulatory needs. Clearly, we must work closely with NIEHS to ensure a
program consisting of projects that are of high priority to both agencies.
Following the Research Planning Workshop on Health Effects of Oxidants
(January 28-30, 1980; documented by this volume), our next step will be to
meet again with some of the National Laboratories investigators. A revised
straw man document will be published. The SAB subcommittee will hold another
public meeting on March 11-12, 1980, at-which we will present the final or
semifinal set of proposed changes. We hope to complete the final 1981 program
decisions in April or May of 1980.
Implementation and Management
Aside from the planning exercise and the resultant program changes, an
equally important question is how these changes will be implemented and the
entire Energy Health Effects program managed. With respect to management by
the Energy Effects Division, our very small staff represents a major
13
-------
2. ENERGY HEALTH EFFECTS PROGRAM PLANNING Frietsch
constraint: it is not possible to run an $18-million program with one or two
individuals. In addition, now that the Energy Health Effects program is more
closely coordinated with the ORD base program, it is increasingly apparent
that these efforts mesh and that most of the technical knowledge (for Theme 1,
at least) exists within ORD at Research Triangle Park, North Carolina (RTP).
Thus, our current plans are to assign project officers from the Health Effects
Research Laboratory at ORD-RTP where possible.
CONCLUDING REMARKS
A similar planning effort is being mounted for Theme 2 (a ~$7.5 million
research effort). Dr. Alan Moghissi from Dr. Gage's office will lead this
planning exercise. Once again we have requested SAB's assistance. To provide
continuity, we have asked Dr. Whittenberger to chair the Theme 2 SAB
subcommittee. Of course, the subcommittee membership will be different (a
different expertise is needed).
Major program changes are always difficult, and the Theme 1 planning
exercise could not have proceeded this far without tremendous cooperation from
all of the participants: the ORD base program, SAB, and particularly the
National Laboratories. We appreciate the feedback we have received, and we
will continue our efforts to acknowledge that feedback in proposals for
program revisions. With everyone's help, the Theme 1 planning exercise will
be drawn to a successful conclusion.
14
-------
2. ENERGY HEALTH EFFECTS PROGRAM PLANNING Frietsch
REFERENCE
Miller, F. J., J. A. Graham, M. Hazucha, C. Hayes, and W. Riggan. 1979.
Preliminary Relevance Review of EPA-DOE Projects on Health Effects of
Criteria and Non-Criteria Pollutants from Fossil Fuel Combustion (Theme
1). Unpublished document, Health Effects Research Laboratory, Office of
Research and Development, U.S. Environmental Protection Agency, Research
Triangle Park, North Carolina, December 12.
WORKSHOP COMMENTARY
D. C. Borg; As someone at one of the National Labs, I'd like to ask if there
is any intention to involve us in the "takeoff" as well as the "landing."
Will you involve us in the planning exercise or will you just tell us its
results?
W. Frietsch; Yes, we do plan to involve you in the "takeoff." In fact, we
plan to hold a workshop in February 1980 at which all Theme 2 participants
will give verbal presentations. Also present will be representatives of our
base program and (if assembled) the SAB subcommittee.
As we go through this planning exercise, we're learning that it would be
very desirable to involve you more in the "up front" end of it. That's
exactly our intent in having a workshop to kick off the whole planning
exercise. We will also require (at the meeting) some written material, as we
did in the first Theme. But we definitely want people to give verbal
presentations and answer questions, etc. about their work.
15
-------
3. ROLE OF THE SCIENCE ADVISORY BOARD IN RESEARCH PLANNING AND REVIEW
James L. Whittenberger
Department of Physiology
School of Public Health
Harvard University
665 Huntington Avenue
Boston, MA 02115
INTRODUCTION
This report briefly describes the role of the Science Advisory Board
(SAB) in the ongoing revisions to Theme 1 of the Energy Health Effects program
(see Chapter 2 of this volume). It is important to bear in mind that SAB's
role is advisory; we are assisting EPA in redirecting Theme 1 to make it more
responsive to the needs of the Office of Air Quality Planning and Standards
and to achieve a better balance among the components of relevant animal
toxicological, clinical, and epidemiologic research.
Having participated in the review of the criteria document leading to
revision of the National Ambient Air Quality Standard for photochemical
oxidants, and having participated in the Health Effects Research Review Group
which, at the request of Congress, examined health effects research in
relation to various EPA programmatic needs, this author feels very strongly
about the importance of providing EPA regulatory personnel with the very best
16
-------
3. ROLE OF THE SCIENCE ADVISORY BOARD Whittenberger
possible data. Those who work with criteria documents and who consider the
problem of translating scientific information (particularly health effects
information) into standards realize that, among the large number of studies
cited in a given criteria document, usually only a very small number are
really appropriate for standard-setting. It is thus extremely important that
those few studies be of high quality. Making such assessments of scientific
merit is perhaps the most difficult problem in planning research.
COMMITTEE MAKEUP AND ORIENTATION
SAB participation in the Theme 1 planning exercise dates from September
12, 1979, when Mr. Frietsch and I met to discuss SAB's role. The SAB
subcommittee assembled as a result of that meeting consists of Dr. Jack
Hackney, Dr. Ian Higgins, Dr. Daniel Menzel, Dr. Richard Riley, and myself as
Chairman.
Most of the subcommittee members have a background in reviewing research
protocols as study section members; thus, we prefer to know a lot about each
project, including the methods to be used, the techniques of data analysis and
interpretation, etc. Accordingly, in the Theme 1 exercise, assessing
scientific merit has remained (for. us, at least) an essential part of our job
even though the official subcommittee charge relates to research planning
rather than evaluation. Because of this orientation, we have experienced a
number of problems in reviewing both the written material given to us as well
as presentations at the meetings we have attended. For example, we understand
17
-------
3. ROLE OF THE SCIENCE ADVISORY BOARD Whittenberger
that there will be two new epidemiologic studies: one in El Paso and one in
the Southern California Air Basin. We have not been able to find out much
about either of these studies; therefore, we feel we can express no judgment
whatsoever about them.
STEPS IN THE REVIEW PROCESS
Between the November 13-14 and December 18-19 (1979) public meetings (see
Chapter 2 of this volume), EPA conducted its own review of the Theme 1
projects. The EPA staff members involved in this exercise were Dr. Fred
Miller, Dr. Judith Graham, Dr. Milan Hazucha, Dr. Carl Hayes, and Dr. Wilson
Riggan. These EPA scientists understood that they were to make a relevance
review rather than a scientific merit review. Personally, I question whether
scientific merit can be separated from relevance. All too often I have heard
that a particular study is subject to various significant criticisms and may
be of doubtful validity, but "cannot be ignored in standard-setting." My view
is: Pending replication, such a study not only can be but should be ignored
in standard-setting.
At the December 18-19 meeting our subcommittee received the EPA staff
report (Miller et al. 1979), which included both general recommendations and
specific recommendations on individual projects. In general, the subcommittee
agreed with these recommendations. We could not determine, however, the
extent to which the various projects had undergone peer review on an
individual basis. We knew that the National Laboratories employ laboratory
18
-------
3. ROLE OF THE SCIENCE ADVISORY BOARD Whittenberger
reviews (program reviews); we could not ascertain the extent to which they
employ project reviews.
RECOMMENDATIONS
Following the December meeting, the SAB subcommittee assembled our own
tentative recommendations which we consider to be consistent with the EPA
recommendations. First, we recommend that rigorous external as well as
internal project review be undertaken for fiscal year 1981 and thereafter. We
strongly endorse the exchange between investigators of protocols for clinical
research. We find certain projects to be of low priority (at least, as EPA
priorities have been defined to us), and we recommend a major reprogramming of
effort. Redirection of Theme 1 projects can lead, we believe, to an increase
in clinical and epidemiologic work. At present, our subcommittee lacks
sufficient information to determine what that balance should be. One of the
principal goals of future meetings will be-to obtain such information.
CONCLUDING REMARKS
With regard to comments by Frietsch (Chapter 2 of this volume) concerning
the needs of other participating agencies, it is not clear why the needs of
any agency other than EPA should be of primary concern. In this exercise it
is EPA requirements that are at issue. EPA operates under laws that require
its research to be relevant to standard-setting and other forms of regulation;
if that isn't the case (regardless of whether the studies are long-term or
19
-------
3. ROLE OF THE SCIENCE ADVISORY BOARD Whittenberger
short-term), the research is simply not appropriate for EPA's purposes. We in
the SAB subcommittee view the Research Planning Workshop on Health Effects of
Oxidants (January 28-30, 1980; documented by this volume) as an opportunity to
become much better informed about the individual projects that comprise Theme
1 of the Energy Health Effects program.
REFERENCE
Miller, F. J., J. A. Graham, M. Hazucha, C. Hayes, and W. Riggan. 1979.
Preliminary Relevance Review of EPA-DOE Projects on Health Effects of
Criteria and Non-Criteria Pollutants from Fossil Fuel Combustion (Theme
1). Unpublished document. Health Effects Research Laboratory, Office of
Research and Development, U.S. Environmental Protection Agency, Research
Triangle Park, North Carolina, December 12.
20
-------
4. OVERVIEW OF REGULATORY AND COMPLIANCE REQUIREMENTS
Michael H. Jones
Strategies and Air Standards Division, MD-12
Office of Air Quality Planning and Standards
U.S. Environmental Protection Agency
Research Triangle Park, NC 27711
INTRODUCTION AND OVERVIEW
This report attempts to provide some insight into the regulatory process
and into how EPA's Office of Air Quality Planning and Standards (OAQPS), as a
customer for the output of the scientific community, uses information
generated by scientists in developing standards and regulations. The key
point, perhaps, is that setting a standard is not strictly a scientific
enterprise. Certainly at the heart of a standard's development is the
scientific information upon which a choice must be based. It is important to
recognize, however, that there are other ingredients in setting National
Ambient Air Quality Standards (NAAQS) (or any other standards, for that
matter).
Interpretation of the Clean Air Act and the legislative history is not an
activity of OAQPS alone. We rely on EPA's Office of General Counsel and their
attorneys to provide us with information on how far to go (if at all) in
21
-------
4. REGULATORY AND COMPLIANCE REQUIREMENTS Jones
considering economics, for example, when selecting a standard. Do we consider
the issue of attainability? Although we're on record on most of these issues,
we nonetheless must rely on the guidance of the Office of General Counsel and
their attorneys to give us the proper direction. As an example, probably the
key procedural issue during review of the photochemical oxidants standard was
the so-called "Shy Panel" that was convened early in the regulatory process.
Critics argued that procedural irregularities were associated with that
meeting. This controversy flags the importance of working closely with our
attorneys to ensure that our actions are proper and in accordance with the
law.
There is, of course, the obvious issue of interpretation of the health
evidence. This is certainly the area where we spend most of our time and
where we need the most communication with the scientific community.
Control effectiveness is another scientific element in development of a
standard. In the case of ozone, for example, atmospheric chemistry bears
importantly on the effectiveness of control options.
Determining the numbers of persons at risk is yet another scientific
enterprise. This information is of interest to the Administrator in choosing
the level of a standard.
An understanding of economic impacts is also important, even if that
information is not used in setting a standard level. While it is our
22
-------
4. REGULATORY AND COMPLIANCE REQUIREMENTS Jones
interpretation of the Clean Air Act that economics not be considered in
setting NAAQS, it is also our position that economic impacts be estimated and
made clear to the public so that the electorate can, if it sees fit, act
through elected officials to modify the way the law is structured. In other
words, if the ozone standard is set at 0.12 ppm and provides a certain amount
of protection, it is important that the public realize the economic cost
associated with that protection. If the electorate feels so inclined, it can
make choices for new directions. Therefore, we strive to delineate economic
impacts as accurately and definitively as possible.
The decisionmaker (the Administrator of EPA) represents the
Administration. His task involves a judgmental element and is not, as noted
earlier, a strictly scientific enterprise. There is a margin of safety. How
that margin of safety is characterized and what magnitude of risk the
decisionmaker will accept are important issues, and we must do our best to
clearly present the relevant information to him. Concerns in this area
include marginal choices on what constitutes an adverse health effect. In the
case of ozone, there is certainly disagreement not only among policymakers but
also within the scientific community as to what constitutes an adverse effect.
It is unlikely that disagreements of this sort will ever be totally resolved.
None the less, judgments must be made that reflect our best understanding of
science and of the health policy of our elected officials.
23
-------
4. REGULATORY AND COMPLIANCE REQUIREMENTS Jones
REGULATORY OPTIONS UNDER THE CLEAN AIR ACT
NAAQS are not the only regulatory tool available to EPA for control of
air pollutants. Section 112 of the Clean Air Act provides for National
Emission Standards for Hazardous Air Pollutants; these exist for mercury,
beryllium, asbestos, and vinyl chloride. Standards of Performance for New
Stationary Sources (Section 111) are technology-based standards that are, in
many cases, designed to support NAAQS. Emission Standards for Moving Sources
(Section 202) are likewise tied in and support the attainment of NAAQS.
Other options include the regulation of fuels and fuel additives,
inspection maintenance programs, and related control strategies. Also, the
Emergency Powers clause (Section 303) permits the Administrator to take
immediate action in certain situations if there exists significant risk to
individuals.
All of these regulatory options are available to the Administrator to
control air pollutants. The Clean Air Act provides clear guidance as to how
he can and cannot use each option.
DEVELOPMENT OP NATIONAL AMBIENT AIR QUALITY STANDARDS
In the 1970 Clean Air Act, EPA is tasked to list those ubiquitous
pollutants which, in the Administrator's judgment, have adverse effects on
public health and welfare. Primary standards protect public health; secondary
24
-------
4. REGULATORY AND COMPLIANCE REQUIREMENTS jones
standards protect the public welfare against known or anticipated adverse
effects. The second step is to issue criteria documents containing the latest
scientific knowledge on identifiable effects of pollutants on public health
and welfare. After publication of each criteria document, a standard is
proposed. Next is a period of public comment, followed by promulgation.
Finally, there is a requirement for periodic review and (where appropriate)
revision of each criteria document and standard.
These provisions are changed and updated in the 1977 Clean Air Act
Amendments. With regard to NAAQS, the 1977 Amendments set a timetable for
review of all the standards by the end of the 1980 calendar year. The
Amendments further require (1) review on a five-year basis of all NAAQS, (2)
issuance of nitrogen dioxide criteria for a short-term standard, and (3)
establishment of a scientific committee to review criteria and standards.
Determination of a primary standard level entails definition of the
adverse effect against which the standard is to protect, definition of the
population or group which is most sensitive to the particular effect, and
evaluation of all existing medical evidence set forth in the criteria
document. In addition, outstanding uncertainties must be defined and
reasonable provision made for scientific and medical knowledge not yet
acquired.
A number of judgmental choices need to be made in conjunction with the
scientific community. EPA interprets the Clean Air Act and the legislative
25
-------
4. REGULATORY AND COMPLIANCE REQUIREMENTS Jones
history to indicate that Congress has made a policy judgment in favor of
prevention of harm and in favor of exercising caution with regard to the
public health.
A margin of safety is necessary to protect against hazards which research
has yet to define. This is a legal as well as a scientific issue. In my
view, a simple absence of evidence does not confirm that safety exists. While
this is a very difficult issue, it is nonetheless something that the
Administrator must consider in selecting the final level of a standard.
To protect the most sensitive groups in the population, the Administrator
must weigh risks to public health at pollutant levels below those convincingly
shown to impact health. Once again, assessment of risk to public health is
based on scientific evidence but not based on fact alone. In judging what
constitutes an adequate margin of safety, uncertainties in the evidence,
conflicting evidence, extrapolation, trends of known facts, and projections
from imperfect data must all be weighed. We simply cannot ignore incomplete
or inconclusive evidence.
Scientific information must be generated and then presented and compiled
into a criteria document. Subsequently there is involvement of the scientific
community and review groups to ensure the adequacy of that document and that
we have the best possible basis upon which to set a standard.
26
-------
4. REGULATORY AND COMPLIANCE REQUIREMENTS Jones
Since the 1977 Amendments, a staff paper has been employed to signal to
the public, prior to proposal, the way in which EPA interprets the medical
evidence in the criteria document. The staff paper is a public document which
is subject to public review. In the carbon monoxide staff paper, we did our
best to present to the Clean Air Scientific Advisory Committee (CASAC) of the
Science Advisory Board (SAB) how we plan to use the criteria document
information in the regulatory process.
The next step is an internal review process within the Agency. Then the
Administrator makes a decision on the standard. Following the Administrator* s
decision, a standard is proposed.
To indicate the length of time involved, our meeting with CASAC(SAB) to
discuss the carbon monoxide staff paper was held on June 11, 1979; hopefully
we'll be able to propose a standard in February or March of 1980. Development
of NAAQS is not a speedy process. In the case of the original oxidants
standard, it was probably a 30-month period from the time the scientific
information was compiled and presented in the first draft of the criteria
document until we proposed or promulgated the standard.
ISSUES IN DEVELOPMENT OF THE OZONE STANDARD
The 1977 Clean Air Act Amendments required EPA to review the 1971
National Ambient Air Quality Standard for Photochemical Oxidants. Today, the
revised standard (the "Ozone" standard) is probably the most expensive
27
-------
4. REGULATORY AND COMPLIANCE REQUIREMENTS Jones
environmental program in the country, with an estimated $5-10 billion annual
cost. Most of that cost is associated with the emission standards on
automobiles.
Of course, there was (and is) a wide divergence of medical opinion with
regard to the scientific evidence on ozone. Therefore, the scientific
community's assistance was essential in providing our Administrator with the
kind of information that he needed in order to make a reasoned judgment.
Besides medical questions, a number of other issues surfaced during
development of the ozone standard. One such issue was severity of effects. A
reinterpretation of evidence from the study upon which the original oxidants
standard was based (the Schoettlin/Landau study) led to designation of a new
estimated effect level in the revised criteria document. There was a
continuing debate with the attorneys as to whether that action was appropriate
or inappropriate. Review of the criteria document by CASAC(SAB) pointed up
differences of opinion as to the quality of the document.
Another issue centered on measurement methods for singlet oxygen. The
State of Texas brought to our attention what it considered to be an important
finding: that all the clinical studies were simply invalid due to artifact
production by the ozone generators. This issue had to be resolved prior to
promulgation of the standard.
28
-------
4. REGULATORY AND COMPLIANCE REQUIREMENTS Jones
The Council of Economic Advisors, after reviewing the standard, suggested
that the level could be relaxed above what was proposed. The Council felt
that the Federal Government should consider cost across public health and
safety fields: For example, are health dollars spent to protect people from
ozone as cost-effective as those that could be spent on highway guard rails?
We simply don't feel that this is an issue, because the Clean Air Act
emphasizes the need for protection of public health from air pollutants.
Other issues included: economics, our modification of the nomenclature
from "photochemical oxidants" to "ozone," the contribution of natural ozone,
and adequate margins of safety.
CURRENT RESEARCH NEEDS
During development of the ozone standard, we flagged several areas in
which a need for additional scientific evidence exists. For example, to help
resolve the issue of what constitutes an adverse health effect, we need to
improve the characterization of the kind and magnitude of various pulmonary
function detriments, biochemical changes, and other responses.
We require a better understanding of the importance of the animal
studies, particularly with regard to increased resistance to bacterial
infection. Although (to the author's knowledge) this effect has not been
demonstrated in clinical studies, we remain concerned due to the seriousness
29
-------
4. REGULATORY AND COMPLIANCE REQUIREMENTS Jones
of any such potential effect. Any information on its relevance to human
health is certainly needed.
Determination of the long-term chronic effects of photochemical oxidants
and of the potential for such damage as accelerated aging of body tissues,
chromosomal damage, mutagenesis, carcinogenesis, etc. is another area in which
good information is sparse. Often we don't understand the significance or
importance of the effects that have been observed.
Identification of particularly sensitive groups and investigation of
adverse effects on materials are additional scientific needs.
REFERENCES
\
Clean Air Act. 42 USC 7401 et seq.
Clean Air Act Amendments of 1977. Public Law 95-95, August 7, 1977.
30
-------
5. CRITERIA DOCUMENT REVIEW AND REVISION
Beverly E. Tilton
Environmental Criteria and Assessment Office, MD-52
Office of Research and Development
U.S. Environmental Protection Agency
Research Triangle Park, NC 27711
INTRODUCTION
The review and revision of criteria documents constitute a broad topic,
but may be considered to consist of three aspects: the framework, the
procedures, and the content. The framework is made up of the Clean Air Act
requirements, the organizational structure of the pertinent EPA organizations,
and the division of responsibilities among those organizations. The
procedures followed are those stipulated by the Clean Air Act and those
established by EPA on the bases of precedent, practice, and practicality. The
content of a criteria document is determined by the Clean Air Act, by the
nature of the pollutant under consideration, and by the intended use of the
document.
31
-------
5. CRITERIA DOCUMENT REVIEW AND REVISION Tiltqn
THE FRAMEWORK
Section 108 of the Clean Air Act as amended in 1977 specifies that the
Administrator issue air quality criteria for each air pollutant
(a)(1)(A) emissions of which, in his judgment, cause or
contribute to air pollution which may reasonably be anticipated to
endanger public health or welfare;
(B) the presence of which in the ambient air results from
numerous or diverse mobile or stationary sources;...
The Act further specifies that
(2) ...Air quality criteria for an air pollutant shall
accurately reflect the latest scientific knowledge useful in
indicating the kind and extent of all identifiable effects on public
health or welfare which may be expected from the presence of such
pollutant in the ambient air, in varying quantities. The criteria
for an air pollutant, to the extent practicable, shall include
information on—
(A) those variable factors (including atmospheric conditions)
which of themselves or in combination with other factors may alter
the effects on public health or welfare of such air pollutant;
(B) the types of air pollutants which, when present in the
atmosphere, may interact with such pollutant to produce an adverse
effect on public health or welfare; and
(C) any known or anticipated adverse effects on welfare.
It should be noted at the outset that the Clean Air Act does not employ
the terminology "criteria document." It specifies the "issuance" of criteria,
but does not specify the form or format in which such criteria are to be
32
-------
5. CRITERIA DOCUMENT REVIEW AND REVISION Tilton
issued or disseminated. Criteria documents as they presently exist are the
product of an evolving process that had its genesis in legislation prior to
the Clean Air Act and that was first manifested in the criteria documents
issued in 1970. Early legislation did specify the publication of information
on the effects of and control of air pollutants, but subsequent legislation
did not refer specifically to publication of criteria.
Not only does it not specify the form in which criteria are to be issued,
the Clean Air Act also does not define "air quality criteria." By dictionary
definition, criteria are "rules, tests, or indices that can be used as a basis
for decisions or judgments." It is clear from the portion of Section 108
cited above that criteria are to be descriptions of the effects of air
pollutants on man and his environment. Combining the dictionary definition
with the definition implicit in Section 108, "air quality criteria" are
descriptions of air pollution effects presented in a manner that renders them
suitable as the scientific basis for National Ambient Air Quality Standards
(NAAQS). Thus, in practice, air quality criteria are descriptions of
cause-and-effect relationships expressed, to the extent possible, as
dose-response functions. Criteria documents, then, while they may serve a
number of worthwhile purposes, are prepared primarily as the vehicle for
issuance of the criteria that serve as the scientific basis for NAAQS. Air
quality criteria, as indicated in Section 108, must reflect the latest
scientific information available that pertains to the effects of an air
pollutant, direct or indirect, on man and his environment. Scientific
information needed to formulate control strategies—such as that useful in
33
-------
5. CRITERIA DOCUMENT REVIEW AND REVISION Tilton
defining atmospheric transformations and source-receptor relationships—is to
be included only to the extent practicable.
The legislative framework that determines the basic philosophy and
content of criteria documents has been discussed above. A brief examination
follows of the organizational framework within which document review and
revision proceed.
Responsibility for the preparation of air quality criteria documents
rests with the Environmental Criteria and Assessment Office (ECAO). Located
at Research Triangle Park, North Carolina, ECAO is one of two such offices
within the Office of Health and Environmental Assessment; the latter is in
turn part of the Office of Research and Development (ORD) (Figure 5-1).
Responsibility for deriving and promulgating NAAQS, and for issuing control
techniques information, lies with the Office of Air Quality Planning and
Standards (OAQPS), which is part of the Office of Air, Noise and Radiation.
Because the Clean Air Act stipulates that standards be proposed and
information on control techniques be issued at the same time that criteria are
issued, close coordination between ECAO and OAQPS is required. Most
day-to-day communication occurs directly between these offices.
THE PROCEDURES
The responsibilities of ECAO versus those of OAQPS in the review and
revision of criteria and standards are shown in Figure 5-2. Segments in which
34
-------
OFFICE OF RESEARCH AND DEVELOPMENT
OFFICE OF RESEARCH PROGRAM
MANAGEMENT
OPERATIONS OFFICE
TECHNICAL INFORMATION OFFICE
u>
en
CENTER FOR ENVIRONMENTAL
RESEARCH INFORMATION
CINCINNATI. OH
ASSISTANT ADMINISTRATOR FOR
RESEARCH AND DEVELOPMENT
OFFICE OF THE PRINCIPAL SCIENCE
ADVISOR
OFFICE OF MONITORING AND
TECHNICAL SUPPORT
PROGRAM OPERATIONS STAFF
NATIONAL WORKFORCE
DEVELOPMENT STAFF
QUALITY ASSURANCE AND
MONITORING SVSTEMS DIVISION
TECHNICAL SUPPORT DIVISION
ENVIRONMENTAL MONITORING
SVSTEMS LABORATORY
RESEARCH TRIANGLE PARK, NC
ENVIRONMENTAL MONITORING
AND SUPPORT LABORATORY
CINCINNATI. OH
ENVIRONMENTAL MONITORING
SVSTEMS LABORATORY
LAS VEGAS. NEVADA
OFFICE OF ENVIRONMENTAL
ENGH. AND TECH.
PROGRAM OPERATIONS STAFF
WASTE MANAGEMENT DIV.
ENERGY PROCESSES DIVISION
INDUSTRIAL AND EXTRACTIVE
PROCESSES DIVISION
MO GRAM INTEGRATION
AND POLICY STAFF
INDUSTRIAL ENVIRONMENTAL
RESEARCH LABORATORY
RESEARCH TRIANGLE PARK. NC
INDUSTRIAL ENVIRONMENTAL
RESEARCH LABORATORY
CINCINNATI. OH
MUNICIPAL ENVIRONMENTAL
RESEARCH LABORATORY
CINCINNATI, OH
SENIOR ORD OFFICIAL.RESEARCH
TRIANGLE PARK. NC
SENIOR ORD OFFICIAL
CINCINNATI. OH
OFFICE OF ENVIRONMENTAL
PROCESSES AND EFFECTS RESEARCH
PROGRAM OPERATIONS STAFF
AGRICULTURE AND NON POINT
SOURCES MANAGEMENT DIVISION
MEDIA DUALITY MANAGEMENT
DIVISION
ENERGY EFFECTS DIVISION
ECOLOGICAL EFFECTS DIVISION
ENVIRONMENTAL SCIENCES
RESEARCH LABORATORY
RESEARCH TRIANGLE PARK. NC
ENVIRONMENTAL RESEARCH
LABORATORY, ATHENS. GA
ROBERT S. KEHR ENVIRONMENTAL
RESEARCH LABORATORY, ADA, OK
ENVIRONMENTAL RESEARCH
LABORATORY, CORVALLIS, OREGON
ENVIRONMENTAL RESEARCH
LABORATORY, DULUTH, MN
ENVIRONMENTAL RESEARCH
LABORATORY, NARRAGANSETT,
Rl
ENVIRONMENTAL RESEARCH
LABORATORY. GULF BREEZE. Fl
OFFICE OF HEALTH RESEARCH
PROGRAM OPERATIONS STAFF
HEALTH EFFECTS DIVISION
HEALTH EFFECTS RESEARCH
LABORATORY
RESEARCH TRIANGLE PARK. NC
HEALTH EFFECTS RESEARCH
LABORATORY CINCINNATI. OH
NATIONAL CENTER FOR
TOXICOLOGICAL RESEARCH
JEFFERSON. AR
OFFICE OF HEALTH AND
ENVIRONMENTAL ASSESSMENT
CARCINOGEN ASSESSMENT
GROUP
EXPOSURE ASSESSMENT GROUP
REPRODUCTIVE EFFECTS
ASSESSMENT GROUP
ENVIRONMENTAL CRITERIA
AND ASSESSMENT OFFICE
RESEARCH TRIANGLE PARK. NC
ENVIRONMENTAL CRITERIA
AND ASSESSMENT OFFICE
CINCINNATI, OH
o
S
a
8
CO
H
i
Figure 5-1. Office of Research and Development, U.S. Environmental Protection Agency.
§
-------
O
B
ORO AND
SCIENTIFIC
COMMUNITY
OAQPS (OANR)
B
ECAO (ORD)
ECAO (ORD)
OAQPS (OANR)
OAQPS (OANR)
REGULATORY IMPACT
ANALYSIS
u>
SCIENTIFIC
RESEARCH
1
l
CRITERIA
DOCUMENT
1 i
i '
t
ORD
RESEARCH
COMMITTEES
STAFF PAPER
INTERPRETING KE>
STUDIES IN
CRITERIA DOCUMEN
1 '
PUBLIC AND
SCIENTIFIC PEER
REVIEW
f
T
REGULATORY
DECISION
PACKAGE
i
\ r
PUBLIC AND
SCIENTIFIC PEER
REVIEW
AGENCY
REVIEW
ADMINISTRATOR'S
DECISION
1 '
PROPOSAL
PUBLIC MEETINGS
AND COMMENTS
AGENCY
REVIEW
REGULATORY DECISION
PACKAGE REFLECTING
PUBLIC COMMENTS
(ORD) • OFFICE OF RESEARCH AND DEVELOPMENT
(ECAO) - ENVIRONMENTAL CRITERIA AND ASSESSMENT OFFICE
(OANR) • OFFICE OF AIR, NOISE, AND RADIATION
(OAQPS) • OFFICE OF AIR QUALITY PLANNING AND STANDARDS
ADMINISTRATOR
DECISION
PROMULGATION
Figure 5-2. National Ambient Air Quality Standard-setting process.
O
a
-------
JL» CRITERIA DOCUMENT REVIEW AND REVISION Tilton
ECAO has primary responsibility or participates are highlighted in bold
outline. ECAO has responsibility for preparing the criteria document, for
reviewing and incorporating (as needed) comments resulting from the public
review process, and for helping OAQPS prepare the staff paper that forms a
significant part of the regulatory information developed for NAAQS proposal
and subsequent promulgation. In addition, ECAO provides technical assistance
to OAQPS by helping brief the Administrator on the scientific data base and
its significance and by attending public hearings on the standard to present
the EPA viewpoint and to answer questions regarding criteria for the pollutant
under consideration.
The recent inclusion of ECAO staff as members of ORD research committees
is a significant development. This involvement, along with the principal role
played on these committees by OAQPS staff, provides a mechanism for pointing
out the research needs and problems identified during criteria and standards
development. This important feedback mechanism will help ORD plan its
research programs and will assist those EPA offices and personnel responsible
for developing timetables and allocating funds. In time, the mechanism will
also help those of us in ECAO and OAQPS who are in constant search of the data
base needed for standard-setting.
Figure 5-2 does not give an adequate view of the effort that must be
expended to prepare a criteria document. A better view of the complexity,
time, and demands of the review and revision process is shown in Table 5-1. As
indicated, the preparation of a criteria document proceeds in six phases.
37
-------
5. CRITERIA DOCUMENT REVIEW AND REVISION Tilton
TABLE 5-1. PHASES IN CRITERIA DOCUMENT PREPARATION
Phase Activity Time
(days)
I document planning and initiation 60-120
II preparation of working draft 60-200
III preparation of external review draft 60-120
(revisions to working draft)
IV public and CASAC(SAB)* review 60-90
of external review draft
V preparation of final document 60-100
(revisions to external review draft)
VI closure by CASAC(SAB) 60-90
on document and publication
Total (months): 13-24
Clean Air Scientific Advisory Committee (CASAC) of the Science Advisory
Board (SAB).
Phase I consists of (1) literature search and acquisition; (2)
publication of a Federal Register notice to solicit information on the
pollutant under consideration; (3) selection of the ECAO project manager and
team members; (4) development of a work plan, document outline, and schedule;
(5) selection of a task force composed of EPA personnel; (6) recruitment of
outside consultants to write the document and additional consultants to review
it.
38
-------
5. CRITERIA DOCUMENT REVIEW AND REVISION Tilton
Phase II consists of (1) reading and analyzing literature; (2) continuing
the acquisition of literature; (3) writing; (4) holding meetings between the
ECAO team and consultants; (5) typing, proofreading, and printing the working
draft; and (6) distributing the draft for internal review by EPA personnel and
consultants.
Phase III begins with the convening of a workshop, which is one of the
best and least time-consuming mechanisms for ensuring that ECAO receives
adequate constructive criticism on its documents early in their preparation.
Follow-up meetings with consultants are held and the working draft is revised,
resulting in an external review draft, the availability of which is announced
to the public in a Federal Register notice.
In Phase IV, the document is reviewed by the public, by the Clean Air
Scientific Advisory Committee (CASAC) of the Science Advisory Board (SAB), by
EPA personnel (including the task force), and by other Federal agencies. The
review culminates in presentation and discussion of the external review draft
at a public meeting of CASAC(SAB).
Phase V entails assessing comments received (1) by mail from the general
public and governmental sectors and (2) at the public meeting from CASAC (SAB)
members and other attendees. All comments are logged in, are reviewed, and as
appropriate are incorporated in revisions to the document. All comments and
responses to those comments are kept as part of the public record for the
document.
39
-------
5. CRITERIA DOCUMENT REVIEW AND REVISION Tilton
In the final phase, the revised draft is recirculated to CASAC(SAB) if
requested. If major changes have been made to it, the document is also
recirculated to the public for a second review and is presented and discussed
at a second public meeting of CASAC(SAB). When the document meets its
approval, CASAC(SAB) submits a letter of concurrence and a written report to
the Administrator of EPA. After further editing for clarity, style, and
format, the document is published.
THE CONTENT
The Clean Air Act stipulates that information on the effects of air
pollutants on the public health and welfare constitute the minimum and primary
content of any air quality criteria document. It recommends, but does not
dictate, the inclusion of additional information, as noted earlier in the
quotation from Section 108. As possible, or where necessary, information on
air quality management is included (e.g., emission inventories, modeling, and
atmospheric transport and transformation). Such information is particularly
appropriate for secondary pollutants whose source-receptor relationships can
only be ascertained through understanding the transformation and transport of
the primary pollutants that serve as precursors.
Criteria documents are prepared for substances that have been clearly
identified as pervasive pollutants posing threats to public health and the
environment. When EPA begins to prepare a criteria document, then, it usually
40
-------
5. CRITERIA DOCUMENT REVIEW AND REVISION Tilton
encounters a sizeable and comprehensive body of literature. Typically, an air
quality criteria document contains the following:
1. Extensive chapters on animal toxicology, human clinical studies,
and human epidemiologic studies, both occupational and
nonoccupational. Such chapters include the pertinent
information on human metabolism and pharmacokinetics, including
uptake, distribution, metabolism or detoxification, and
excretion.
2. One or more chapters on the effects of the pollutant on plants,
animals, acquatic and terrestrial ecosystems, and natural and
man-made materials.
3. One or more chapters that treat fairly exhaustively the
methodology available for measuring the pollutant in ambient
air, for measuring the pollutant in biological tissues and
fluids (where pertinent), and, in a separate chapter or in the
health effects chapters, for measuring observed effects. Where
possible, these chapters include estimates of the precision,
accuracy, and reliability of the methods actually employed to
obtain the ambient air, exposure, and effects data presented in
the document.
4. Condensed chapters dealing, respectively, with chemical and
physical properties; sources and emissions; geographic and
temporal distribution of sources; ambient air concentrations;
the global cycle of the pollutant in all its forms; and
atmospheric transport and transformation of the pollutant.
5. Where possible, a chapter that assesses quantitatively the
potential risk for the most sensitive segment of the population.
6. A summary and conclusions chapter focusing on the interpretation
and significance of the data cited in the document. This
chapter relates ambient air concentrations to observed effects,
"no observed effects," and "least observed effects" levels for
sensitive or at-risk populations.
As this enumeration indicates, air quality criteria result from the
synthesis of essentially two lines of inquiry (Figure 5-3).
41
-------
5. CRITERIA DOCUMENT REVIEW AND REVISION
Tilton
AMBIENT AIR MEASUREMENTS
MEASURED LEVELS
VALIDITY OF MEASUREMENTS
SOURCE IDENTIFICATION
HEALTH/WELFARE EFFECTS OBSERVATIONS
IDENTIFICATION OF EFFECTS
ESTABLISHMENT OF LEVELS/DOSES
IDENTIFICATION OF POPULATIONS AT RISK
AIR QUALITY CRITERIA
AMBIENT AIR LEVELS AT WHICH EFFECTS WILL OCCUR
SIGNIFICANCE FOR POPULATIONS OF CORRELATED
EFFECTS AND LEVELS
Figure 5-3. Lines of inquiry resulting in synthesis of air quality criteria.
Without health or welfare effects we would have no criteria upon which to
base regulations, but it is equally true that the observation of effects is
meaningless in the context of standard-setting if the effects cannot be
correlated with the true levels at which they occur. Thus, it is important to
be able to measure the criteria pollutant with specificity, accuracy, and
precision in the ambient air and in experimental settings. Equally important,
though not shown in Figure 5-3, is the ability to measure the effect itself
with specificity, accuracy, and precision. Qualitative information on
pollutant effects is never disregarded in preparing a criteria document, but
it serves mainly to help establish the validity of available quantitative
data. To be of use in setting an ambient air standard, air quality criteria
should be of a quantitative nature. Air quality criteria, then, are the
synthesis of these two basic lines of inquiry and documentation.
42
-------
5. CRITERIA DOCUMENT REVIEW AND REVISION Til ton
It should be noted that air quality criteria-must be objective statements
of the most recent available scientific evidence. They are to be free from
bias, from speculation stated as fact, from economic considerations relative
to the protection of public health, and from such language and presentation as
might cause them to be construed as standards rather than criteria.
Criteria do not include a margin of safety but, wherever possible, they
include an assessment of any uncertainties associated (1) with the severity
and nature of the effects caused by the pollutant in question and (2) with the
measurement and determination of the concentrations at which such effects
occur.
CURRENT RESEARCH NEEDS
No discussion of the review and revision of criteria and criteria
documents would be complete without brief mention of areas in which further
research and guidance are needed from the scientific community, both within
and outside EPA. The following list identifies recurring or continuing
problems that EPA encounters in preparing criteria documents:
1. What is a health effect? An adverse health effect?
2. Can we extrapolate from animal studies to human health effects?
From clinical studies, with their limited numbers of subjects,
to populations? From high-level, short-term exposures to
low-level, long-term exposures?
3. Can we establish "no observed effects" levels (NOEL'S)? Are
there thresholds? Are there threshold ranges?
43
-------
CRITERIA DOCUMENT REVIEW AND REVISION Tilton
4. Have we overlooked important confounding or intervening
variables in our epidemiologic studies, such that the reported
results are made suspect?
5. Have we overlooked important factors in what may be a
multifactorial etiology?
6. How good are our measurements of ambient air concentrations? Of
doses administered? Of background levels? Of responses
elicited?
7. Does ambient air monitoring of the pollutant in question
adequately reflect actual human exposure? That is, can we
adequately describe source-receptor relationships?
8. What kinds of groups should be excluded from consideration as
sensitive population(s)?
The first question of this list may be unanswerable by scientific
inquiry, since part of the answer may be judgmental. For example, the Clean
Air Act itself allows the Administrator of EPA to make certain decisions
according to his "best judgment." Furthermore, even health researchers may
differ about what constitutes a health effect and particularly about what
constitutes an adverse health effect. Certain kinds of experiments, however,
can at least provide a partial answer to this question. Long-term, low-level
exposure experiments plus specific experiments on adaptation may be of value.
Such low-level chronic exposures may also help resolve whether thresholds
exist and whether the NOEL1s we try to identify are real or only a product of
the insensitivity of present test systems.
Some of the recurring problems listed here arise during derivation of the
standard rather than in preparation of the criteria document itself. The last
question, for example, is usually treated in the staff paper that is part of
44
-------
5 . CRITERIA DOCUMENT REVIEW AND REVISION Tilton
the regulatory package forwarded to the Administrator. If smokers, for
example, were excluded from consideration as the sensitive population to be
protected by a carbon monoxide primary standard, such an exclusion would be a
policy decision rather than a scientific judgment. Writers of a criteria
document, however, must try to identify all groups at risk and to provide
scientific documentation that will aid OAQPS and the Administrator in
determining whether a given group should be considered the sensitive group to
be protected, or whether a given group can in fact be protected by an ambient
air standard.
Finally, broad areas in which ECAO believes research information is
presently inadequate include:
1. Human base-line data defining what is "normal" for certain
pulmonary function tests and biochemical indicators.
2. Human health effects data resulting from long-term, low-level
exposures to air pollutants, especially gas-phase pollutants.
3. Human studies of good design and execution, especially
epidemiologic studies.
4. Studies at the biochemical/mechanistic level for gas-phase
pollutants.
5. Correlation of classic pulmonary function tests with underlying
biochemical/physiological changes.
6. Biochemistry, biophysics, and pharmacokinetics of cardiovascular
and cardiopulmonary systems in general, and of the lung in
particular.
45
-------
5. CRITERIA DOCUMENT REVIEW AND REVISION Tilton
In view of the basic problems and questions remaining, and in view of the
areas in which good scientific data are still lacking, those of us who daily
engage in the writing of air quality criteria and related documents look to
the scientific community for the continuation of high-quality research, both
in the specific areas mentioned above and in the general pathonomy of air
pollution related sickness and disease.
REFERENCES
Clean Air Act. 42 USC 7401 et seq.
Clean Air Act Amendments of 1977. Public Law 95-95, August 7, 1977.
WORKSHOP COMMENTARY
K. M. Schaich; You make recommendations which seem quite at variance with
some of the recommendations of [Miller et al. 1979; see Chapter 2 of this
volume for details of reference]. Where is the kind of mechanistic
information that you seek to fund into your program?
B. E. Tilton; Let me point out that ECAO does not do any funding. I hope
that question can be addressed by the appropriate committees, but it may be
that the needs I have indicated are not the consensus of others involved.
46
-------
6. TOXICOLOGICAL RESEARCH STRATEGIES IN RELATION TO EPA REGULATORY NEEDS
Edward L. Alpen
Lawrence Berkeley Laboratory
University of California
Berkeley, CA 94720
INTRODUCTION
Prior to the Research Planning Workshop on Health Effects of Oxidants,
prospective attendees were sent an EPA staff report known as a "straw man"
document. This document (Miller et al. 1979) proposes certain revisions in
the EPA-funded research program designated as "Theme 1" ("Health Effects of
Criteria and Non-Criteria Pollutants from Fossil Fuel Combustion"). (For
further information on Theme 1 and the entire Energy Interagency Health and
Ecological Effects Program, see Chapter 2 of this volume.) As an
"old-fashioned" toxicologist, I felt strong personal dismay in reviewing the
recommendations contained in the straw man document. In my opinion, the
suggested changes diverge so sharply from toxicological dogma (be it good or
bad) that an examination of their merits is warranted.
The EPA statements and/or recommendations that I wish to challenge are:
47
-------
6. TOXICOLOGICAL RESEARCH STRATEGIES Alpen
1. Data from studies involving exposures of >1 ppm ozone (03) or >5
ppm nitrogen dioxide (NO2) are not significant to EPA. Sulfur
dioxide particulates of >25 mg/m3 have no impact and have no
significance to EPA.
2. All experiments should be conducted at ambient levels.
3. Negative results are as important to EPA as positive results;
environmentally relevant exposure regimens should be used.
4. In vitro studies have essentially no value to EPA.
COMMENTS ON EPA RECOMMENDATIONS
In my opinion, to perform studies at ambient levels and to expect other
than negative data is to concede that control programs have failed.
Obviously, control programs are designed to incorporate margins of safety (see
Chapter 4 of this volume).
Negative studies, in my view, have very little value except in completing
the bottom ends of dose-response curves that are already well developed.
Standard toxicological practice is to range-find first, and to then work down
to the bottom of the dose-response curve. As a researcher, I do not pursue
negative studies because their results are trivial to users.
To me, the fourth statement listed above (regarding _in vitro studies) is
especially strange. Clearly, EPA needs in vitro studies in order to know what
to study in vivoI Due to limited resources of time and money, it is simply
not possible to perform every study in the whole animal. Results from the
48
-------
6. TOXICOLOGICAL RESEARCH STRATEGIES
Alpen
relatively fast and inexpensive jji vitro systems provide important input in
designing fruitful jLn vivo experimentation.
REFERENCE
Miller, F. J., J. A. Graham, M. Hazucha, C. Hayes, and W. Riggan. 1979.
Preliminary Relevance Review of EPA-DOE Projects on Health Effects of
Criteria and Non-Criteria Pollutants from Fossil Fuel Combustion (Theme
1). Unpublished document, Health Effects Research Laboratory, Office of
Research and Development, U.S. Environmental Protection Agency, Research
Triangle Park, North Carolina, December 12.
WORKSHOP COMMENTARY
J. A. Graham; I don't think we're as far apart as some of your statements
imply. It seems to me that we all agree that concentration-response studies
are necessary; it' s just a question of the concentration range over which
those studies are to be performed. The straw man document [Miller et al.
1979] uses, I believe, the wording "near ambient levels." We agree that it
would be silly to begin a study at an ambient concentration; rather, one
should start at a slightly higher concentration and work down. If somebody
else had done some research at, say, 0.5 ppm, one might decide to start there.
The straw man offers broad guidelines. When we speak of starting at a
"high concentration" and going lower, we would consider 1 ppm to be a "high
concentration" for O3. That is definitely not an ambient concentration; 0.1
ppm would be more likely in an urban environment, although higher
concentrations have been reported.
So, in my opinion, we're not so far apart. It's just that our
definitions of "high" may be different. Our problem with high-concentration
studies also relates to the amount, of work involved. In other words, if 100%
of the work effort for a given laboratory is at 20 ppm O3, that represents one
category. If 5% of the effort is at 2 or 3 ppm 03, that's a different case
altogether. We're talking about the majority of the research.
It is indeed true that papers on, 10 ppm O3 are not even reviewed for
inclusion in the criteria document. The reason is that, with the past 03
criteria document, there were so many data available in the literature that it
became necessary to make a cut somewhere. The flavor we got from all those
49
-------
6. TOXICOLOGICAL RESEARCH STRATEGIES Alpen
papers (and for nitrogen oxides toxicology, as well) is that higher
concentrations (e.g., 20-30 ppm) might even cause different types of effects.
For example, I believe several authors have reported O3 studies in which high
concentrations caused echinosis of the red cells while lower concentrations
caused spherocytosis [see Chapter 11 of this volume].
Comment; Those "high concentrations" were 2 to 4 ppm, not 20 to 30.
J. A. Graham; We get in trouble by using words like "high" and "low." The
studies showed that 4 ppm caused a different type of effect.
E. L. Alpen; But that's not what you said. I've been peddling my business to
the government for nearly 30 years, and I've discovered that when somebody who
works for the government writes things down, you're not there to explain
them—
J. A. Graham; That* s the purpose of this [Research Planning Workshop on
Health Effects of Oxidants]. That's why the front page of the straw man uses
the word "preliminary."
E. L. Alpen; Toxicology has proven over and over that even very high level
studies are extremely useful in providing insights on mechanisms. If one
finds high-level studies or dose-dependency curves that go through maxima (and
we know lots of these in toxicology), it is crucial to know what side of the
maximum one is on.
I would point out one other fact: Almost every major epidemiologic
study, whether published or in progress, is not at ambient levels. Such
studies focus on high-risk populations, because that's where investigators
hope to find something. For example, our group is conducting a hydrocarbon
toxicology study for the Department of Energy. We're not out looking at
people on the street. Rather, we're looking at three populations of high-risk
workers: ramp workers at airports, industrial retail petroleum product
deliverers, and petrochemical refinery workers. Most epidemiologic studies
are conducted on high-risk populations, because that's where one finds what to
look for in the low-risk groups. The same generalization applies in
toxicology, too.
J. A. Graham; We in EPA agree.
E. L. Alpen; I don't think we agree.
J. A. Graham; One starts at higher concentrations and works down.
E. L. Alpen; Your position is: If a researcher starts at a high
concentration and finds something, he works to lower concentrations, and other
investigators should not return to the higher level. My position is: They
should, because the first investigator might have missed a lot!
50
-------
6. TOXICOLOGICAL RESEARCH STRATEGIES
Alpen
J. A. Graham; It depends on the concentration you1 re talking about.
E. L. Alpen; I'll just read your own numbers to you: "Any results above 1
ppm 03 are of no relevance or significance to EPA."
J. A. Graham; The straw man states that they don't go into the criteria
document, not that they are "of no relevance or significance."
E. L. Alpen; That doesn't mean they're neither relevant nor significant;
they're both, whether they go into the criteria document or not.
J. A. Graham; Here again, we're in terminology.
E. L. Alpen; Would anybody else like to say something?
Comment; As a toxicologist, I understand Dr. Alpen1 s position very well. But
levels above 1 ppm for O3 and above, say, 5 ppm for NO2 are so far out of the
potential for human exposure under air pollution circumstances that the data
from such studies are unlikely to be important to determining end points of
biological effects. They may be important to understanding mechanisms, but
they1re not likely to be important to understanding the potential hazard to a
population that is commonly exposed at factors tenfold below this.
With respect to the practical issue of setting standards, I do therefore
believe that the limitations recommended in the straw man are appropriate
limitations. They are not limitations on toxicology.
J. A. Graham; With respect to your statement on in vitro studies, we did not
say that they have "no value," period. We said that jLn vitro studies have
value under certain circumstances and that those circumstances are quite
limited. The reason is that j.n vitro data simply cannot be used for
regulatory purposes; they can be used to substantiate an in vivo effect, and
one might choose an jln vitro system to help elucidate the mechanism of a
certain effect. There are thousands of chemicals to be considered, so one
might want to use in vitro screening methods. There are other cases where in
vitro studies are appropriate, but in vitro data in and of themselves cannot
be used for regulation. So when a decision is made to use in vitro exposure
experiments, they must be viewed from that standpoint. In vitro studies are
not of general value; they are of value in certain circumstances, and those
circumstances need to be evaluated.
Question; May I have some clarification of the statement, "cannot be used for
regulation?" For those of us not familiar with EPA's procedures, is that an
internal policy or practice of EPA or is that mandated by some law? Where
does the "cannot" come from?
J. A. Graham; Generally speaking, in a criteria document, when a decision is
made on what levels are harmful for man and what types of effects occur in
man, the emphasis is on the concentration-response data. In other words, one
51
-------
6. TOXICOLOGICAL RESEARCH STRATEGIES Alpen
looks at the human data for 03, let's say, and sees that "X" effects in
pulmonary function are demonstrated at a given concentration. An in vitro
study cannot relate concentration to the human exposure condition as can a
human clinical study. When EPA views a human clinical study on 03, we see
what concentrations cause what effects and can then make decisions about
safety factors to be included. When one is dealing with an in vitro study (no
matter what the pollutant), one cannot really use those data in the regulation
on a microgram per cubic meter basis.
52
-------
7. CORRELATING EPIDEMIOLOGIC, CLINICAL, AND BIOLOGICAL
RESEARCH ON OXIDANTS
David L. Coffin
Health Effects Research Laboratory, MD-70
Office of Research and Development
U.S. Environmental Protection Agency
Research Triangle Park, NC 27711
INTRODUCTION
Developing pollution standards requires us to ask certain basic
questions. Typically, the first question is: "How much is too much?" The
sequel to that is: "How much are we willing to accept because of economic and
other factors?" The first question must be answered by toxicologists, the
second question by politicians and philosophers.
Another basic question that is certainly important to those who have the
unfortunate and rather unrewarding job of setting standards is: "By which
discipline or disciplines do we attack the problem? What can we use to
convince people that this agent is toxic at a certain level or not toxic at a
certain level?" As a toxicologist, I have always been rather sensitive to
that question. My work has generally been with mice; when significant results
are obtained, critics sometimes complain: "Well, a mouse is only that long
and a human being is this tall. How can they correlate?"
53
-------
7. CORRELATING LINES OF RESEARCH Coffin
To accumulate data in support of standard-setting activities, three
general disciplines are available: epidemiology, clinical experimentation,
and biological experimentation. Each carries certain advantages and
disadvantages.
Epidemiologic studies are sometimes appropriate for community problems,
but epidemiology was originally developed and is more appropriate for
relatively simple problems of infectious disease (e.g., when one traces
typhoid to a contaminated well). Environmental epidemiologic investigations
are plagued with two problems. First, the dose is very low. Secondly, there
are innumerable confounding variables. In short, environmental epidemiology
carries great potential but is difficult to apply in practice.
Clinical experimentation has the advantage of yielding data that are
directly applicable to human beings. There may be difficulties in obtaining
dose responses, however, because subjects cannot be exposed to toxicants at
concentrations that are much above ambient levels. Also, because of the small
number of subjects in any given study, there is a tendency to overlook
individual differences in susceptibility or resistance. But such individual
differences may be very important, particularly for oxidant air pollutants.
Biological experimentation offers several advantages: higher doses,
primary detection of effect, better construction of the dose-response curve,
and exposure of relatively large numbers of individuals of defined genetic
pattern. Under strictly controlled conditions, it is usually possible to
54
-------
7. CORRELATING LINES OF RESEARCH Coffin
determine whether susceptibility or resistance is. the result of genetic
structure or is acquired. Of course, the disadvantages are also great.
Extrapolation of the obtained experimental data to human beings may be
difficult because of discrepancies in lung size, pulmonary frequency, etc.
Also, there may be certain biochemical differences between the experimental
animals and humans.
It appears, therefore, that the ideal strategy would be to apply these
three disciplines in correlative studies that yield unifying answers. The
problem is how to do this. How do we detect primary human effect? How do we
identify susceptible or resistant subgroups in the human population? How do
we develop a dose-response curve? Obviously, dose-response information is
best obtained in animals or other biological systems permitting a sufficient
spread of dose. If the investigator is limited to ambient or near-ambient
levels, the top dose is so small that there is very little room to work.
PREVIOUS RESEARCH
In the case of oxidants, few correlative parameters have been
successfully pursued from animal biological through clinical experimental and
epidemiologic studies. This is an area deserving more attention.
Eye irritation is a primary problem of oxidant air pollution. This
parameter cannot be studied in animals; it is, apparently, a peculiarly human
affair. When studied experimentally using artificial smog, eye irritation
55
-------
7. CORRELATING LINES OF RESEARCH Coffin
does not appear to correlate with ozone level or other obvious measurements of
oxidant pollution. It is associated, however, with the oxidizing atmosphere.
Pulmonary function is another parameter that has been examined in humans.
Pulmonary functions are easily obtained with experimental subjects, and
epidemiologic data can be acquired. Also, some information is available on
individual differences. For example, in Bates and Hackney's switch
experiments, subjects from Canada were found to be more susceptible to oxidant
exposure as measured by pulmonary function than subjects living in Los
Angeles. Such findings probably reflect some sort of acquired resistance.
A number of parameters measurable in animals may be applicable to human
beings. Pulmonary pathology is one of the most sensitive and relatively
successful parameters in animals. Oxidant exposure results in definite
disease of the small airway. When animals are exposed to appreciable amounts,
the resultant lung changes, which include loss of recoil, appear to be rather
profound. It appears that no one has attempted comparable measurements in
humans; perhaps larger doses would be required. Positive results would
suggest that oxidants contribute to pulmonary emphysema in man; already there
may be some evidence of this.
Another successful animal parameter is activation £f infection. Several
studies have shown that animals react to both nitrogen dioxide and ozone at
relatively low levels and become more susceptible to bacterial infection of
the deep lung, probably as a result of oxidant action on the macrophages and
56
-------
7 . CORRELATING LINES OF RESEARCH Coffin
other lung defense systems. Thus, activation of infection would appear to be
a highly appropriate model with which to "crosswalk" to humans. With humans
there are certain problems, however. First, bacterial pneumonia has become an
uncommon disease in man. When it does occur, the disease is frequently
secondary to viral infection. Insofar as certain viral infections affect the
macrophage system as profoundly as (or more profoundly than) oxidant exposure,
human data might be difficult to interpret. Secondly, epidemiologic data for
Los Angeles show no correlation between the oxidant season and the "pneumonia
season" (if there is such a thing). Perhaps some other area (e.g., Phoenix or
Houston) would show a better correlation. Finally, as mentioned earlier,
there are problems of species differences and human individual differences.
PROMISING AVENUES FOR FUTURE RESEARCH
One goal for future epidemiologic studies might be to screen subjects to
find a susceptible subgroup. Such screening could be based, perhaps, on a
correlation between individuals who are exceedingly susceptible to eye
irritation and those who demonstrate lung-reactivity. Experimental work in
Cincinnati showed that some individuals are highly susceptible to eye
irritation while others are positively resistant to it. There may be some
theoretical objections to using this parameter as a screen: the eye
irritation is probably due to other agents within photochemical smog (rather
than the oxidant itself). But, as noted earlier, eye irritation does
correlate roughly with the oxidizing atmosphere.
57
-------
7. CORRELATING LINES OF RESEARCH Coffin
Another potentially fruitful line of research would be to investigate
whether human "acquired resistance" to oxidants resembles the "tolerance" seen
in animals. Over 20 years ago, Stokinger and others showed that low-dose
exposures to one gas will protect animals against exposures to large doses and
even against exposures to other gases (e.g., phosgene will protect against
ozone, or vice versa). In our laboratory, we showed that such results are
associated with protection or inhibition of edema. We postulated that this is
a laboratory artifact that probably does not occur in nature: the
"edemagenic" exposures do not obtain in ambient air, and our cellular studies
did not reflect any tolerance phenomenon. Still, clinical experiments have
shown that humans do acquire some sort of pulmonary functional resistance to
oxidants, and this remains a potentially productive line of inquiry.
One of the obvious places to look for correlations would be in pulmonary
function studies. Involved here are noninvasive tests that can be applied
across the board in epidemiology, clinical experimentation, and animal
experimentation. We already know, of course, that the human response to
oxidants entails such physiological responses as constricted airway. The
question is: Are these responses significant—are they health-impairing—or
do they represent a physiological response that is easily accommodated? The
other question is: Does a 4-h acute exposure in animals result in a
measurable pulmonary physiological response? In studies performed in the
early 1960's, Murphy demonstrated certain changes, including increased
frequency of respiration. But he could not convincingly demonstrate any
changes in pulmonary resistance, and subsequent studies have been equally
58
-------
7. CORRELATING LINES OF RESEARCH Coffin
unsuccessful. Perhaps the problem is one of primitive techniques:
anesthetizing the animals may wipe out these minor changes. To effectively
employ pulmonary resistance as a "crosswalk" between animals and humans, then,
will likely require us to improve our techniques in animals. There are
techniques under development that do not require anesthetization of the
animal; this may enhance the response.
CONCLUDING REMARKS
In conclusion, I would suggest that the oxidants problem be reassessed as
a "new" problem. Are oxidants a problem? Are the present standards adequate?
Rather than reapplying techniques and parameters that have been used in the
past, investigators should devote more effort to innovation. We need to
develop parameters that provide "crosswalks" from species to species. Only in
this way can we develop a coherent body of knowledge with which to approach
standard-setting.
59
-------
8. OVERVIEW OF CURRENT AND PLANNED RESEARCH
AT BROOKHAVEN NATIONAL LABORATORY
Robert T. Drew
Medical Department
Brookhaven National Laboratory
Associated Universities, Inc.
Upton, NY 11973
LIFE SCIENCES RESEARCH AT BROOKHAVEN NATIONAL LABORATORY
Brookhaven National Laboratory is located approximately 70 miles east of
New York City in the geographic center of Long Island. While the major focus
at Brookhaven has been in high energy physics, there are a number of
departments which conduct life sciences research. The Safety and
Environmental Protection Division devotes a portion of its efforts to studying
health effects of radiation. The relatively new Department of Energy and
Environment (DEE) has several programs associated with health effects
research. One group is assessing health effects of population exposures at
the national level and of a variety of energy alternatives. A second group in
DEE is developing the ability to assess mutagenic and carcinogenic effects of
chemicals using short-term in vitro systems. Within the Biology Department,
the major focus, of course, is on biological research of all kinds. One
effort that relates to oxidants is a program developed by the late Dr. Arnold
Sparrow to investigate the mutagenic effects of air pollutants on the higher
60
-------
8. BROOKHAVEN NATIONAL LABORATORY Drew
Plant Tradescantia. This study is underway; trailers have been set up within
and adjacent to industrial complexes in the Northeast.
BROOKHAVEN INHALATION TOXICOLOGY FACILITY
Inhalation toxicology is a fairly new effort within the Medical
Department. Before this author joined Brookhaven in 1976, the only oxidants
research being conducted was that of Drs. Schaich and Borg (see Chapter 11 of
this volume) . The majority of my time at Brookhaven has been spent in
developing a laboratory for exposure of rodents to any compound, regardless of
the hazard associated with that compound. This facility, which is now
operational, is described below.
Facility Layout
The first two figures illustrate the layout of the new Brookhaven
Inhalation Toxicology Facility. The inset in Figure 8-1 shows the conceptual
layout of the space. On the right-hand side is nonregulated space; on the
left-hand-side, regulated space. The chambers and glove boxes, indicated by
"C" in the inset, are maintained within the regulated space. The air handling
system for the nonregulated space is a conventional system with common supply
and local exhaust ventilation. The air handling system for the regulated
space is completely separate and independent from that of the nonregulated
space and has a common exhaust system providing several degrees of filtration
for exiting air. A third separate air handling system is provided for the
61
-------
00
V
V
1
to
i i
REGULATED
AIR SUPPLY
NON-REGULATED
AIR SUPPLY
V
c
R
r
NR
H
i
Figure 8-1. Outline of Brookhaven Inhalation Toxicology Facility.
handling systems.
Inset: Separation of air
o
H
-------
8. BROOKHAVEN NATIONAL LABORATORY Drew
inhalation chambers and glove boxes. The air systems for the regulated space
and the chambers are totally redundant. In operation, the regulated space is
maintained at ~0.08 inH^O pressure less than the nonregulated space. Chambers
are operated at -0.5 to -1.0 inH2O with respect to the regulated space.
The actual layout of the building is shown in Figure 8-2. Within the
nonregulated space are the electron microscopy suite, two conventional
laboratories, two offices, and a glass washing area. The regulated space
consists of two large chamber rooms separated by a bank of small laboratories;
total separation between the two chamber rooms is accomplished by closing one
door. We intend to use Chamber Room A for highly hazardous materials and
Chamber Room B for more conventional compounds. Two small animal rooms are
adjacent to Chamber Room B. Access to the regulated space is through three
small vestibules, indicated in Figure.8-1, which are ventilated at ~1 air
change/min with clean, fresh air. Two large viewing windows allow observation
of operations behind the barrier (Figure 8-3). Thus, visitors can view the
exposure facilities without entering the regulated space. Catwalks over the
inhalation chambers provide space for contaminant generation and for chamber
monitoring (Figure 8-4).
Figure 8-5 is a close-up of one of the 5-ft inhalation chambers. The
chambers are installed in pits and arranged so that a rack of animals can be
wheeled in. Each rack houses 12 cage units, each capable of holding 8 rats in
individual cages. Thus, we can expose up to 96 rats in 1 rack, or a total of
192 rats in 1 inhalation chamber. The cages are capable of housing either 1
63
-------
8. BROOKHAVEN NATIONAL LABORATORY
Drew
u
I C
-*
INUALATI0N -TOXICOLOGY FACILITY
SCALt- J4"« t'-Q*
, a
=r c=r
t
=]
Figure 8-2. Floor plan of Brookhaven Inhalation Toxicology Facility.
64
-------
00
H
s
«M>
1
I
Figure 8-3. Chamber Room B (through the viewing window).
-------
8. BROOKHAVEN NATIONAL LABORATORY
Drew
Figure 8-4. Generation and monitoring equipment located above the chambers.
66
-------
Figure 8-5. Close-up of caging unit.
i
-------
8. BROOKHAVEN NATIONAL LABORATORY Drew
rat, 2 hamsters, or 3 mice; thus, this design is capable of exposing up to 576
mice in 1 system.
These facilities came on-line in December 1979 and are now in operation.
Chamber Room A will be used to expose animals to highly hazardous compounds,
such as polycyclic aromatic hydrocarbons and other known carcinogens. For
these compounds, we intend to employ chambers that will house and expose
animals in a totally attached unit. Design of these chambers^ is underway.
Instrumentation and Techniques
Our goal is developing this research effort was to build a team capable
of measuring functional or physiological changes, anatomical changes, and
biochemical changes in small animals. To that end, we recruited a small
animal respiratory physiologist. Dr. Daniel Costa, who trained under Dr. Mary
Amdur. He has purchased and assembled equipment necessary for measuring the
physiological status of the rat pulmonary system in terms of spirometry and
mechanical function. We are using plethysmography and gas dilution techniques
We believe these methods represent the state of the art.
We will employ a recently acquired scanning electron microscope to assess
the anatomical structure1 of lungs of animals exposed to a variety of airborne
materials. Conventional pathologic evaluation will be included, and we intend
to develop morphometric techniques at the level of the light microscope. In
addition to histopathologic evaluation, we will also measure a number of lung
68
-------
8. BROOKHAVEN NATIONAL LABORATORY Drew
biochemical indices which indicate response to challenges of environmental
agents.
Current and Planned Studies
Projects that are underway include: a study of the deposition and
translocation of sized glass fibers, supported in part by the Thermal
Insulation Manufacturers Association; a comparison of functional change versus
anatomical change after exposure of rodents to a series of known pulmonary
toxicants (ozone was one of the first compounds chosen as a model pulmonary
toxicant); a study of the interaction between hypertension and air pollution
(see Chapter 10 of this volume); the use of bleomycin to produce a model of
fibrosis in the rat; and exposure of animals to coal dust in order to develop
a rodent model of coal workers' pneumoconiosis.
Since epidemiologic evidence shows 'that people who are already sick are
at a much greater risk during episodes of high air pollution, we intend to
develop animal models of various aspects of chronic lung disease. Having
developed these models, we hope to then superimpose exposure to some of the
common air pollutants (such as nitrogen dioxide and ozone) in an attempt to
understand why already-stressed persons are at a greater risk during episodes
of high air pollution.
69
-------
9. INCORPORATING OXIDANTS IN ASSESSMENTS OF ENERGY-RELATED HEALTH EFFECTS
Samuel C. Morris
Biomedical and Environmental Assessment Division
Brookhaven National Laboratory
Associated Universities, Inc.
Upton, NY 11973
THE NATURE OF A HEALTH EFFECTS ASSESSMENT
A health effects assessment is a documented, quantitative description of
knowledge and uncertainty regarding the potential health impact of a program
or policy action. Such assessments are useful in evaluating or formulating
research and development in both the energy and environmental areas.
An assessment considers the potential health implications of an installed
future energy industry, for example, based on current knowledge of the
technology, its environmental residuals, and effects on workers and the
general population. Computer modeling and simulation studies form the
framework of such an evaluation. Figure 9-1 is the schematic outline of a
model for air pollution effects. One or more scenarios are developed, plants
are sized and sited, and other activities (e.g., automobile traffic) are
distributed geographically. Based on technological characterizations,
emission estimates are made. Air transport, dispersion, and chemical modeling
70
-------
9. OXIDANTS IN ASSESSMENTS OF ENERGY-RELATED HEALTH EFFECTS
Morris
Scenario Development
Population Exposure
Estimates
Population
Projections
Dose-Response
Functions
Health Effect Estimates
Health Costs
Figure 9-1.
Model for assessing the health effects of energy-related air
pollution.
71
-------
9. OXIDANTS IN ASSESSMENTS OF ENERGY-RELATED HEALTH EFFECTS Morris
are performed. The results of these activities are combined with population
projections to estimate population exposure. Such factors as population
time-activity patterns are often included. In some cases, detailed lung and
metabolic models are used to translate exposure into dose. Using
dose-response or health damage functions, the population exposure or dose
estimates are used to estimate health damage. In some instances, health
damage is translated into monetary terms.
SPECIAL CHALLENGES OF ENERGY-RELATED ASSESSMENTS
One major problem in assessing energy-related health effects is our
near-total lack of health damage functions for the kinds of low-level
exposures expected from most new energy developments. We are dealing in an
area of great uncertainty. The uncertainty is compounded when we project
emission characteristics of a new technology for which only pilot plant
measurements may exist. In fact, the degree of uncertainty may itself be a
more important consideration than our best estimate of effect*
To the extent possible, we must deal explicitly with this uncertainty.
One way is to express parameters as probability density functions (pdf's)
rather than as "best estimates." As an example, Figure 9-2 displays the
probability that one particular coefficient will fall in a given range. It
takes some judgment to come up with such density functions, but they are
important when one needs to combine several uncertain independent parameters.
It would be very unlikely that all would be at the high (or low) end of the
72
-------
9. OXIDANTS IN ASSESSMENTS OF ENERGY-RELATED HEALTH EFFECTS
Morris
PERCENT/HOUR
Figure 9-2. A probability density function.
0.00
°
POPULATION EXPOSURE, 106 PERSON -wi/m*
Figure 9-3. Result of combination of probability density '«"<*»• *>
characterize total population exposure (after Morgan et al. 1978),
73
-------
9. OXIDANTS IN ASSESSMENTS OF ENERGY-RELATED HEALTH EFFECTS Morris
range at once. Figure 9-3 illustrates the result of combining several pdf's
to yield the likelihood of various population exposure levels.
Another way to use assessment models is through sensitivity analysis,
where we test the effects of various hypotheses of environmental transport,
dose-response functions, etc. on the final estimate of damage. This can be
particularly useful in finding the importance of one line of research over
another.
Ideally, we would prefer a complete dose-response function for man in the
range of exposure. Certainly this is what investigators should aim for.
Realistically, we settle for much less, and it becomes important to draw
information from many sources to form a hypothesis of the function. High-dose
information helps define the shape of the curve and identify mechanisms. In
vitro studies may also provide clues to mechanisms. In addition to helping
guide the design of ambient level studies, these data lead to hypotheses that
can be directly applied in assessment models.
INCORPORATING THE CONSIDERATION OF OXIDANTS
With regard to oxidants, what has been accomplished in energy-related
health assessments? Very little. There have been a number of assessments of
the health effects of various energy systems:
74
-------
9. OXIDANTS IN ASSESSMENTS OF ENERGY-RELATED HEALTH EFFECTS Morris
An Assessment of National Consequences of Increased Coal Utilization
(U.S. Department of Energy 1979a)
Regional Issue Identification and Assessment (U.S. Department of
Energy 1979b)
Energy in Transition, 1985-2010 (National Research Council 1980)
The Environmental Impacts of Production and Use of Energy (Part I,
Fossil Fuels) (United Nations Environment Programme 1979)
None of these includes quantitative assessments of the health effects of
oxidants. Estimates have generally been based on sulfur and particulates;
this is a major shortcoming. For example, in the first study listed above,
increasing emission controls were projected to lead to decreasing sulfur
dioxide and sulfate levels despite a considerable increase in coal
consumption. Nitrogen oxide emissions and presumably ozone levels, however,
would increase. These were not included for two reasons: First, a credible
atmospheric model having a large regional basis is not available. Thus, it is
impossible to make estimates of population exposure. Second, no acceptable
health damage function is available. The first problem is expected to be
resolved by 1982; the latter is of more interest here.
There have been several difficulties in developing an oxidant health
damage function. Epidemiologic studies are generally negative or
inconclusive. Clinical and animal studies show strong effects, but these
effects generally involve subclinical physiological factors. Changes in lung
compliance, pulmonary function, and blood chemistry are difficult to relate
directly to morbidity and mortality. We have designated substantive disease
75
-------
9. OXIDANTS IN ASSESSMENTS OF ENERGY-RELATED HEALTH EFFECTS Morris
or death as the only end points qualifying for consideration. This stems, in
part, from the fact that the health effects documented in our assessments are
in some sense "weighed" (by others) against the cost of controls. We do not
feel likely to convince anyone to enter into a multi billion dollar control
program on the basis of predictions of subclinical effects. Even large
numbers of headaches and respiratory symptoms do not constitute a strong
counterweight•
RESEARCH NEEDS
What can be done? First, we need more information on the contributions
of oxidants to chronic disease. As we all know, there are some suggestions of
chronic respiratory disease. One way to attack this problem is through animal
models of chronic disease* Another avenue is further epidemiology. Still
another possibility is the indirect route of linking changes in pulmonary
function, changes in blood chemistry, biochemical effects, and minor
respiratory symptoms with chronic disease.
Secondly, we can form hypotheses on the relation of oxidants to chronic
respiratory disease and test the implications of those hypotheses in
assessment models. Sensitivity testing can provide some notion of the degree
to which changes in the hypotheses affect predictions of overall effect. It
appears that we will have to build models that are much more specific about
sensitive subpopulations, physical activity, and (perhaps) change of exposure
level with time.
76
-------
9. OXIDANTS IN ASSESSMENTS OF ENERGY-RELATED HEALTH EFFECTS Morris
ACKNOWLEDGMENT
This research was supported by the Health and Environmental Risk Analysis
program, Office of Health and Environmental Research, U.S. Department of
Energy.
REFERENCES
Morgan, M. G., S. C. Morris, A. K. Meier, and D. L. Shank. 1978. A
probabilistic methodology for estimating air pollution health effects
from coal-fired power plants. Energy Systems and Policy, 2:278-309.
National Research Council (Committee on Nuclear and Alternative Energy
Systems). 1980. Energy in Transition, 1985-2010. W. H. Freeman, San
Francisco.
United Nations Environment Programme. 1979. The Environmental Impacts of
Production and Use of Energy. Part I. Fossil Fuels. United Nations
Environment Programme, Nairobi, Kenya.
U.S. Department of Energy (Office of Technology Impacts, Division of
Technology Assessments). 1979a. An Assessment of National Consequences
of Increased Coal Utilization. TID-2945, Vol. 2. U.S. Department of
Energy, Washington, D.C.
U.S. Department of Energy (Office of Technology Impacts, Regional Assessment
Division). 1979b. Regional Issue Identification and Assessment. First
Annual Report. U.S. Department of Energy, Washington, D.C.
77
-------
10. INTERACTIONS BETWEEN HYPERTENSION AND OXIDANT AIR POLLUTANTS
Robert T. Drew
Daniel L. Costa
Sonja Haber
Junichi Iwai
Medical Department
Brookhaven National Laboratory
Associated Universities, Inc.
Upton, NY 11973
INTRODUCTION
In the 1960's when the word "relevance" was not so important, one very
interesting project at Brookhaven National Laboratory was an investigation of
the causes of hypertension. By developing an animal model of hypertension,
the late Dr. Lewis K. Dahl proved that both genetic and environmental factors
are involved in the development of hypertension. Over the ensuing 15 years or
so at Brookhaven, we have refined this model, selectively breeding two lines
of rats from the same initial breeding stock. One of the lines is susceptible
(S) to salt-induced hypertension, and the other is resistant (R) to
salt-induced hypertension (Figure 10-1). We have used this model to
investigate interactions between commonly encountered air pollutants and the
development of hypertension. Since the experiment is loaded in favor of
producing a positive effect, a negative finding (i.e., one in which no
increased effects occur subsequent to a significant challenge by the pollutant
78
-------
10. INTERACTIONS BETWEEN HYPERTENSION AND OXIDANTS
Drew et al.
in question) can be interpreted favorably. This paper summarizes our studies
investigating the interaction of two air pollutants, sulfur dioxide (802) and
ozone (03), with the hypertensive model.
UJ
a:
5> 200
UJ °>
(Y ~^~
f\ P
| 160
Q C
0
0
m 120
1 1 1 1 1 1 1 1
m " • s-
_ • s-
D R -
OR-
_ * • *
-8 9 o °
0 o
1 1 1 1 1 1 1 1
1 1
HIGH -
LOW _
HIGH
LOW ~~
—
1 1
10 20 30 40 50
WEEKS ON DIET
MALE RATS ON 4% SALT
Figure 10-1. Blood pressure of S and R male rats on high- and low-salt diets
as a function of time.
METHODS
Both studies were designed as 2 x 2 x 2 matrix studies with 3 variables:
air vs. pollutant, high dietary salt vs. low dietary salt, and S vs. R line of
rats. This design required 8 groups of 10 rats each.
79
-------
10. INTERACTIONS BETWEEN HYPERTENSION AND OXIDANTS Drew et al.
Sulfur Dioxide Study
For studies with SC>2, the high-salt diet consisted of 4% salt; the
low-salt diet, 0.4% salt. Forty male S rats and 40 male R rats were weaned at
21 d, arranged in groups of 10 animals each, and immediately established on
either the high- or low-salt diet.
Exposures to S(>2 started the following week. All exposures were done in
27- x 27-in stainless steel and glass chambers. The animals were exposed to
SO2 at 50 ppm for 6 h/d, 5 d/week, for 31 weeks. Control animals were exposed
to air for the same 6-h regimen.
Blood pressures were measured under ether anesthesia using a tail cuff
method after the fourth week of exposure and on alternate weeks thereafter.
Animals were allowed food and water ad libitum (except during exposure) and
were housed under a 12-h on/off light cycle. The measurements were made
immediately after exposure to SC>2 on Thursday and Friday.
Ozone Study
Ozone exposures were carried out in similar chambers at a concentration
of 2 ppm for 6 h/d, 5 d/week, for 20 weeks. In this study, however, female
rather than male rats were used. The animals were allowed 1 week from the
time of weaning until establishment on the low- or high-salt diet. The
high-salt diet was 8% salt (rather than 4% salt as in the S(>2 experiment). In
80
-------
10. INTERACTIONS BETWEEN HYPERTENSION AND OXIDANTS Drew et al.
addition, because of the limited availability of sibling animals, each group
was divided into subgroups of 5 animals. Thus, 20 S females and 20 R females
were weaned on 2 consecutive weeks prior to random assignment to 1 of the 8
experimental groups. Exposures of the first sibling group (half of the
animals) began 1 week prior to exposures of the second group. Subsequently,
all animals were exposed simultaneously. Blood pressure was measured after
the fourth week of exposure and on alternate weeks thereafter. Data were
accumulated on the basis of exposure duration.
RESULTS AND DISCUSSION
Sulfur Dioxide Study
Figures 10-2 and 10-3 show blood pressure as a function of time for SC>2
animals on low and high dietary salt, respectively. The figures clearly show
that the initial blood pressures in S animals were markedly higher than in R
animals. It is also evident (Figure 10-2) that there were minimal differences
over the course of the experiment between S(>2- and air-exposed rats fed
low-salt diets. The mean blood pressure of S animals maintained on high-salt
diets increased as expected (Figure 10-3). In this case, S02-exposed animals
had blood pressures higher than those of air-exposed counterparts. This
difference was statistically significant at certain time points while not
significant at others. However, all differences disappeared after the last
exposure to
81
-------
CO
to
-,200
,
UJ
ct:
z>
C/)
CO
UJ
180
160
140
Q
O
q 120
GO
100
S
R
LOW SALT
LAST S02
EXPOSURE
AIR
SEM FOR ALL ANIMALS -1-3
i I i I i
10 14 18 22 26
TIME (weeks)
30 34
Figure 10-2. Blood pressure of S and R male rats exposed to air or S02 and maintained on low
dietary salt.
3
H
CO
0)
W
co
O
x
H
CO
(D
ft
Ql
-------
00
OJ
220-
E200-
E
LU
o:
^
c/)
a.
o
o
o
«j
CD
180
160
140
120
100
R
1 1 T
HIGH SALT
—\ r
LAST S02
EXPOSURE
*
S02
SEM FOR R RATS 1-3
46 10 14 18 22 26
TIME (weeks)
30 34
Figure 10-3. Blood pressure of S and R male rats exposed to air or S0_ and maintained on high
dietary salt.
H
CO
CO
H
O
O
H
(D
ft
-------
10. INTERACTIONS BETWEEN HYPERTENSION AND OXIDANTS Drew et al.
In these studies, the percentages of dietary salt and the concentration
of SC>2 were chosen in order to severely challenge the animals. Based on
previous studies, it was anticipated that the S animals given high dietary
salt would succumb to hypertension in 4 to 6 mo. This did not happen. Only 4
S rats on high salt, 2 from the exposed group and 2 from the control group,
died during the SC>2 study. Furthermore, no differences in growth rates were
observed between any of the groups in question. Although the exposed S rats
on high-salt diets always had slightly higher blood pressures, any apparent
relationship between SC>2 and hypertension is tenuous. Thus, we conclude from
these studies that exposure to 50 ppm SC>2 does not alter the development of
salt-induced hypertension.
Ozone Study
The results of our 03 exposures are displayed in Figures 10-4 and 10-5,
which plot blood pressure as a function of time for the animals on low and
high dietary salt-f—respectively. It is apparent from these figures that the
Oj-exposed animals in both diet groups had lower blood pressures than their
air-exposed counterparts. This difference is significant at the probability
level of 0.05.
Of greater interest is the effect of 03 on mortality (Figure 10-6). The
level of dietary salt was sufficient to cause a 50% mortality in the S animals
exposed to air on high dietary salt; no other air-exposed animals died. The
03 challenge (2 ppm for 6 h/d, 5 d/week) was sufficient to cause 3 and 2
84
-------
00
160
o»
X
140
UJ
a: 120
ID
en
CO
LJ 100
cr
a.
a 80
o
_J
m 60
S-AIR-LOW
S-03-LOW
^R-AIR-LOW
R-03-LOW _
1
1
2 4 6 8 10 12 14 16 18 20 22
WEEKS OF EXPOSURE
Figure 10-4. Blood pressure of S and R female rats exposed to air or O_ and maintained on low
dietary salt.
§
TO
td
3
i
D
0>
-------
H
160
140
c/)
UJIOO
a:
a.
§ 80
o
60
S-0-3-HIGH
*j
1
S-AIR-HIGH
R-AIR-HIGH _
R-0* -HIGH
i
i
68 10 12 14 16 18 20 22
WEEKS OF EXPOSURE
H
1
W
CO
H
§
8
CO
Figure 10-5. Blood pressure of S and R female rats exposed to air or 0 and maintained on high
dietary salt.
-------
CO
XR-03-HIGH
S-0,-HIGH
18 20 22
WEEKS OF03 EXPOSURE
Figure 10-6. Mortality of S and R female rats exposed to air or 0 and maintained on high or low
dietary salt.
CO
u
M
CO
H
§
H
O
0>
PI
-------
10. INTERACTIONS BETWEEN HYPERTENSION AND OXIDANTS Drew et al.
deaths in the R rats on high or low salt, respectively. In contrast, all S
animals died within 16 weeks of exposure to 03, regardless of the presence or
absence of salt in the diet. These deaths appear to be unrelated to
hypertension.
What we may have observed is a susceptibility to 03 that is independent
of dietary salt and blood pressure. We are excited about these results, since
this difference in 03 susceptibility with regard to S vs. R animals may be
very useful in understanding how the lung interacts with 03 from a
biochemical/physiological standpoint. There is, however, the possibility of a
general difference in susceptibility to all stresses. If the susceptibility
is unique to 03, this model will allow us to evaluate the biochemical changes
resulting from 03 exposure and perhaps to intervene with various treatment
regimes to ameliorate the response to 03.
WORKSHOP COMMENTARY
Question; Was this systolic pressure?
R. T. Drew; Yes, we measured systolic blood pressure using a tail cuff
method.
Question; Was there any change in diastolic pressure?
R. T. Drew; We cannot measure diastolic pressure with this technique.
Question; Were these changes in the pulmonary epithelium as a result of the
high salt in the diet?
R. T. Drew; Not to my knowledge, but we have not yet systematically evaluated
all of these studies. The tissue has been prepared and is being evaluated at
the present time. I'm not aware of anything in the past literature or in any
88
-------
10. INTERACTIONS BETWEEN HYPERTENSION AND OXIDANTS Drew et al.
of Dahl1s previous work that suggests that there would be differences in the
respiratory epithelium.
J. D. Hackney; Is it possible that these animals are susceptible to any kind
of stress?
R. T. Drew; Yes, it is possible. This is one of the reasons that I'm a
little reserved when I say that we may have a biochemical model; there is a
suggestion that S rats are generally susceptible to stress. I don't know
exactly what this means.
Question; Did (^-exposed animals lose more weight?
R. T. Drew; In animals exposed to SO2, there were no differences. The week
that the exposure started, the SO2-exposed group did not gain weight. From
then on, all groups gained at comparable rates. The S02~exposed group never
caught up and was always a little bit lighter. With O3, there were some
differences at the start of the study. This is unfortunate, but when you're
working with a limited number of animals you really can1t select for uniform
weight as with a larger number of animals. However, in every case, O3-exposed
animals weighed less than their air-exposed counterparts.
Question; Did you identify the factor causing hypertension?
R. T. Drew; We believe it is a genetic factor.
Question; Is it a kidney or a vascular problem? What* s causing the animals
to be more susceptible to high salt?
K. M. Schaich; There have been very few studies to characterize these rats in
terms of physiology and biochemistry. There were some early studies in which
hormonal effects were sought, but none were observed that could cause the
increased blood pressure. Presently, we are working on biochemical
characterization of these rats, but we don't yet know how they differ. There
is some evidence that liver function may be different in the S vs. R rats, but
we don't have detailed information yet.
Question; Is there any difference in the pathology of these animals? Do they
have more extensive lesions? Do the lesions appear earlier? Obviously at 2
ppm the animals will develop lesions.
R. T. Drew; Dr. Slatkin, a colleague, is working on this evaluation now. I
can't answer the question. The animals that died in the 03 study clearly died
a respiratory death. It was not a hypertensive death; on gross pathology, the
lungs were deeply involved (as one would expect). The micropathology is yet
to come.
89
-------
10. INTERACTIONS BETWEEN HYPERTENSION AND OXIDANTS Drew et al.
We've done respiratory function studies on the R animals in this group
(there are no S animals left). Our results appear consistent with small
airway disease.
Comment; With that concentration, you might be getting some effect on the
converting enzyme. Have you looked at that?
R. T. Drew; No, we haven't. This is an exciting observation and we will
certainly pursue it.
Comment; I don't know the data on ordinary species of rats. Is it possible
that you have a very resistant strain as opposed to a sensitive one? Was the
difference because one strain is extraordinarily resistant to this very high
level of 03?
R. T. Drew; I don't believe so. I don't believe my data on 03 toxicity are
in conflict with the literature on mortality alone.
Comment; That's what I'm asking. What is the experience with other strains?
R. T. Drew; Well, I'm presently exposing Fisher 344 rats to 03 for the
National Toxicology Program. We are exposing animals toO, 0.2, 0.8, and 2
ppm 03 and are beginning to see deaths in the 2 ppm group 8-9 weeks after
beginning the exposures. So, the results are consistent.
Question; Are you saying that your S rat is more susceptible to the same
doses of 03 than other strains of rats?
R. T. Drew: Yes, I think that is correct.
90
-------
11. THE EFFECTS OF OZONE ON RAT ERYTHROCYTES AFTER EXPOSURE IN VIVO
Karen M. Schaich
Donald C. Borg
Medical Department
Brookhaven National Laboratory
Associated Universities, Inc.
Upton, NY 11973
INTRODUCTION
For some time, concern has been expressed that toxic effects of ozone
(03) may not be limited to the lung but may extend to other body tissues.
Most biochemical investigations of systemic toxicity have focused on
erythrocytes: the cells which have first contact with the exposed lung and
which, along with plasma, presumably will be the carriers of cytotoxic
potential to other sites. Observations from a large number of studies in
several species have been published (Table 11-1), but the issue of
extrapulmonary effects of 03 reflected in red blood cells is anything but
clear. As can be seen in Table 11-1, a variety of end points have been
monitored on blood, including hematological/morphological, physiological, and
biochemical/metabolic. There has been little effort, however, to integrate
different types of effects into an overall pattern of response. In terms of
cellular biochemistry, most statistically significant changes have been noted
only when rats were made tocopherol (vitamin E) deficient before exposure or
91
-------
TABLE 11-1.
Species
rat
rat
Sprague-Dawley
male
3 mo
± 45 ppm tocopherol
mouse
Caesarian-del ivered- 1
male
adult
vo
to mouse
rat
Spr agvie-Dawl ey
male
mouse
rat
Sprague-Dawley
male
5 mo
± 45 ppm tocopherol
rat
Spr ague-Da wl ey
male
2 mo
EFFECTS IN RED BLOOD CELLS FOLLOWING IN VIVO EXPOSURE TO OZONE
Exposure Conditions Observationsa Reference
1.5 ppm x 3 d no effect SOD, GPx, K+ flux Larkin et al. 1978
6 ppm x 4 h tretics. (all levels)
8 ppm x 4 h tHb, Hct, echinocytes II s III (6 & 8 ppm)
( echinocytes correlated with petechia in
lungs, indicative of vascular endothelial
damage)
0.8 ppm continuous x 7 d -Vit. E: (-KGSH, 4/GPx, tPK Chow and Kaneko 1979
+O3 : *GSH, tGPx, tPK
tLDH, no effect SOD,
catalase, TBA, G6PD
MetHb, reticulocytes
+(Vit. E + 03): no 03 effects
0.85 ppm x 4 h tHeinz bodies Menzel et al. 1975
(4/with continued exposure)
8 ppm x 4 h +AChE, (+JGSH, tHb Goldstein et al. 1968
5 ppm x 1.5 h tH2O2, ~50% 4- catalase Goldstein 1973
6.7 ppm x 1.5 h tH2O2, ~16% 4- catalase Goldstein 1973
1.7 ppm x 1 . 5 h no effects
1.0 ppm x 4 h no effects
0.8 ppm x 7 d +O3 : tGPx, 4-GSH (both groups) Chow et al. 1976
-Vit. E: 4-GPx, GSH
tG6PD & 2,3-DPG
0.5 ppm x 8 h/d x 7 d lung: tGSH, GPx, GRase, G6PD Chow et al. 1975
(20-25%)
RBC's: nonsignificant tGPx,
4- GSH
(continued)
q
i
01
0
"a
O
O
M
50
T*
1-3
M
3
3
S
g
K<
K!
CO
01
&
01
H*
O
V
I
0
-------
TABLE 11-1 (continued)
Species
Exposure Conditions
Observations*
Reference
rhesus monkey
male
2 yr
rat
HLA-Greenacres
male
0.5 ppm x 8 h/d x 7 d
1.5-12.5 ppm x 1-5 h
lung: tGSH, GPx, GRase, G6PD
(10-15%)
RBC's: nonsignificant +GPx,
4-GSH
tneutrophil:lymphocyte ratio
Chow et al. 1975
Bobb and Fairchild 1967
3
CO
O
O
rat
C.R. Fischer
male
1-2 ppm x 2 or 7 d
no blood chemistry changes
Cavender et al. 1977
guinea pig
C.R. Hartley
male
1-2 ppm x 2 or 7 d
no blood chemistry changes
Cavender et al. 1977
VO
w
squirrel monkey
0.75 ppm x 4 h/d x 4 d
lung: tlipid oxidation
+ tocopherol
tG6PD, GRase, LDH
no effect MDH, SOD
RBC's: tRBC fragility (H2O2>
+GSH, (AChE)
no effect LDH, G6PD
(paired analyses)
Clark et al. 1978
CQ
rabbit
0.2 ppm x 4 h
fosmotic fragility
spherocytosis
Brinkman et al. 1964
rabbit
New Zealand White
male and female
~1 kg
rabbit
0.4 ppm, 1 ppm
6 h/d x 5 d/wk x 10 mo
10 ppm x 1 h/wk x 6 wk
after ~105 d:
+serum albumin
to- S Y-globulin
no change total protein
^appetite last 3 mo
(effects greater in males)
production of serum Ab's
that reacted with ozonized
egg albumin but not native
ovalbumin
P'an and Jegier 1976
Scheel et al. 1959
«
u
I
( continued)
d
-------
Species
TABLE 11-1 (continued)
Exposure Conditions
Observations3
Reference
Homo sapiens
male
Homo sapiens
Homo sapiens
male
Homo sapiens
male (20)
female (2)
asthmatic
0.15 ppm x 1 h ± exercise
0.30 ppm x 1 h ± exercise
0.2 ppm x 30-60 min
0.5 ppm x 165 min
0.25 ppm x 2 h ± exercise
no effect NPSH, G6PD
6PGD, GRase, Hb
spherocytosis
tosmotic fragility
+ AChE, GSH, G6PD
+serum TEA, tocopherol, LDH
+H2O2 fragility
+GSH, AChE
tG6PD, LDH
no effect GRase, GPx,
23-OPG, Hct
DeLucia and Adams 1977
Brinkman et al. 1964
Buckley et al. 1975
Linn et al. 1978
Definitions of abbreviations: Ab's, antibodies; AChE, acetylcholinesterase; 2,3-DPG, 2,3-diphosphoglycerate; GPx,
glutathione peroxidase; GRase, glutathione reductase; GSH, reduced glutathione; G6PD, glucose-6-phosphate dehydrogenase;
VO Hb, hemoglobin; Hct, hematocrit; LDH, lactate dehydrogenase; MDH, malate dehydrogenase; MetHb, methemoglobin; NPSH,
* nonprotein sulfhydryl; PK, pyruvate kinase; RBC's, red blood cells; 6PGD, 6-phosphoglycerate dehydrogenase; SOD, superoxide
disrautase; TBA, thiobarbituric acid reactive products.
n
W
0)
O
n
o
-------
11. EFFECTS OF OZONE ON RAT ERYTHROCYTES Schaich and Borg
when, in primate studies, paired comparisons (each animal or human subject
serves as its own control) were used in analyzing the data. These important
points will be discussed later.
Our involvement in this kind of research began with the publication of
research (Zelac et al. 1971) showing strikingly high levels of chromosome
aberrations in circulating lymphocytes of Chinese hamsters exposed to
approximately industrial tolerance levels of 03. These findings raised
questions regarding the carcinogenic potential of 63 as well as the mechanisms
that could be operative in affecting lymphocyte chromatin. Within the Medical
Department at Brookhaven, C.J. Shellabarger's program offers the
Sprague-Dawley rat mammary tumor model (Huggins et al. 1959) calibrated for
tumor production by various radiations and chemical carcinogens (Shellabarger
1971, 1972; Shellabarger and Soo 1973; Shellabarger and Straub 1972). Hence,
this model seemed suitable for a direct test of possible extrapulmonary
carcinogenicity from 03. Furthermore, mammary tissue seemed an appropriate
"target tissue" considering our expectations that oxidizing lipids or their
products might play a major role in amplifying any systemic cytotoxicity
(Goldstein et al. 1969; Roehm et al. 1971). One such oxidation product,
malonaldehyde, shows a weak mutagenic potential in the Ames screening test
(Mukai and Goldstein 1976). In conjunction with the tumor studies, then, we
began to look for biochemical markers indicating either (1) passage of
oxidative products through the lung barrier or (2) other systemic changes
attributable to 03 (Borg and Shellabarger 1978).
95
-------
11. EFFECTS OF OZONE ON RAT ERYTHROCYTES Schaich and Borg
INITIAL STUDIES
Using an O3 exposure system constructed from a modified controlled
atmosphere chamber, we first focused on finding evidence of lipid oxidation
and directly related oxidative damage in red blood cells and mammary tissue:
destruction of red cell membrane integrity, appearance of thiobarbituric acid
reactive products (TEA) or fluorescence characteristic of iminopropene
compounds, and changes in thin layer chromatography (TLC) patterns of
extracted lipids. Subsequently, analyses for reduced glutathione (GSH),
enzymes, and hemolysis were added to the list of parameters monitored. We
found great variability in the resistance of individual rats and different
batches of rats. ("Resistance" is here defined as the absence of observable
changes in red cells; all exposed rats showed pulmonary lesions typical of 03
toxicology.) Changes in the various parameters following acute exposures
(e.g.: 3-8 ppm, 2-6 h, 1-3 d) were seldom observed in old rats (~>80 d),
whereas rats exposed at ~40 d of age for only 1 d often showed dramatic
responses in iminopropene fluorescence (Figures 11-1 through 11-3), oxidation
products apparent on thin layer chromatographs (Figure 11-4), and depression
of acetylcholinesterase activity. When such effects were manifested in red
cells, they were short-lived (no more than 2 d), thereafter declining to
approximately control level, then gradually increasing again when exposures
were extended.
96
-------
11. EFFECTS OF OZONE ON RAT ERYTHROCYTES
Schaich and Borg
A. ERYTHROCYTES EMISSION
EXCITATION
EXPOSED
CONTROL
300 320
360 400
FLUORESCENCE (nm)
440
480
B. MAMMARY TISSUE
300 320
360 400
FLUORESCENCE (nm)
440
480
Figure 11-1. Typical iminopropene fluorescence patterns of erythrocyte (A)
and mammary tissue (B) lipids from female Sprague-Dawley rats
exposed to 2 ppm 03 for 5 h.
97
-------
M
00
D VITAMIN E DEFICIENT
O VITAMIN E ENRICHED
A PURINA
30 CALENDAR DAY
EXPOSURE
(ppm 03 for 5 h/d)
Figure 11-2. Iminopropene fluorescence in erythrocyte lipids from female Sprague-Dawley rats exposed
chronically to O-.
I
2
M
a
I
M
CO
CO
8-
o
01
a.
-------
10
D VITAMIN E DEFICIENT
O VITAMIN E ENRICHED
A PURINA
M
*a
3
9
1
w
a
0
1—2-
8 10 12 14 16 18 20 22 24 26 28 30 CALENDAR DAY
I-H EXPOSURE
I 2 1 1—2 1 I 2 1 (ppm 03 for 5 h/d)
Figure 11-3. Iminopropene fluorescence in mammary tissue lipids from female Sprague-Dawley rats
exposed chronically to 0 .
O
H
Kq
-------
11. EFFECTS OF OZONE ON RAT ERYTHROCYTES
Schaich and Borg
SOLVENT
FRONTS
A
o
< \
0
'-..
6
SOLVENT
FRONnr
^
t>
o
o
•i
'0
SOLVENT
rOrtftlT* ,
FRONT
a
Vh
Cl
n
a
©
0
r >
O
0
y I
o i
'-'Jl>l
VIA L,l
" 1
• 1
[/
O
1
"o~
D
O
_ f» 1
A
J^'l
F 1
g,'U
S. ,1
(J 'l
i
S
o
0
0
e 1
1
c=>
O
1
A
l"3! O <=
1 O •=•
1 is A
Q ^>
0 Q
-L,l
© C c:
2 4 7
B
a o <=»
O 0
_
n o 0
c> **
0 0 ey
*~*
C: ,-, o
247
C
<£Z^
s S
o
m fh m
0 o n
o
° (i *
\j
247
0 I
•s,i
°
^=>
i <
CM
gIM,
~^~
1
(ffh, C
p '
1
1
«=> 1
1
w N 1
9
^^ 1
®S-'
1
0 1
0 1
^ 1
/.* M I
9 '
Figure 11-4,
OZONE EXPOSURE DAY
Thin layer chromatograms of erythrocyte lipids from Sprague-
Dawley rats exposed chronically to 2 ppm 03 x 5 h/d. Solvent is
hexane:ethyl etherracetic acid = 80:20:1. Spot visualization
(see right margin if unlabeled): I - iodine vapor; S - short
UV; L - long UV; F - fluorescent; N - ninhydrin. Cross-hatching
indicates UV light. Diets: Purina (A), vitamin E enriched (B),
vitamin E deficient (C).
100
-------
11. EFFECTS OF OZONE ON RAT ERYTHROCYTBS Schaich and Borq
EFFECT OF DIETARY TOCOPHEROL
The difficulties we encountered in inducing measurable oxidative effects
in red cells of rats fed the standard Purina Laboratory chow led us to
consider dietary and antioxidant factors. Lab chow contains a relatively high
level of tocopherol (for females more than twice and for males ~8 times the
level necessary for maintaining reproductive capacity), so rats subsisting on
it should be well protected against oxidants. Therefore, to find a more
sensitive population in which potential systemic effects of 03 would be more
readily observed and defined, we used a synthetic "Draper" diet (Draper et al.
1964) to induce tocopherol deficiency in some rats before exposure. Other
rats were maintained on the Draper diet supplemented with a-tocopherol (75
mg/kg chow, ~10 times the tocopherol level of standard chow).
Although the Draper diet is the classic diet most commonly employed in
tocopherol deficiency studies, we found it to be less than satisfactory as a
supporting diet. Females transferred to the diet (with or without
supplementary tocopherol) immediately after weaning never grew properly or
matured sexually, and hence could not be used for tumor studies (mammary tumor
production in our model is hormone-dependent), and relatively mild acute
exposures to 03 (e.g., <2 ppm for 2 h) were occasionally lethal. As a
compromise protocol, weanling rats were fed for 3 weeks on standard laboratory
chow (by which time most had reached estrus) before transfer to one of the
synthetic diets for 7 to 10 d (7 d was sufficient to attain a deficiency
state, as measured by hemolytic susceptibility) and subsequent exposure to O3.
101
-------
11. EFFECTS OF OZONE ON RAT SRYTHROCYTSS Schaich and Borg
Even so, deficiencies (or toxicants) other than tocopherol existed in the
synthetic diets, and led to some anomalous response behavior, especially in
the tocopherol-supplemented rats. A comparison of diet effects alone on
control values shows some biochemical parameters (e.g., GSH, fragility,
methemoglobin) to be consistent with antioxidant protection of extra
tocopherol, whereas patterns of enzyme activities cannot be attributed
entirely to tocopherol deficiency or supplementation (Table 11-2).
In extending our studies from acute to short-term chronic exposures, we
included analyses of some red cell metabolic enzymes. Dietary factors
represented a second set of variables. Our goal was to gain a clearer
understanding of sequences of 03 effects (if any) and to determine whether the
apparent absence of lipid oxidation and glutathione destruction were real as
opposed to reflecting enhanced intracellular decomposition and regeneration,
respectively. We exposed rats on each of the three diets (Purina, Draper ±
tocopherol) to 2.0 ppm 03 for 5 h/d for 1 and 10 d. The animals were 50 d of
age at first exposure.
As shown in Table 11-3, acute exposures under these conditions induced no
changes in the red cell parameters measured, nor were effects much more marked
after 10 d of exposure. What is interesting about the pattern of effects in
the chronic exposures is that all statistically significant changes occurred
in rats on the Draper synthetic diet—both with and without tocopherol (Table
11-4): decreased GSH (± tocopherol), increased TEA values and decreased
osmotic fragility (+ tocopherol), decreased glucose-6-phosphate dehydrogenase
102
-------
Test
Parameter3
GSH (mg %)
TBAb
osmotic
fragility0
G6PD (units)
AChE (units)
o
w MetHb (%)
TABLE 11-2. DIET EFFECTS ON BIOCHEMICAL PARAMETERS
IN RED BLOOD CELLS OF UNEXPOSED SPRAGUE-DAWLEY RATS
Diet Probability ( 1-p) That
Vitamin E vitamin E Difference Between Diet
Purina Enriched Deficient Groups Is Not Due to Chance
(P) (E) (D) P-E P-D E-D
75.16 ± 0.86 88.13 ± 3.24 68.47 ± 2.32 1.00 1.00 1.00
6.71 ± 0.76 8.27 ± 0.98 8.97 ± 1.46 0.99 0.99 0.82
0.476 ± 0.003 0.461 ± 0.006 0.470 ± 0.007 0.96 0.81 0.85
13.27 ± 0.04 12.42 ± 0.62 10.09 ± 0.44 0.98 1.00 1.00
11.86 ± 0.48 11.56 ± 0.21 11.35 ± 0.77 0.88 0.90 0.72
4.81 ± 0.34 2.72 ± 0.35 3.60 ± 0.62 1.00 ~1.00 0.99
aDefinitions of abbreviations: TBA, thiobarbituric acid reactive products; MetHb, methemoglobin ; GSH,
glutathione; AChE, acetylcholinesterase ; G6PD, glucose-6-phosphate dehydrogenase . ^felues given are mean
± standard deviation.
bMoles x 10~6
thiobarbituric acid products/ml packed cells.
c% NaCl solution giving 50% hemolysis.
^
. EFFECTS OF OZONE
§
1
M
i
1
1
CO
g.
a
I
0
3
Qt
W
O
1
Q
-------
TABLE 11-3. BIOCHEMICAL CHANGES IN RAT RED BLOOD CELLS FOLLOWING ACUTE EXPOSURE TO OZONE ^
Diet —
Test Purina Vitamin E Enriched Vitamin E Deficient Jjj
Parameter" Control Exposed p° Control
TBAd 6.64 ± 0.53 7.14 ± 0.91 NS 6.23 ± 1.16
HetHb (%) 4.35 ± 0.60 4.49 ± 0.30 NS 3.88 ± 1.16
osmotic
fragilitye — -_ 0.526 ± 0.020
GSH (mg %) 79.3 ± 14.8 82.8 ± 13.0 NS 89.77 ± 6.22
AChE (units) 13.6 ± 1.5 14.2 ± 2.3 NS 14.1 ± 3.7
G6PD (units) 8.2 ± 0.9 8.2 ± 0.9 NS 8.6 ±1.5
aPurina diet: 3 ppm 03 x 4 h. Vitamin E Enriched and Deficient diets:
Exposed pc Control Exposed pc (•}
3
5.73 ± 0.82 NS 5.77 ± 0.48 5.85 ± 0.69 NS O
3.53 ± 0.53 NS 4.27 ± 0.67 4.57 ± 0.60 NS Q
§
0.503 ± 0.013 NS 0.533 ± 0.05 0.527 ± 0.03 NS W
85.86 i 3.44 0.06 115.3 ± 2.8 116.4 ± 7.6 NS 3
13.2 ± 2.5 NS 12.4 ± 1.6 12.5 ± 2.0 NS V*
9.0 ± 1.2 NS 11.5 ± 0.6 11.7 ± 1.1 NS RJ
2 ppm 03 x 4 h. S
Q
Definitions of abbreviations: TEA, thiobarbituric acid reactive products; MetHb, methemoglobin j GSH, reduced glutathione; AChe, O
acetylcholinesterasei G6PD, glucose-6-phosphate dehydrogenase. Values
°NS - not significant (p > 0.10 in one-tailed t-test) .
dMoles x 10~* thiobarbituric acid reactive products/ml packed cells.
e% NaCl solution giving 50% hemolysis.
given are mean ± standard deviation. 2
m
^
01
01
o
3-
h«.
n
3-
(h
a
Qi
a
o
a
-------
Ul
TABLE 11-4. BIOCHEMICAL CHANGES IN
RAT RED BLOOD CELLS FOLLOWING CHRONIC EXPOSURE TO OZONE3 ^
Diet M
Test Purina
Parameter^ Control Exposed p=
TBAd 6.71 ± 0.76 7.43 ± 1.21 NS
MetHb (%) 4.81 ± 0.34 4.28 ± 0.76 0.10
osmotic
fragility8 0.476 -± 0.03 0.474 ± 0.002 NS
GSH (mg t) 75.16 ± 0.86 75.35 ± 3.31 NS
AChE (units) 11.86 ± 0.48 12.10 ± 0.35 NS
G6PD (units) 13.27 ± 0.04 14.49 i 0.47 NS
Heinz
bodies (%) 0.10 ± 0.00 0.10 ± 0.00 NS
a2 ppm 03 x 4 h/d x 10 d.
3
Vitamin E Enriched Vitamin E Deficient HI
Control Exposed pc Control Exposed p° O
CO
8.27 ± 0.98 9.39 ± 1.49 0.06 8.97 ± 1.46 9.48 ± 2.10 NS O
"g
2.72 ± 0.35 2.77 ± 0.30 NS 3.60 ± 0.62 3.81 ± 0.47 NS Q
O
0.461 ± 0.006 0.449 ± 0.006 -(0.06) 0.470 ± 0.008 0.471 ± 0.000 NS M
88.13 ± 3.24 69.63 ± 5.27 0.005 68.47 ± 2.32 62.21 ± 4.13 0.005 §
11.56 ± 0.20 11.77 ± 0.99 NS 11.35 ± 0.77 11.11 ± 1.15 NS g
12.42 ± 0.62 12.24 ± 1.08 NS 10.09 ± 0.44 9.21 ± 0.63 0.01
3
0.50 i 0.28 0.60 ± 0.36 NS 1.40 ± 0.28 2.20 ± 0.00 0.03 J|
I
^Definitions of abbreviations: TBA, thiobarbituric acid reactive products; MetHb, methemoglobin ; GSH, reduced gluta thione ; AChE, acetylcholinesterase; 3
G6PD, glucose-6-phosphate dehydrogenase . Values given are , mean ± standard deviation. (Q
°NS = not significant (p > 0.10).
dMoles x 10"6 thiobarbituric acid reactive products/ml packed
e%NaCl solution giving 50% hemolysis.
cells.
a
u
j.
•f
u
3
Q*
to
3
-Q
-------
11. EFFECTS OF OZONE ON RAT ERYTHROCYTSS Schaich and Borg
(G6PD) and increased Heinz bodies (- tocopherol). The seemingly anomalous
Decrease in fragility after 03 exposure in the supplemented rats may be
explained by observations of substantial hemolysis in blood drawn from vitamin
E supplemented rats; washed cells used for this test could be a residual, more
resistant population.
DISCUSSION
From this inconsistent pattern of changes alone (not to mention much of
our early experience with acute and other short-term chronic exposures) it is
tempting to conclude that cytotoxic or oxidative potential from 03 is not
transferred from lung tissue to circulating red cells. However, we are not
ready to dismiss the issue of systemic effects for three primary reasons:
1. TLC patterns of lipids extracted from red cells and mammary
tissue indicate oxidative degradation of lipids;
2. Unpublished data from our lab and Dr. Drew's bring into
question the utility of red cell studies in screening for
oxidant systemic effects;
3. Rats, and perhaps all rodents, are unsuitable as models
for human response to atmospheric oxidants.
The following paragraphs consider each of these points in more detail.
106
-------
11. EFFECTS OF OZONE ON BAT ERYTHROCYTES Schaich and Borq
Evidence for Oxidative Degradation of Lipids
After nearly every short-term chronic exposure, some degree of oxidative
response was apparent, but the pattern of response varied between batches of
animals and with age, exposure level, etc. The only effect that occurred
repeatedly and consistently with all diet groups, batches, and exposures
(including high-level acute exposures) was changes in TLC patterns of
extracted lipids.
New TLC spots typical of those expected from a variety of oxidation
products appeared in the nonpolar fractions of lipids extracted from red cells
(Figure 11-5). The classes of lipids that seem to have been involved are
cholesteryl esters, diglycerides and triglycerides, and free fatty acids (and
perhaps their esters), indicating glyceride hydrolysis and fatty acid
oxidation. Tocopherol showed little effect.
In contrast, in mammary tissue the phospholipid and polar lipid fractions
were most affected (Figure 11-6), and the TLC changes were cumulative with
exposure time (Figure 11-7). From comparisons with pure standards, it seems
that, in mammary tissue, the phospholipids phosphatidylethanolamine,
phosphatidylcholine (lecithin), phosphatidylserine, and phosphatidylinositol
are the most likely targets. The changes are consistent with fatty acid
oxidation, hydrolysis to lysolecithins and phosphatic acids, and formation of
iminopropene fluorescent products from reaction of fatty acid carbonyls with
amino groups.
107
-------
11. EFFECTS OF OZONE ON RAT ERYTHROCYTES
Schaich and Borg
SOLVENT
FRONT
CONTROL EXPOSED CONTROL EXPOSED CONTROL EXPOSED
| 1 | 1 ) ,
PURINA
VITAMIN E ENRICHED VITAMIN E DEFICIENT
Figure 11-5. Thin layer chromatograms of erythrocyte lipids from female
Sprague-Dawley rats exposed to 03 (2 ppm x 4 h x 10 d).
Solvent is hexane:ethyl ether:acetic acid = 80:20:1. Spot
visualization: I - iodine vapor? S - short UV; L - long UV;
N - ninhydrin.
108
-------
11. EFFECTS OF OZONE ON RAT ERYTHROCYTES
Schaich and Borg
SOLVENT
FRONT
IS
\
\J
0
a i
N
A
0
A i
i
0 0
ON
N.I
''L'F
ON
9 NN''
N
CONTROL EXPOSED CONTROL EXPOSED CONTROL EXPOSED
1 1 I 1 1 1
DIET: PURINA
VITAMIN E ENRICHED VITAMIN E DEFICIENT
Figure 11-6. Thin layer chromatograms of mammary tissue lipids from female
Sprague-Dawley rats exposed to 03 (2 ppm x 4 h x 10 d). Solvent
is chloroform:methanol:acetone:acetic acid:water =
65:10:20:10:3. Spot visualization: I - iodine vapor; S - short
UV; L - long UV; F - fluorescent; N - ninhydrin.
109
-------
11. EFFECTS OF OZONE ON RAT ERYTHROCYTSS
Schaich and Borg
SOLVENT
0 0
0 N-'
o o
o
1
I
N I
'
oN,l o
SOLVENT
0 0 0
1
Figure 11-7.
OZONE EXPOSURE DAY
Thin layer chromatograms of mammary tissue lipids from female
Sprague-Dawley rats exposed chronically to 2 ppm 03 x 5 h/d.
Solvent is chloroform:methanol:acetone:acetic acid:water =
65:10:20:10:3. Spot visualization (see right margin if
unlabeled): I - iodine vapor; S - short UV; L - long UV;
F - fluorescent; N - ninhydrin. Cross-hatching indicates UV
light. Diets: Purina (A), vitamin E enriched (B), vitamin E
deficient (C).
110
-------
11. EFFECTS OF OZONE ON RAT ERYTHROCYTES Schaich and Borg
These changes show clearly that oxidative potential initiated in the lung
can be carried to distant sites. Whether these specific changes are
functionally meaningful is another matter, since they cannot be correlated
with any of the other parameters measured. Nevertheless, they do indicate
latent damage which, under our exposure conditions, is either (1) not
sufficiently extensive to affect cell functioning, or (2) rapidly repaired
(e.g., by the spleen's clearing of damaged red cells).
The Dynamic Nature of the Systemic Response to Oxidants
The concept of latent damage brings us to some unreported observations in
our laboratory suggesting that systemic responses to oxidants and related
secondary stress are dynamic. From these observations we can infer that
"spot-checking" a few parameters for evidence of deleterious effects might not
yield an accurate picture of what is happening in vivo.
A few reports in the literature have considered red cell morphology to be
affected by O3, noting spherocytosis and crenation of red cells (Larkin et al.
1978) as well as reticulocytosis and changes in the proportional distributions
of blood cell types (Bobb and Fairchild 1967). We have noted but not yet
quantified similar changes. In our studies, spherocytosis seemed most common
for low-level exposures whereas formation of echinocytes (crenated or spiny
cells) was quite marked at high-level acute exposures or after chronic
exposures. Each of these types of cells denotes latent damage that alters
response to the osmotic environment without, for the most part, affecting
111
-------
11. EFFECTS OF OZONE ON RAT ERYTHROCYTES . Schaich and Borg
cell function. We have also seen a variable reticulocytotic response. When
reticulocytosis seemed most marked (with attendant increases in hematocrit),
biochemical measures showed the greatest change and the animals themselves
showed the greatest respiratory distress. Thus, we are left with several
questions: Is the reticulocytosis a direct response to red cell damage jLn
vivo, or is it a secondary reaction to pulmonary pathology? Since enzyme
activities in young erythrocytes are generally higher than in mature cells, is
it possible that the apparent lack of observable changes results from larger
numbers of reticulocytes counterbalancing decreased activities of defective
cells? To what extent do these types of latent damage indicate the presence
of a rat response threshold below which "systemic" effects of 03 are
detectable only in hypersensitive populations, such as in tocopherol or other
nutrient deficiencies? We are currently looking for answers to these
questions.
Consider, also, that few studies have focused on tissues other than
blood, presumably on the assumption that any extrapulmonary effects will be
first noticeable in the bloodstream. In some preliminary studies by Dr.
Drew's group (Costa and Drew 1980), rats exposed to sulfur dioxide for only a
few hours showed a reproducible destruction of sulfhydryl groups in the lung
and liver but not in blood. While extrapolation would be premature, these
findings certainly raise questions about the utility of red cell studies in
screening for systemic effects of oxidants.
112
-------
11. EFFECTS OF OZONE ON RAT ERYTHROCYTES Schaich and Borg
Nonsuitability of Rats as Models for Human Response to Oxidants
This brings us to the last point: the nonsuitability of rats (and
perhaps all rodents) as models for human response, either pulmonary or
systemic, to 03 and oxidant gases. The difficulties we have encountered in
producing in rats a consistent pattern of measurable systemic damage from 03
similar to that observed in other species, coupled with known physical and
biochemical/metabolic characteristics of the rat, lead us to contend that the
rat is in fact a poor surrogate for humans. Three lines of reasoning support
this contention.
First, there are significant differences in respiratory structure between
rodents and primates. Rats are obligate nasal breathers; thus O3 uptake in
the upper airway is greater than if breathing were by mouth (as in, for
example, humans who are exercising or who have nasal obstruction). Unlike
primates, rats do not have well develope'd respiratory bronchioles (the
transition phase between the terminal bronchiole and alveolar ducts) (Mellick
et al. 1977; Castleman et al. 1977). Both of these factors contribute to the
decreased distribution of oxidant gases to respiratory exchange tissues in the
rodent. Miller et al. (1978) developed mathematical models of transport and
removal of oxidant gases (03 and nitrogen dioxide) in lungs of different
species. These authors predicted the following relative respiratory
bronchiolar doses (assuming tracheal doses of >0.05 ppm): man, 100; rabbit,
80; guinea pig, 40. Relative comparisons of lung morphology and structure
would place rats on the lower end of this scale. The net implication is that,
113
-------
11. EFFECTS OF OZONE ON RAT ERYTHROCYTES Schaich and Borg
for identical closes, much less 03 reaches peripheral lung cells and the
gas/blood barrier in rats than in humans. In addition, the GSH peroxidase
system for detoxifying and preventing accumulation of oxidative products in
lung tissue normally operates at a much higher level in rats than in
primates—e.g., four times that in monkeys (Chow et al. 1975)—and is also
more readily stimulated by oxidant challenge. Thus, neither primary oxidant
(03) nor secondary toxic products such as peroxides would be expected to pass
into blood at levels comparable to those in humans and other primates.
"Antioxidative" defense mechanisms in red cells are also much more active
in rats than in man. Kurian and Iyer (1977), for example, have shown that
rats and other rodents are much less susceptible than man and other primates
to stress with the oxidant acetylphenyl hydrazine (APH) (Figure 11-8). Heinz
body production is also correspondingly decreased. Higher levels of GSH, GSH
peroxidase, and G6PD protect against even intense oxidative stress while
efficiently regenerating active levels of intracellular reducing agents
(Table 11-5). Thus, in comparison to human red cells, rat red cells should be
better able to cope with any 03 stress.
Finally, there are problems with statistical treatment of data necessary
to show "significant" differences between exposed and control values. In
studies citing significant 03 effects on red cell enzyme levels in monkeys
(Clark et al. 1978) and humans (Buckley et al. 1975; Linn et al. 1978), the
actual differences in activities were small. (For example, Table 11-6
presents some results of the Linn et al. study.) However, the data were
114
-------
11. EFFECTS OF OZONE ON RAT ERYTHROCYTES
Schaich and Borg
100-
CAT
DOG
MAN
COW
MONKEY
GOAT
RAT
RABBIT
GUINEA PIG
Figure 11-8.
30 40
TIME (min)
Species differences in sensitivity to oxidative (APH) stress.
A 10% suspension of washed erythrocytes in Ringer phosphate was
treated with APH (2 mg/ml) at 38°C; at 15-min intervals 0.5-ml
aliquots were withdrawn, diluted with 5 ml ice-cold water, and
MetHb estimated. Data from: Kurian and Iyer (1977).
TABLE 11-5. SPECIES DIFFERENCES IN LEVELS OF SOME PROTECTIVE ENZYMES
FOLLOWING STRESS WITH ACETYL PHENYLHYDRAZINEa
Parameter^
Rat
Man
Monkey
SOD (units)
GPx (yM GSH/g
Catalase (K x
Hb/min)
103)
52
122
8.6
30
12
36.5
28
5
38.7
aData from: Kurian and Iyer (1977).
^Definitions of abbreviations: GPx, glutathione peroxidase; GSH, reduced
glutathione; Hb, hemoglobin; SOD, superoxide dismutase.
115
-------
TABLE 11-6. EFFECTS OF OZONE IN ASTHMATICS*1
variable11
Hb (g/100 ml)
Hct (%)
HjO2 fragility
GSH
AChE
G6PD
LDH
GRase
GPx
2,3-DPG
a Adapted from:
bDefinitions of
Sharac
14.99 ± 0.96
1
44.4 ± 2.9
24.1 ± 7.0
I
30.8 ± 5.4
M
1
22.2 ± 2.5
I
1
5. 12 ± 1.06
t
|
104.0 ± 12.8
f
2.64 ± 0.79
10.87 ± 3.05
14.30 t 1.67
Odor-Shamc
14.97 ± 0.94
i
44.6 t 2.7
25.0 ± 6.9
t
28.7 ± 5.4
I
21.5 ± 2.4
I I
5.45 ± 1. 16
t
108.4 ± 16.1
2.66 ± 0.78
10.67 ± 2.73
14.52 ± 1.78
Linn et al. (1978) (Table 6). Exposure - 0.25 ppm O3
abbreviations : AChE,
GSH, reduced glutathione (red blood
red blood cell
fragility expressed a
acetylcholinesterase ; 2,3-DPG, 2
cell) ; G6PD, glucose-6-phosphate
s % hemolysis when incubated with
03 ExposureC F(2,42)d pd Change (%)
14.88 ± 0.99 5.2" 0.009 0.7
| |
I
44.2 ± 2.7 1.83 NS 1
26.8 ± 7.2 7.46 0.002 11
i !
i
29.3 ± 5.0 5.53 0.007 5
I
20.9 ± 2.5 24.94 <0.001 6
I t
5.67 ± 1.20 9.98 <0.001 11
i
111.4 ± 17.0 4.82 0.013 7
I
2.68 ± 0.74 0.89 NS
10.99 ± 2.54 0.57 NS 3
14.58 ± 2.53 0.50 NS 2
x 2 h. N = 22.
, 3-diphosphoglycerate ; GPx, glutathione peroxidase; GRase, glutathione reductase;
dehydrogenase; Hb, hemoglobin concentration; Hct, hematocrit; H202 fragility.
2% hydrogen peroxide solution; LDH, lactate dehydrogenase.
c values shown are mean * S.D. Pairwise significant differences are indicated by solid arrows for p < 0.05; by dashed arrows for p < 0.01.
dOverall variation is indicated by the F statistic (degrees of freedom
in parentheses) and the p value. NS = not significant.
^
3
a
Q
CO
o
o
i
tt
3
d
3
3
in
CO
o
o
rr
91
s
Oi
td
0
n
-------
11. EFFECTS OF OZONE ON RAT ERYTHROCYTES Schaich and Borg
obtained as paired comparisons (comparisons of measures on the same subject
before and after exposure) rather than as comparisons between exposed and
nonexposed populations. We took the post-exposure mean and standard deviation
data from some paired-comparison studies and applied several other standard
statistical tests of group differences. As we had suspected, under these
conditions—conditions commonly used in small animal studies—we found no
statistical differences, thus illustrating the power of paired comparisons as
a statistical tool.
In small animals, the problems are twofold: (1) paired comparisons are
not physically possible; (2) differences in individual animal responses give
group variances that are too large to allow statistical detection of small but
probably real group differences. If, then, the issue of 03 effects in red
cells is largely a "numbers game," the rules are predesigned for "no effects"
in any small animal study that doesn't involve very large numbers of animals
or show large differences in group mean -values. With respect to the latter,
comparisons of exposed and control lung tissue (the first site of interaction
with oxidant gases) in rats and other species have shown 25% or greater
differences in, for example, enzyme or GSH levels. But comparable effects in
red blood cells—especially the red blood cells of rodents—would not be
expected and in fact have not been seen, even in the presence of superimposed
stress such as tocopherol deficiency.
This analysis does not imply that the use of rats and other rodents in
oxidants research should be discontinued. However, the considerations
117
-------
11. EFFECTS OF OZONE ON RAT ERYTHROCYTES Schaich and Borg
outlined here urge strong caution in the use of rats as surrogates for human
beings. Especially in the case of extrapulmonary effects, it is extremely
doubtful whether "no effect" observations in rats can be credibly extrapolated
to predict "no" human responses.
SUMMARY AND OVERVIEW OF CURRENT RESEARCH
In summary, acute and chronic exposures of rats to relatively high levels
of 03 have failed to provide evidence for statistically significant
extrapulmonary effects in red cells except for oxidation in component lipids.
Hematological observations and trends in GSH and enzyme changes, however,
suggest the possibility of a response threshold not reached in animals exposed
under our experimental conditions. Observations of damage to lipid and GSH in
mammary and liver tissues, respectively, indicate the potential even in rats
for systemic effects at sites distant from the lung—effects that have been
largely unexplored.
Finally, although laboratory rats are quite useful for research purposes,
they may not be appropriate or adequate models for directly predicting human
response to gaseous oxidative pollutants. We must identify an animal having a
respiratory structure comparable to that of the human but which is less
expensive than primates and large enough to allow sequential tissue sampling
and paired before-and-after comparisons.
118
-------
11. EFFECTS OF OZONE ON RAT ERYTHROCYTES Schaich and Borg
Efforts to respond to these research needs are currently underway within
the Medical Department at Brookhaven, where we are (1) exposing rats to 03 for
longer periods (up to 67 d) and correlating measures of lung function,
biochemistry, and pathology with red cell biochemistry and metabolism,
hematology, serum measures of hepatic function, and cytogenetic end points;
and (2) exposing sheep to oxidant gases and investigating mechanisms of a wide
variety of effects. In the latter system, paired before-and-after comparisons
as well as selective exposure of single lobes (using another nonexposed lobe
in the same animal as the control) are possible. Also, the species is
sufficiently large to allow simultaneous studies of systemic functional or
biochemical, immunological, and pulmonary function and biochemistry studies in
a single animal. Such coordinated studies should provide considerable insight
into the total and overall effects of oxidative pollutants, the mechanisms
responsible, and the relationship between effects on different body systems.
REFERENCES
Bobb, G. A., and E. J. Fairchild. 1967. Toxicol. Appl. Pharmacol., 11:558.
Borg, D. C., and C. J. Shellabarger. 1978. Internal Report 24400 (EPA
Contract IAG-D5-0410), Brookhaven National Laboratory, Upton, New York.
Brinkman, R., H. B. Lamberts, and T. S. Veninga. 1964. Lancet, 1:133.
Buckley, R. D., J. D. Hackney, K. Clark, and C. Posin. 1975. Arch. Environ.
Health, 30:40.
Castleman, W. L., W. S. Tyler, and D. L. Dungworth. 1977. Exp. Mol. Pathol.,
26:384.
Cavender, F. L., W. H. Steinhagen, C. E. Ulrich, et al. 1977. J. Toxicol.
Environ. Health, 3:521.
119
-------
11. EFFECTS OF OZONE ON RAT ERYTHROCYTES Schaich and Borg
Chow, C. K., and J. J. Kaneko. 1979. Environ. Res., 19:49.
Chow, C. K., M. G. Mustafa, C. E. Cross, and B. K. Tarkington. 1975.
Environ. Physiol. Biochem., 5:142.
Chow, C. K., W. M. Aufderheide, J. J. Kaneko, et al. 1976. Fed. Proc.,
35:741.
Clark, K. W., C. I. Posin, and R. D. Buckley. 1978. J. Toxicol. Environ.
Health, 4:471.
Costa, D., and R. Drew (Medical Department, Brookhaven National Laboratory).
1980. Unpublished data.
De Lucia, A. J., and W. C. Adams. 1977. J. Appl. Physiol.: Resp. Environ.
Exercise Physiol., 43:75.
Draper, H. H., J. G. Bergan, M. Chin, A. S. Csallany, and A. V. Boaro. 1964.
J.-Nutrit., 84:395.
Goldstein, B. D. 1973. Arch. Environ. Health, 26:279.
Goldstein, B. D., B. Pearson, C. Lodi, et al. 1968. Arch. Environ. Health,
16:648.
Goldstein, B. D., C. Lodi, C. Collison, and O. J. Balchum. 1969. Arch.
Environ. Health, 18:631.
Huggins, C., G. Briziarelli, and H. Button. 1959. J. Exp. Med., 109:25.
Kurian, M., and G. Y. N. Iyer. 1977. Can. J. Biochem., 55:597.
Larkin, E. C., S. L. Kinzey, and K. Siler. 1978. J. Appl. Physiol.: Resp.
Environ. Exercise Physiol., 45:893.
Linn, W. S., R. D. Buckley, C. E. Spier, et al. 1978. Amer. Rev. Resp.
Dis., 117:835.
Mellick, P. W., D. L. Dungworth, L. W. Schwartz, and W. S. Tyler. 1977.
Lab. Invest., 36:82.
Menzel, D. B., R. J. Slaughter, A. M. Bryant, and H. O. Jauregui. 1975.
Arch. Environ. Health, 30:296.
Miller, F. J., D. B. Menzel, and D. L. Coffin. 1978. Environ. Res., 17:84.
Mukai, F. H., and B. D. Goldstein. 1976. Science, 191:868.
P'an, A. Y. S., and Z. Jegier. 1976. J. Am. Assoc. Ind. Hyg., 37:329.
120
-------
11. EFFECTS OF OZONE ON RAT ERYTHROCYTES Schaich and Borg
Roehm, J. N., J. 6. Hadley, and D. B. Mengel. 1971. Arch. Environ. Health,
23:142.
Scheel, L. D., O. J. Dobrogorski, J. T. Mountain, et al. 1959. J. Appl.
Physiol., 14:67.
Shellabarger, C. J. 1971. Proc. Soc. Exp. Biol. Med., 136:1103.
Shellabarger, C. J. 1972. Cancer Res., 32:883.
Shellabarger, C. J., and V. A. Soo. 1973. Cancer Res., 33:1567.
Shellabarger, C. J., and R. F. Straub. 1972. J. Natl. Cancer Inst., 48:185.
Zelac, R. E., H. L. Cromrory, W. E. Bolch, Jr., B. G. Dunavant, and H. A.
Bevis. 1971. Environ. Res., 4:262, 325.
WORKSHOP COMMENTARY
Question; Did you test for hemolysis in these studies?
K. M. Schaich; Yes.
Question; And you did not find hemolysis?
K. M. Schaich; We tried a number of hemolysis stress tests and finally used
osmotic hemolysis—differential response to salt—because it gave the most
consistent results. However, this is perhaps not the right measure for
hemolysis. Osmotic hemolysis reportedly is related to the sphericity of the
cells (the amount that they are puffed up). It results from pressure from
the inside. Thus, in the absence of overt damage to the cell membrane or
gross alterations to intracellular chemistry, observations of osmotic
stress effects are unlikely.
Also, the rat red cells seem to be much more resistent to osmotic stress
than human erythrocytes.
Question; What is the possibility that you have a red cell subpopulation
which is so sensitive that it breaks down, and is accordingly not tested?
K. M. Schaich; That is precisely why I mentioned our findings with
tocopherol-supplemented rats. We have not yet been able to do the
hematological studies in a controlled quantitative way at the same time as the
biochemical studies. However, in the supplemented rats we could never draw
blood without substantial hemolysis; thus, we suspect a supersensitive
121
-------
11. EFFECTS OF OZONE ON RAT ERYTHROCYTES Schaich and Borg
population of cells in those rats. The same kind of hemolysis did not occur
in handling the other rats, but those animals' spleens may have cleared the
damaged cells more efficiently and rapidly.
I can't give you a confirmed answer. I suspect that we may in fact be
looking at a more resistant, residual population, but until we finish the
current study in which we are monitoring hematological and biochemical
changes, I cannot say anything more definite.
Question; In your interesting results on extrapulmonary effects, are you able
to control for lung injury as a possible nonspecific kind of extrapulmonary
effect?
K. M. Schaich; We cannot. That is one of the problems of this type of animal
study. To my knowledge, in the literature there have been no reports directly
comparing and correlating pulmonary and extrapulmonary effects of oxidant
gases. Recently, there have been some studies which monitored what was
happening morphologically in the lung with what was happening biochemically
in the bloodstream, but I think the correlations were more limited than the
implications of your question. We have always been concerned that the changes
we observe in red cells may be secondary responses to infection rather than
primary effects of O3. It is precisely this consideration that led us, in our
current studies, to use specific pathogen free animals (thereby eliminating
endemic pulmonary infections) and to simultaneously monitor hematological
changes (if any), red cell and serum biochemistry, and lung function and
pathology. In this way we hope to be able to differentiate primary and
secondary effects.
Question; What was the age of the rats when you started the studies?
K. M. Schaich; In the chronic studies, the rats were ~50 d of age when we
started the exposures, and that bothers me. Age at initial exposure is one
reason why we handled our deficient diets (the Draper diets) in a manner
different from that usually reported in the literature. Normally, rats are
maintained on lab chow for a month or two to allow them to grow and establish
approximately adult weight before transferal to the deficient diets.
Considering our earlier results, however, we were afraid we would miss
potential early effects if we followed such a procedure. We would have
preferred to maintain rats on the Draper diets from the weanling age (19-21 d)
so that they would be deficient by 40 to 42 d—the age at which we saw the
early responses in rats on lab chow. Regardless of diet, animals did not show
"first day" or acute effects if they were older than ~45-48 d.
There is a marked age dependence in the response of rats to various
radiations and chemicals. This age dependence is at least partly related to
hormone levels. Shellabarger showed in his mammary tumor model that there is
a critical period between ~45 and ~60 d where hormonal changes proceed so
122
-------
11. EFFECTS OF OZONE ON RAT ERYTHROCYTES Schaich and Borg
rapidly that the overall response of the rats to systemic challenges is
altered.
I cannot pinpoint how potential responses were affected by our regimen.
H. P. Witschi; Would you expand on your comment that rats are not appropriate
experimental animals for pulmonary studies?
K. M. Schaich; I don't think rats are adequate models for human response,
since blood studies in humans have consistently demonstrated some effects in
red cells and red cell enzymes while rat studies have not. Compared to
humans, I don't expect as much 03 or oxidative potential to get through the
lung to the blood because of the rat lung structure and because of the
biochemical protective mechanisms which are much more active in rat lung.
H. P. Witschi; Oxygen readily diffuses through membranes. How do you know
that 03 isn't acting similarly, without reacting in the membrane?
K. M. Schaich; I don't think that 03 itself gets into the blood; the more
likely effective toxic agent(s) is some breakdown product of 03 or some
secondary oxidation product. There has been a long-standing dispute regarding
the disposition of 03 in the respiratory tree and what happens to 03 when it
traverses membranes (if it does). Ozone is so reactive that when it contacts
anything (e.g., surfactant), it decomposes or interacts immediately with a
sensitive group. Therefore, in the rat it is unlikely that 03 itself, having
to traverse the nasal passages as well as the upper respiratory tree, would
ever get into distal portions of the lung. Oxidation products such as
superoxide anion may get through to the air/blood interface, but that species
is also very reactive. I think it is more likely that secondary oxidation
products of lipids in the membranes—hydroperoxides or epoxides or product
aldehydes—provide the mobile and more stable oxidative potential that may be
picked up by the blood and carried on to initiate subsequent damage.
Comment; Ozone is more reactive than oxygen, and oxygen doesn't diffuse very
well through the upper airway- That region, therefore, is not an important
locus for gas exchange. As others have shown in terms of the depth to which
one can see damage in the lung following cytotoxic 03 exposures, damage hardly
gets below the basal membrane, so in thick tissues not much gets through.
Ozone and gas diffusion in the lungs is quite different than irn cell
suspensions in vitro.
J. A. Graham; Using a different systemic parameter and agents other than O3,
we've found that the female is much more sensitive than the male. Were you
using males or females?
K. M. Schaich; We were using females in the beginning; we are using males
now. We started with Sprague-Dawley rats and now we're using Fischer rats.
For the first couple of years we used "scrap rats" from wherever we could get
them.
123
-------
11. EFFECTS OF OZONE ON RAT ERYTHROCYTES Schaich and Borg
When we did have male rats, our impressions were that they were less
sensitive. But they were usually also older, so that confounds the issue.
Comment; Some of these parameters are reminiscent of aging red cells. Have
you determined the age distributions of the red cells you sampled? Were they
comparable?
K. M. Schaich; No, we had no hematological backup in the early studies. You
can see, however, there is a long list of things that we're beginning to
follow up now. In a study that is just beginning, we will look directly at a
variety of hematological parameters in order to understand the red cell
population being monitored. If we don't have the same red cell populations
and type distributions in exposed and normal rats, then obviously we should
see some differences in the enzyme responses. We are not presently equipped,
however, to measure red cell kinetics and turnover, which would provide a more
direct answer to your question.
E. Hu; Are you testing the effects of oxidants on the tumorous animal, too?
K. M. Schaich; Shellabarger did those studies and we worked in conjunction
with him. Those studies were essentially negative. There were some
borderline positive responses but nothing consistent.
Question; There are two forms of glutathione peroxidase. Which form were you
talking about?
K. M. Schaich; I don't know that we differentiate; we use standard
glutathione peroxidase assays for the red blood cells as outlined by E.
Beutler [1975. Red Cell Metabolism: A Manual of Biochemical Methods. Grune
and Stratton, New York].
Question; How do you measure? Do you use hydrogen peroxide?
K. M. Schaich; No, t-butyl hydroperoxide.
Question; Because there are different levels of the peroxidase and other
materials that might decompose 03 in the blood, you're assuming that it is in
fact the 03 that's causing the problem. But maybe the decomposition products
are effecting the damage.
K. M. Schaich; I don't think the O3 is causing damage directly in the red
cells. As I said before, I believe it to be more likely that ozonides or
other oxidation products carry the damage potential from the lungs to other
tissues. That is why we initially looked only for oxidation effects,
especially in lipids.
Question; Since peroxidase activities are higher in rats than in humans or in
primates, is it possible that you saw even more effect, because breakdown
products enter the bloodstream faster?
124
-------
11. EFFECTS OF OZONE ON RAT ERYTHROCYTES Schaich and Borg
K. M. Schaich; No. Hydroperoxides are decomposed by peroxidases to products
with lower oxidation potential and generally more hydrophobia character.
These products, therefore, are less likely to be picked up by the bloodstream,
and are less damaging if they are. More rapid peroxide decomposition,
furthermore, means that lower concentrations of the toxic hydroperoxides will
be available for transport by the bloodstream. Hydrogen peroxide and lipid
hydroperoxides that do get into circulation provide quite active substrates
for the iron in the blood, forming an in vivo Fenton-like system which
produces, at a minimum, the more reactive and cytotoxic hydroxy and alkoxy
radicals. Thus, in terms of potential damage it should be of advantage to
keep hydroperoxide levels as low as possible.
guestion; These agents occur in the blood, and your measurements were in the
blood?
K. M. Schaich; Yes. Let me restate my reasons for contending that rats are
poor models for providing data which may be directly extrapolated to man—at
least in terms of pulmonary and systemic oxidant effects. First: Rat lung
tissue has a substantially greater capacity than does the human lung for
decomposing oxidation products formed during normal metabolism or after
exposure to gaseous or other oxidants. Thus, lower concentrations of the most
potently damaging products are available to be picked up and circulated by
the blood. Consequently, initial challenge to circulating red blood cells
is lower in rats than in man. Second: Levels of glutathione peroxidase in
red cells are much higher in rats than in man. Since glutathione peroxidase
is located at the membrane and acts on organic peroxides as well as hydrogen
peroxide, effects in rat red cells will be counteracted at the membrane
level—at the site of initial oxidant reaction. In contrast, in man higher
levels of catalase alone are ineffective in counteracting sustained oxidant
stress, especially at the membrane level. Damage to the membrane allows
cytotoxic agents to be absorbed or otherwise taken up by the cell, thereby
affecting metabolic machinery and other components inside the cell. Thus,
the site of major reactions and red cell damage are quite different for rats
as compared to man. As a consequence, patterns of response to oxidant
challenge would not be expected to be comparable.
125
-------
12. PATHOGENESIS OP CHRONIC LUNG DISEASE:
THE ROLE OF TOXICOLOGICAL INTERACTION
Hanspeter Witschi
Biology Division
Oak Ridge National Laboratory
Post Office Box Y
Oak Ridge, TN 37830
INTRODUCTION
It is known that simultaneous or successive exposure to two or possibly
several toxic agents will sometimes elicit a biological response that is
considerably more severe than the effects produced by one chemical alone
(National Research Council 1980). To consider toxicological interactions
should be an important element in attempts to estimate and evaluate the health
effects of air pollution. These attempts require an understanding of the
nature and mechanism of each interaction.
Our group recently developed an experimental model to help describe and
analyze how an acute interaction between two toxic agents produces a chronic
form of lung damage. The two compounds are the antioxidant butylated
hydroxytoluene (BHT) and oxygen (02). We hope that it will be possible to
examine whether the principles governing the interaction between BHT and ©2
126
-------
t2. TOXICOLOGICAL INTERACTIONS IN CHRONIC LUNG DISEASE Witschi
are of general significance and might be applied .to better understand some
health effects of oxidants.
BACKGROUND
In mice BHT produces extensive and uniform death of the Type I alveolar
cells. Within 24 h after a single administration of BHT, necrotic Type I
cells disintegrate and slough off, leaving large areas of denuded basement
membrane. If the animals are left undisturbed, the changes in the alveolar
zone are successfully repaired within the next 2 to 6 d. Recovery is
accomplished, as in many other forms of acute toxic lung injury, by
proliferation of Type II alveolar cells which eventually transform into Type I
cells. A normal air/blood barrier is restored ~1 week after BHT, and the
lungs of BHT-exposed animals look essentially normal again. Therefore, a
single acute episode of toxic lung damage appears to be without any serious
long-term consequences (Hirai et al. 1977; Adamson et al. 1977).
It is, however, possible to interfere with tissue recovery. Once Type II
alveolar cells begin to divide following BHT-induced lung injury, they appear
to become quite susceptible to elevated concentrations of ©2 in the inspired
air and may be killed. Oxygen concentrations as low as 50% have been found to
have an adverse effect on epithelial cell division in lung. Interestingly,
dividing interstitial or dividing capillary endothelial cells appear resistant
to the cytotoxic action of O2. This observation led us to predict that the
presence of O2 following lung injury would compromise reepithelialization of
127
-------
12. TOXICOLOGICAL INTERACTIONS IN CHRONIC LUNG DISEASE Witschi
the alveolar zone without, however, interfering with the proliferation of
fibroblasts. A possible consequence would be the development of fibrosis
t >
(Witschi and Cote 1977a). The experiments summarized below confirmed this
hypothesis.
METHODS AND RESULTS
Animals treated with BHT and exposed to O2 during the phase of epithelial
cell proliferation developed extensive and diffuse fibrosis within 1 to 2
weeks. Quantitative determination of lung hydroxyproline showed that exposure
to 50% to 80% 02 alone for up to 6 d did not increase lung collagen.
Injection of BHT alone produced only slight fibrosis. However, combined
treatment with BHT and 02 increased total lung collagen 2 to 3 times above
controls. The combined effects of BHT and 02 were synergistic (Haschek and
Witschi 1979). Histopathology showed the animals to have developed diffuse
interstitial fibrosis with many ultrastructural features common to the human
disease known as Hamman-Rich syndrome (Brody 1980). Similar observations were
made when epithelial cell proliferation was inhibited by low and nonfibrogenic
doses of x-rays (<200 rads) (Haschek et al. 1980). Animals exposed to 02 or
to x-rays once reepithelialization of the alveolar zone was complete showed no
development of fibrosis, however (Haschek and Witschi 1979; Haschek et al.
1980).
In subsequent studies, we used lung hydroxyproline as a quantitative end
point to further study this interaction between a bloodborne lung toxic agent
128
-------
12. TOXICOLOGICAL INTERACTIONS IN CHRONIC LUNG DISEASE Witschi
and inhaled 02. Several observations were made. Abnormally high levels of
total lung collagen persisted for up to 6 mo (possibly longer) after a single
episode of exposure to BHT followed by 02 or x-rays (Witschi et al. 1980).
Interestingly, lung morphology appeared to return to practically normal,
although lung collagen levels remained almost twice as high as in controls.
The influence of persisting high collagen concentrations on lung function
remains to be established. Increased hydroxyproline was also found if
exposure to 70% 02, a concentration which in itself does not produce fibrosis,
was limited to 48 h only. The maximum fibrogenic response developed when
animals were exposed to 02 immediately after BHT. Delay of 02 exposure for 48
or even 96 h still produced fibrosis, but its development was much diminished.
Significant fibrosis also developed if animals given BHT were exposed for 2 d
to O2 concentrations of >70%, for 3 d to >60%, or for 6 d to >50%; only 40% O2
had no significant effect. The most frequently used dose of BHT (400 mg/kg)
itself produced some accumulation of lung hydroxyproline. If the dose of BHT
was lowered to 200 or 100 mg/kg, no fibrosis developed. Yet exposure to 70%
02 following a nonfibrogenic dose of BHT was accompanied by the development of
fibrosis (Witschi et al. 1980b).
DISCUSSION
From these experiments we draw several conclusions. A form of chronic
lung damage—interstitial fibrosis—may result from an acute interaction of
two toxic agents in the lung. Timing is of crucial importance for the
occurrence of this interaction. Fibrosis only develops if O2, or possibly any
129
-------
12. TOXICOLOGICAL INTERACTIONS IN CHRONIC LUNG DISEASE Witschi
other toxic agent, is present during the phase of epithelial cell
proliferation following lung injury. Oxygen will not produce fibrosis once
reepithelialization is complete. If it develops, however, fibrosis persists
for up to 6 mo or longer. The interaction between (>2 and BHT does not occur
because one agent directly enhances the action of the other (Williamson et al.
1978). Rather, 02 adversely affects a cell population that is different from
the original BHT target cells and which has been called into action to repair
the initial damage.
It is conceivable that this scenario for the interaction of BHT and 02 in
mouse lung reflects a general principle underlying the development of chronic
lung damage. Ample experimental evidence documents that many toxic inhalants
and numerous bloodborne agents will produce Type I alveolar cell deaths
followed by Type II cell proliferation. Several chemicals are also known to
interfere with cell proliferation in the lung (Witschi and Cote 1977b). In
principle, fibrosis might therefore develop under anv circumstances in which
exposure to a first agent causing alveolar cell death is followed by exposure
to a second cytotoxic agent, provided a critically ordered sequence of
exposure is observed. It does not seem likely that the temporal relationships
needed to produce fibrosis (according to our proposed mechanism) always occur.
However, if the right conditions are met, the resulting lesion will likely
persist. Although single episodes might easily go unnoticed, repeated
episodes would eventually result in diffuse and recognizable lung damage.
130
-------
12. TOXICOLOGICAL INTERACTIONS IN CHRONIC LUNG DISEASE Witschi
CONTINUING STUDIES
Attempts have begun to verify our hypothesis with studies of agents other
than BHT and 02. Thus far, we have not been able to produce fibrosis by
exposing animals first to 2 ppm ozone (O3) followed by 70% O2. There are
three possible explanations: (1) our proposed mechanism for pathogenesis of
lung fibrosis is not as universal as assumed; (2) the cell kinetics following
exposure to 03 are different from those following exposure to BHT, and our
timing of 02 exposure following O3 was wrong; or (3) the initial lesion
produced by exposure to 03 was too small and comparable only to lesions
produced by low doses of BHT (50 mg/kg), which have not produced fibrosis.
The third possibility needs to be studied in more detailed experiments; such
studies could well lead to information on no-effect or negligible levels of 03
exposure. On the other hand, we have found that most animals treated with 400
mg/kg BHT will die if subsequently exposed for 48 h to 1 ppm 03. The
survivors develop extensive fibrosis. Animals with acutely damaged lung are
thus more susceptible to 03. This illustrates that the adverse effects of a
common oxidant on a diffusely damaged lung may be much more serious than the
effects on a healthy organ. Finally, it will be possible to study the effects
of airborne toxic agents on animals in which diffuse interstitial fibrosis has
been previously produced by exposure to BHT and 02. These studies will permit
examination of the effects of oxidants in an animal model which mimics a not
uncommon form of human lung disease, chronic interstitial fibrosis.
131
-------
12. TOXICOLOGICAL INTERACTIONS IN CHRONIC LUNG DISEASE Witschi
REFERENCES
Adamson, I. Y. R., D. H. Bowden, M. G. Cote, and H. P. Witschi. 1977. Lung
injury produced by butylated hydroxytoluene. Cytodynamic and biochemical
studies in mice. Lab. Invest., 36:26-32.
Brody, A. R. 1980. Personal communication.
Haschek, W. M., K. R. Meyer, R. L. Ullrich, and H. P. Witschi. 1980.
Potentiation of chemically induced lung fibrosis by thorax irradiation.
Int. J. Radiat. Oncol. Bio. Phys., 6:449-455.
Haschek, W. M., and H. P. Witschi. 1979. Pulmonary fibrosis—A possible
mechanism. Toxicol. Appl. Pharraacol., 51:475-487.
/^ *
Hirai, K. I., H. P. Witschi, and M. G. Cote. 1977. Electron microscopy of
butylated hydroxytoluene-induced lung damage in mice. Exp. Mol. Pathol.,
27:295-308.
National Research Council. 1980. Report of Panel on Evaluation of Hazards
Associated with Maritime Personnel Exposed to Multiple Cargo Vapors.
Washington, D.C., National Academy of Sciences (in press).
Williamson, D., P. Esterez, and H. P. Witschi. 1978. Studies on the
pathogenesis of butylated hydroxytoluene-induced lung damage in mice.
Toxicol. Appl. Pharmacol., 43:577-587.
Witschi, H. P., and M. G. Cote. 1977a. Inhibition of butylated
hydroxytoluene induced mouse lung cell division by oxygen: Time-effect
and dose-effect relationships. Chem.-Biol. Interactions, 19:279-289.
Witschi, H. P., and M. G. Cote. 1977b. Primary pulmonary responses to toxic
agents. CRC Grit. Rev- Toxicol., 5:23-66.
Witschi, H. P., W. M. Haschek, K. R. Meyer, R. L. Ullrich, and W. E. Dalbey.
1980a. A pathogenetic mechanism in lung fibrosis. Chest, 78:3953-3998.
Witschi, H. P., W. M. Haschek, A. J. P. Klein-Szanto, and P. J. Hakkinen.
1980b. Potentiation of diffuse lung damage by oxygen: Determining
variables. Am. Rev. Resp. Dis. (in press).
132
-------
12. TOXICOLOGICAL INTERACTIONS IN CHRONIC LUNG DISEASE Witschi
WORKSHOP COMMENTARY
D. L. Coffin; I gathered from the paper that there seems to be a correlation
of this fibrosis with the degree of epithelial reaction. In view of the
apparently strong interaction between macrophages and fibroblastic
hyperplasia, have you examined the status of the macrophage as an alternate
means of mechanism?
H. P. Witschi; That's a very good point. No, we haven't looked very much at
the macrophages. One of the reasons is that we were able to see only
comparatively few macrophages in the damaged lungs. Dr. Haschek performed the
histology and can confirm this.
I do not think, in this particular instance, that macrophages are
involved. Rather, I think that Q£ prevents epithelialization, which results
in the formation of fibroblasts. There are some experimental studies showing
exactly the same thing. Some very convincing results were reported by
Terzaghi [Terzaghi, M., P. Nettesheim, and M. L. Williams. 1978.
Repopulation of denuded tracheal grafts with normal, preneoplastic and
neoplastic epithelial cell populations. Cancer Res., 38:4546-4553]. Denuded
tracheal grafts, when implanted, filled up with fibroblasts in no time.
Inoculation with epithelial cells of a certain critical number resulted in no
fibrosis.
Question; Did you try altering the order in which you gave BHT and 03, to see
if the 03 set up a system for BHT damage?
H. P. Witschi; No, we have not done this yet.
Question; You mentioned that you thought this lesion might disappear after 6
mo to 1 yr. If it's fibrosis, why would you expect it to disappear?
H. P. Witschi; What persist are increased levels of hydroxyproline, which
probably indicate collagen. Interestingly enough, the change in
histopathology at 6 mo is not very impressive, although the lung still has a
large amount of collagen.
Question; Could this be just a .matter of turnover: you have more of a
turnover in your damaged lung than in the undamaged lung? If so, it isn't
fibrosis. It's just that more turnover results in more residual
hydroxyproline.
H. P. Witschi; No, we have increased synthesis and diminished degradation.
Question; So you had total hydroxyproline increase?
H. P. Witschi; Yes.
133
-------
12. TOXICOLOGICAL INTERACTIONS IN CHRONIC LUNG DISEASE Witschi
Comment; I don't quite understand. Let me put it this way: If you were to
do, say, connective tissue stains, would that correspond to your
hydroxyproline? In other words, you're saying that you don't have increased
collagen tissue but that you do have increased hydroxyproline. That's
confusing.
H. P. Witschi; It's merely a question of semantics. What is measured after
hydrolyzing the lung is hydroxyproline, or an index for collagen. I think
~90% of all lung hydroxyproline is in collagen.
134
-------
13. OVERVIEW OF CURRENT AND PLANNED RESEARCH
BY THE RESPIRATORY UNIT, OAK RIDGE NATIONAL LABORATORY
PART 1. RAT TRACHEAL TRANSPLANT SYSTEM
Ann C. Marchok
Respiratory Unit
Oak Ridge National Laboratory
Post Office Box Y
Oak Ridge, TN 37830
INTRODUCTION AND BACKGROUND
One overriding theme of the research effort in the Respiratory Unit at
Oak Ridge National Laboratory (ORNL) is the study of the morphogenesis of lung
cancer. A major part of this effort is directed towards developing and
utilizing experimental animal model systems to identify agents which might
enhance tumor induction or progression in the respiratory tract.
Currently, three approaches are used. The first approach is that of
inhalation experiments in which animals pretreated with carcinogens are
exposed to irritant gases. The tracheal washing technique developed at ORNL
by Schreiber et al. (1975) and further developed there by Yarita et al. (1978,
represents a second approach. With this system, we plan to precondition the
respiratory epithelium before exposure to aerosols. It should be emphasized
that this model is particularly amenable to studying any interrelationship
135
-------
13. RAT TRACHEAL TRANSPLANT SYSTEM Marchok
between the nutritional state of the animal and susceptibility. Studies with
both of these systems are discussed by Dalbey (Chapter 14 of this volume).
A third approach is the rat tracheal transplant system developed in our
laboratory by Kendrick et al. (1974). The rat trachea was chosen as a model
because of its structural similarity to the human bronchus. This report
discusses current and proposed studies with the rat tracheal transplant, most
of which fall under the EPA umbrella.
An inherent limitation in systems in which the carcinogen is introduced
into the respiratory tract by inhalation or the more common intratracheal
injection is the unpredictability of the site of tumor development. It is
also impossible to determine either the carcinogen dose or the duration of
carcinogen exposure for any given site of the respiratory tract. Both of the
model systems developed in our laboratory (i.e., the tracheal washing
technique and the tracheal transplant) have the advantage that the target site
for the carcinogen as well as for the cocarcinogen is well defined. An
accurate dose of the agent(s) can be delivered, the site of tumor development
can be predicted (Griesemer et al. 1977; Nettesheim et al. 1977; Pal et al.
1978), and multiple exposures of the same or different agents can be carried
out.
136
-------
13. RAT TRACHEAL TRANSPLANT SYSTEM Marchok
CONTROLLED DELIVERY OF THE TEST AGENT(S)
Obviously, delivery of the agent is a key issue in this model system.
Ideally, one would like to test very low levels of substances and have a high
degree of control over the rate and duration of exposure. The first tumor
induction studies were carried out with polycyclic hydrocarbons (PCH) released
from beeswax pellets. The rate of PCH release was determined by measuring the
material remaining in the pellet at different time intervals following
placement in the transplant.
B. Pal of our laboratory has developed an in vitro assay system to
pretest agent release rates prior to placement of pellets in transplants.
This method consists of shaking the carcinogen pellets in a flask containing
fetal calf serum at 37°C. The serum is changed at time intervals and the
carcinogen released into the serum is determined by radio assay. As shown in
Figure 13-1, the release rates in vitro"and _in vivo are
concentration-dependent (i.e., as the concentration in the pellet diminishes
with time, the amount released per time diminishes) . The rate of in vivo
release from pellets containing 100 yg PCH is faster (100% released in 2
weeks) than that from pellets containing ,1000 yg PCH. At low concentrations
of PCH, in vivo and jLn vitro release rates are similar. At high PCH
concentrations, in vivo rates are markedly slower than in vitro rates.
These studies prompted exploration of other possibilities for altering
and reducing PCH release. The first to be explored was the effect of adding
137
-------
13. RAT TRACHEAL TRANSPLANT SYSTEM
Marchok
100 {A
10 20
TIME (days)
30
Figure 13-1. Cumulative release of benzopyrene (BP) from beeswax pellets.
A, pellets containing 100 yg BP. o, in vitro release; •, i.n
vivo release in tracheal grafts. Each point represents the
mean observation from 4 pellets _in vitro ± S.D. and 6 pellets
in vivo ± S.D. B, pellets containing 1000 yg BP. o, ±n vitro
release; •, jln vivo release in tracheal grafts. Each point
represents the mean observation from 3 pellets in vitro ± S.D.
and 12 pellets in vivo ± S.D. C, pellets containing 1000 yg BP
implanted s.c. Each point represents the mean observation from
6 pellets ± S.D. From: Pal et al. (1978).
138
-------
13. RAT TRACHEAL TRANSPLANT SYSTEM Marchok
cholesterol or cholesterol laurate to achieve beeswax:cholesterol proportions
ranging from 1:1 to 1:9. As seen in Figure 13-2, cholesterol was very
effective in retarding BP release. At a beeswax:cholesterol concentration of
1:9, the average amount of BP released in vitro within 1 week from pellets
containing 100 ug BP was ~13% as compared to ~90% released during the same
time from pure beeswax pellets. A similar retardation of BP release was
observed in the jln vivo studies.
Our current effort follows similar lines. Pal is attempting to modify
the cholesterol and is pursuing many other kinds of substances, including
various polymers and biodegradable materials, as matrices to control the
release of agents having a wide range of chemical properties. The goal is to
examine the inductive or cocarcinogenic effects of such agents on the
respiratory epithelium. Since the model is still under development, we
currently use commonly known carcinogens and cocarcinogens. We already have
an indication that slow release of a particular carcinogen dose is much more
tumorigenic than rapid release. One reason for this, most likely, is a
reduction in the initial toxic effect.
INCREASED SPECIFICATION OF END POINTS
The common end point in testing agents for induction and promotability of
cancer is actual development of a tumor in animal survival studies. There is
a great need for end points that are less costly and time-consuming. Our
laboratory is attempting to reduce this end point in time and actually use
139
-------
13. RAT TRACHEAL TRANSPLANT SYSTEM
Marchok
100-
0
10 20
TIME (days)
30
Figure 13-2. Cumulative release of BP from pellets with modified matrix, at
100 yg BP per pellet. A, _in vitro studies, o, beeswax pellets;
A, beeswax:cholesterol laurate pellets at a ratio of 1:9; Q,
beeswax:cholesterol pellets at a ratio of 1:1; H, beeswax:
cholesterol pellets at a ratio of 1:3; I, beeswax:cholesterol
pellets at a ratio of 1:9. Each point represents the mean of 2
to 5 observations ± S.D. B, in vivo studies in tracheal grafts.
o, beeswax pellets; <», beeswax:cholesterol pellets at a ratio
of 1:3; •, beeswax:cholesterol pellets at a ratio of 1:9. Each
point represents the mean of 4 to 6 observations ± S.D. From:
Pal et al. (1978).
140
-------
13. RAT TRACHEAL TRANSPLANT SYSTEM Marchok
numbers and kinds of lesions identified by ways other than and/or in addition
to morphology- The design of these experiments is as follows:
1. Expose transplants to 200 yg dimethylbenzanthracene (DMBA).
2. At set times, sample transplants and cut into 2 x 3 ml
explants.
3. Place into organ culture. After 24 h, collect media and
prepare the exfoliated cells for cytopathologic diagnosis.
Fix explant in one set for pathologic diagnosis. Compare.
Place second set of explants into outgrowth culture to
determine other markers of cellular alterations.
In studies funded by the National Cancer Institute, we are beginning to
follow the fate of cell populations from specific lesions. The properties to
be determined are:
1. Rate of epithelial outgrowth
2. Maintenance of primary cultures in suboptimal media
(increased jLn vitro growth capacity)
3. Focal morphological changes in the primary cultures
4. Subculturability of the primaries into cell lines
5. Markers of transformation in the cell lines:
(a) growth in agar
(b) formation of tumors
6. Correlation of lesion severity as determined
morphologically to the time of appearance of markers
of transformation
141
-------
13. RAT TRACHEAL TRANSPLANT SYSTEM Marchok
The basic methodology for these studies has been described previously
(Marchok et al. 1977, 1978). So far, we have observed the following:
1. A very early marker of transformation is an increased
capacity to survive and grow in vitro.
2. The brief in vivo carcinogen exposure "initiated" some
cells, and the processes initiated in vivo proceeded
in vitro, leading to the emergence of some neoplastic
cell populations.
3. Since the primary cell cultures and most cell lines were
not malignant when first tested, but some became so later,
progression of the neoplastic disease must have occurred
in vitro.
4. There were dose-related effects:
(a) More primary cultures and cell lines could be
established from tracheas exposed to a high
(640 vg DMBA) than a low (165 yg DMBA) dose.
(b) There were differences in the in vitro "latency"
periods (i.e., the time cells are maintained before
becoming oncogenic).
(c) Tumor induction time (i.e., the time from cell
inoculation to development of a palpable tumor)
correlated in most cases with j.n vitro latency.
Tumor induction times were 9 to 60 d for cell lines
with short in vitro latency periods and from 100 to
250 d for cell lines with long jLn vitro latencies.
(d) The combined in vitro-in vivo tumor latency time
was similar to tumor induction times found in the
tracheal transplants in vivo.
With this experimental approach, we should achieve the following:
1. Direct determination of the fate of specific lesion
types in terms of:
142
-------
13. RAT TRACHEAL TRANSPLANT SYSTEM Marchok
(a) acquisition of the neoplastic state
(b) tumor latency
2. Identification of other markers (end points) for
determining the tumorigenic potential of cells
3. Identification of cell populations at risk in the
respiratory system of experimental animals
4. Use of these end points to test the carcinogenic
potential of environmental agents on the respiratory
system
5. Application of the same criteria to the cytopathology
and to biopsies of human tissue
Looking ahead, it should be emphasized that this approach can be utilized
regardless of the form of exposure or type of agent(s) tested.
REFERENCES
Griesemer, R. A., P. Nettesheim, D. H. Martin, and J. E. Caton, Jr. 1977.
Quantitative exposure of grafted rat tracheas to
7,12-dimethylbenz(a)-anthracene. Cancer Res., 37:1266-1271.
Kenrick, J., P. Nettesheim, and A. S. Hammons. 1974. Tumor induction in
tracheal grafts: A new experimental model for respiratory carcinogenesis
studies. J. Natl. Cancer Inst., 52:1317-1325.
Marchok, A. C., J. C. Rhoton, R. A. Griesemer, and P. Nettesheim. 1977.
Increased in vitro growth capacity of tracheal epithelium exposed in vivo
to 7,12-dimethylbenz(a)anthracene. Cancer Res., 37:1811-1821.
Marchok, A. C., J. C. Rhoton, and P. Nettesheim. 1978. In vitro development
of oncogenicity in cell lines established from tracheal epithelium
preexposed in vivo to 7,12-dimethylbenz(a)anthracene. Cancer Res.,
38:2030-2037.
Nettesheim, P., R. A. Griesemer, D. H. Martin, and J. E. Caton, Jr. 1977.
Induction of preneoplastic and neoplastic lesions in grafted rat tracheas-
continuously exposed to benzo(a)pyrene. Cancer Res., 37:1272-1278.
143
-------
13. RAT TRACHEAL TRANSPLANT SYSTEM Marchok
Pal, B. C., D. C. Topping, R. A. Griesemer, F. R. Nelson, and P. Nettesheim.
1978. Development of a system for controlled release of benzo(a)pyrene,
7,12-dimethylbenz(a)anthracene, and phorbol ester for tumor induction in
heterotopic trachea1 grafts. Cancer Res., 38:1376-1383.
Schreiber, H., K. Schreiber, and D. H. Martin. 1975. Experimental tumor
induction in a circumscribed region of the hamster trachea: Correlation
of histology and exfoliative cytology. J. Natl. Cancer Znst.,
54:187-197.
Yarita, T., P. Nettesheim, and M. L. Williams. 1978. Tumor induction in the
trachea of hamsters with N-nitroso-N-methylurea. Cancer Res.,
38:1667-1676.
WORKSHOP COMMENTARY
D. L. Coffin; In the work that showed an interaction between DMBA and
asbestos, how was the asbestos applied to the implanted tracheas?
A. C. Marchok; The asbestos was released from gelatin pellets.
D. L. Coffin; And the DMBA was in the beeswax?
A. C. Marchok; Yes. They were exposed sequentially.
D. L. Coffin; Which came first?
A. C. Marchok; The DMBA came first as the initiator; asbestos was evaluated
as a promoter.
D. L. Coffin; How long after DMBA was the asbestos administered?
A. C. Marchok; The gelatin pellets containing asbestos were inserted into the
tracheas 4 weeks after the start of DMBA exposure.
144
-------
14. OVERVIEW OP CURRENT AND PLANNED RESEARCH
BY THE RESPIRATORY UNIT, OAK RIDGE NATIONAL LABORATORY
PART 2: TRACHEAL WASHING SYSTEM AND OXIDANT INHALATION EXPERIMENTS
Walden E. Dalbey
Respiratory Unit
Oak Ridge National Laboratory
Post Office Box Y
Oak Ridge, TN 37830
INTRODUCTION
This report provides an overview of current Oak Ridge National Laboratory
studies on inhalation of oxidant gases. Three main areas are discussed: (1)
irritant gases as cofactors in nitrosamine-induced tumorigenesis, (2)
modulation of tumor incidence using the trachea! washing model for tumor
induction, and (3) the role of frequency and duration of oxidant gas exposure
in pulmonary toxicity.
Since man is rarely exposed to doses of chemical carcinogens large enough
to result in tumors by themselves, investigations of factors which could
influence tumor incidence are of great practical importance. Such potential
enhancement is the focus of the first two studies described below.
145
-------
14. TRACHAEL WASHING SYSTEM AND OXIDANT INHALATION EXPERIMENTS Dalbey
IRRITANT GASES AS COFACTORS IN NITROSAMINE-INDUCED TUMORIGENESIS
In this study, subcutaneous administrations of diethylnitrosamine (DEN)
were performed to induce tumors in the respiratory tracts of male Syrian
golden hamsters. This use of a systemic carcinogen was designed to decrease
the variability in tumor response that is apparently inherent with
intratracheal instillation of carcinogens.
Hamsters received 10 weekly injections of 0.5 mg DEN. In addition,
groups of animals were exposed to either nitrogen dioxide (N<>2) or
formaldehyde (HCHO) for life. These gases were chosen as model lower and
upper respiratory tract irritants, respectively. The irritant treatment was
further subdivided so that irritant exposures began at one of two times in
relation to DEN treatment: (1) 48 h prior to each weekly DEN injection, so
that the respiratory epithelium was actively proliferating during the time of
carcinogen treatment, or (2) 2 weeks after the final DEN injection. Irritant
gas exposures were given weekly for the lifetime of each animal. Ancillary
data indicated that such weekly exposures would result in epithelial necrosis
and proliferation subsequent to each exposure analogous to that occurring
after the initial exposure. In other words, the animals would demonstrate no
adaptation with weekly exposures.
After each animal's death, the incidence of respiratory tract tumors was
evaluated by means of a clearing technique rather than histology. Fixed
tissue was stained with Wright1s stain and rendered semitransparent by
146
-------
14. TRACHAEL WASHING SYSTEM AND OXIDANT INHALATION EXPERIMENTS
Dalbey
dehydration and eventual transfer into methyl salicylate. Areas of dense cell
aggregation (tumors) stained darkly and were readily visible under a
dissecting microscrope. Observations of tumors at this subgross level
correlated well with subsequent microscopic evaluation, except that several
tumors found at the subgross level would not have been observed during routine
histological procedures. There were no false positive or negative tumor
identifications during the histological evaluation. All observed tumors were
adenomas; no invasive tumors were observed. As of this writing, the
experimental data are still tentative. We have not yet fully evaluated the
incidence of tumors in control animals.
We obtained data on tumor incidence in animals exposed to DEN and either
10 ppm NC>2 or 30 ppm HCHO. The incidence of nasal tumors was low and not
influenced by irritant exposures. Most of the observed tumors were in the
larynx and trachea, with a smaller percentage occurring in the lung. With the
DEN dose used, the percentage of tumor-bearing animals was high, and no
differences in percentage of tumor-bearing animals were observed between
experimental groups. However, the number of tumors per tumor-bearing animal
was significantly increased in the tracheas of animals that were concurrently
exposed to HCHO and DEN. Also, there was a significant increase in total
respiratory tract tumors in animals concurrently receiving N02 exposures and
DEN treatment, although there was no significant increase in either the
larynx, trachea, or lung individually.
147
-------
14. TRACHAEL WASHING SYSTEM AND OXIDANT INHALATION EXPERIMENTS Palbey
It appeared from these observations that both irritant gases acted to
enhance tumor incidence when the carcinogen was administered at a time of
epithelial proliferation resulting from irritant gas exposure. No "promoting
activity" was observed.
Partly in view of this lack of observed promoting activity, and also in
view of the relative paucity of demonstrated promoting activity in the
respiratory tracts of whole animals, we entered into the second major area of
work: modulation of tumorigenesis using the tracheal washing technique for
tumor induction.
TRACHEAL WASHING SYSTEM
Marchok (Chapter 13 of this volume) has discussed the development of the
tracheal washing technique at our laboratory. The anesthetized hamster is
placed on its back and a small cannula with a double lumen is inserted through
the larnyx. A buffered solution containing N-methyl-N-nitrosourea (NMU) is
injected via a syringe pump through the outer lumen of the cannula. The
solution moves 5 mm down the tracheal epithelium before reaching the elongated
tip of the inner tube of the cannula, where suction draws the solution out of
the trachea. Thus, by placing the cannula tip at a standard location within
the trachea, a particular site of tracheal epithelium can be repeatedly washed
with NMU solution.
148
-------
14. TRACHAEL WASHING SYSTEM AND OXIDANT INHALATION EXPERIMENTS Daibey
This model of tumor induction offers several advantages over
intratracheal instillation or systemic administration of carcinogen for
whole-animal studies. These advantages include: more quantitative delivery
of the carcinogen to a specific site of respiratory epithelium; a small,
predictable location of tumor development to facilitate histological sampling
of both cancerous and precancerous lesions; and induction of carcinomas in
addition to noninvasive lesions. We believe the full potential of this system
has not yet been realized, and we are currently utilizing it in multifactorial
studies. The first is a promotion study in which the tracheal epithelia of
NMU-treated animals are also exposed to a known tumor promoter, croton oil.
An aerosol of croton oil provides a noninvasive means to repeatedly treat the
epithelium. We also plan tumorigenesis studies involving modification of the
physiological state of the animal (namely, vitamin A deficiency). The system
is amenable to studies with other agents or physiological alterations.
EFFECTS OF INHALED NITROGEN DIOXIDE ON PULMONIC LESIONS IN RATS
A third area of work in our laboratory concerns the role of duration and
frequency of exposure to oxidant gases in the severity of resulting pulmonic
lesions. Much of this work relates to the biological effects of acute peaks
of oxidant concentration. So far, most of these have involved exposure of
specific pathogen free (SPF) rats to NO2-
As observed in other laboratories, we noted a dose-related increase in
incorporation of tritiated thymidine into pulmonary DNA after acute NO2
149
-------
14. TRACHAEL WASHING SYSTEM AND OXIDANT INHALATION EXPERIMENTS Dalbey
exposure. The peak incorporation occurred ~24 h after exposure and was linear
with the concentration of exposure. The limits of detection were rather high
(~10 ppm NO2 for a 5-h exposure) when the biochemical method was used to
determine thymidine incorporation. Also, we observed an increase in
phospholipids after single NO2 exposures, with a peak in pulmonary
phospholipids at ~48 h after exposure. The disaturated species of
phosphatidylcholine increased preferentially over unsaturated species,
indicating that the increase in phospholipids may be in part due to increased
biosynthesis.
We conducted additional experiments on the influence of vitamin A
deficiency on the response to acute NO2 exposures, but the data are incomplete
at this time. Vitamin A is important in maintenance of the normal epithelium.
We thought it would be of interest to examine further the relative
importance of exposure duration with single acute exposures to NO2. The end
points used were thymidine incorporation, amount of phospholipids, and
morphology. We compared 1- and 5-h exposures. Unlike the linear increases in
thymidine incorporation and amount of phospholipids in relation to NO2
concentration, the values for these end points appeared to plateau as exposure
duration increased. Presumably the lungs were adapting during the exposure.
We subsequently found that only one 1- or 5-h exposure to NO2 was sufficient
to adapt the animals to a similar exposure on the following day. That is,
after the second exposure the animals failed to respond with similar evidence
of cell necrosis or proliferation.
150
-------
14. TRACHAEL WASHING SYSTEM AND OXIDANT INHALATION EXPERIMENTS Dalbey
Adaptation to NO2 therefore appears to occur rapidly. We subsequently
investigated the duration of adaptation. Animals were reexposed either 1, 3,
or 7 d after adaptation had been initiated, and thymidine incorporation was
monitored subsequent to the second exposure. Data were returned demonstrating
that adaptation was complete with a 1-d interval, intermittent with a 3-d
interval, and nonexistent after 7 d between exposures. We can assume,
therefore, that once-a-week exposures would result not in adaptation to N(>2
but in occurrence of the same lung events observed after the initial
exposure.
To our knowledge, the emphysematous lesions that have been reported with
chronic exposures of rats were observed under daily exposure regimes in which
the animals were probably in the adaptive state. We asked whether similar
emphysematous lesions could be produced with weekly exposures of nonadapted
rats. SPF male Fischer 344 rats were exposed once per week, 5 h/d, to 20 ppm
NC>2. This high concentration was used to ensure an observable effect on the
lungs. The experiment has not been completed, but some data on pulmonary
function are available.
Some animals were killed after 6 and 18 mo of exposure; additional
animals remain in a recovery period. Pulmonary function tests were performed
immediately before killing the animals; we obtained data on pulmonary
compliance, lung volume, pulmonary resistance, nitrogen washout, and flow
volume. These data did not indicate any change in pulmonary function
following N02 exposure. After the pulmonary function tests, lungs from the
151
-------
14. TRACHAEL WASHING SYSTEM AND OXIDANT INHALATION EXPERIMENTS Dalbey
animals were fixed under constant pressure and paraffin sections prepared. No
change in mean linear intercept was observed after 6 mo of exposure; data on
the 18-mo exposure are not yet available.
If subsequent data obtained in this study are similar, the negative
results may represent a significant finding that the emphysematous lesions
previously reported to occur with chronic NO2 exposure do not occur under
conditions of sporadic peak exposures. Our laboratory is currently planning
similar work with ozone, but these plans are readily adaptable to the needs of
EPA.
152
-------
15. EARLY STAGES OF RESPIRATORY TRACT CANCER: A REVIEW
Carol A. Heckman
C. C. Scotta
A. C. Olson
F. Snyderb
Biology Division
Oak Ridge National Laboratory
Post Office Box Y
Oak Ridge, TN 37830
INTRODUCTION
About 90% of all human cancers appear to originate in epithelia. The
epithelial linings of several organs, including the respiratory tract and gut,
are topologically continuous with the exterior environment. These linings
make up the body's first line of defense•against the entry of noxious
environmental agents. Simply by reason of their spatial location, epithelia
may be particularly susceptible to the toxic and carcinogenic effects of
environmental agents. Cumulative effects and delayed effects are of
particular concern, because their detection by epidemiologic methods may not
be possible until decades after the initial exposure has taken place. An
increased understanding of delayed effects—particularly of carcinogenicity—•
would be useful for identifying and controlling suspect agents.
Presently at Department of Nutrition, Harvard University School of Public
Health, 665 Huntington Avenue, Boston, MA 02115.
bOak Ridge Associated Universities, Oak Ridge, TN 37830.
153
-------
15. EARLY STAGES OF RESPIRATORY TRACT CANCER Heckman et al_.
To define changes that are specifically correlated with carcinogenesis,
we need to identify features that distinguish between normal and neoplastic
epithelium at the cellular level. For technical reasons having to do with the
growth requirements of cultured cells, the features best understood at the
cellular level have been studied in fibroblasts and embryonic mesenchymal
cells. While these models are far from ideal for studying carcinogensis/ they
have the advantage of yielding rapid and quantitative data for a number of
experimental end points. These systems have been used to define differences
in the surface structure of the cells (Rapin and Burger 1974; Porter et al.
1973; Borek and Fenoglio 1976), contact inhibition of growth (Aaronson and
Todaro 1968; Holley and Kiernan 1968), anchorage-independence (Jones et al.
1976; Shira et al. 1975), loss of the protein fibronectin from cell surfaces
(Hynes 1974; Yamada and Weston 1975), and increased plasrainogen activator
activity (Ossowski et al. 1973; Unkeless et al. 1973). All of these
differences have been described in the last 10 to 12 years.
Until quite recently, many of us who work on carcinogenesis had
considered that at least some of the features that discriminate between normal
and transformed fibroblasts would also discriminate between normal and
neoplastic epithelium. Such features would provide a "tag" for neoplastic
cells that could possibly be applied to a practical problem: defining
criteria to use in short-term carcinogenesis assays and in early clinical
detection of cancer.
154
-------
15. EARLY STAGES OF RESPIRATORY TRACT CAKCER Heckman et al.
Unfortunately, the prospect of directly applying what has been learned
from fibroblast models to studies of epithelium now seems less promising.
Recent studies have reported no changes in the amounts of fibronectin on cell
surfaces for bladder or for salivary or mammary gland epithelium (Wigley and
Summerhayes 1979; Yang et al. 1980). No difference in plasminogen activator
activity has been found for normal and neoplastic mammary epithelium (Wigley
and Summerhayes 1979). Both our studies (Heckman and Olson 1979) and those of
McGrath and Medina on the mammary gland (Voyles and McGrath 1976; Butel et al.
1977) have shown little consistent difference in contact inhibition of growth.
Our group has also shown the density of surface features to correlate poorly
with the tumorigenicity of cell lines from rat tracheal epithelium (Heckman
and Olson 1979). Finally, anchorage-independence and lectin-mediated
agglutinability, although somewhat more promising than the other criteria,
have not been very reliable in discriminating between normal and tumor-derived
human cells (Franks 1979). These two markers are currently being studied by
many other research groups.
A few years ago, when we began to look for markers to use in studies of
progressive changes in the respiratory tract, we considered that many or most
of the biochemical differences that had been described for fibroblast
transformation models would relate to their differentiated functions and would
therefore be specific to mesenchymal cells. Recent reports by other
investigators indicate that this may be the case. Based on this assumption,
we decided to concentrate our efforts on a few criteria which had been shown
to discriminate between normal and neoplastic epithelium. One of these was
155
-------
15. EARLY STAGES OF RESPIRATORY TRACT CANCER Heckman et al.
the accumulation of ether-linked lipids, which had been described for whole
tumors by Snyder and Wood (1968) but studied rarely since then. These lipids,
particularly the alkyldiacylglycerols, are present only in trace amounts in
normal tissues of most organs. A second marker was described by Montesano and
Sanford (Montesano et al. 1977) for normal and neoplastic cell lines grown in
culture. These investigators compared several lines and found differences in
the morphology and cytology of cells in colonies.
THE ETHER LIPID MARKER
Before studying sequential changes in carcinogen-treated respiratory
tract tissues, we wanted to confirm that the lipids occur in tumors from the
respiratory tract but not in normal tissues. For this, four transplantable
rat tumors obtained after benzo(a)pyrene (BaP) treatment were studied.
Thin-layer chromatography showed the alkyldiacylglycerols to be present in the
squamous cell carcinomas but lacking in the normal lung (Figure 15-1). All
four transplantable tumors had high levels of alkyldiacylglycerols. We wanted
to determine how soon after carcinogen treatment this marker could be detected
in respiratory tract epithelium. To do this, we used a lipid precursor,
1-3H-hexadecanol, specifically incorporated into the ether linkage (Topping et
al. 1978), since mass determinations of the lipids would have required large
numbers of animals for quantification, particularly in the case of normal
epithelium.
156
-------
15. EARLY STAGES OF RESPIRATORY TRACT CANCER
Heckman et al,
Figure 15-1.
Thin-layer chromatogram of lipids extracted from normal lungs
(Lanes 1-2) and transplantable carcinomas (Lanes 3-6).
TABLE 15-1. 1-3H-HEXADECANOL INCORPORATION INTO TRIGLYCEROLS AND
ALKYLDIACYLGLYCEROLS OF TRACHEAL IMPLANT EPITHELIUM (DPM X 10~3)
Treatment
Time After Precursor Delivery (d)
21
16-week reversal)
42
untreated
control
DMBA
(4-week exposure)
DMBA
(4 -week exposure,
3.01
±1.94 SE
18.70
±12.94
8.34
±2.86
0.72
±0.36/ SE
7.58
±6.12
4.62
±1.16
0.50
±0.06 SE
5.99
±2.84
2.36
±1.88
0.12
±0.05 SE
0.33
±0.29
0.20
±0.13
0.13
±0.02 SE
0.21
±0.19
-
157
-------
15. EARLY STAGES OF RESPIRATORY TRACT CANCER Heckman et al.
The animal model used for studies of the lipid marker was the tracheal
implant (Marchok, Chapter 13 of this volume) after treatment with the potent
carcinogen 7,12-dimethylbenz(a)anthracene (DMBA). To review the sequence of
morphological changes obtained with the tracheal implant model, after a 4-week
exposure to the carcinogen we observe a squamous metaplastic response mixed
with atrophic epithelium. By 2 mo after the exposure is terminated, we begin
to see focal areas of squamous metaplasia with atypia. Lesions with marked
atypia can be found by 4 mo after termination of the exposure.
A recent paper dealing with the morphological changes induced by various
polyaromatic hydrocarbons (Lumb and Snyder 1971) demonstrated that the effects
of the carcinogenic hydrocarbons DMBA and BaP are slow to be reversed in
comparison to those of the noncarcinogenic or weakly carcinogenic
hydrocarbons. Although the lesions obtained with the two carcinogens differ
in morphology, they are similar in that they persist in the epithelium. Thus,
the concept of reversibility appears to be important in this model system.
Table 15-1 shows the levels of tritium incorporated from 1-3H-hexadecanol
into the alkyldiacylglycerol and triglycerol fractions at various times after
delivery of the precursor to tracheal implants. The amount of label
incorporated into the normal implants was minimal and declined relatively
rapidly. The effect of a 4-week exposure to DMBA was striking. The average
level of labeled alkyldiacylglycerols was 6 times higher than in the normal
epithelium, and the turnover of the label appeared to be slower. Precursor
incorporation was also studied at 4 mo after termination of carcinogen
158
-------
15. EARLY STAGES OF RESPIRATORY TRACT CANCER Heckman et al.
treatment (i.e., at the time when atypical lesions can first be detected).
The level of label incorporated at this time was 2 to 3 times higher than in
the normal epithelium, and the turnover again appeared slower.
With the ether lipid marker, then, we succeeded in identifying a
biochemical change that resembles previously studied morphological changes in
that it is slow to be reversed. The accumulation of ether lipids,
particularly the alkyldiacylglycerols, is also somewhat remarkable in that it
occurs very early after carcinogen treatment.
We also wanted to understand the biochemical basis of ether lipid
accumulation. To study this problem, we cultured normal rat tracheal
epithelium in parallel with cells from one of our transplantable tumor lines.
In particular, we wanted to know whether the turnover of the ether lipids was
impaired in carcinogen-treated epithelium. If this were the case, then
absence of the cleavage enzyme might provide a more direct marker for
preneoplastic changes. However, when grown in vitro, the normal epithelium
synthesized levels of alkyl lipids were equivalent to those of the tumor
cells. Incorporation in alkyldiacylglycerols of labeled palmitate (the
precursor) was similar for the two cell types,, as was the rate of turnover
(Scott et al. 1979a).
The unexpected similarity of lipids from normal and neoplastic cells
grown in vitro suggested that alkyl lipid accumulation might be related to
glucose levels, which are high in the culture medium and which are known to be
159
-------
15. EARLY STAGES OF RESPIRATORY TRACT CANCER Heckman et al.
elevated in many neoplastic cells. In a second series of experiments, we
showed this to be the case. By growing neoplastic cells from two of our
transplantable cell lines in media supplemented with varied levels of glucose,
we could induce the accumulation of correspondingly high or low levels of the
triglycerides and their ether analogues (Table 15-2). The effect of glucose
is related to levels of the precursors needed for biosynthesis of the ether
lipids. The precursor for the glycerol backbone is dihydroxyacetone
phosphate, which is formed as a direct glycolytic intermediate, according to
the Embden-Meyerhof-Parnas scheme. It is not clear how synthesis of the other
necessary precursors, the long-chain fatty alcohols, might be related to
glucose levels. Our evidence indicates that their biosynthesis is regulated
by an as yet unidentified product of glycolysis (Scott et al. 1979b). In any
case, these results suggest the feasibility of a more direct assay relating to
glycolytic metabolism for application to the problem of detecting
preneoplastic changes in the respiratory tract.
THE CELL SHAPE MARKER
We undertook experiments similar to previously described studies using
liver cells (Montesano et al. 1977) to examine the possibility that
morphological features could serve as markers for early stages of cancer in
the respiratory tract. The model system was an ^.n vitro system comparable to
the DMBA tracheal implant model. Two cell lines derived from the tracheal
epithelium after 2 weeks of carcinogen exposure (Marchok et al. 1978) were
used. Figure 15-2 diagrams the development of oncogenicity in these lines.
160
-------
15. EARLY STAGES OF RESPIRATORY TRACT CANCER
Heckman et al
TABLE 15-2. VARIATION OF TRIGLYCEROL AND ALKYLDIACYLGLYCEROL CONTENT
WITH LEVELS OF GLUCOSE IN CULTURE MEDIAa
Cell
Line
B2-1
BP3-0
Glucose
Concentration
(mM)
0
0.2
1.0
5.5
0
0.2
1.0
5.5
Triglycerol
Content
(yg/mg protein)
4.8
12.3
61.6
109.6
21.1
22.2
22.8
35.6
Alkyldiacylglycerol
Content
(yg/mg protein)
5.7
14.6
41.8
75.5
6.7
9.5
13.0
34.6
aData from Scott et al. (1979b).
INITIATION TO TIME OF IN VITRO
H VITRO CULTURE CULTURE
y looow
* •
p— T
tOOO WT
f
I TIME OF
IN VIVO
GROWTH
ess
BP 3-0
—T
Figure 15-2.
Diagrammatic representation of in vitro derivations and
experimental sampling points (*) for three rat tracheal
epithelial lines.
161
-------
15. EARLY STAGES OF RESPIRATORY TRACT CANCER Heckman et al.
The 1000 W line, upon becoming oncogenic, formed tumors in all
immune-suppressed hosts tested. As positive and negative controls, a cell
line that formed tumors in relatively few sites (165 S) and a highly oncogenic
line from a transplantable tumor (BP 3-0) were used.
To ensure that we identified changes linked to the oncogenic status of
the line rather than with the amount of time it was carried in in vitro
culture, a tumor-derived subline of 1000 W was also studied. In this way, we
could examine an additional oncogenic 1000 W population that had been
maintained in culture for only a brief time.
There were no significant differences in kinetic characteristics of cells
in colonies from the early and the tumor-derived 1000 W populations. For
example, Figure 15-3 plots labeling index versus the number of cells in the
colonies. In colonies from the early passage of the line, the cells tended to
be flat and regular. However, in colonies from late passages and from the
tumor-derived subline, the cells tended to be more elongated (although many
were still regular in shape). A quantitative analysis of the shape criterion
confirmed this visual impression (Heckman and Olson 1979).
It is important that we identify the cellular mechanism underlying the
changes in cell shape. Particularly if these changes involve an extracellular
constituent, they could provide a second marker for in vivo changes related to
carcinogenesis. One way to approach the problem, in terms of the cellular
biology of the model, would be to perturb the cells from nononcogenic
162
-------
15. EARLY STAGES OF RESPIRATORY TRACT CANCER
Heckman et al
LABELING INDEX
100
80-
LJJ
60
X
CO
CD
Z
<
oc
O
a_
cc
O
O
40
3
_i
LU
O
20
234567
POPULATION DOUBLINGS (number)
8
Figure 15-3.
Percentage of cells incorporating 3H-thyniidine in colonies from
the 1000 W (E) and tumor-derived 1000 W (T) lines.
163
-------
15. EARLY STAGES OF RESPIRATORY TRACT CANCER Heckman et al^
populations and see if they could be induced to mimic the oncogenic
populations. To accomplish this obviously requires a more quantitative assay
for cell shape, which we developed using a computerized image analysis method.
Initial tests demonstrated the method's success in discriminating among cell
lines (Olson et al. 1980) (Figure 15-4). In preliminary studies of the 1000 W
cell line, we learned that differences in cell shape can be detected in single
cells as opposed to colonies. It is also possible to calculate the mean for
individual cells in a population and to use that value to estimate the length
of time the population has been carried in vitro. While at present it is not
clear how this particular end point might be applied to problems of in vivo
detection of preneoplastic changes, it remains a promising area for future
development.
ANOMALIES IN RESPIRATORY TRACT CARCINOGENESIS
The final section of this review discusses some findings that we consider
anomalous in terms of the usual conceptualization of lung cancer etiology.
Hopefully, an awareness of these findings will at least prevent us from
becoming too complacent about our current understanding of carcinogenesis in
the respiratory tract.
Several years ago, we had the good fortune to collaborate with Dr. Walden
Dalbey in a chronic inhalation study of the effects of tobacco smoke on the
lung. The studies were performed with Fisher 344 rats; thus, the findings may
only be indicative of the processes which take place in the lungs of larger
164
-------
1000W
165S
2C5
Figure 15-4.
Jk_
•if
0
0
26
T
0
0
17
™
1
28
4
T
0
7
3
™
49
0
0
T"
0
6
0
K
CO
CO
H
1
o
o
I
Hierarchical cluster analysis of cell shape data input for 50-cell samples from three
rat tracheal epithelial cell lines.
a
n>
o
m
rt
-------
|5. EARLY JTAGES OF RESPIRATORY TRACT CANCER Heckman et al.
animals. However, the results were useful in determining the fate of
particulates in the lung in a situation of particulate overloading.
Particulates in the tobacco smoke we studied ranged from 0.1 to 0.6 ym in
actual diameter but were accumulated into larger masses within the
macrophages. By 3 d after cessation of smoking, nearly all of the
particulates were found within macrophages. With chronic exposure, the
macrophages were present not only in the alveolar space but also in the
interstitial connective tissue. They were particularly common in the
connective tissue at the terminal bronchiole and at the pleura. These sites
suggested a relationship with the patterns of lymphatic drainage in the rat
lung. However, there was nothing more than this circumstantial evidence to
implicate the lymphatics.
The major type of lesion found in the lung was typically near the
alveolar duct. Most frequently, these lesions developed near sites where the
ducts terminated on the pleura. In serial sections, the alveolar duct lesions
were often adjacent to a branch of the pulmonary vein which had an
accumulation of macrophages in the perivascular adventitia. Because of this
juxtaposition, it seems likely that perivascular lesions may contribute to the
development of lesions in the parenchyma surrounding the adjacent alveolar
duct.
At the cellular level, it seems anomalous that the vast majority of deep
lung particulates occur in macrophages while the macrophage does not itself
166
-------
15. EARLY STAGES OF RESPIRATORY TRACT CANCER Heckman et al
seem to be a target cell for carcinogens in this, organ. We know that the
macrophage can metabolize BaP, and that it should be able to form reactive
metabolites. At one time, cell biologists also thought that the alveolar
macrophage did not divide and could not, therefore, be a target for
transformation. We now know that these cells are capable of division.
A second apparent anomaly involves the sites in which macrophages
accumulate in the lung parenchyma. The most common sites are in the
perivascular adventitia, yet the cells of the vasculature do not seem to be
target cells, either.
These questions may be resolved when there is better insight into the
biological and metabolic characteristics of different cell types in the deep
lung. Meanwhile, they should serve to remind us that the field of pulmonary
carcinogenesis still poses a number of unresolved questions.
REFERENCES
Aaronson, S. A., and G. J. Todaro. 1968. Basis for the acquisition of
malignant potential by mouse cells cultivated in vitro. Science,
162:1024-1026.
Borek, C., and C. M. Fenoglio. 1976. Scanning electron microscopy of surface
features of hamster embryo cells transformed in vitro by X-irradiation.
Cancer Res., 36:1325-1334.
Butel, J. S., J. P. Dudley, and D. Medina. 1977. Comparison of the growth
properties in vitro and transplantability of continuous mouse mammary
tumor cell lines and clonal derivatives. Cancer Res., 37:1892-1900.
Franks, L. M. 1979. Personal communication to C. M. Heckman.
167
-------
15. EARLY STAGES OF RESPIRATORY TRACT CANCER Heckman et al.
Heckman, C. A., and A. C. Olson. 1979. Morphological markers of oncogenic
transformation in respiratory tract epithelial cells. Cancer Res.,
39:2390-2399.
Holley, R. W., and J. A. Kiernan. 1968. "Contact inhibition" of cell
division in 3T3 cells. Proc. Natl. Acad. Sci. U.S.A., 60:300-304.
Hynes, R. O. 1974. Role of surface alterations in cell transformation: The
importance of proteases and surface proteins. Cell, 1:147-156.
Jones, P. A., W. E. Lang, A. Gardner, C. A. Nye, L. M. Fink, and W. F.
Benedict. 1976. In vitro correlates of transformation in C3H/10T 1/2
clone 8 mouse cells. Cancer Res., 36:2863-2867.
Lumb, R. H., and F. Snyder. 1971. A rapid isotopic method for assessing the
biosynthesis of ether linkages in glycerolipids of complex systems.
Biochim. Biophys. Acta, 244:217-221-
Marchok, A. C., J. C. Rhoton, and P. Nettesheim. 1978. In vitro development
of oncogenicity in cell lines established from tracheal epithelium
preexposed in vivo to 7,12-dimethylbenz(a)anthracene. Cancer Res.,
38:2030-2037.
Montesano, R., C. Drevon, T. Kuroki, L. Saint Vincent, S. Handleman, K. K.
Sanford, D. De Feo, and I. B. Weinstein. 1977. Test for malignant
transformation of rat liver cells in culture: Cytology, growth in soft
agar, and production of plasminogen activator. J. Natl. Cancer Inst.,
59:1651-1656.
Olson, A. C., N. M. Larson, and C. A. Heckman. 1980. Classification of
cultured mammalian cells by shape analysis and pattern recognition.
Proc. Natl. Acad. Sci. U.S.A., 77:1516-1520.
Ossowski, L., J. C. Unkeless, A. Tobia, J. P. Quigley, D. B. Rifkin, and E.
Reich. 1973. An enzymatic function associated with transformation of
fibroblasts by oncogenic viruses. II. Mammalian fibroblast cultures
transformed by DNA and RNA tumor viruses. J. Exp. Med., 137:112-126.
Porter, K. R., G. T. Todaro, and V. A. Fonte. 1973. A scanning electron
microscope study of surface features of viral and spontaneous
transformants of mouse BALB/3T3 cells. J. Cell. Biol., 59:633-642.
Rapin, A. M. C., and M. M. Burger. 1974. Tumor cell surfaces. General
alterations detected by agglutinins. Adv. Cancer Res. (G. Klein, S.
Weinhouse, and A. Haddow, eds.), 20:1-91.
Scott, C. C., C. A. Heckman, P. Nettesheim, and F. Snyder. 1979a. Metabolism
of ether-linked glycerolipids in cultures of normal and neoplastic rat
respiratory tract epithelium. Cancer Res., 39:207-214.
168
-------
15. EARLY STAGES OF RESPIRATORY TRACT CANCER Heckman et al
Scott, C. C., C. A. Heckman, and F. Snyder. 1979b. Regulation of ether
lipids and their precursors in relation to glycolysis in cultured
neoplastic cells. Biochim. Biophys. Acta., 575-215-224.
Shira, S., V. H. Freedman, R. Risser, and R. Pollack. 1975. Tumorigenicity
of virus-transformed cells in nude mice is correlated specifically with
anchorage independent growth in vitro. Proc. Natl. Acad. Sci. U.S.A.,
72:4435-4439.
Snyder, F., and R. Wood. 1968. The occurrence and metabolism of alkyl and
alk-1-enyl ethers of glycerol in transplantable rat and mouse tumors.
Cancer Res., 28:972-978.
Topping, D. C., B. C. Pal, D. H. Martin, F. R. Nelson, and P. Nettesheim.
1978. Pathologic changes induced in respiratory tract mucosa by
polycyclic hydrocarbons of differing carcinogenic activity. Am. J.
Pathol., 93:311-322.
Unkeless, J. C., A. Tobia, L. Ossowski, J. P- Quigley, D. B. Rifkin, and E.
Reich. 1973. An enzymatic function associated with transformation of
fibroblasts by oncogenic viruses. I. Chick embryo fibroblast cultures
transformed by ovian RNA tumor viruses. J. Exp. Med., 137:85-111.
Voyles, B. A., and C. M. McGrath. 1976. Markers to distinguish normal and
neoplastic mammary epithelial cells _in vitro: Comparison of saturation
density, morphology and concanavalin A reactivity. Int. J. Cancer,
18:498-509.
Wigley, C. B., and I. C. Summerhayes. 1979. Loss of LETS protein is not a
marker for salivary gland or bladder epithelial cell transformation.
Exp. Cell, Res., 118:394-398.
Yamada, K. M., and J. A. Weston. 1975. The synthesis, turnover, and
artificial restoration of a major cell surface glycoprotein. Cell,
5:75-81.
Yang, N.-S., W. Kirkland, T. Jorgensen, and P. Furmanski. 1980. Absence of
fibronectin and presence of plasminogen activator in both normal and
malignant human mammary epithelial cells in culture. J. Cell. Biol.,
84:120-130.
WORKSHOP COMMENTARY
E. Hu: [Inaudible]
169
-------
15. EARLY STAGES OF RESPIRATORY TRACT CANCER Heckman et al.
C. A. Heckman: The study I discussed involved measurements only of the
alkyldiacylglycerols and ether-linked phospholipids. The gangliosides have
not yet been studied in these model systems.
170
-------
16. CARDIOVASCULAR EFFECTS OF OZONE AND CADMIUM INHALATION IN THE RAT
Nathanial W. Revis
T. Major
Walden E. Dalbey
Biology Division
Oak Ridge National Laboratory
Post Office Box Y
Oak Ridge, TN 37830
INTRODUCTION
At present there is widespread interest and concern about the effects of
environmental pollutants upon human health. Much of this concern stems from
recent experimental studies showing a variety of environmental pollutants to
be carcinogenic. Information on the relationship of pollutants and
cardiovascular diseases, however, remains to be determined. Diseases of the
cardiovascular system may be grouped into three major categories:
hypertension, atherosclerosis (including heart attack and stroke), and chronic
heart failure. Together, these three categories represent the major cause of
death in North America.
171
-------
16. CARDIOVASCULAR EFFECTS OF OZONK AND CADMIUM Revis et al.
BACKGROUND
The research interests of our group relate to the effects of
environmental pollutants (in particular, energy-related pollutants) in
cardiovascular disease. In recent studies, we showed that lead and cadmium
(energy-related pollutants) induce hypertension and atherosclerosis in the
White Carneau pigeon. Similar results were observed in the rat. These
observations support epidemiologic studies which show that these elements
correlate with hypertension and coronary artery disease (i.e., coronary
atherosclerosis). We also obtained preliminary results showing that the
chemical carcinogen benzo(a)pyrene induces an increase in the number and size
of aortic atherosclerotic plaques in the pigeon. These results show quite
clearly that environmental pollutants are involved in cardiovascular disease.
The major routes of intake of environmental pollutants in humans are the
gastrointestinal tract and respiratory tract. To obtain the results just
described, pigeons and rats were exposed to the pollutants via drinking water
(i.e., uptake occurred via the gastrointestinal tract). However, information
on the effects of these pollutants in cardiovascular disease following
inhalation exposures (i.e., uptake via the respiratory tract) remains to be
determined. With the exception of carbon monoxide (an air pollutant
associated experimentally with atherosclerosis) we know little about the
effects of air pollutants in cardiovascular disease. Nevertheless, oxidants
such as ozone (03) and the sulfur oxides have been shown to induce reversible
electrical changes in the heart as measured by electrocardiographic methods.
172
-------
16. CARDIOVASCULAR EFFECTS OF OZONE AND CADMIUM Revis et al.
In addition, Trams et al. (1972) demonstrated that the activity of monoamine
oxidase (MAO) and catechol O methyltransferase (COMT) in the dog brain is
significantly reduced after exposure to 1 ppm O3 for 18 mo. These two enzymes
affect blood pressure through their catabolic effect on the neurotransmitter
norepinephrine. Thus, a decrease in the activity of these enzymes may
increase the half-life of norepinephrine, ultimately provoking an increase in
blood pressure.
In previous studies, we exposed rats to 5 ppm Cd via drinking water for 6
mo. After exposure, the animals were anesthetized, and the femoral artery
pressure and blood norepinephrine measured. Prior to killing the animals, %
norepinephrine was injected into the femoral vein, and blood pressure and 3fl
norepinephrine levels were measured over a period of 1 h. The results of this
study showed both blood pressure and norepinephrine to be significantly
increased in the Cd-treated rats. Furthermore, the half-life of 3n
norepinephrine was significantly increased in the Cd-treated animals. As
previously noted, the half-life of norepinephrine is in part regulated by MAO
and COMT. The half-life is also regulated by cellular uptake (both neuronal
and extraneuronal). We previously showed the activity of these two catabolic
enzymes to be decreased in the aorta of rats exposed to 5 ppm for 6 mo. Thus
the increase in blood pressure observed in the Cd-treated rats may be the
result of an increase in the half-life of norepinephrine due to Cd effects on
these two catabolic enzymes. However, Cd may also affect the uptake of
norepinephrine. In any event, a decrease in the activity of these enzymes can
effect a change in blood pressure.
173
-------
16. CARDIOVASCULAR EFFECTS OF OZONE AND CADMIUM Revis et al.
METHODS AND RESULTS
To expand the observations of Trains et al. (1972), we exposed Fischer
rats as follows:
(1) O3, 0.6 ppm, 5 h/d, for 3 d
(blood pressure recorded on day 4)
(2) Cd, 3 mg/m3, for 1 h
(blood pressure recorded on day 8)
(3) Cd, 3 mg/m3, for 1 h
4 d later: 03, 0.6 ppm, 5 ii/d, for 3 d
(blood pressure recorded on day 8)
After these exposures, femoral artery pressure, electrocardiographic data, and
tissue concentrations of Cd (i.e, aorta, heart, and lungs) were obtained.
Table 16-1 shows the effects of these treatments on the cardiovascular
system. Systolic pressure and heart rate were significantly increased in the
treated animals. However, diastolic pressure and mean pressure were not
significantly changed. The relation of maximal velocity of pressure increase
in the femoral artery in the isometric phase of contraction (dpp/dt) was
significantly increased by Cd but not by 03. Note, also, that the effects of
O3 and Cd were not additive with respect to the changes in systolic pressure,
heart rate, and dpp/dt. In fact, the changes observed in the Cd plus 03
treated rats more closely resembled those observed in the Cd-treated animals.
174
-------
TABLE 16-1. EFFECTS OF CADMIUM AND OZONE INHALATION ON THE CARDIOVASCULAR SYSTEM IN THE FISCHER RAT
Group3
Systolic Pressure
(mm Hg)b
Diastolic
Pressure
(mm Hg)b
Mean Arterial
Pressure
(mm Hg)b
Heart Rate
(beats/min)b
H
3
M
01
O
Cd (5)
Cd (5)
03 (5)
Cd +03 (5)
116 ± 22
139 ± 10C
143 ± 9C
139 ±
95 ± 25
81 ± 3
86 ± 4
82 ± 2
102 ± 20
100 ± 7
105 ± 3
100 ± 5
370 ± 80
456 ± 3QC
414 ± 55d
456 ± 40°
2500 ± 400
3300 ± 50QC
2700 ± 200
3050 ± 700°
aNumber of animals per group is given in parentheses.
bMean ± S.E.M.
^h
01 cp < 0.01.
ap < 0.05.
a
-------
16. CARDIOVASCULAR EFFECTS OF OZONE AND CADMIUM Revis et al.
The significant increase in the QRS interval and in the amplitude of the
R wave (Table 16-2) suggests that these treatments may have affected the
permeability of the myocardial membranes or the metabolism of the
neurotransmitter norepinephrine. The increase in heart rate would support the
latter suggestion. However, an effect on the permeability cannot be ruled
out.
Table 16-3 shows the tissue distribution of Cd. As might be expected,
lung Cd was significantly increased (by a factor of 5) following Cd exposure.
Aortic Cd increased significantly, while the concentration of Cd in the heart
did not significantly change following Cd treatment. In the Cd plus 03
treated rats, Cd was decreased in the lung and increased in the aorta (as
compared to rats treated only with Cd, suggesting that 03 may increase Cd
clearance from the lung.
These results demonstrate that Cd and 03 have similar effects on the
cardiovascular system. The results also suggest that the cardiovascular
effects may be ^exerted~~thf ough an effect on norepinephrine metabolism. With
respect to effects on the cardiovascular system, no interactions between 03
and Cd were evident.
176
-------
TABLE 16-2. EFFECTS OF CADMIUM AND OZONE INHALATION
ON THE ELECTRICAL PROPERTIES OF THE HEART IN THE FISCHER RAT
Groupa
Heart Rate
(beats/min)b
PR Interval
(s)b
QRS Interval
(s)b
Amplitude of
P Wave
(mV)b
Amplitude of
R Wave
(mV)b
H
3
Control (5)
ca (5)
03 (5)
Cd +03 (5)
370 ± 80
456 ± 30°
414 ± 85d
456 ± 40C
0.047 ± 0.003
0.047 ± 0.007
0.047 ± 0.001
0.044 ± 0.001
0.021 ± 0.009
0.034 ± 0.003d
0.038 ± 0.0009°
0.039 ± 0.005C
0.0060 ± 0.0004
0.0034 ± 0.0006C
0.0076 ± 0.001C
0.00ft4 ± 0.008°
0.054 ± 0.009
0.070 ± 0.006°
0.067 ± 0.004d
0.069 ± 0.006°
M
W
O
N
O
p
aNumt>er of animals per group is given in parentheses.
bMean ± S.E.M.
°p < 0.01.
d,
lp < 0.05.
CO
(D
-------
16. CARDIOVASCULAR EFFECTS OF OZONE AND CADMIUM Revlsetal.
TABLE 16-3. TISSUE DISTRIBUTION OF CADMIUM IN THE FISCHER RAT FOLLOWING
CADMIUM AND OZONE INHALATION
Level of Cd in Tissue
(yig/g dry weight)^
Groupa Aorta Heart Lung
Control (5)
Cd (5)
03 (5)
Cd + 03 (5)
0.44 ± 0.09
0.82 ± 0.07°
0.50 ± 0.008
1.29 ± 0.13d
0.24 ± 0.04 1.9 ± 0.7
0.31 ± 0.03 11.3 ± 1.4d
0.29 ± 0.07 2.4 ± 0.8
0.27 ± 0.05 10.3 ± 2.3d
aNumber of animals per group is given in parentheses.
bMean ± S.E.M.
cp < 0.01.
dp < 0.001.
RECOMMENDATIONS FOR FURTHER RESEARCH
Based on these preliminary studies, our group offers the following
recommendations for future research:
(1) Determine if O3 has a dose-dependent effect on the
cardiovascular system
(2) Determine the effects of O3 on norepinephrine metabolism
(3) Determine the effects of other oxidants on the
cardiovascular system
(4) Determine if oxidants interact with other air pollutants
in effecting a change in the cardiovascular system
178
-------
16. CARDIOVASCULAR EFFECTS OF OZONE AND CADMIUM Revis et al,
(5) Determine the effects of various air pollutants in
animals with cardiovascular disease
WORKSHOP COMMENTARY
J. L. Whittenberger; In measuring the effects of 03 exposure on heart rate
and blood pressure, did you measure pulmonary ventilation? I wonder whether
hypoventilation might have played a role in producing the cardiovascular
effects.
N. W. Revis; The effects may very well be due to hypoventilation. They may
as well be attributable to the change in the pulmonary system rather than the
secondary effect; I'm not sure and cannot give a specific answer to your
question. The results suggest to me that the effect is not simply an effect
on the sympathetic nervous system required to stimulate the heart rate, thus
affecting systolic pressure.
J. L. Whittenberger; I was talking only about the level of ventilation, not
about the damage to the lungs. What dose of 03 was administered?
W. S. Dalbey; The dose was 0.6 ppm for 5 h for 3 d in a row. We measured
pulmonary resistance for totally other purposes during the O3 exposure, but
the present animals were not examined until the following day. We didn't do
the measurements until ~24 h after the last day of exposure.
179
-------
17. EFFECTS OF NITROGEN DIOXIDE AND 3-METHYLFURAN INHALATION
ON THE SMALL AIRWAYS IN THE MOUSE
Wanda M. Haschek
Biology Division
Oak Ridge National Laboratory
Post Office Box Y
Oak Ridge, TN 37830
INTRODUCTION
Small airways, i.e., the smallest bronchi and bronchioles, are one of the
most vulnerable regions in the lung for damage by many inhaled irritants.
Small airways are a major site of airflow obstruction in chronic bronchitis,
bronchiectasis, and, probably to a large extent, emphysema (Macklem et al.
1971; Hogg et al. 1968). Despite this fact, few details are available
concerning the response of the epithelial components to injury in this region;
most attention has centered on the alveolar zone of the lung.
The cell population of the small airways consists of the ciliated cells
(responsible for the movement of small particles up the mucociliary escalator)
and the nonciliated or Clara cells (the progenitors of the ciliated cells).
Although the exact function of the Clara cells is still under investigation,
they are secretory cells and are rich in mixed function oxidases. The Clara
cells are thus capable of activating drugs and other compounds to highly
180
-------
17. SMALL AIRWAY DISEASE Haschek
reactive and toxic metabolites, and of inactivating others. Small numbers of
goblet cells, responsible for mucous secretion, are also present in the small
bronchi.
The ciliated cells of the terminal airways are readily damaged by the
oxidants ozone (O3) and nitrogen dioxide (NO2). Following a single exposure
to either 03 or NO2, the airway epithelium is rapidly repaired due to
proliferation of nonciliated bronchiolar epithelial cells which subsequently
differentiate into ciliated cells. Since 03 and NC>2 are the major oxidants
present during peak traffic hours, their potential contribution to small
airway disease in man must be considered.
Exposure of man to very high NC>2 concentrations has been reported to
result in bronchiolitis obliterans. Bronchiolitis obliterans is a
particularly severe form of small airway disease in response to local injury
of the wall. Initially, cellular granulation tissue more or less fills the
bronchiolar lumen. This tissue then undergoes organization and takes a
polypoid form, the final fibrosis extending to the musculoelastic and
peribronchial regions. Partial to complete obstruction of the bronchiolar
lumen occurs. Bronchiolitis obliterans has also been reported to follow
exposure to sulfuric acid, ammonia, and war gases. In a study of O3 exposure
in rats (Last et al. 1979), a focal polypoid thickening of alveolar duct walls
was observed. This lesion was partially reversible on removal from O3.
181
-------
17. SMALL AIRWAY DISEASE Haschek
Nonciliated bronchiolar epithelial or Clara cells were recently shown to
be the target cells for several toxic furans, including 4-ipomeanol and
3-methylfuran (3MF) (Boyd 1980). 3MF has been identified in city smog and is
believed to be formed by photoxidation from naturally occurring terpenes which
originate from deciduous trees. The Clara cell activates these compounds to
their toxic metabolites, causing cell death. With a single exposure to a low
or moderate dose of 3MF, complete regeneration of the bronchiolar epithelium
takes place within a few days; following a high dose, regeneration is not
fully complete even after 3 weeks (Haschek, unpublished observations).
The study described below was designed to examine whether damage to the
progenitor cell—the nonciliated cell—would interfere with successful
recovery of the injured bronchiolar epithelium.
METHODS
NO2 was chosen to produce necrosis of the ciliated bronchiolar cells; a
single exposure to 20 ppm for 24 h was used. Inhalation of 3MF at a dose of
2.5 pi/liter for 1 h was employed to damage the nonciliated bronchiolar cells.
Young adult male Balb/C mice were divided into five groups and treated as
follows:
182
-------
17. SMALL AIRWAY DISEASE Haschek
Group Treatment
Day 1 Day 2 Day 3
A
B
C
D
E
NO2
NO2
N02
-
-
-
3MF
3MF
3MF
3MF
-
-
3MF
-
3MF
Half the animals in each group were killed on the 6th day; the other half
were killed on the 10th day. Animals were killed by cervical dislocation, and
lungs were fixed in situ by intratracheal instillation of 10% buffered
formalin. Lung tissue was processed, embedded in paraffin, sectioned at 3-4
ym, and stained with hematoxylin and eosin.
RESULTS
A large proportion of the mice exposed consecutively to 3MF died?
therefore, groups C and E will not be discussed further. Lungs from animals
exposed to NO2 alone (group A) showed minimal changes consisting of mild
hypercellularity around bronchioles and alveolar ducts at 5 d. By 10 d, the
lungs appeared virtually normal.
183
-------
17. SMALL AIRWAY DISEASE Haschek
Animals exposed to 3MF only (group D) had marked loss of nonciliated
bronchiolar cells. In small bronchioles, segmental areas of denudation were
interspersed with squamous and nonciliated cuboidal cells. In larger
bronchioles, normal ciliated cuboidal and small groups of columnar nonciliated
cells were present. After 10 d, regeneration was evident: bronchioles were
lined by varying numbers of undifferentiated cuboidal-to-columnar cells, as
well as some normal ciliated and nonciliated cells.
Lung changes in animals exposed to W>2 and subsequently to 3MF (group B)
were similar to those seen following exposure to 3MF alone, with the exception
that only small numbers of normal ciliated cells remained in the large
bronchioles. Scattered small fibrocellular polypoid masses, covered by
squamous to cuboidal epithelium, as well as sessile fibrous thickenings of the
bronchiolar wall, protruded into the lumen. After 10 d, regeneration was
evident, with many undifferentiated nonciliated cuboidal cells, some
dome-shaped Clara cells, and small numbers of ciliated cells. Only a few
polypoid lesions remained.
DISCUSSION
This study demonstrated an interaction between NO2 and 3MF in the small
airways. While the ciliated cell injury produced by NO2 was virtually
repaired by 5 d, and the effect of 3MF was primarily on the nonciliated
bronchiolar cells, the combined effect of these two agents greatly retarded
regeneration of ciliated cells and produced focal thickening and polypoid
184
-------
17. SMALL AIRWAY DISEASE Haschek
lesions in the bronchiolar wall. The retardation of ciliated cell
regeneration is attributed to damage of the nonciliated progenitor cells by
3MF. Sustained loss of ciliated cells might severely impair the defense
mechanisms of the ciliary escalator.
The pathogenesis of the bronchiolar polyps can only be conjectured. The
polyps seen in this study were similar in both appearance and reversibility to
those produced in the rat by exposure to 03 (Last et al. 1979). The polyps
produced by 03 were located in the alveolar ducts, whereas in the present
study they were located in the bronchioles. The difference in location can be
explained by the difference in principal site of injury between these agents.
Since simple epithelial damage is usually repaired without the formation of
such polyps, one may ask whether damage to underlying elements (such as
basement membrane or subepithelial tissue) may play a role in this phenomenon.
In bronchiolitis obliterans, where granulation tissue polyps partially or
completely occlude the bronchiolar lumen, damage to the bronchiolar wall is
implicated. We produced a lesion similar to that of bronchiolitis obliterans
by intratracheal instillation of a coal liquefaction distillate in the rat
(Haschek et al. 1981). Extensive necrosis of the airways epithelium allowed
leakage of the distillate (visualized by fluorescence microscopy) through the
mucosa into the underlying tissue and even, in areas, into the surrounding
parenchyma. Proliferation of the subepithelial connective tissue resulted in
polypoid lesions in the airways and alveolar ducts. These lesions became
somewhat condensed with time but were still present at 60 d. Whether such
lesions persist or not may depend on the severity of initial damage, the
185
-------
17. SMALL AIRWAY DISEASE Haschek
degree of the proliferative response, or, as in the case of the distillate, on
the persistance of the agent within the lesion.
In conclusion, despite our recognition of small airway disease as the
underlying cause of airflow obstruction in chronic obstructive lung disease
and our knowledge that small airways may be severely affected by inhaled
oxidants (e.g., N(>2 and 03) as well as naturally occurring toxins (e.g., 3MF),
little is known about the pathogenesis of small airway disease. The response
of the small airways to various types and severities of injury, as well as to
various combinations of naturally occurring irritants and toxicants, needs to
be examined in detail before an understanding of small airway disease can be
reached.
REFERENCES
Boyd, M. R. 1980. Biochemical mechanisms in chemical-induced lung injury:
Roles of metabolic activation. CRC Grit. Rev. Toxicol., 7:103-176.
Haschek, W. M., M. E. Boling, M. R. Guerin, and H. P. Witschi. 1981.
Pulmonary toxicity of a coal liquefaction distillate product. In: 19th
Annual Hanford Life Sciences Symposium on Pulmonary Toxicology of
Respirable Particles, Richland, Washington, Oct. 22-24, 1979 (in press).
Hogg, J. C., P- T. Macklem, and W. M. Thurlbeck. 1968. Site and nature of
airway obstruction in chronic obstructive lung disease. New Engl. J.
Med., 25:1355-1360.
Last, J. A., D. B. Greenberg, and W. L. Castleman. 1979. Ozone induced
alterations in collagen metabolism of rat lungs. Toxicol. Appl.
Pharmacol., 51:247-258.
Macklem, P. T., W. M. Thurlbeck, and R. G. Fraser. 1971. Chronic obstructive
disease of small airways. Ann. Int. Med., 74:167-177.
186
-------
17. SMALL AIRWAY DISEASE Haschek
WORKSHOP COMMENTARY
Comment; Putting fairly toxic materials into the lung either by inhalation or
by intratracheal administration will induce more than a mere destruction of
epithelial cells. The subsequent organization of the inflammatory process
(which can be reversible) might be the source of such a fibroblast
proliferation. This would agree with Dr. Witschi's observations on the
fibrosis that follows inhibition of epithelial cell regeneration. Hitting
small airways so heavily as to induce such severe pathologic changes should
not be interpreted as unique to the agent, but rather to the degree of damage
that is produced.
W. M. Haschek; I agree. I think the airways have very limited ways in which
they can respond, and that the response is due to the degree of damage and not
to the agent.
R. P. Sherwin; There1s quite a distinction between the polyps of your inhaled
agents and the ones produced by the distillate. One was edema, raising the
bronchiolar surface. This is why we should be very careful to distinguish
between cellular damage with edema and spindly epithelial cells as opposed to
true proliferative fibro-collagenous depositions. In humans there are many
counterparts to these so-called "benign fibrous polyps."
W. M. Haschek; I agree with most of your comments. However, it wasn't simply
edema that raised the surface of the bronchiole. There was cell
proliferation—proliferation of the underlying fibrous tissue—within these
polyps.
R. P. Sherwin; I didn't mean that it was simply edema, but that any fibrous
proliferation may be a small part of the lesion. There is a normal turnover
of interstitial cells, and a temporary increase in turnover following injury
may be occurring. While you may be right about the fibrosis, should it
disappear, it would be contrary to the usual human fibrosis, where destruction
and fibrous tissue replacement are characteristically persistent. The key
question is what absolute amount of collagen is found. We must bear in mind
-that several studies of lung fibrosis have failed to show increased
concentrations of collagen. There is a controversy about this—
H. P. Witschi; There is no controversy; that has been resolved. If you have
a fibrotic lung and take an aliquot only for analysis, not only total collagen
but also lung weight has increased. If you divide collagen by weight, the
"specific activity of collagen," so to speak, stays the same. If you measure
collagen per total lung, the value increases.
R. P. Sherwin; The last statement pertaining to total collagen content of the
lung is the critical point. Are we dealing with an absolute or relative
increase of collagen? A very simple and real explanation for some if not most
187
-------
17. SMALL AIRWAY DISEASE Haschek
of the collagen increase in human lungs with fibrosis is the phenomenon known
as condensation fibrosis. When sensitive epithelium is lost, the more
resistant interstitial tissues of the alveolar walls become stacked; i.e., the
alveolar walls collapse and fuse. Also, alveoli are completely lost and in
part are replaced by bronchioles with normally thicker walls. The key
question, again, is how to distinguish between relative and absolute collagen
increases. Is there more absolute collagen?
H. P. Witschi; Yes. If you calculate collagen per total lung, there is more.
However, if you just take a few grams of lung, determine collagen, and
calculate collagen for every gram of lung, you do not find more.
R. P. Sherwin; Ron Crystal of the National Heart, Lung and Blood Institute
(Pulmonary Branch) remarked on this interesting paradox in his own studies
(i.e., a discrepancy between the histological and biochemical data). At a
meeting we jointly attended, I proposed to him that condensation fibrosis
would be a plausible explanation. At any rate, there is a problem that has
not been completely resolved.
Comment; Ron Crystal's problem with increased lung collagen is supposedly in
idiopathic pulmonary fibrosis. It may or may not be unique to the patients
he's investigated. Some investigators do see an increase in whole lung
collagen. Whole lung collagen, as I understand it, is measured by analysis of
hydroxyproline, directed to the total lung. I think this is what Dr. Witschi
mentioned. If there is an increase in whole lung hydroxyproline, there is (in
all probability) some form of fibrosis. Some forms are visible as scars; I
think some are visible with special collagen stains. What Dr. Witschi
described is diffuse or interstitial fibrosis. Some of the collagen is not
visible but still apparent as an increase in whole lung collagen.
Question; How much human lung disease is idiopathic fibrosis?
Comment; Quite a lot of it.
Comment; It is a common problem. There are other kinds of fibrosis, however,
which constitute a very significant portion of the disease.
W. M. Haschek; The reversibility of lung fibrosis was mentioned. This is a
very important question. Human fibrosis does not appear to be reversible.
However, human cases of fibrosis frequently are not detected until a late,
chronic stage. With bleomycin there is initially fibrosis, but after a year
there is no detectable increase in lung collagen. Therefore, there is some
indication that very early so-called "fibrosis" may be reversible. Certainly
this aspect needs investigation.
[Note: Pickrell (Chapter 24 of this volume) reports a finding of increased
collagen but no histological evidence of fibrosis, and thinning of the
alveolar walls.]
188
-------
18. OVERVIEW OF RESEARCH AT LAWRENCE BERKELEY LABORATORY
Edward L. Alpen
Lawrence Berkeley Laboratory
University of California
Berkeley, CA 94720
INTRODUCTION
This report summarizes two oxidants research projects that were conducted
at Lawrence Berkeley Laboratory under EPA sponsorship. The first project
examined cocarcinogenic effects of nitrogen dioxide and sulfur dioxide in mice
(Dr. White, principal investigator). The ^second examined the effects of
low-level ozone exposure on serum lipoproteins in guinea pigs (Drs. Lindgren
and Shu, principal investigators).
COCARCINOGENIC EFFECTS OF NITROGEN DIOXIDE AND SULFUR DIOXIDE IN THE MOUSE
This project was designed to investigate whether the toxic gases nitrogen
dioxide (NO2) and sulfur dioxide (SO2) are tumorigenic by themselves, whether
they promote tumorigenesis, or whether they are in fact cocarcinogenic.
189
-------
18. LAWRENCE BERKELEY LABORATORY Alpen
Selection of Model System and Test Gases
For our model system we employed the lung adenoma initiated by the drug
urethan. Urethan is carcinogenic in the mouse lung without promotion but can
also be promoted by a number of suitable agents. The advantage of the urethan
system is that the incidence of lung tumors is lineally dependent upon dose.
Also, urethan1s action is very quick (95% is excreted in ~8 h), so the time of
exposure is very precise. Finally, the tumors can very easily be counted by
gross methods quite similar to those employed at Oak Ridge National
Laboratory, where the whole lung is dissected. The nodules in the lung are
counted using a dissecting microscope.
We chose to test NO2 because it has been shown to influence a number of
Type II alveolar epithelial cells in the mouse lung. The fact that the
majority of source cells for mouse lung adenomas appear to be Type II alveolar
epithelial cells suggested the possibility of a preconditioning phenomenon
that would increase the number of lung adenomas from urethan or other
appropriate carcinogens. We chose to also test 802 because there are at least
two reports in the literature indicating that SO2 is a promoting cocarcinogen
(in other words, that SO2 exposure after an initiating event with an
incomplete carcinogen will increase the yield of tumors).
190
-------
18. LAWRENCE BERKELEY LABORATORY Alpen
Experimental Protocol
Randomized CF1 mice were treated with the test gases either before or
after exposure to urethan. Exposures to NO2 or SO2 were performed in standard
chambers, 24 h/d, except for a very brief break for removal of feces and
change of food. Chamber gas levels were continuously monitored and recorded.
The NO2 exposures were 20 ppm for 2 d, 10 ppm for 4 d, and 5 ppm for 8 d.
Air controls were in the chamber for 6 d. The SO2 exposures were: 40 ppm for
3 d, 20 ppm for 6 d, and 10 ppm for 12 d. Controls again consisted of urethan
alone and gas alone. All animals survived these exposure regimens.
The mice were serially sacrificed at predetermined times, and the
necessary fixing and counting of tumors were completed. Part of each animal's
lung was histologically prepared for the counting of Type II cells.
The resulting multivariate data were analyzed using the Cox Relative Risk
Model for dose-dependent or time-dependent events that are quantal in nature.
It is extremely important, in multivariate studies, to have independent
controls for each of the treatment groups, to avoid a multiple comparison with
a single control group. As noted, our protocol included such controls.
191
-------
18. LAWRENCE BERKELEY LABORATORY Alpen
Results; Nitrogen Dioxide
The level of adenoma production by ITC>2 alone (without urethan) was
essentially zero (~0.02 adenoma per mouse over the very large sample of
control animals). Thus NO2 alone had no carcinogenic activity.
Analysis of the data also indicated no difference between the effects of
NO2 given before treatment with urethan and given after treatment (same
exposure levels) . In other words, NO2 before urethan was the same as NC>2
after urethan. There was no significant difference even when the data were
pooled. NO2 demonstrated neither promotion nor anticarcinogenic effects.
Variation in control values over replications reinforced this assessment: the
range of control values in the three replications was just about the range of
the data.
At present, Type II cell counts are not complete for this data set.
Results; Sulfur Dioxide
When given without urethan, SO2 (like NO2) demonstrated no effect on the
production of adenomas.
In contrast to IK>2, however, there were notable differences in the
results of SO2 exposure before and after administration of urethan. All
concentrations of SO2 depressed the incidence of tumors when given after
192
-------
18. LAWRENCE BERKELEY LABORATORY _^_ Alpen
urethan and increased the incidence of tumors when given before urethan. The
tumor incidence in animals exposed to 20 ppm and 40 ppm was essentially
identical, while the effects of 10 ppm differed markedly for "before" and
"after."
When SO2 was given before urethan, significant increases in the incidence
of tumors were seen at 20 ppm and 40 ppm SO2. Such increases in the number of
tumors produced do not qualify as "promotion," since initiation had not taken
place. The effect may be due to some change in the cellular population which
is ultimately exposed to urethan and in which the tumors are produced. At the
other exposure level (10 ppm), there was an insignificant difference from
control values.
When SO2 was given after urethan (i.e., the standard initiation and
promotion model), we found essentially the opposite effect. For all levels of
SO2, there was a significantly suppressed incidence of urethan-produced
adenomas in the lung. Such an effect is termed "suppression" (or
"anticarcinogenic effect").
Type II cells counts are complete for this data set. There was no
significant difference from control values throughout the exposure regimen.
193
-------
18. LAWRENCE BERKELEY LABORATORY Alpen
Ongoing and Planned Studies
Due to the apparent lack of effect by NO2, we have discontinued all work
with that gas. An ongoing study with SO2 ties in to the previous work by
employing one exposure level—20 ppm for 6 d—for which data were collected in
the first experiment. The next group of animals will be exposed to 6.7 ppm
for 18 d, and we will again examine both pre- and post-exposure effects.
Depending on the outcome of the 6.7 ppm study, we may proceed to a
carcinogen other than urethan. (Urethan, a carbamate, is a highly artificial
carcinogen that is not known to be carcinogenic in man.) Host likely, we will
employ benzo(a)pyrene (or another carcinogen), lower the SO2 exposure level,
and increase the exposure time. We do believe that we are "in the right
ballpark" in terms of exposure criteria.
EFFECTS OF OZONE ON SERUM LIPOPROTEIN CONCENTRATIONS IN THE GUINEA PIG
Due to the oxidative nature of ozone (03), many interrelated aspects of
lung metabolism involving lipids might easily be influenced by this very
reactive substance. For example, the Type II cells that proliferate after 03
exposure use exogenous lipids during surfactant synthesis. Prostaglandins,
which require essential fatty acids for synthesis, also seem to be involved in
O3 toxicity. Protection against 03-induced peroxidation of unsaturated fatty
acids has been reported by Donovan and Menzel. In addition, Brown and others
have shown that human deaths due to cardiovascular disease are decreased
194
-------
18. LAWRENCE BERKELEY LABORATORY Alpen
during periods of lower urban pollution. This accumulation of experimental
and epidemiologic evidence led us to consider whether the lipoproteins (which
have a very significant influence on atherosclerosis and cardiovascular
disease) are in fact changed by exposure to 03.
As mentioned earlier, Drs. Lindgren and Shu served as principal
investigators for this study. Their complete report, "Serum Lipid and Lipid
Protein Concentrations Following Exposure to Ozone," has been accepted for
publication in the Journal of Environmental Pathology and Toxicology.
Experimental Protocol
Hartley guinea pigs were exposed to 03 for 22 d. The nominal exposure
level was 1 ppm; in retrospect, we believe the actual concentration to have
been closer to 0.8 or 0.85 ppm. As in the N02/SO2 study, standard exposure
chambers were employed. We monitored all of the usual toxicological
indicators, such as body weight, growth, food consumption, and mortality. In
addition, serum lipoprotein values were obtained at the beginning of the
study, at the end of the study, and for 30 d after exposure.
Results; Toxicological Indicators
As expected, we observed a very marked sex-dependent effect of O3
exposure. There were 2 deaths in the male group but none in the female group.
195
-------
18. LAWRENCE BERKELEY LABORATORY Alpen
Other investigators (in our laboratory and elsewhere) have noted that the
female is much more resistant than the male for any end point in 03 exposure.
As predicted from our rat studies, there was a sharp decrease in food
intake. This decrease in food intake was accompanied, surprisingly, by no
change in body weight. Body weight was maintained at the control levels
despite a 35% drop in food intake. Retrospectively, in light of other 1 ppm
exposure studies showing (1) a sharp drop in thyroid hormone levels and
thyroid function and (2) sharp reductions in growth hormone and thyrotropin
from the pituitary (Clemens and Garcia), it's not surprising that the
metabolic rate of the animals was sharply reduced.
Results; Lipoprotein Levels
We examined high-density lipoproteins, low-density lipoproteins, very low
density lipoproteins, triglycerides, and cholesterol.
There was a very large increase in the cholesterol levels of the exposed
animals as compared with normal controls. In males, this increase was 200%;
in females, 100%. Comparisons with pair-fed controls (to remove any dietary
effect) revealed smaller increases in cholesterol: 50% in males and 30% in
females.
Triglycerides and high-density lipoproteins were essentially unaffected
compared to either pair-fed or normal controls. There was, in fact, no
196
-------
18. LAWRENCE BERKELEY LABORATORY Alpen
difference between pair-fed and normal animals for these parameters. On the
other hand, low-density lipoproteins were increased by 100% (males) and 50%
(females) in comparison to pair-fed controls. The very low density
lipoproteins were increased by 100% (males) in comparison to pair-fed
controls.
The probability values for these observations are well below 0.001, so
the findings are highly significant.
Thirty days after the exposure was terminated, the changes in
lipoproteins and cholesterol were still evident. Histologically, the lungs
had returned to normal.
Ongoing and Planned Studies
We recently conducted a few more studies in surviving animals from this
study. At 8 to 10 mo following cessation of exposure, the anJjnals are
returning to normal and probably are not different from normal. We have no
intervening values. Depending on EPA interest, we will presumably repeat this
study at 0.2 ppm.
197
-------
19. OVERVIEW OF RESEARCH AT THE UNIVERSITY OF CALIFORNIA - DAVIS
Marvin Goldman
Laboratory for Energy Related Health Research
University of California
Davis, CA 95616
INTRODUCTION
This report provides a brief overview of oxidants research within the
Laboratory for Energy Related Health Research at the University of
California - Davis* Our interpretation of the concepts of scientific
relevance and merit is presented. Elsewhere in this volume, Dungworth
(Chapter 20) discusses our work with ozone, and Raabe (Chapter 21) describes
our research on rodent and primate reparative and adaptative mechanisms to
coal fly ash and sulfur compounds.
Our research is not performed in isolation: we have an interdisciplinary
program that bridges several of the colleges and falls under the aegis of
various organized research units as well as individual scientists. Our sister
research unit, the California Primate Research Center, is involved in major
research efforts on the effects of particulate and gaseous pollutants in
animal systems.
198
-------
19. UNIVERSITY OF CALIFORNIA - DAVIS Goldman
APPROACH TO RESEARCH
All of our Laboratory's research takes a basic toxicological approach.
In our view, however, there is no such a thing as a "toxic agent"—only toxic
levels. Our goal is to understand the spectrum of each dose-response curve
and its relevance (if any). A major problem, of course, is to determine the
proper experimental model for a given study. We choose each model with
judicious care before proceeding to the stage of study design.
In attempting to extrapolate from the animal to the human situation, our
Laboratory seeks to understand the nature of the dose-response curve and of
the dosimetry. In our view, such an understanding is central to a valid risk
assessment model. Without knowledge of the nature of the dose-response curve,
the statistical significance of an exposure level versus a nonexposure level
may fall on the "wrong side" of the extrapolation curve and thus be very
misleading. With regard to the comments by Alpen (Chapter 6 of this volume),
we feel that an understanding of the mechanism and an understanding of the
dose-response spectrum are parallel requirements; both are essential to a
valid risk assessment model.
CURRENT EFFORTS
With regard to oxidant air pollutants, our Laboratory's involvement began
in 1974 as part of a major Department of Energy project that is primarily
concerned with fossil fuel combustion (conventional fuels as well as the new
199
-------
19. UNIVERSITY OF CALIFORNIA - DAVIS Goldman
synthetic fuels). Our research focuses on airborne particulates from coal
combustion. We employ unique and innovative approaches to collecting samples
in the field. Laboratory activities include the generation of aerosols for
animal exposure, physical and chemical categorization of samples, and
determination of the sample components that carry biological importance in
either the pristine (as captured) or degraded (following an interaction with
biological membranes) state. The latter activity involves a series of
multistage tests. These extend from the more conventional Ames assays (for
bacterial mutation) through tests of macrophage function and toxicity and
tests of cell transformation (in vivo experimentation followed by in vitro
testing of the transformed or affected cells). A major portion of our program
relates to progenitor cells rather than end effect cells; we examine possible
interactions with the immunologic system and attempt to relate hormonal and
immunologic injury to longer-term carcinogenic potential.
200
-------
20. PULMONARY EFFECTS OF OZONE IN THE RAT AND MONKEY
Donald L. Dungworth
Department of Pathology
School of Veterinary Medicine
University of California
Davis, CA 95816
INTRODUCTION
Our work on pulmonary effects of oxidants emphasizes the phenomenon of
adaptation (tolerance). Two other primary areas of focus are the evolution of
chronic damage resulting from low-level exposure and the effects on lung
growth.
CHOICE OF ANIMAL MODELS
Frequently there is much discussion about the appropriateness of a
particular animal model for a particular study. There is no such thing as a
"perfect model"; in choosing a model, the most important issue is: What
questions will be asked of the model? We use two primary species: rats (for
statistical purposes and cost effectiveness) and monkeys (for structural/
functional similarities to man and a variety of other anthropomorphic
201
-------
20. PULMONARY EFFECTS OF OZONE IN THE RAT AND MONKEY Dungworth
features). We develop our idea with rats and look to monkeys for the more
definitive studies.
Regarding lung structural correlates between experimental animals and
man, scanning electron micrographs effectively illustrate that the monkey has
well developed respiratory bronchioles (like man) whereas the rat does not
(Castleman et al. 1975). The importance of this is that the respiratory
bronchiole of the monkey is the pulmonary region that is most damaged by low
levels of ozone (03). At 0.2 ppm, damage is essentially limited to this small
airway (Mellick et al. 1977). We are reasonably confident that the bronchiole
is also where damage caused by high ambient levels of 03 occurs in man. We
also prefer the monkey because of its usefulness for studies of pulmonary
function. We can do repeated experiments; also, the difference in chest wall
compliance between the monkey and dog favors the monkey as the species for
definitive work.
EXPOSURE REGIMENS
Exposure concentrations are 0.2, 0.5, and 0.8 ppm 03. These
concentrations are measured against the neutral buffered KI calibration. We
can factor them to the absolute (UV) standard, but for sake of continuity we
use the older calibration. The duration of exposure depends on the purpose of
the particular study.
202
-------
20. PULMONARY EFFECTS OF OZONE IN THE RAT AND MONKEY Dungworth
STUDIES OF ADAPTATION DURING CHRONIC LOW-LEVEL EXPOSURE
At 7 d of exposure in normal rats, we find the approximate morphologic
no-effect level to be ~0.1 ppm 03 (Plopper et al. 1979). The effect of
prolonged exposure to various 03 concentrations is dose-related, but also
depends on the balance between adaptation and smoldering irritation. Our
general hypotheses are commonly held in the field. The first hypothesis is
that the centriacinar lesion produced by low-level chronic 03 exposure can
cause and/or exacerbate chronic obstructive pulmonary disease. The outcome
depends on the interplay between adaptation and enhancing factors such as
infectious and immunologic processes. The concept is straightforward, but the
mechanisms involved are complex and largely unknown.
The second hypothesis is that O3 has a low-level initiation or promotion
capability for neoplasia of pulmonary epithelium. On current evidence, we do
not believe that O3 is a significant carcinogen or cocarcinogen. It does
produce a larger pool of proliferating cells in small airways in the monkey
(and in the mouse), however; thus, a larger cell population may well be at
risk. This hypothesis requires further testing.
The terms "adaptation" and "tolerance" are often used in different ways.
In the most general sense, adaptation is the process by which tolerance
occurs. We prefer to use "adaptation" when referring to changes in pulmonary
epithelium. First of all, this usage emphasizes the dynamics of the situation
(i.e., what we are trying to understand). Secondly, in O3 toxicology
203
-------
20. PULMONARY EFFECTS OF OZONE IN THE RAT AND MONKEY Dungworth
literature, "tolerance" has been largely preempted by the edemagenic tolerance
model, which is not involved in low-level insult. There is another reason for
using the term "adaptation." In general pathology, the term refers to cell
changes in response to an altered environment, thereby implying that the cells
are there to change. "Adaptation" does not apply to situations in which
maximal damage occurs initially and cells are either killed or incapable of
further response a short time later. Our use of "adaptation" denotes a
lessening of detectable damage in spite of continuing insult. In this
context, there is a higher capacity for adaptation in the rat than in the
bonnet monkey. Here is yet another example of important species differences
in biology.
The morphologic manifestation of adaptation is well illustrated by
comparing scanning electron micrographs of the rat lung after 7 and 90 d of
exposure to 0.2 ppm 03. Qualitatively, the lung appears to have returned to
normal by 90 d, although morphometric studies reveal slightly higher than
normal numbers of macrophages. Comparing 7 and 90 d exposure to 0.8 ppm 03
shows a considerable decrease in the number of macrophages in the lumen, but a
reorganization of the centriacinar region (Boorman et al. 1980). This is seen
as an intermediate zone resembling a respiratory bronchiole (normally not
present in the rat at this age).
Because the damage we see is limited to focal (centriacinar) regions of
the lung, an entirely new approach to sampling is needed. Randomizing counts
204
-------
20. PULMONARY EFFECTS OF OZONE IN THE RAT AND MONKEY Dungworth
on pieces of minced pulmonary parenchyma is not satisfactory. Methods to
sample specifically but without introducing bias must be devised.
With respect to adaptation over a 90-d exposure, the most obvious
differences between the rat and the bonnet monkey are the smaller decrease in
intraluminal macrophages in affected respiratory bronchioles seen in the
monkey and the persistent hyperplasia and hypertrophy of cuboidal, nonciliated
bronchiolar epithelial cells also present in the monkey.
Our main current study is a 12-mo exposure of bonnet monkeys to 0.3 ppm
03, with evaluation of one group of exposed and control monkeys at the end of
exposure and of a second group after a 3-mo recovery period. Assessment
during the course of exposure is by pulmonary function, including acoustic
response (Jackson and Olson 1980), and by pulmonary lavage parameters. These
pulmonary function and lavage studies are important both to monitor changes
occurring during the 12-mo exposure and "to provide a basis of comparison for
what might happen in man. Both types of measurement are feasible in humans
that have been experimentally or naturally exposed to low concentrations of
03. The principal terminal assessments will be biochemical and morphometric.
OTHER STUDIES
We have also initiated studies of the effects of O3 on lung growth and
aging in both the rat and monkey. Morphometric evaluations being developed
205
-------
20. PULMONARY EFFECTS OF OZONE IN THE RAT AND MONKEY Dungworth
by Dr. Walter Tyler will serve as the most important measures in these
studies.
COMPARISON OF EXPERIMENTAL AND EPIDEMIOLOGIC DATA FOR OZONE AND SULFUR OXIDES
One important consideration in inhalation toxicology is that the
dose-response curve varies both with time and with the chemical agent. There
is some suggestion that the curves for 03 and sulfur oxides (SOX) differ
considerably. With ©3, experimentally obtained concentrations in polluted
urban air cause damage which levels off or lessens with time.
Epidemiologically, it is difficult to establish that these levels are in
general harmful to exposed human populations. With SOX, on the other hand,
levels that are many times ambient levels are needed to cause an experimental
effect, yet the epidemiologic evidence of harm to exposed populations is more
satisfying than for 03. Perhaps the recruitment of a number of different
modes of subtle damage occurs gradually until a significant lesion develops.
Copollutants in SOX smogs might also enter into this equation.
REFERENCES
Boorman, G. A., L. W. Schwartz, and D. L. Dungworth. 1980. Pulmonary effects
of prolonged ozone insult in rats. Morphometric evaluation of the
central acinus. Lab. Invest., 43:108-115.
Castleman, W. L., D. L. Dungworth, and W. S. Tyler. 1975. Intrapulmonary
airway morphology in three species of monkeys: A correlated scanning and
transmission electron microscopic study. Am. J. Anat., 142:107-122.
206
-------
20. PULMONARY EFFECTS OF OZONE IN THE RAT AND MONKEY Dungworth
Jackson, A. C., and D. E. Olson. 1980. Comparison of direct and acoustical
area measurements in physical models of human central airways. J. Appl.
Physiol.: Resp. Environ. Exercise Physiol., 48:896-902.
Mellick, P. W., D. L. Dungworth, L. W. Schwartz, and W. S. Tyler. 1977.
Short-term morphologic effects of high ambient levels of ozone on lungs
of rhesus monkeys. Lab. Invest., 36:82-90.
Plopper, C. G., C. K. Chow, D. L. Dungworth, and W. S. Tyler. 1979.
Pulmonary alterations in rats exposed to 0.2 and 0.1 ppm ozone: A
correlated morphological and biochemical study. Arch. Environ. Health,
34:390-395.
WORKSHOP COMMENTARY
D. L. Coffin; As a pathologist, I was quite impressed with Dr. Dungworth1 s
presentation. Many of these things have been examined by other investigators,
but hia group is putting them together in a very nice way.
Question; Is it neutral buffered KI or unbuffered KI that you use for 03
characterization?
D. L. Dungworth; Neutral buffered KI is used to calibrate Dasibi photometric
analyzers. The Dasibi analyzers are used for routine monitoring and the data
are handled by a computer. We also send the Dasibi analyzers away for
occasional checks against absolute UV photometric standards.
Question; How were tissues prepared for scanning electron microscopy? Most
of the specimens [shown on slides during the oral presentation] were devoid of
any mucous layer in the airways, for example.
D. L. Dungworth; They were fixed by intratracheal perfusion of modified
Karnovsky's fixative at 25 cm fluid pressure.
Question; For obtaining the information you presented, how much better is
that than, say, optical microscopy of those sections?
D. L. Dungworth; The scanning electron micrograph affords the ability to
examine a large block of tissue rapidly and also provides high resolution with
large depth of field. This means that one can examine in detail large amounts
of airway and alveolar surface.
Question; You made many statements about cell identification and the like.
However, you're dealing with a topographic analysis of those cells.
207
-------
20. PULMONARY EFFECTS OF OZONE IN THE RAT AND MONKEY Dungworth
D. L. Dungworth; The scanning electron micrographs were used purely to
highlight our studies. We routinely correlate findings by conventional light
microscopy, scanning electron microscopy, light microscopy of large 1-ym
sections, and transmission electron microscopy of specifically selected
regions of pulmonary parenchyma.
208
-------
21. BIOLOGICAL EFFECTS OF FLY ASH FROM COAL COMBUSTION
Otto Raabe
Laboratory for Energy Related Health Research
University of California
Davis, CA 95616
INTRODUCTION
The Laboratory for Energy Related Health Research has maintained a
special interest in the airborne particles released from power plants of
various types. Our interest extends to particulates from combustion of coal,
oil, synthetic fuels, and coal-oil mixtures. This report discusses an
EPA-supported study of the fly ash emitted in combination with sulfur dioxide
from coal-burning power plants. The study is a collaborative effort with the
California Primate Research Center; the large team of investigators is drawn
from the fields of biochemistry, pathology, respiratory physiology, and so on.
This author's primary interest and involvement are in the dosimetric and
exposure phases; the following report emphasizes these aspects but also
touches on some of the salient biological results obtained to date.
209
-------
21. BIOLOGICAL EFFECTS OF FLY ASH Raabe
SAMPLE COLLECTION
In some cases, we directly sample the particles leaving a smoke stack.
More frequently, a sample is taken from the hopper of an abatement system
(e.g., an electrostatic precipitator). For laboratory exposures, the
collected materials are re-aerosolized (see "Exposure Techniques," below).
SAMPLE CHARACTERIZATION
Prior to biological testing, we perform thorough characterization of both
the untreated and the re-aerosolized fly ash. Because the particles'
aerodynamic behavior is the primary characteristic controlling deposition in
the respiratory airways, we give close attention to the parameters that are
implicated in aerodynamic behavior. We look very carefully at the size and
shape of the particles. Comparisons of untreated versus re-aerosolized fly
ash indicate no significant differences in these parameters.
The distribution of electrostatic charge on the particles can also be
quite important. In most of our experiments, we reduce the charge to
Boltzmann equilibrium before animals are exposed.
Chemical characterization is a major focus. For this, we employ atomic
absorption analysis, neutron activation analysis, and par tide-induced X-ray
analysis. Basically, we find that no two particles have exactly the same
chemical composition. Even two particles of the same size will typically show
210
-------
21. BIOLOGICAL EFFECTS OF FLY ASH Raabe
different chemical compositions, particularly with respect to the trace metals
present on the particle surface. Certain elements (e.g., zinc, chromium, and
arsenic) tend to be concentrated in particles of smaller size due to a primary
association with the surface. Comparisons of untreated versus re-aerosolized
fly ash indicate no significant differences in chemical composition.
Whether the particles are deliquescent or hygroscopic also has a marked
effect on behavior and on particle surface chemical reactions.
BIOLOGICAL TEST SYSTEMS
Our program assesses the biological activity of these materials using
short-term bioassay systems (e.g., the Ames assay) as well as inhalation
exposures in whole animals. Whole animals include the mouse, rat, monkey, and
beagle dog. With regard to whole-animal experimentation, almost all current
work with fly ash is performed in rodents; these less expensive species
suffice for the necessary range-finding experiments.
EXPOSURE TECHNIQUES
For inhalation studies we generate fly ash using the Wright dust feed
mechanism. For several decades this mechanism has been used to expose animals
to particles of relatively insoluble materials. Our Laboratory has modified
the system by installing a cyclone separator at the exhaust so that larger
particles (>2.5 ym in aerodynamic size) are not an appreciable part of the
211
-------
21. BIOLOGICAL EFFECTS OF FLY ASH Raabe
exposure aerosol. Prior to inhalation experiments, the sample material is
always size classified to avoid particles that are outside the respirable
range for the experimental species. After this initial size separation, the
material is loaded as a packed dust cake into the Wright dust feed mechanism.
The feeder provides a continuous flow of aerosol as well as a second size
separation. Only particles having an aerodynamic equivalent size of <2.5 ym
pass into the exposure chamber. A Crypton 85 discharge unit assures that the
exposure aerosol is reduced to Boltzmann equilibrium with respect to
electrostatic charge.
Fairly large exposure chambers (4 m3) are available at the California
Primate Research Center. Rodents are typically arranged in a monolayer
configuration in these chambers. To date, the maximum exposure concentration
has been 200 mg/m3.
A key point about these experiments is that the particles are small
enough to be readily inhaled, resulting in fairly large deposition in the
lower respiratory tract. At the same time, the conditions are not "dusty" in
the sense that the animals are covered with fly ash or the chamber is covered
with fly ash. Aerodynamically, the particles are quite stable in this size
range.
212
-------
21. BIOLOGICAL EFFECTS OF FLY ASH Raabe
PRELIMINARY RESULTS
In early screening with the Ames assay (TA-98 strain of Salmonella), the
mutagenicity of fly ash collected from the stack of a coal-burning power plant
varied with particle size. In general, the larger particles showed a lower
number of revertents per milligram of material used in the test. The two
smaller size groups (median size of ~2.2 and ~3.5 ym) tended to have higher
numbers of revertents in this test.
When we examined the effect of temperature on three strains of
Salmonella, we found that heating the samples to >250°C resulted in
disappearance of most of the mutagenic activity. Efforts to explain this
phenomenon are continuing. Obviously, in the course of combustion fly ash is
heated far in excess of 250°C. Probably the as yet unidentified mutagens are
formed as the fly ash is released from the stack.
More recently, extensive studies on fly ash collected from the hoppers of
electrostatic precipitators (same power plant) showed no mutagenic activity at
any particle size. Thus, there is a very definite biological difference
between fly ash released from the stack and fly ash collected in the abatement
system.
With regard to whole-animal studies, one long-term inhalation exposure (4
mg/m3, 8 h/d, 180 d) resulted in no remarkable biological changes.
213
-------
21. BIOLOGICAL EFFECTS OF FLY ASH Raabe
CURRENT STUDIES
Presently we are involved in studies of combinations of sulfur dioxide
(50 ppm) with fly ash. In the first experimental series, animals were exposed
for ~14 d; there were not many remarkable biological effects. We plan to
repeat these experiments at longer time periods.
WORKSHOP COMMENTARY
G. Rausa; A recent article in Science states that the mutagenie activity of
fly ash comes from methyl sulfate; the association with temperature appears to
be about the same. Do you have any technique for ascertaining whether those
results are correct?
O. Raabe; One of the coauthors of that paper, Lee Hansen, has been in our
Laboratory for about six months on sabbatical leave. I don't believe he has
been able to find any methyl sulfate in our aerosols.
G. Rausa; Do you have any conjecture as to what it might be?
O. Raabe; We are working quite diligently to identify the mutagen; it is a
very important project to us.
Question; What about polycyclic hydrocarbons? At a conference a couple of
summers ago, Paul Morrow pointed out that benzopyrene, which is the surrogate
for the rest of the polycyclic hydrocarbons, comes out of the stack in a
volatile state and then condenses on the particles as the plume cools. Not
finding it in the hopper and finding it on the true fly ash would seem to fit
with that.
O. Raabe; To be mutagenie, benzopyrene must be activated. In our results, no
activation effect is apparent. So it may be something "from" benzopyrene, but
it isn't "just" benzopyrene.
The problem with our mutagenic fly ash is that it is very rich in organic
materials, and polycyclic aromatic hydrocarbons are not a major part of that
organic constituency. Dr. Kimball of our Laboratory has been working with
this material; she finds, I believe, that there is so much organic material
(in comparison to polycyclic hydrocarbons) that it becomes very difficult to
identify polycyclic hydrocarbons as the culprit agent.
214
-------
21. BIOLOGICAL EFFECTS OF FLY ASH Raabe
Question; Was the ash studied by the Utah group collected way down where it
was cool, or was it more reflective of the stack?
0. Raabe; I think they collected it somewhere downstream, outside the power
plant. Ours was collected from the base of the smoke stack. Thus, there is
quite a difference.
Question; How close is the geometric surface to the gas absorption surface on
these spheres?
0. Raabe; That's a good question; unfortunately, I don't have the data with
me. Our measurements show that the gas absorption surface is greater than the
geometric surface, but it's not remarkably greater. For the very large
particles, there are quite a few hollow spheres and spheres that are tightly
packed with little particles. These features are not seen in particles of the
small respirable range, however. The small particles have a rough surface but
no obvious pores.
215
-------
22. OVERVIEW OF RESEARCH AT THE INHALATION TOXICOLOGY RESEARCH INSTITUTE
Joe L. Mauderly
Inhalation Toxicology Research Institute
Lovelace Biomedical and Environmental Research Institute
Post Office Box 5890
Albuquerque, NM 87115
INTRODUCTION
This report presents a brief overview of the Inhalation Toxicology
Research Institute. Included are descriptions of the Institute's resources
and of the current research program. Chapters 23 through 26 of this volume
present more detailed information on four projects relating to oxidant air
pollutants.
OPERATION AND FUNDING
The Inhalation Toxicology Research Institute is operated by the Lovelace
Biomedical and Environmental Research Institute for the Department of Energy
under the Assistant Secretary of the Environment. The parent foundation is
the Lovelace Medical Foundation. Most (currently, 80%) of the Institute's
funding comes from the Department of Energy, but several projects are funded
by other governmental agencies. Funding from these other agencies (EPA, the
216
-------
22. INHALATION TOXICOLOGY RESEARCH INSTITUTE Mauderly
Nuclear Regulatory Commission, and the Department of Defense) is made possible
through interagency agreements with the Department of Energy.
FACILITIES AND STAFF
The Institute is located on a 138-acre site on Kirtland Air Force Base
south of Albuquerque, New Mexico. Facilities include ~215,000 ft2 of floor
space; the current replacement value is somewhere on the order of $20 million.
All support services are provided onsite; the Institute is essentially
self-contained.
There is maintenance housing for ~15,000 small animals; the exact number
housed, of course, depends on the mix between mice and larger species. At
present, there are facilities for housing ~2,000 dogs. The Institute
maintains a breeding colony of beagle dogs; mice, rats, and Chinese hamsters
are also bred. The current production capability is ~4,000 mice and ~2,000
rats per month, and ~400 dogs per year. Actual production depends on research
needs at any given time.
In general, the Institute at full strength employs ~250 people. The
present total staff of 227 includes ~50 doctoral-level personnel and a mix of
disciplines: ~50% in the life sciences (biology or medical-related degrees),
~25% in biophysics and biomedical engineering, and ~25% in the physical and
engineering sciences (including mathematical modeling and risk assessment).
217
-------
22. INHALATION TOXICOLOGY RESEARCH INSTITUTE Mauderly
RESEARCH CAPABILITIES
For several years the Institute has maintained strong capabilities in the
area of inhalation exposure to both radioactive and nonradioactive materials
and aerosols. There is also considerable analytical capability/ not only in
chemistry but also in radioanalysis.
Another Institute strength is the diversity of biological end points that
can be evaluated. Capabilities encompass not only the clinical end points
(e.g., physical examinations, radiography, clinical chemistry, and hematology)
but also organ function (e.g., respiratory function, mucociliary clearance,
and lung defense mechanisms). In other words, it is possible to evaluate
organ function as well as whole body function.
Deposition and retension of inhaled materials (including aerosols,
vapors, and gases) continue to be important end points in determining critical
doses to tissues. In the area of immunology, there is a strong program
focusing on the impairment in immune mechanisms caused by inhaled
environmental pollutants (particularly in the lung-associated lymph nodes) and
on determination of the immune pathways involved.
Mutagenesis and carcinogenesis testing capabilities, of course, are
related, and one of the Institute's major strengths is the ability to go from
in vitro mutagenicity studies to whole-animal carcinogenicity studies.
Current protocols involve bacterial and mammalian cell mutagenicity testing
218
-------
22. INHALATION TOXICOLOGY RESEARCH INSTITUTE Mauderly
followed by long-term studies of cancer development in animals. In the field
of cytogenetics, there is a group looking directly at the genetic
transformations that take place in mammalian cells and scoring chromosome
damage.
Mortality, morbidity, and pathology are common end points in any
long-term animal study. The Institute employs not only gross morphology and
light microscopy, but also transmission and scanning electron microscopy.
Quantitative morphometric measurements are also commonly used.
The Institute's capabilities in evaluating biochemical end points are
particularly strong. One focal area involves lung lipid chemistry, Type II
cell metabolism and injury, and lung surfactant studies. Another strong area
is the biochemistry of lung connective tissue.
CURRENT PROJECTS
Approximately 50% of the Institute's current research program relates to
nuclear power production. Studies involve both the fission products and the
transuranics such as plutonium and other alpha emitters. Both short- and
long-term studies are ongoing in these areas.
The Fossil Fuel program evaluates effluents from stationary sources:
coal combustion and coal gasification. In the area of coal combustion, both
219
-------
22. INHALATION TOXICOLOGY RESEARCH INSTITUTE Mauderly
conventional combustors and fluidized bed combustors are undergoing
evaluation.
In the Mobile Sources program, the main effort currently relates to the
study of diesel exhaust. A five-part program begins with characterization of
the exhaust material and extends through animal life-span studies.
The Solar program is the only area involving studies that are not
primarily inhalation-oriented. Because of the possibility for heat transfer
fluids to enter the potable water supply. Institute researchers are studying
the toxicology of these materials. Studies focus on the ingestion and contact
toxicology of heat transfer fluids.
The Conservation program* s main effort relates to the study of fibers
(both natural and man-made) used in insulating processes.
The spectrum of materials that can be analyzed ranges from solid
particles to droplets, vapors, and gases. Materials currently analyzed
include conventional combustor and fluidized bed combustor fly ash, diesel
particulates, ammonium sulfate (as either a solid particle or droplet),
sulfuric acid, and irritant or oxidant vapors and gases. Most of the current
studies are oriented towards single agents. As the need is recognized, the
Institute will undertake studies of combinations of these materials, bearing
on the problem of what happens when ambient mixtures are inhaled.
220
-------
22. INHALATION TOXICOLOGY RESEARCH INSTITUTE Mauderly
Studies can also progress in the opposite direction. For example,
efforts in the diesel exhaust study started with diluted raw exhaust but now
involve some of the individual components.
In subsequent chapters of this volume, other authors from the Inhalation
Toxicology Research Institute report results of studies that have been
concluded, review ongoing efforts, and emphasize proposed directions for
future research. Henderson (Chapter 23) discusses a project in which the
primary goal was to develop and demonstrate the usefulness of cellular damage
indicators for lung injury. The techniques developed in this recent, exciting
work show much promise as early detectors for lung cell injury and death.
Pickrell (Chapter 24) outlines a program on inhaled nitrogen dioxide that
includes both short- and long-term studies. Silbaugh (Chapter 25) discusses a
program to examine the structural and functional changes associated with
inhaled sulfates, again including both short- and long-term studies.
Finally, Wolff (Chapter 26) reviews studies primarily related to the
clearance of inhaled particles in the lung and the effect of inhaled sulfates
on the pattern (velocity or rapidity) of clearance of inhaled materials. This
is actually a study of lung defense mechanisms.
221
-------
23. CELLULAR (IN VIVO) AND BIOCHEMICAL CHANGES FOLLOWING INHALATION
OF ACID SULFATES: RAPID SCREENING TESTS TO DETERMINE THE PULMONARY
RESPONSE TO COMBINED POLLUTANT EXPOSURES
Rogene F. Henderson
Inhalation Toxicology Research Institute
Lovelace Biomedical and Environmental Research Institute
Post Office Box 5890
Albuquerque, NM 87115
PROBLEM
A major lack in the information required to regulate air pollutants is
knowledge of how various pollutants interact to produce synergistic, additive,
or negating effects in the exposed populace. To expose experimental animals
to single component aerosols on a long-term basis is expensive; adding double
or triple components increases the cost even more. One approach to the
problem is to use rapid in vitro screening tests, which cost less and yield
results quickly. However, for studying pollutant interactions, an in vitro
approach does not take into account the interactions that affect site of
deposition, clearance, retention, and (consequently) pollutant dose to tissue.
Epidemiology assesses real-world pollutant mixtures but cannot isolate the
effects of individual components of those mixtures. What is needed is a rapid
bioassay that forms a bridge between the _in vitro studies and the long-term _in
vivo and epidemiologic studies.
222
-------
23. PULMONARY EFFECTS OF POLLUTANT COMBINATIONS Henderson
APPROACH
Our laboratory uses bioassays that involve short-term _in vivo inhalation
exposures followed by rapid evaluation of pulmonary response by a variety of
indicators of early lung damage. Our past use of these bioassays to measure
pulmonary response to single pollutants is summarized below.
Lavage Fluid Analysis
Previous work showed that analysis of saline lung washings of exposed
animals can be used to detect an inflammatory response in the lung (Henderson
et al. 1978a, 1978b, 1979a, 1979b). Extracellular lactate dehydrogenase in
the airway fluid indicates increased cell membrane permeability; lysosomal
enzymes indicate phagocytic activity? increased sialic acid indicates an
irritant response in the upper airway? increased soluble protein in the
airways indicates damage of the capillary-alveolar barrier; and an influx of
neutrophils is consistent with an inflammatory response. Biochemical and
cytological changes in lavage fluid correlated well with morphological
indicators of an acute inflammatory response in Syrian hamster lungs exposed
to a nonionic surfactant (Henderson et al. 1978a) and to metal salts
(Henderson et al. 1979a, 1979b). The most sensitive indicator of the
multifocal lung damage (terminal bronchiolitis) characteristic of oxidants was
an increase in the polymorphonuclear cells in lavage fluid from hamsters
exposed to nitrogen dioxide (NO2) (DeNicola et al. 1979). Rats exposed to
sulfuric acid (H2S04) mist showed an increase in the sialic acid content of
223
-------
23. PULMONARY EFFECTS OF POLLUTANT COMBINATIONS Henderson
the lavage fluid as well as speeding of mucociliary clearance (Henderson et
al. 1978c).
1^C-Thymidine incorporation in Lung Cells
Pulse labeling of Chinese hamsters with 3H-thymidine demonstrated an
increased uptake of radiolabel in the lungs of animals exposed to N©2 (Hackett
1979). Tissue oxidizer analysis of 3H retention by the lung could be used as
a rapid screen for cellular damage. In animals showing the greatest response,
autoradiographic techniques could be applied to portions of the tissue to
determine the actual site of damage.
Lipid Synthesis
One indicator of cellular damage is an increased synthesis of membrane
lipids during repair processes, fln increased uptake of ^C-palmitate into
membrane lipids was detected in lung cells from animals exposed to 5 ppm NO2
for 8 h (Pfleger and Rebar 1978). The same label would also detect increased
synthesis of the surfactant lipid, dipalmitoyl lecithin. This lipid is
synthesized by the Type II epithelial cell, which is known to proliferate in
response to the most common type of lung injury—loss of the Type I epithelial
cell.
224
-------
23. PULMONARY EFFECTS OF POLLUTANT COMBINATIONS Henderson
PLANNED RESEARCH
The bioassays described above have been used to detect pulmonary response
to single pollutants. We propose to extend their use to detect the effects of
combinations of pollutants.
We plan to use the Charles River Hartley guinea pig because of our
previous experience with this animal's response to H2S04 (Silbaugh et al.
1978). Animals will be exposed to four combinations of NO2 and H2S04 (5 ppm
N02 plus 1.0 mg/m3 H2SO4; 1 ppm NO2 plus 1.0 mg/m3 H2S04? 5 ppm NO2 plus 10
mg/m3 H2S04; 1 ppm NO2 plus 10 mg/m3 H2SO4> and to the same levels of the
individual components. Control animals will be exposed to room air.
Exposures will be for 6 h on 2 subsequent days. The animals will be evaluated
at 48 h after initiation of the exposure, which has been identified as the
time of peak response to NO2 (DeNicola et al. 1979; Hackett 1979; Pfleger and
Rebar 1978). The end points to be measured include thymidine and palmitate
uptake into lung tissue as well as lavage fluid parameters. The lavage fuid
parameters will include lactate dehydrogenase, acid phosphatase, sialic acid,
soluble protein, glutathione peroxidase and reductase, fibrin degradation
products, and total and differential cell counts. In addition to these end
points, we will perform light microscopic evaluation of slices of deep lung
tissue, trachea, larynx, and nasal turbinates. In animals shown by the screen
to be most affected by exposure, autoradiographs of the 3H-thymidine labeled
tissue and scanning electron micrographs of the tissues will be obtained.
225
-------
23. PULMONARY EFFECTS OF POLLUTANT COMBINATIONS Henderson
Upon completion of this study, we will examine the effects of combined
exposure to H2S04 (1 mg/m3) and fly ash (5 mg/m3) from fluidized bed
combustion of coal. The final experiment will examine the effects of combined
exposure to all three pollutants: NO2 (5 ppm), H2SO4 (1 mg/m3), and fly ash
(5 mg/m3).
This research should aid in validating the screening system; the higher
pollutant levels have been included to meet this validation need. The major
result of the experiment will be to determine if any synergistic effects occur
in combined acute exposures to low levels of the described pollutants. This
determination should contribute to the groundwork for any regulatory decisions
based on combined effects from pollutant mixtures.
ACKNOWLEDGMENTS
This research was supported in part by the U.S. Environmental Protection
Agency via Interagency Agreement Number EPA-IAG-D5-E61 under U.S. Department
of Energy (DOE) Contract Number EY-76-C-04-1013 and in part by the U.S. DOE
under Contract Number EY-76-C-04-1013 and conducted in facilities fully
accredited by the American Association for Accreditation of Laboratory Animal
Care.
REFERENCES
DeNicola, D. B., A. H. Rebar, and R. F. Henderson. 1979. Early indicators of
pulmonary damage in Syrian hamsters exposed to N02« In: Annual Report
of the Inhalation Toxicology Research Institute, 1978-1979. Publication
LF-69, Lovelace Medical Foundation, Albuquerque, New Mexico, pp.
524-528.
226
-------
23. PULMONARY EFFECTS OF POLLUTANT COMBINATIONS Henderson
Hackett, N. A. 1979. Proliferation of lung and airway cells induced by
nitrogen dioxide. J. Toxicol. Environ. Health, 5:917-928.
Henderson, R. F., E. G. Damon, and T. R. Henderson. 1978a. Early damage
indicators in the lung. I. Lactate dehydrogenase activity in the
airways. Toxicol. Appl. Pharmacol., 44:291-297.
Henderson, R. F., B. A. Muggenburg, J. L. Mauderly, and W. A. Tattle. 1978b.
Early damage indicators in the lung. II. Time sequence of protein
accumulation and lipid loss in the airways of beagle dogs with beta
irradiation of the lung. Radiat. Res., 76:145-158.
Henderson, R. F., R. K. Wolff, A. H. Rebar, D. B. DeNicola, and R. L. Beethe.
1978c. Early indicators of lung damage from inhaled sulfuric acid mist.
In: Annual Report of the Inhalation Toxicology Research Institute,
1977-1978. Publication LF-60, Lovelace Medical Foundation, Albuquerque,
New Mexico, pp. 352-355.
Henderson, R. P., A. H. Rebar, and D. B. DeNicola. 1979a. Early damage
indicators in the lung. IV. Biochemical and cytologic response of the
lung to lavage with metal salts. Toxicol. Appl. Pharmacol., 57:129-135.
Henderson, R. F., A. H. Rebar, J. A. Pickrell, and G. J. Newton. 1979b.
Early damage indicators in the lung. III. Biochemical and cytological
response of the lung to inhaled metal salts. Toxicol. Appl. Pharmacol.,
50:123-136.
Pfleger, R. C., and A. H. Rebar. 1978. Pulmonary alveolar Type II cells and
macrophage damage following an NO2 inhalation exposure. In: Annual
Report of the Inhalation Toxicology Research Institute, 1977-1978.
Publication LF-60, Lovelace Medical Foundation, Albuquerque, New Mexico,
pp. 419-424.
Silbaugh, S. A., D. G. Brownstein, R. K. Wolff, and J. L. Mauderly. 1978.
Airway response of the Hartley guinea pig to sulfuric acid aerosol. In:
Annual Report of the Inhalation Toxicology Research Institute, 1977-1978.
Publication LF-60, Lovelace Medical Foundation, Albuquerque, New Mexico,
pp. 360-363.
WORKSHOP COMMENTARY
J. D. Hackney; In the 3H-thymidine labeling, what was labeled, and how? Was
this by autoradiography or was this extracted from the lung?
227
-------
23. PULMONARY EFFECTS OF POLLUTANT COMBINATIONS Henderson
R. F. Henderson; This was by an autoradiographic technique which showed
labeling of epithelial cells. But autoradiography is not a screening tool.
It is a very slow procedure. For a screening tool we propose to use the first
step of our autoradiography studies in which a tissue oxidizer is used to
determine which animals have actually taken up label. The tissue oxidizer
oxidizes the 3H and 14C in the tissues to 3H2O and 14CO2, which can be
collected and counted in a liquid scintillation spectrometer.
D. E. Gardner; In your studies showing increases in polymorphonuclear cells,
did you look at the functioning of these polymorphonuclear cells, their
phagocytic capabilities, or the macrophages—
R. F. Henderson; No. The macrophages appeared activated (that is, they were
enlarged and had large inclusion bodies), but we made no measurements other
than to observe the morphological changes.
D. E. Gardner; Did you find a decrease in macrophages with the increase in
polymorphonuclear cells following NO2 exposure?
R. F. Henderson; Both the macrophages and the polymorphonuclear cells in the
lavage fluid increased.
228
-------
24. RESPIRATORY TOXICOLOGY OP NITROGEN OXIDES
John A. Pickrell
Inhalation Toxicology Research Institute
Lovelace Biomedical and Environmental Research Institute
Post Office Box 5890
Albuquerque, NM 87115
PROBLEM
Assessing the health effects of nitrogen oxides (NOX) provides
information that is necessary in the regulatory process. Although short-term
inhalation of high concentrations of nitrogen dioxide (NO2) is known to damage
both pulmonary alveolar macrophages and Type I epithelial cells, the health
effects of inhaling environmental levels of NO2 for long periods of time have
not been adequately defined. Little is known concerning the health effects of
inhaling NO2 in combination with other pollutants such as ozone (O3) or fly
ash. Also, the effects of inhaling other oxides of nitrogen (nitric oxide,
nitrates, nitrites, peroxyacyl nitrate, nitrous or nitric acid, nitrosamines)
have not been studied extensively.
Lung tissue exposed to a variety of N02 levels seems to adapt to this
insult. The mechanism and extent of the adaptation and its relation to the
animal's health status are poorly understood. Indicators for a variety of
229
-------
24. RESPIRATORY TOXICOLOGY OF NITROGEN OXIDES _ Pickrell
changes are required to assess the extent of pulmonary injury, adaptation, and
residual injury. Biochemical indicators of damage, although quite sensitive,
must be related to cellular and morphological changes to understand lung
injury following NC>2 inhalation. Of special importance are those parameters
which persist long after pulmonary injury has ceased. These are the
biomedical determinants of the resulting pulmonary injury.
The experiments described below were directed toward defining the NO2
dose causing acute effects and the consequence of those effects for
development of chronic pulmonary injury. Nitrogen dioxide damage to pulmonary
alveolar Type I cells is reflected in cell necrosis, pulmonary edema, an
influx of polymorphonuclear (PMK) leukocytes, and an increased pulmonary
distensibility. This damage leads to turnover of extracellular matrix,
heightened immune response in the regional lymph nodes from increased access
of antigen to interstitial tissue, and a proliferation of Type II alveolar
pneumocytes. Most indicators of damage return to normal and the lung seems to
"adapt" to the NO2 injury. Spindle cell proliferation, continued low-level
inflammation, continued increased pulmonary distensibility, and increased mean
linear intercepts and pulmonary collagen suggest that incomplete pulmonary
adaptation to N©2 exposure may lead to chronic pulmonary injury.
Future work will define the relation of incomplete adaptation to
development of chronic pulmonary injury from VfO2 levels relevant to
smog-filled urban atmospheres. The relation of NO2 and 03 to pulmonary injury
will be explored (again at levels relevant to environmental exposures).
230
-------
24. RESPIRATORY TOXICOLOGY OF NITROGEN OXIDES Pickrell
Finally, acute studies will assess the toxicities.of other oxides of nitrogen
in comparison to NO2.
APPROACH
The effect of inhaling IK>2 at several levels was studied in Chinese and
Syrian hamsters and Fischer 344 rats. Chinese hamsters were used in
cytokinetic studies so that the resulting data could be correlated with data
from parallel analyses of lung chromosomes in the same species. Chinese
hamsters were selected for the lung chromosome analysis because of the
species' simple chromosome pattern. Syrian hamsters had been determined to be
free of long-term lung disease in earlier parallel studies, and were therefore
selected. Later, Fischer 344 rats were used because of a longer lifespan in
comparison to the Syrian hamsters. Where comparisons between species were
made, IK>2 seemed to induce similar series of morphological, cytological, and
biochemical events. The significance of each change is discussed below.
Exposure of Fischer 344 rats to 20 ppm NO2 for 48 h led to damaged
alveolar Type I cells (Pickrell et al. 1978). These changes were reflected by
increased airway lactate dehydrogenase (LDH), suggesting pulmonary cell
damage, and by increased airway alkaline phosphatase, suggesting pulmonary
cell necrosis. Increased airway protein and trypsin inhibitory capacity
suggested pulmonary edema. Pulmonary edema in the airways suggested both
increased vascular permeability and an interruption of the tight junction of
alveolar Type I cells due to cell damage. This Type I cell damage led to an
231
-------
24. RESPIRATORY TOXICOLOGY OF NITROGEN OXIDES Pickrell
influx of PMN leukocytes into the airway and an increased turnover of collagen
of the extracellular matrix by 14 d. This change reflected collagenolysis and
proteolysis of pulmonary interstitium probably from exposure to airway enzymes
from inflammatory cells (macrophages and PMN leukocytes). Such exposure was
possible, since damage to alveolar Type I cells had occurred. Increased
parenchymal alkaline phosphatase and neutral protease were consistent with the
terminal bronchiolitis observed at that time. Results from other studies in
rats and Syrian hamsters suggested damage to alveolar Type I cells.
In a second study, exposure of adult male Fischer 344 rats to 26 ppm NC>2
for 24 h increased the number of anti sheep red blood cell plaques formed by
lung-associated lymph nodes (LALN). This change may have reflected an altered
load to these nodes due to pulmonary Type I cell damage. In another study,
Syrian hamsters exposed to 12 ppm NO£ for 48 h had a mild necrotizing terminal
bronchiolitis with increased airway PMN leukocytes. In another experiment,
Syrian hamsters exposed to 5 ppm NO2 for 8 h showed increased quantities of
viable airway granulocytes by 48 h after initiation of exposure, suggesting
damage to alveolar Type I cells and recruitment of inflammatory cells.
Necrosis of pulmonary alveolar Type I cells led to hyperplasia of
pulmonary alveolar Type II cells (Evans et al. 1977). In Chinese hamsters
exposed to 15 ppm N(>2 for 24 h, an increased labeling index for 3H-thymidine
was observed by 24 h and persisted to 3 weeks after exposure (Hackett 1979).
Type II cells were observed twice as frequently in the terminal bronchioles as
in other pulmonary areas. The Type II cell cycle time was reduced from 26 d
232
-------
24. RESPIRATORY TOXICOLOGY OF NITROGEN OXIDES Pickrell
to 3 d. These changes were consistent with terminal bronchiolitis. Type I
cell death, and Type II cell proliferation. Epithelial cells lining small
airways and alveoli were more susceptible to NO2 exposure than were bronchial
and tracheal epithelia (Hackett 1979). Syrian hamsters exposed to 12 ppm NO2
for 48 h showed bronchiolar epithelial hyperplasia. Fischer 344 rats exposed
to 20 ppm NC>2 for 48 h had terminal bronchiolitis and changes suggestive of
Type II cell hyperplasia (Pickrell et al. 1978). Both studies were consistent
with damage to alveolar Type I cells in terminal bronchiolar areas leading to
Type II cell hyperplasia.
Type II alveolar cell hyperplasia was associated with alterations in
synthesized lipids probably destined for cell membranes. Cells from an
"enriched Type II cell fraction" of lungs of Syrian hamsters exposed to 15
(Pfleger et al. 1980), 5, or 1 ppm for 8 h had increased incorporation of
radiolabel into unsaturated phosphatidyl choline, reflecting increased
production of membrane lipid. These changes were consistent with increased
production of cell membrane lipid by Type II cells (exposed to 1 ppm N02 for 8
h), a step which must precede hyperplasia of Type II alveolar pneumocytes.
When rats were exposed to 20 ppm N02, most indicators of damage returned
to normal by 2 mo after initiation of exposure (Pickrell et al. 1978). The
lung seemed to "adapt" to continued exposure. Return of airway granulocytes,
tissue alkaline phosphatase and acid protease, and tissue LDH to control
levels by 2 mo after initiation of exposure suggested that parenchymal tissue
inflammation was no longer present. Return of airway LDH and alkaline
233
-------
24. RESPIRATORY TOXICOLOGY OF NITROGEN OXIDES _ Pickrell
phosphatase to control levels suggested that parenchymal cell damage and
necrosis were no longer occurring. Finally/ return of airway collagen to
control levels suggested that disruption of the extracellular matrix had
ceased. This was consistent with a reduction in Type I alveolar cell necrosis
and an "adaptation" of the lung to continued NC>2 exposure. Lungs of Syrian
hamsters exposed to 12-22 ppm NC>2 had returned to normal by 21 d after
initiation of a 48-h exposure. By 7 d after exposure of Fischer 344 rats to
26 ppm NO2 for 24 h, the ability of LALN to form plaques following challenge
with sheep red blood cells had returned to the control level, suggesting that
the antigen load to LALN had returned to normal. This was consistent with
cessation of damage to Type I alveolar cells and with lung adaptation to
Several parameters remained altered (relative to control animals) in
Fischer 344 rats exposed to 20 ppm NC>2 for 2 mo. These changes suggested that
any "recovery" to that exposure was incomplete by 10 mo after stopping
exposure. Chronic pulmonary injury had presumably occurred. Pulmonary
distensibility at 20 cml^O was increased relative to controls at 6.5 mo but
not at 12 mo after initiation of NC>2 exposure. This morphometric parameter
slowly adapted, suggesting that pulmonary inflammation or alveolar distension
lessened. Persistence of increased neutral protease throughout the study
suggested that chronic low-level inflammation continued 10 mo after N(>2
exposure ceased. The increase of mean linear intercept observed 2-12 mo after
initiation of exposure was not accompanied by histological evidence of septal
destruction or emphysema. The thinning of walls and increased alveolar size
( histologically within normal limits) observed 12 mo after initial NC>2
234
-------
24. RESPIRATORY TOXICOLOGY OF NITROGEN OXIDES Pickrell
exposure were consistent with increased linear intercept and may have
represented a reduction in elastic recoil. The degree to which elastin
quantity or metabolism were altered was not determined. By 14 d after
initiation of NC>2 exposure, alveolar walls were subtly thickened by apparent
spindle cell proliferation suggestive of early fibrosis, although collagen
content was normal. Later, collagen content was increased from 2-12 mo after
initiation of NC>2 exposure, although no histological evidence of fibrosis was
present. One may speculate that these changes were consistent with collagen
replacement of alveolar elastic tissue, loss of elastic recoil, and increased
mean linear intercepts. If the replacement was sufficiently diffuse, no
histological evidence of wall destruction might have been noted. Certainly
these changes suggest that incomplete "adaptation" to pulmonary injury from
N02 exposure may lead to chronic pulmonary injury not resolved 10 mo after
cessation of exposure. Such chronic pulmonary injury might occur even in the
absence of septal destruction. The relation of pulmonary "adaptation" and
chronic pulmonary injury to lower levels of IK>2 with intermittent "spike"
levels relevant to concentrations encountered in smog should be of
considerable interest.
PLANNED RESEARCH
Future work assessing the toxicity of inhaled NOX will bear on important
scientific questions relevant to the goals of EPA. Further work with NO2 will
address the still unanswered questions of (1) a minimal effects or low-risk
level, (2) adaptation to intermittent NO2 spikes simulating urban levels, and
235
-------
24. RESPIRATORY TOXICOLOGY OF NITROGEN OXIDES Pickrell
(3) the effects of NO2 in older animals or in animals with preexisting
disease. The most important and relevant question is the mechanism, extent,
and consequences of adaptation to intermittent spikes of NO2 superimposed upon
a 3 to 5 fold lower maintenance level of NC>2«
A second important area for which there is little information is the
interaction of NO2 with such other pollutants as 03, peroxyacyl nitrate, fly
ash, nitric oxide, and oxides of sulfur. The most frequent interaction of NO2
is with 03; these pollutants frequently coexist in smog. Both have
considerable potential for producing chronic pulmonary injury. The pulmonary
system seems to at least partially "adapt" to either pollutant.
Another important scientific need is an assessment of the relative
toxicities of other oxides of nitrogen, peroxyacyl nitrate, nitrosamines,
nitrates and nitrites, and nitric and nitrous acid. Peroxyacyl nitrate is a
potent oxidant. Preliminary reports suggest that it has the same approximate
relative toxicity as ©3 and greater relative toxicity than NC>2, although
ambient atmospheric conditions seem quite low. Nitrosamines are known
carcinogens, but ambient atmospheric conditions appear considerably lower than
even those of peroxyacyl nitrate.
The mechanism, extent, and consequences of pulmonary adaptation to NOX
will be studied using a maintenance level of 1 ppm NO2 for 24 h/d, 5 d/week.
A spike of 5 ppm N02 will be superimposed on the 1 ppm level in a second group
of animals. This spike will occur 2 times/d, to simulate urban pollution
236
-------
24. RESPIRATORY TOXICOLOGY OF NITROGEN OXIDES Pickrell
conditions. These groups will be compared to a third group of animals exposed
to 5 ppm NC>2 5 d/week and to a control group housed in exposure chambers.
Sacrifices will occur from 2 weeks to 2 yr after the initiation of exposures.
The study will employ ~250 Fischer 344 rats.
End points will relate to the general areas of pathology, morphometry,
physiology, and biochemistry. The pathological assessment will include
histopathology and electron microscopy (both scanning and transmission
electron microscopy). Morphometric end points will include pulmonary
distensibility (displaced volume of lung fixed at 24 cmH2O/kg body weight) and
mean linear intercept. Internal surface areas will be calculated. Pulmonary
physiology will include measurement of static compliance in excised lungs and
measurement of other respiratory functions in selected animals. Biochemical
measurements will include lipoperoxidation, glutathione peroxidase and
reductase, acid and alkaline phosphatase, LDH and glucose-6-phosphate
dehydrogenase, protein, trypsin inhibitory capacity, protease activity as a
function of pH, and collagen and elastin metabolism.
The effect of inhaling NC>2 and an associated pollutant, 03, is an
important scientific question. Fischer 344 rats will be used to establish a
dose response and interaction. The end points to be studied are described
above. Six groups will be exposed to NO2, O3, or both: (1) 1 ppm NO2, (2) 5
ppm N02, (3) 0.25 ppm 03, (4) 1 ppm O3, (5) 1 ppm NO2 plus 0.25 ppm 03, and
(6) 5 ppm N02 plus 1 ppm O3. An equal sized group of controls will be housed
in chambers and exposed to ambient filtered air. Each group will contain -140
237
-------
24. RESPIRATORY TOXICOLOGY OF NITROGEN OXIDES Pickrell
rats. Serial sacrifices will occur from 1 mo to 3 yr after initiating
exposure. Approximately 60 rats will be retained for lifespan studies.
Assessement of the relative toxicities of other oxides of nitrogen will
be accomplished in short-term acute experiments. These dose-response studies
will employ at least two levels, one quite high. Responses will be observed
in days to weeks as opposed to years (except for known carcinogens, for which
observations may continue for up to 1 yr). The relative toxicity information
will be combined with estimates of environmental concentrations in an effort
to predict relative hazard. The compound with the greatest relative hazard
will be selected for a long-term toxicity study. We consider peroxyacyl
nitrate and nitrosamines to be the most relevant pollutants for initial
assays. Peroxyacyl nitrate is reported to be a potent oxidant: preliminary
work suggests its toxicity to exceed that of NC»2 and to approach that of 03.
To date, peroxyacyl nitrate has received little attention because its ambient
concentrations appear to be extremely low. Nitrosamines are known to be
potent carcinogens. However, their ambient concentrations appear to be orders
of magnitude below even those of peroxyacyl nitrate. Methods of detecting and
quantifying peroxyacyl nitrate and nitrosoamines continue to be improved, and
care must be taken in interpreting measured atmospheric levels.
ACKNOWLEDGMENTS
This research was supported in part by the U.S. Environmental Protection
Agency via Interagency Agreement Number EPA-IAG-D5-E61 under U.S. Department
of Energy (DOE) Contract Number EY-76-C-04-1013 and in part by the U.S. DOE
238
-------
24. RESPIRATORY TOXICOLOGY OF NITROGEN OXIDES Pickrell
under Contract Number EY-76-C-04-1013 and conducted in facilities fully
accredited by the American Association for Accreditation of Laboratory Animal
Care.
REFERENCES
Pickrell, J. A., A. H. Rebar, E. G. Damon, R. L. Beethe, R. C. Pfleger, and C.
H. Hobbs. 1978. Exposure to NC>2 as a model for producing emphysema:
Early results. In: Annual Report of the Inhalation Toxicology Research
Institute, 1977-1978. Publication LP-60, Lovelace Medical Foundation,
Albuquerque, New Mexico, pp. 428-432.
Evans, M. J., L. J. Cabral-Anderson, and G. Freeman. 1977. Effects of NO2 on
the lungs of aging rats. II. Cell proliferation. Exp. Mol. Pathol.,
27:366-376.
Hackett, N. A. 1979. Proliferation of lung and airway cells induced by
nitrogen dioxide. J. Toxicol. Environ. Health, 5:917-928.
Pfleger, R. C., A. H. Rebar, and J. A. Pickrell. 1980. Biochemical effects
of nitrogen dioxide on pulmonary alveolar Type II cells and pulmonary
alveolar macrophages from Syrian hamsters. Toxicol. Appl. Pharmacol., in
press.
WORKSHOP COMMENTARY
P. E. Morrow; You mentioned something about the antigenic burden of the lymph
nodes. Would you expound on that?
J. A. Pickrell: Dr. David Bice of our laboratory performed a study on
pulmonary immunology and NO2. Rats were exposed to 26 ppm NO2 for 24 h. Use
of sheep red blood cells to stimulate the lung immune response in regional
lymph nodes led to increased plaque formation in LALN. The increase in plaque
formation peaked at 2 d, was reduced at 3 d, and had returned to normal by 7 d
after exposure. His interpretation was that the increased load to the LALN
was due to interruption of the Type I alveolar pneumocyte barrier, that the
reduction represented Type II cell proliferation and phagocytosis °f
peroxyacyl nitrate, and that normal pulmonary architecture had been restored
by 1 week after exposure.
P. E. Morrow; What does that have to do with the Type I cell barrier?
239
-------
24. RESPIRATORY TOXICOLOGY OF NITROGEN OXIDES Pickrell
J. A. Pickrell: The sheep red blood cells were introduced into the lung and
the lymph nodes were stimulated. Bice interpreted this to mean that the
epithelial integrity of the lung was interrupted; I think he had some
histopathology to back this up. He concluded that more sheep red blood cells
were therefore available to the lymph nodes and that the lymph nodes were
simply responding to an increased load in a normal immunologic manner.
P. S. Morrow; I follow the antigenic stimulation concept but I don't think
the Type I integrity follows.
There is another point that I wish to clarify. In all of the studies in
which you've done thymidine incorporation and have looked for proliferative
changes and so on, you did not mention the Clara cell in these rodents. I
thought that this was a very prominent feature of small airway effects: that
there would be damage in that area and there would be Clara cell
proliferation. But it hasn't been mentioned; instead, the discussion focused
on Type II cells and basal cells. Did you look specifically at the Clara
cell?
J. A. Pickrell; It was looked at by Dr. Nora Hackett of our laboratory; the
predominant response she saw was a Type II alveolar pneumocyte proliferation.
P. E. Morrow; Isn't the Clara cell alleged to be the progenitor cell in the
small airway epithelium?
J. A. Pickrell; Yes.
R. F. Henderson; I don't believe Dr. Hackett distinguished the Clara cell,
but merely indicated nonciliated airway epithelium.
Question; Did Dr. Hackett differentiate what you're calling "Type II" from
the Clara cell?
J. A. Pickrell; She differentiated Type II cells from nonciliated and
ciliated airway lining cells.
Question; What is your understanding of the ambient concentrations of 03 and
N(>2? You mentioned 1 and 5 ppm as "approaching ambient concentrations."
J. A. Pickrell; In the Los Angeles Basin, 63 might approach 0.3 ppm to 0.5
ppm; NC>2 seldom exceeds 1 ppm and is usually ~0.5 ppm. The Los Angeles Basin,
I believe, displays one of the higher concentrations in the country. Does
that answer your question?
Comment; Yes. My comment would be that 5 ppm is not anywhere near ambient
concentration.
(
J. A. Pickrell; That's correct. 1 ppm would be.
240
-------
24. RESPIRATORY TOXICOLOGY OF NITROGEN OXIDES Pickrell
Comment; I don't believe that there are any monitoring data anywhere close to
1.5 for NO2—
J. A. Pickrell; I believe in some of the NO2 reports I've seen it has
approached 1, but you're right: it's more nearly 0.5.
D. E. Gardner; I'm sure you're familiar with Dr. Gus Freeman's work with NO2
in which rats and monkeys were exposed for a lifetime. Freeman very
methodically described many of the effects that you have reported, not only
with NO2 alone but also with NO2 plus 03 (related to what you're proposing).
Is the purpose of your studies to confirm his? What have you added to
Freeman' s studies?
J. A. Pickrell; I don't think he used spikes of NO2 to simulate urban
conditions.
D. E. Gardner; I'm referring to your NO2-O3 study.
J. A. Pickrell; Freeman's NO2-(>3 study, if my recollection is correct, used
somewhat higher levels than ours.
D. S. Gardner; The lowest level was 2 ppm.
J. A. Pickrell; Freeman's NO2 level was 2 ppm, but the NO2-03 study was done
at a somewhat higher level, if I remember correctly.
Comment; The 03 level was 0.6 ppm.
S. V. Dawson; In view of your use of high and low levels, what do you
consider to be high and low levels for the peroxyacyl nitrate and nitric acid
exposures that you're planning? Also, which type of peroxyacyl nitrate will
be used, and how will you obtain it?
J. A. Pickrell; I haven't decided what type of peroxyacyl nitrate will be
used.
When they're detected at all, peroxyacyl nitrate levels are in the ppb
range. The effects levels that have been reported so far are in the ppm
range. Therefore, we propose to start where effects levels have been reported
and work down toward ambient levels. As our high level, we will use a
concentration just below the concentration at which effects have been
reported. For our low level, we will use a level an order of magnitude
lower.
D. E. Gardner; Have you begun to generate and monitor peroxyacyl nitrate?
J. A. Pickrell: No.
241
-------
24. RESPIRATORY TOXICOLOGY OF NITROGEN OXIDES Pickrell
D. E. Gardner; Our recent experience suggests that you may encounter real
problems. Peroxyacyl nitrate is not very toxic. In most model systems it
presents real problems. You might talk with the Illinois Institute of
Technology Research Institute [Life Science Division, 10 West 35th St.,
Chicago, IL 60616]; they did the work for us, and talking with them might
give you a head start in this area.
J. A. Pickrell; Thank you.
242
-------
25. ACUTE-CHRONIC LOSS OF LUNG FUNCTION FOLLOWING INHALATION
OF ACID SULFATES
Steven A. Silbaugh
Inhalation Toxicology Research Institute
Lovelace Biomedical and Environmental Research Institute
Post Office Box 5890
Albuquerque, NM 87115
PROBLEM
The existence, nature, and importance of health effects resulting from
the inhalation of acid sulfate aerosols (alone and in combination with other
fossil fuel combustion byproducts) are being addressed. This project provides
information relevant to the setting of sulfate and fine particulate
standards.
APPROACH
Toxicity of Sulfuric Acid Aerosols in the Guinea Pig
We conducted dose-response studies of the mortality of Hartley guinea
pigs exposed for 8 h to sulfuric acid (H2SO4) particles of either 0.4 or 0.8
Um in diameter. A total of 96 animals were divided into groups and exposed to
43, 83, or 109 mg/m3 of the 0.4-um aerosol or 21, 32, or 43 mg/m3 of the
243
-------
25. LUNG FUNCTION LOSS FOLLOWING INHALATION OF ACID SULFATES Silbauqh
0.8-ym aerosol. Marked differences in lethality of the two aerosols were
observed. The LCso for the °«8-Vm aerosol was ~30 mg/m3, while an LC50 for
the 0.4-ym aerosol was not reached even at 109 mg/m3.
Effects of Sulfuric Acid Aerosols on Pulmonary Function in the Guinea Pig
Using Hartley guinea pigs, we completed studies of the acute respiratory
function effects of inhaled 1.0-pm 112804 particles. A total of 57 animals
were exposed for 1 h to concentrations of 0, 1.2, 1.3, 14.6, 24.3, or 48.3
mg/m3; respiratory patterns and breathing mechanics were measured during
exposure. The observed effects differed from the graded dose-response
relationship previously reported by Amdur et al. (1978). Functional responses
reflecting airway constriction were either absent (nonresponsive animals) or
overwhelming (responsive animals). The proportion of responsive to
nonresponsive animals increased with exposure concentration, but the magnitude
of pulmonary function change was similar for responsive animals regardless of
concentration. Our results suggest that the Hartley guinea pig reacts to
inhaled H2SO4 with an essentially all-or-none airway constrictive response,
and that there is considerable variation in the concentrations at which
different animals develop the response.
Animal Strain Differences and Adaptation to Acute Sulfuric Acid Exposure
Twelve guinea pigs of each of the following strains were exposed to 30-36
mg/m3 of 1.0-ym ^804 aerosol: Charles River Hartley, Camm Hartley, Camm
244
-------
25. LUNG FUNCTION LOSS FOLLOWING INHALATION OF ACID SULFATES Silbaugh
English Shorthair, and NIH Strain 13. Half of the exposures involved a
gradual increase to the above concentration over a 1-h period, followed by a
1-h exposure at a steady concentration. The remaining exposures were
conducted for 1 h with an immediate change from clean air to the desired
concentration. The degree of labored breathing during exposure was scored
visually- The NIH Strain 13 guinea pigs developed more severely labored
breathing than the other strains, but these animals were later discovered to
have respiratory infections. All strains had less labored breathing when
H2SO4 concentrations were increased gradually rather than suddenly, suggesting
partial adaptation.
Sulfuric Acid and Nitrogen Dioxide Induced Alterations
in the Guinea Pig* 8 Sensitivity to Histamine Aerosol
Hartley guinea pigs exposed for 1 h to an ^804 aerosol of 1 urn in mass
median aerometric diameter (MMAD) or to nitrogen dioxide (NO2> gas were
examined for alterations in airway responsiveness to inhaled histamine.
Concentrations of H2SO4 ranged from 4 to 40 mg/m3; concentrations of NO2
ranged from 7 to 146 ppm. One group of animals exposed to filtered room air
served as the control. Animals were exposed to stepwise increased
concentrations of ~0.6-wm histamine aerosol 2 h prior to pollutant exposure
and at 0, 2, and 19 h after exposure. The histamine concentration required to
produce a 50% decrease from pre-challenge dynamic lung compliance was used as
a measure of airway sensitivity. All animals exposed to NO2 exhibited an
increase in histamine sensitivity immediately after exposure; the magnitude of
245
-------
25. LUNG FUNCTION LOSS FOLLOWING INHALATION OF ACID SULFATES Silbaugh
this increase was concentration dependent. We observed a dramatic return of
the sensitivities of most animals toward base-line values at 2 h after
exposure; however, the sensitivities of several animals remained elevated at
19 h after exposure. Animals exposed to 1*2804 exhibited major increases in
histamine sensitivity only if labored breathing developed during exposure. We
concluded that both NC>2 and 112804 alter airway sensitivity in the guinea pig,
but by apparently different mechanisms.
PLANNED RESEARCH
Alterations in the Guinea Pig* s Sensitivity to Histamine
Following Acute Nitrogen Dioxide Exposure at 5 ppm
The above results demonstrated that major alterations in the guinea pig's
airway responsiveness to inhaled histamine could be produced by 1-h exposures
to 40 ppm NO2» Airway-sensitizing effects were noted at lower concentrations,
but more information is needed at concentrations that are nearer ambient
levels. This study will examine the acute alterations in the guinea pig's
histamine responsiveness induced by 1-h exposures to 5 ppm N02« Two groups of
animals will be exposed: one to room air (n = 10) and one to 5 ppm NO2 (n =
10). One control and one NO2 animal will be exposed on each exposure day.
Histamine responsiveness will be measured before and immediately after
exposure, using methods similar to those described in the previous section.
246
-------
25. LUNG FUNCTION LOSS FOLLOWING INHALATION OF ACID SULFATES Silbaugh
Pollutant Interaction Studies; Nitrogen Dioxide and Sulfuric Acid
Several epidemiologic studies have suggested a positive correlation
between air pollutant levels and frequency of asthmatic attacks. Although
H2SO4 produces an asthma tic-like response in the guinea pig, it does so only
at concentrations that are far above ambient concentrations; the lowest
concentration at which we have observed this asthmatic-like response in
healthy guinea pigs is 15 mg/m3. If H2SC>4 does act to promote airway
c.onstrictive responses in human populations, other predisposing factors are
probably prerequisite. If airway sensitivity to H2SO4 is, like histamine
sensitivity, increased by oxidant gas exposures, then NC>2 might act as a
predisposing factor. In this study, we will determine if an NC>2 exposure
known to produce altered histamine sensitivity in the guinea pig also results
in increased airway sensitivity to H2SO4. An initial study will use high
concentrations of NO2 and H2S04 to determine whether such an interaction
exists. For each of three exposure days, one group of guinea pigs will be
exposed for 1 h to 40 ppm NO2. A second group of animals will be exposed for
1 h to filtered room air. Immediately after exposure, both control and
N02-exposed animals will be randomly assigned to cages within an H2S04
exposure chamber. Animals will be exposed for 2 h to an H2SO4 concentration
of 20 mg/m3 and 1.0 ym in MMAD. We will use excised lung volume as an
indicator of airway constrictive response, since our previous experience has
indicated that the lungs removed from animals which develop an airway
constrictive response are consistently hyperinflated. Animals that die during
H2S04 exposure will be removed from the exposure chamber through a
247
-------
25. LUNG FUNCTION LOSS FOLLOWING INHALATION OF ACID SULFATES Silbaugh
pass-through box; their lungs will be excised immediately and lung volume
measured by water displacement. Animals that survive exposure will be
sacrificed immediately after exposure with an euthanasia solution; the lungs
will be excised and lung volume measured. If this preliminary study indicates
a positive interaction, studies will be conducted at lower NO2 and H2S04
concentrations. Combination as well as sequential exposures are of interest
and will be examined in subsequent studies.
Pulmonary Effects of Chronic Exposure to Sulfuric Acid, Fly Ash, and
Their Combination on Normal and Elastase-Treated Guinea Pigs
This study will examine the response of normal and impaired
(elastase-treated) guinea pigs to long-term exposure to 112804, fly ash, and
their combination. The study is designed to model "sensitive" segments of the
human population: of various laboratory animals, the guinea pig may possess
airway responses most similar to those of the human asthmatic population;
elastase treatment will impose an impairment with structural and functional
similarities to human emphysema. We plan to conduct lung function and
biochemical tests plus morphological evaluations that have been carefully
selected to detect subtle pulmonary effects.
Five groups of 60 animals each will be exposed for 6 h/d, 5 d/week, for 3
yr. Of each group of 60 animals, 30 will be treated with porcine pancreatic
elastase ~3 weeks prior to exposure. The remaining 30 will be untreated.
Present plans are to expose the five groups of animals to (1) filtered room
248
-------
25. LUNG FUNCTION LOSS FOLLOWING INHALATION OF ACID SULFATES Silbaugh
air, (2) low-level H2S04 (in the range of 0.5-1 mg/m3), (3) high-level H2SO4
(in the range of 1-5 mg/m3), (4) fly ash alone (one level), and (5) high-level
H2S04 in combination with fly ash. The H2S04 aerosol will be ~0.4 urn in MMAD.
The fly ash will be a well characterized sample obtained from a fluidized bed
coal combustor.
Ten normal and 10 elastase-treated animals will be randomly selected from
each exposure group. The airway histamine sensitivity of each selected animal
will be measured prior to exposure and after 1, 3, and 6 mo and 1, 1.5, 2,
2.5, and 3 yr of exposure. Pulmonary function measurements, including
measurements of dynamic compliance and total pulmonary resistance, will be
obtained prior to each airway sensitivity measurement. This group of animals
will be sacrificed at 3 yr after the start of exposure. Half of the remaining
normal and elastase-treated guinea pigs in each exposure group will be
sacrificed after 1 and 2 yr of exposure. Other evaluations will be performed
on these animals (serially or prior to sacrifice), including measurements of
mucociliary clearance and lung function measurements sensitive to small airway
dysfunction. The nature and schedule of these evaluations will depend on the
outcome of preliminary feasibility studies. At sacrifice, one lung from each
animal will be processed for histopathologic and morphometric evaluation. The
other lung will be washed for analysis of cytoplasmic and lysosomal enzymes
present in the airways. Cytological assays will also be performed on airway
*
fluid samples. The tissue of the washed lung will be analyzed for collagen
and elastin content and enzymatic activity.
249
-------
25. LUNG FUNCTION LOSS FOLLOWING INHALATION OF ACID SITUATES Silbaugh
ACKNOWLEDGMENTS
This research was supported in part by the U.S. Environmental Protection
Agency via Interagency Agreement Number EPA-IAG-D5-E61 under U.S. Department
of Energy (DOE) Contract Number EY-76-C-04-1013 and in part by the U.S. DOE
under Contract Number EY-76-C-04-1013 and conducted in facilities fully
accredited by the American Association for Accreditation of Laboratory Animal
Care.
REFERENCE
Amdur, M. O., M. Dubriel, and D. A. Creasia. 1978. Respiratory response of
guinea pigs to low levels of sulfuric acid. Environ. Res., 15:418-423.
WORKSHOP COMMENTARY
0. Raabe; Could you give us more details about the chemical and physical
characteristics of the fly ash aerosols?
S. A. Silbaugh; The aerosols will come from a fluidized bed combustor at the
Morgantown Energy Research facility- If requested, I can provide the details
of trace element analysis.
O. Raabe; What is the size distribution, and so on?
S. A. Silbaugh; The size distribution is ~3 ym MMAD, I believe. Maybe Dr.
Henderson can comment on that.
R. F. Henderson; The sample we ran previously was 3 ym in MMAD, with a
geometric standard deviation of 1.7. In contrast to your fly ash, it's a low
temperature, highly porous, and very polydispersed ash.
J. D. Hackney; In the histamine responsiveness studies you used dynamic
compliance measurements, correct? •
S. A. Silbaugh; Yes.
J. D. Hackney; Did you also look at resistance?
250
-------
25. LUNG FUNCTION LOSS FOLLOWING INHALATION OF ACID SULFATES Silbaugh
S. A. Silbaugh; Yes, we looked at resistance, but we found (along with the
group of Douglas, Dennis, and the late Dr. Bouhuys at Yale) that the best
indicator of the airway constrictive response in the histamine-exposed guinea
pig is a decrease in dynamic compliance. Resistance changes are more
variable.
As soon as the compliance drops by 50%, the animal is removed from the
histamine; we don't want any residual effects caused by the spasm itself. All
I can say is that the resistance at that time is variable, and we haven't
followed it beyond the time at which the compliance drop appears.
P. E. Morrow; You stated that your H2S(>4 aerosol was 0.4 ym in aerodynamic
size. How do you vary the concentration and keep that droplet size constant?
Have you done it over the range of 0.1 to 100 mg/m3?
S. A. Silbaugh; We have a system whereby we can keep the particle size
constant. Dr. Wolff has been more involved with that than I have. Dr.
Carpenter (Inhalation Toxicology Research Institute) has worked with a system
in which dry nitrogen is used to pick up sulfur trioxide, which is then mixed
with humid air. Varying the dilutions of the sulfur trioxide and the water
vapor as well as the amount of aging permits us to control the particle size
of the aerosol.
R. K. Wolff; We have no problem in varying the concentration and keeping the
particle size constant. We have generated 0.4-ym particles at concentrations
ranging from 100 ug/m3 to 100 mg/m3.
Question; What is the droplet concentration at 100 mg/m3?
R. K. Wolff; The concentration is ~5 x 106 particles/cm2—the coagulation
limit. I'll allude to this a little later.
Question; On all of your function measurements in the guinea pig, were you
able to precharacterize your responses (i.e., determine preexposure factors
relating to individual animal sensitivity to H2SO4)?
S. A. Silbaugh; We did some work with that. In one study, we had two
exposure levels that resulted in something like 40/60% responders/
nonresponders. We took those two exposure groups and looked for preexposure
characteristics that would tend to segregate animals. It does appear that the
responsive animals tend to have higher resistance values and lower compliance
values prior to exposure. There also appear to be differences in the
relationships between, for instance, transpulmonary pressure and resistance.
So there do appear to be detectable differences that are related to whether an
animal tends to be responsive or nonresponsive.
In an article published a few years ago, Dr. Mary Amdur also reported
something like this. Amdur also observed that animals with higher resistance
values tended to be more reactive when exposed to various aerosols.
251
-------
25. LUNG FUNCTION LOSS FOLLOWING INHALATION OF ACID SULFATES _ Silbaugh
Comment; We're doing K^SC^ aerosol studies in humans and we use the sulfur
trioxide system of aerosol generation. We've found that [inaudible] and
humidity in our chamber determine the particle sizes; we're running 0.1 to
0.3. The concentration we're using is 100 mg/m3.
In your studies of ^804 alone, you used 10 to 40 mg/m3. Have you
decided on the concentrations to be used in the fly ash/H2SO4 studies?
S. A. Silbaugh; We have not yet decided on the H2SO4 concentrations, but
we're thinking along the lines of 0.5 to 1 mg/m3 for the low level and
somewhere between 1 and 5 mg/m3 for the higher level. We can only pick one
level for the fly ash; we're thinking of 5 mg/m3.
Question; Have you permitted the aerosols of acid and fly ash to age
together, to examine any possible interactions?
S. A. Silbaugh; No, this would be a preliminary part of the study. The main
study would not start until later in the fiscal year, and we have quite a bit
of developmental work yet to do. That would be one part of the developmental
work.
Question; Regarding the mechanics of the exposure facility, what provisions
would be required? Have you any way to pool the two as a single atmosphere,
or to provide a means for the aging to occur? Do you think there is any
advantage to it?
S. A. Silbaugh; That is not the realm of my expertise. Dr. Wolff, would you
care to comment?
R. K. Wolff; It would depend on how much aging you wanted. We could fairly
easily go to a few minutes of aging in a relatively large chamber. To go
beyond that would involve considerable space and expense. It could be done if
it were deemed worthwhile.
Question; What plans have you for varying humidity?
R. K. Wolff; Previously we varied humidity from 40% and 80%. In the planned
studies we will probably use 40 to 50% relative humidity. We have not seen
striking relative humidity effects. That is another issue we can still
address .
M. J. Wi ester; The data you showed did not indicate very much of an increase
in hi stamina response up to very high concentrations of N(>2 —
S. A. Silbaugh; Are you referring to both N<>2 and ^804?
M. J. Wiester: I'm not sure.
252
-------
25. LUNG FUNCTION LOSS FOLLOWING INHALATION OF ACID SOLFATSS Silbaugh
S. A. Silbaugh; The NO2 response is quite variable. We're still looking at
different ways of analyzing this. For instance, if you consider a particular
animal to be an outlier, it may be that a straight line relationship is not
the best relationship: it may be that with more animals at higher
concentrations we'd see another kind of response. When NO2 concentrations are
expressed as logarithms, the straight line fit does become slightly better.
M. J. Wiester; What I'm saying is that you don't increase your sensitivity of
the response at the ambient level—
S. A. Silbaugh; This is what I have outlined for a future study. We need to
expose more animals at levels lower than 7 ppm to really get any information
regarding that region. The only area where we can say for sure that there is
a response is the region of 40 ppm or so. With more animals we should be able
to find out.
M. J. Wiester; [inaudible]
S. A. Silbaugh; We found no correlation of H2SO4 concentration and histamine
sensitivity for the nonresponsive animals. There was a trend at the 2-h point
for histamine sensitivity to decrease with increasing H2SO4 concentration.
Dr. Wolff may be able to shed some light on the implications of these results
for future studies. The indications that 112804 may stimulate mucous
production raise the possibility of a mechanism by which increasing 1^804
concentration can actually desensitize animals.
D. E. Gardner; In your NO2 studies, did some of the animals die?
S. A. Silbaugh; No. We obtained our animals from Charles River in six
shipment groups over the course of the study. The shipment groups reported
here all responded fairly uniformly. One shipment group (five animals) had no
response at concentrations of up to ~120 ppm. We contacted the breeder but
could not detect any preshipment differences. Our guinea pigs may actually be
less sensitive than rats to NO2«
M. Goldman; The way the data are plotted on a log scale, aren't you really
looking at a two-phase reaction? Anything between 50% and 500% is normal, and
then you have a few there—
S. A. Silbaugh; Right.
M. Goldman; Drawing a line with regard to concentrations, it seems to me that
you have a uniform response with sensitive animals [inaudible].
S. A. Silbaugh; With NO2?
M. Goldman; Right. In other words, anything over -20 gave about the same
response.
253
-------
25. LUNG FUNCTION LOSS FOLLOWING INHALATION OF ACID SULFATES Silbaugh
[Simultaneous discussion]
M. Goldman; If you try to scale it, you pick up the same data linearly and
you have what's here.
S. A. Silbaugh; If these points are omitted, I think what you're saying is
true.
M. Goldman; I was looking at all of the studies in which you're using this
end point. It seems to me that you've just got one group of sensitive
animals. The other groups are not sensitive. There1s no concentration
effect.
T. Crocker; I'm struck by the way in which you're using the relationship of
NC>2 to the histamine responsiveness. Heretofore, the histamine response has
been used to assist in identifying very low level effects of NC>2; here you're
getting a histamine effect and then trying to poison it with a very high level
of NO2- I don't quite understand the rationale for that approach, but it
doesn* t seem to me to improve our understanding of the responsiveness of the
respiratory tract at very limited concentrations.
S. A. Silbaugh; We plan to do those exposures at lower levels. Our histamine
challenge procedure is a recent development, the initial purpose of which was
to simply quantitate a dose-response relationship for NOj. We wanted to check
our system, you might say. These exposures have served several purposes.
First, we did find some information concerning the dose-response relationship.
We may be "poisoning it" (as you say) at higher levels, but we did want to see
responses in the initial exposures. We planned to go down from there.
We agree that the low concentration range is the important range to
study. One thing we did obtain is an indication of a wide variation in
individual animal responses. One is thus more encouraged to go back to lower
levels and do a large number of animals. We may see no responses in quite a
few animals, but there may be animals that are sensitized at very low levels.
Starting out with the lower levels might have caused us to become discouraged
early, because we wanted to see an effect in this system.
Also, we didn't know what to expect with the H2SO4 exposure, and we felt
that the NO2 exposures served as a positive control since others have
demonstrated a sensitization to airway constricting agents with exposure to
NO2-
T. Crocker; Yes, but you didn't use NO2 in the range of 1 to 3 ppm, which is
near the limit of that gas' effectiveness in producing sensitization and where
histamine makes the measurement of that sensitization more relevant (or at
least more detectable, I should say). The issue, really, is that if you go to
very high concentrations of NO2, you may be poisoning the very receptor
capability of the system that subsequently also responds to histamine. I'm
concerned that we may be overusing the toxic power of NO2 when what we're
254
-------
25. LUNG FUNCTION LOSS FOLLOWING INHALATION OF ACID SULFATES Silbaugh
really looking for is the ability to identify an increase in sensitivity.
[inaudible] N<>2 in the concentrations you're dealing with would be expected
to wipe out a large part of the cell population, especially near the alveoli/
the [inaudible] of the alveoli junction. N(>2 in these concentrations would
also affect cells elsewhere in the respiratory tract, and might very well
eliminate, in fact, the animals that respond.
S. A. Silbaugh; I see your point.
D. E. Gardner: I think Dr. Crocker's comments are very valid and pertinent,
especially in regard to the discussion yesterday [Chapter 6 of this volume] on
the relevance of high doses and how they can perhaps be misleading when you're
going to look at low levels of concentration.
255
-------
26. LONG CLEARANCE MECHANISMS FOLLOWING INHALATION OF ACID SOLFATES
Ronald K. Wolff
Inhalation Toxicology Research Institute
Lovelace Biomedical and Environmental Research Institute
Post Office Box 5890
Albuquerque, NM 87115
PROBLEM
Sulfur oxidation products result from the burning of fossil fuels and are
widely present in urban atmospheres. Acid sulfates inhaled in the urban
atmosphere may have deleterious health effects, and definitive information is
needed to set air quality standards. Sulfuric acid (H2SO4) mist may be the
most irritating of these products, at least with respect to pulmonary function
effects (Amdur 1971; Amdur et al. 1975); thus, the study of potential health
effects is of considerable interest.
The mucoclliary clearance system provides a sensitive indicator of lung
injury following exposure to irritant gases and aerosols. Alterations in
mucociliary clearance have been observed in man, donkeys, and dogs for H2SO4
mist exposures at industrial threshold limit values and below (Leikauf et al.
1979; Schlesinger et al. 1978; Wolff et al. 1978, 1979a, 1979b). Impairments
in this important lung defense mechanism may contribute to the development of
256
-------
26. LUNG CLEARANCE FOLLOWING INHALATION OF ACID SULFATES Wolff
chronic lung disease or the exacerbation of existing disease. Impairment of
mucous clearance might result in increased residence times of toxic materials
in the lung, thereby increasing susceptibility to bacterial infection.
Inhibition of mucous clearance could also lead to the plugging of smaller
airways, followed by atelectasis and a chain of events contributing to chronic
obstructive pulmonary disease.
The project reported here explored dose-response relationships between
impairment of tracheal mucous clearance and acid sulfate concentrations in
acute and chronic exposures of dogs, rats, and guinea pigs. Acute studies
were carried out in dogs for reasons of anatomic similarity to humans. Other,
smaller animals were used in efforts to identify a chronic animal model that
best mimics the human response.
APPROACH
Methods have been developed to measure tracheal mucous clearance in the
awake state in dogs, rats, guinea pigs, and rabbits. Microliter quantities of
radiolabeled material are instilled in the tracheas of these animals using
halothane anesthesia. The animals are allowed to regain consciousness in
restrainers and the movement of labeled material is measured using external
detection (gamma camera or slit-collimated Nal scanner).
257
-------
26. LUNG CLEARANCE FOLLOWING INHALATION OF ACID SULFATES Wolff
Dogs
Eight beagle dogs were exposed for 1 h to concentrations of 1.0 and 0.5
mg/m3 H2S04 mist. The particles were 0.9 pm in mass median aerodynamic
diameter (MMAD) with a geometric standard deviation (Og) of 1.4 as measured by
cascade impaction. Temperature was 74°F and relative humidity was 80%.
Tracheal mucous velocities were measured 1 week prior to exposure and also 0.5
h, Id, and 1 week after exposure. The method was to anesthetize each dog
with halothane, insert a fiber-optic bronchoscope into the trachea, turn off
the anesthetic, and deposit a 10-yl droplet containing ~20 yCi of 99mTc
macroaggregated albumin (MAA) as the dog started to regain consciousness.
Subsequently, gamma camera scintiphotos were taken at 1-min intervals for 25
min in the awake dog to measure the velocity of the labeled material moving up
the trachea. Measurements were made of discrete spot velocities and of the
velocity of the leading edge of moving activity.
For the 1.0-mg/m3 exposures, tracheal mucous velocities were
significantly depressed after 0.5 h (26% reduction, p < 0.05), 1 d (40%, p <
0.01), and even after 1 week (30%, p < 0.05). Velocities returned to the
control range after 5 weeks. Also, a much greater proportion of material
remained at the initial deposition site 1 week after exposure than in the
control preexposure experiments (47% vs. 14%).
The results for the 0.5-mg/m3 exposures were somewhat different.
Nonsignificant increases in clearance were seen after 0.5 h (35%) and 1 d
258
-------
26. LUNG CLEARANCE FOLLOWING INHALATION OF ACID SULFATES Wolff
(8%). However, a statistically significant depression in clearance was
observed 1 week after exposure (35%, p < 0.05). These observed abnormalities
indicate impairment of an important lung defense mechanism for a prolonged
period following acute exposures to relatively low levels of H2S04 mist.
Subsequent exposures of the same 8 dogs to an H2S(>4 aerosol of smaller
size (0.3 ym MMAD) at levels of 1 mg/m3 and even 5 mg/m3 showed no significant
effects on mucous clearance. Sham experiments were carried out under the same
exposure and measurement schedule described above. Velocities were very
consistent; there were no significant differences at any time.
Thus, despite the relatively small difference in the size of these two
aerosols, a wide disparity in effects was observed. The same pattern was
observed in acute toxicity studies using guinea pigs: 0.8-0.9 ym MMAD aerosol
was estimated to be 5 times more toxic than 0.3-ym aerosol. The studies
reported in this paper indicated a difference of at least a factor of 5, since
no effect was observed at 5 mg/m3 for the 0.3-ym aerosol compared to the
decided effects at 1.0 mg/m3 for the 0.9-ym aerosol.
Our accumulated data indicate that upper airway reflex mediated
bronchoconstriction is the predominant mechanism of action of 112804 mist.
Therefore, if more material is deposited in upper airways, a greater effect
may be elicited. Deposition studies carried out at the Inhalation Toxicology
Research Institute confirm that there is greater upper airway deposition of
0.8-0.9 ym MMAD H2SO4 aerosols compared to 0.3-0.4 ym MMAD aerosols.
259
-------
26. LUNG CLEARANCE FOLLOWING INHALATION OF ACID SULFATES Wolff
Rats
Groups of rats were exposed to 1, 50, and 100 mg/m3 H2S04 for 0.5 h and
to 1, 10, and 100 rag/m3 H2S04 for 6 h.
Trachael clearance was measured in four males and four females in each
exposure group 1 week prior to exposure and 1 d, 1 week, and 3 weeks following
exposure. (The 3-week measurement was omitted for the 0.5-h exposure.) Rats
were anesthetized with 5% halothane to a level of deep anesthesia sufficient
to allow intratracheal instillation of 99mTc MAA. The rats were placed in
plastic restraining tubes, allowed to regain consciousness, and driven past a
slit-collimated Nal detector to produce a profile scan of the labeled material
in the trachea. Profile scans were taken atO, 2, 4, 6, 8, 10, 15, 20, 30,
40, 50, and 60 min after initial instillation. The percentage of activity
remaining at the instillation site after 1 h was the principal measure of
clearance. Tracheal mucous velocities were also determined by measuring the
movement of the leading edge of material. The two types of measurements
correlated well but the retention measurements were more consistent.
Two-tailed paired t-tests were used to detect statistically significant
changes in clearance.
Speeding in clearance was observed following the 0.5-h exposure. No
changes were observed for the control group, but dose-related changes were
observed for the exposed groups. Significant speeding of clearance was
observed after both 1 d and 1 week following the 100-jag/m3 exposure. For the
260
-------
26. LUNG CLEARANCE FOLLOWING INHALATION OF ACID SOLFATES Wolff
50-mg/m3 exposure, a slight depression of clearance was observed at 1 d, while
moderate speeding was observed at 1 week. For the 10-mg/m3 exposure, a
speeding in clearance was observed only at 1 week after exposure.
Following 6-h exposures, dose-related speeding in clearance was again
observed, with the most striking results at the highest exposure level (100
mg/m3). Clearance was enhanced for the 100-mg/m3 exposure at 1 and 3 weeks
following exposure; for the 10-mg/m3 exposure at 1 d; and for the 1-mg/m3
exposure at 3 weeks. No significant changes were seen in the control group,
and clearance results were highly reproducible. Clearance effects were
consistent for the two sexes except at 1 d after the 100-mg/m3 exposure. At
that time speeding was observed in females but not in males.
Biochemical response in the lavage fluid and in the lung tissue was
minimal. In lavage fluid, the only detectable response to any level of
exposure at 1 d was an elevation in sialic acid, a marker for acid mucous
glycoproteins. As for clearance, this elevation occurred at the 100-mg/m3
level in females only. Sialic acid levels were always increased either
preceding or in coincidence with increases in mucous clearance. Scanning
electron micrographs also indicated increased mucous secretions in the
trachea.
Speeding in mucous clearance has been observed in dogs, man, and donkeys
but only during or soon after low-level exposure «1 mg/m3) (Leikauf et al.
1979; Schlesinger et al. 1978; Wolff et al. 1978, 1979a, 1979b). The speeding
261
-------
26. LUNG CLEARANCE FOLLOWING INHALATION OF ACID SULFATES
Wolff
in clearance that was observed in rats after both 0.5- and 6-h exposures to
H2SO4 mist occurred at levels of up to 100 mg/m3—100 times the industrial
threshold limit value. This speeding in clearance was somewhat surprising,
and the reasons for it are not clear. Figure 26-1 conceptualizes the observed
disparities between the rat and the other extensively studied species.
Speeding was observed in rats at all levels; in the other species, however,
clearance was generally increased at low levels but depressed at higher levels
(>1 mg/m3). These results are quoted for aerosols in a similar size range
(0.5-0.9 ym MMAD).
200
IU
>
CO
D
O
U
RAT
I
1 10
H2 SO4 CONCENTRATION (mg/m3)
100
Figure 26-1. Mucous velocity by species following exposure to 112804.
262
-------
26. LUNG CLEARANCE FOLLOWING INHALATION OF ACID SULFATES Wolff
PLANNED RESEARCH
Results from the rat model do not appear to agree well with results from
human or dog experiments. Therefore, the rat may be a poor model for chronic
exposure studies. However, the guinea pig has been demonstrated to show
pulmonary function responses to ^804 similar to those of humans.
Accordingly, we have initiated mucous clearance studies with guinea pigs.
Recently obtained base-line data show mucous clearance patterns similar to
those in rats and rabbits. In these studies, tracheostomies were used to
deposit the labeled markers. We are starting to investigate clearance in
guinea pigs using instillation via the intratracheal route. This has been
difficult because of the guinea pig's sensitivity to laryngospasm. However,
this problem has been solved and studies are about to begin. The advantage of
intratracheal instillation is that animals can be used as their own controls
(as demonstrated in the previous dog and rat studies).
Acute Exposures
For guinea pigs, we plan 1-h exposures to ^804 at levels of <1 mg/m3
with particle sizes of 0.3-0.4 ym and 0.8-0.9 ym MMAD. Following completion
of these studies, we plan 1-h exposures to (NH4)2SO4 and NH4HS04 at levels of
<1 mg/m3 and particle sizes of 0.3-0.4 ym MMAD. Also, 1-h exposures to 5
mg/m3 fly ash and 5 mg/m3 fly ash + 1 mg/m3 H2SO4 will be carried out. These
studies will be a prelude to chronic exposures. For dogs, we plan 1-h
exposures to (1*114)2804 and NH4HS04 at <1 mg/m3.
263
-------
26. LUNG CLEARANCE FOLLOWING INHALATION OF ACID SULFATES Wolff
For all acute exposures, if no effects are initially noted at 1 mg/ra3, we
will conduct more extended or repeated exposures at this level.
Chronic Exposures
We plan chronic exposures to H2SO4 and H2S(>4 + fly ash according to the
schedule outlined under project RPIS Number 4054, Acute-Chronic Loss of Lung
Function Following Inhalation of Acid Sulfates. Mucociliary clearance
measurements performed on the same schedule will serve as pulmonary function
measurements in normal and elastase-treated guinea pigs.
ACKNOWLEDGMENTS
This research was supported in part by the U.S. Environmental Protection
Agency via Interagency Agreement Number EPA-IAG-D5-E61 under U.S. Department
of Energy (DOE) Contract Number EY-76-C-04-1013 and in part by the U.S. DOE
under Contract Number EY-76-C-04-1013 and conducted in facilities fully
accredited by the American Association for Accreditation of Laboratory Animal
Care.
REFERENCES
Amdur, M. O. 1971. Aerosols formed by oxidation of sulfur dioxide—Review of
their toxicology. Arch. Environ. Health, 23:459-468.
Amdur, M. 0., J. Bayles, V. Urgo, M. Dubriel, and D. W. Underbill. 1975.
Respiratory response of guinea pigs to sulfuric acid and sulfate salts.
Presentation at: Symposium on Sulfur Pollution and Research Approaches.
Durham, North Carolina, Duke University Medical Center, May 27-28.
Leikauf, 6., D. B. Yeates, K. Wales, and M. Lippmann. 1979. Effect of
inhaled sulfuric acid mist on tracheobronchial mucociliary clearance and
264
-------
26. LUNG CLEARANCE FOLLOWING INHALATION OF ACID SULFATES Wolff
respiratory mechanics in healthy nonsmokers (abstract). Am. Rev. Resp.
Dis., 119:227.
Schlesinger, R. B., M. Lippmann, and R. E. Albert. 1978. Effects of
short-term exposures to sulfuric acid and ammonium sulfate aerosols upon
bronchial airway function in the donkey. J. Am. Assoc. Ind. Hyg.,
39:275-286.
Wolff, R. K., M. Dolovich, G. Obminski, and M. T. Newhouse. 1978. Effect of
TLV levels of SO2 and H2SO4 on bronchial clearance in exercising man.
Arch. Environ. Health, 33:24-32.
Wolff, R. K., B. A. Muggenburg, and S. A. Silbaugh. 1979a. Effect of
sulfuric acid mist on tracheal mucous clearance in awake beagle dogs
(abstract). Am. Rev. Resp. Dis., 119:243.
Wolff, R. K., S. A. Silbaugh, R. L. Carpenter, and J. L. Mauderly. 1979b.
Toxicity of 0.4 and 0.8 ym sulfuric acid aerosols in guinea pigs. J.
Toxicol. Environ. Health, 5:1037-1047.
WORKSHOP COMMENTARY
Comment; You examined the clearance of particles over a fairly short period.
In the first phase of clearance, you did have some residual particles; of
course, these disappear radioactively- Is anybody looking at the long-term
clearance of residual particles?
R. K. Wolff; Currently, a number of studies at the Inhalation Toxicology
Research Institute are investigating long-term clearance of particles. We may
do some longer-term clearance studies in connection with those outlined here.
265
-------
27. OVERVIEW OF CURRENT AND PLANNED RESEARCH
BY THE INHALATION TOXICOLOGY BRANCH
Donald E. Gardner
Health Effects Research Laboratory, MD-82
Office of Research and Development
U.S. Environmental Protection Agency
Research Triangle Park, NC 27711
INTRODUCTION
This report presents an overview of current and projected research on
oxidants by the Inhalation Toxicology Branch (Environmental Toxicology
Division, Health Effects Research Laboratory, Office of Research and
Development, U.S. Environmental Protection Agency, Research Triangle Park,
North Carolina). Elsewhere in this volume, Graham (Chapter 28) and Miller
(Chapter 29) discuss some of these studies in greater detail.
STAFF, FACILITIES, AND BUDGET
The Inhalation Toxicology Branch (ITB) is staffed by ~22 permanent,
full-time employees and ~13 part-time employees. The Branch is divided into
three sections: Pharmacology and Microbiology, Physiology, and Inhalation
Exposure and Dosimetry.
266
-------
27. INHALATION TOXICOLOGY BRANCH Gardner
Each Section employs individuals with a wide .variety of scientific
backgrounds. For example, in the Microbiology and Pharmacology Section there
are several microbiologists, pharmacologists, and a number of biologists;
there are also entomologists who are working in the area of unconventional or
biological pesticides (i.e., using viruses, bacteria, and protozoans as
pesticides). The Physiology Section includes individuals with expertise in
cardiovascular disease, physiology, biochemistry, and organic chemistry. The
Inhalation Exposure Section is composed of inhalation engineers, electronic
technicians, physical scientists, and a biostatistician.
In addition to these individuals who are directly employed by ITB, our
program is supported by ~22 in-house contractor personnel from Northrop
Services, Inc. - Environmental Sciences. The majority of these individuals
provide direct support for the inhalation exposure and engineering needs of
the Branch. They are responsible for maintaining the exposure facilities,
which includes generating and monitoring a large number of chemicals of
interest to EPA. Northrop also maintains individuals trained in the areas of
physiology, microbiology, and pharmacology who perform research in certain
areas of our base program.
Comparing the allocation of funds for 1979 and 1980 indicates how our
research program can shift from one year to the next. In fiscal year (FY)
1979, 41% of our total budget was devoted to the energy program, whereas in FY
1980 there was no money to either continue or start any studies in that area.
In FY 1980 there was a significant increase (5% to 21%) in criteria research
267
-------
27. INHALATION TOXICOLOGY BRANCH Gardner
fvoiding, and a large jump (19% to 51%) in non-criteria pollutant research. We
are also involved in one transportation program whose objective is to compare
the carcinogenicity of diesel exhaust and various diesel extracts with roofing
tar, cigarette smoke condensate, and coke oven emissions. This work is being
done under a grant and represents ~15% of our total budget. There is also a
small program (5%) in pesticides and toxic substances.
For FY 1980, ~35% of our total budget is allocated to in-house programs;
the balance (~65%) is funded extramurally under grant programs, cooperative
agreements, or contracts.
GOALS FOR CRITERIA POLLUTANT RESEARCH
In order to make our research most useful for the setting of criteria, we
incorporate certain broad objectives in our research planning. First, we're
interested in making animal data more relevant to man. Others have stressed
that "a mouse is not a man" (e.g.: Chapters 4, 5, and 7 of this volume).
What we want and need to do is to develop a scientifically sound data base
that can be used to develop modeling systems that will help us extrapolate
animal data to humans. For example: If a mouse receives a certain
concentration of a pollutant gas at its nose and subsequently shows a certain
health effect, what concentration would have to be at a man's nose to show a
similar effect, assuming the same kinds of target organs or target tissues
were affected? Such information would greatly improve the usefulness of
animal toxicology data.
268
-------
27. INHALATION TOXICOLOGY BRANCH Gardner
Secondly, we are also interested in confirming and understanding many of
the important toxicological studies that presently comprise the data base for
establishing a scientifically sound standard for ozone (03) and nitrogen
dioxide (NO2) and other criteria pollutants.
Our program is interested in interactions. It's time to get away from
studying single pollutants. Other chapters in this volume demonstrate that
there are already many studies examining the effects of such combinations as
03 with NO2; sulfuric acid (H2SO4) with 03; and H2SO4 with NO2. As time goes
on, we will expand these mixture studies to include other gaseous and
particulate pollutants (such as hydrocarbons, sulfur dioxide, etc.).
It is imperative that we continue to identify any new effects or end
points having potential utility in human studies. For example, during the
past few years, we have established a very comprehensive program specifically
devoted to the study and development of new models for measuring in small
animals various sensitive pulmonary functional parameters similar to those
measured in humans. Our program continues to seek noninvasive techniques that
will hopefully provide indicator systems useful to both the epidemiologists
and to the clinical investigators.
Lastly, we are interested in developing new animal models that mimic the
special human subpopulation that is most sensitive to environmental chemicals.
A number of new animal models of human diseases are now available for
application by toxicologists in environmental studies.
269
-------
27. INHALATION TOXICOLOGY BRANCH Gardner
POLLUTANTS STUDIED
To date we have investigated the effects of exposure to 03, NO2, and
combinations of 03 with NO2» with particulates, etc. Some work has been
conducted under contract to determine the health effects of exposure to
peroxyacetyl nitrate. We continue to study the effects of complex mixtures,
including the combination of 03, SC>2/ and trans-2-butene. Our research
program also considers pollutants that are beyond the scope of this volume
(e.g., pesticides, toxic substances, non-criteria pollutants).
MODEL SYSTEMS IN USE
Biological indicators of health effects include both interaction with
infectious microorganisms (e.g., bacteria, virus, Mycoplasma) as well as
specific alterations in a number of specific host defenses. Included here are
functional and morphological alterations in alveolar macrophages, various
humoral and cell-mediated immunities (including B and T cell transformation),
and antibody titers.
Other studies investigate the effects of various environmental chemicals
on upper respiratory clearance mechanisms. In this model system we examine
changes in ciliary function by measuring the beating rate and frequency, and
we attempt to relate these changes to histological abnormalities of the cilia.
270
-------
27. INHALATION TOXICOLOGY BRANCH Gardner
J. O'Neil, M. J. wiester, and J. A. Raub of our Branch have developed a
battery of pulmonary functional tests that should be very useful in measuring
various changes in respiratory rate and other functional parameters in small
animals.
With 03 and N(>2, current studies model the pulmonary deposition and the
fate and absorption of these gases in the lung. Although we are primarily
interested in the lung as the target site, we investigate other organ and
target tissues for any extrapulmonary effects that might result from inhaling
these environmental chemicals.
Animal models are employed to measure cardiovascular effects of 03, NO2/
and other compounds of interest to EPA. The end points include identification
of lipid profiles, EKG measurements, and determination of other physiological
parameters.
Another model focuses on pentobarbital-induced sleeping time. Early
results indicate a significant sex difference: females are much more
susceptible to the actions of NC>2 and 03 than males. Graham (Chapter 28 of
this volume) expands on these studies.
When indicated, we complement all our studies with examination of various
lung and serum biochemical end points and/or performance of necessary
hiatopathology.
271
-------
27. INHALATION TOXICOLOGY BRANCH Gardner
In addition to our in-house research efforts, we are collaborating with
investigators at the University of North Carolina in a study of how the
microorganism Mycoplasma pneumonias interacts with normal systems and of how
airborne pollutants may alter the microbial action upon the upper airway.
This model system is discussed further in the following section.
MODEL SYSTEMS UNDER DEVELOPMENT
Several model systems are presently under development. In our acute
infectivity system, bacteria are given after exposure to a gaseous pollutant.
In a normal animal or an animal that is not affected by the pollutant, the
bacteria are all rapidly removed or killed within 6-8 h; the animal's lungs
become normal again (i.e., free of the inhaled bacteria). This very sensitive
model is excellent for mimicing acute infectious diseases but fails to
resemble a chronic infectious disease.
A chronic respiratory model is being developed by E. Hu of our Branch.
This model uses the bacterium Mycoplasma pneumoniae. In these studies, the
animals exhibit a low-grade infection which mimics chronic bronchitis. We
plan to soon examine the effects of various pollutants using this chronic
respiratory disease model.
A viral model system similar to our acute bacterial infectivity system
but employing viruses instead of bacteria is being developed by M. J. Selgrade
of our Branch. This project investigates the effects of environmental
272
-------
27. INHALATION TOXICOLOGY BRANCH Gardner
chemicals on the reactivation of latent viruses and the interaction of viruses
and bacteria.
There is also interest in better defining the susceptible human
subpopulation. One study at Duke University will attempt to determine the
effects of pollutants on the young developing lung. This study will resemble
the work by Dungworth (Chapter 20 of this volume) but will involve exposures
to NO2 and to 03 for longer periods of time in hope of correlating the
morphological effects with various pulmonary functional changes. Under this
cooperative agreement study, we plan to perform detailed morphometrics as well
as identification of various biochemical changes in the lung.
There is also a project whose objective is to examine individuals with
glucose-6-phosphate dehydrogenase (G6PD) deficiency and to determine how they
respond to 03 assault. The investigators have bred G6PD-deficient mice and
sheep and are attempting to find out if they are more susceptible to oxidants
than normal animals.
An"animal model of emphysema is being developed. Animals which are
exposed to elastase develop emphysema-like lesions; our program is attempting
to correlate and quantitate the amount of emphysema using pulmonary functional
techniques. Elastase-exposed animals will be exposed to various pollutants
and their responses compared to proper controls. This model mimics an
individual in the population who has emphysema and is exposed to a pollutant.
273
-------
27. INHALATION TOXICOLOGY BRANCH Gardner
To complement this study, we plan to breed and expose a strain of mouse (the
blotchy mouse) that is genetically prone to develop emphysema.
Finally, there is interest in how diet deficiency may alter
susceptibility to 03 and to N(>2« The biological end points include various
barrier functions, the production of edema, the breaking of tight junctions in
the upper airway, and similar biochemical and morphological indicators of
damage.
STUDIES TO BE REPLICATED
There are some important studies that we plan to replicate and/or extend.
One of these is a study by Bartlett et al. (1974) which examined the effects
of oxidant pollutant (0.2 ppm) on the young developing lung. These authors
concluded that the developing lung is more susceptible to 03 than the adult
lung.
Using a very sensitive model, Sherwin (Chapter 36 of this volume)
detected edema in the lung at a very low level (0.4 ppm) of NO2« We plan to
extend and replicate this study.
Some years ago, 03 was found to accelerate aging (Stokinger 1957). There
is a need to repeat this study to better define those results. Unfortunately,
the funds are not available for FY 1980.
274
-------
27. INHALATION TOXICOLOGY BRANCH Gardner
Finally, it was recently reported that exposure to 0.2 ppm NC>2 causes a
decrease in metabolism of prostaglandin E (Menzel et al. 1976; Menzel 1980).
These results also demand replication to better define the mechanism of action
and possible consequences.
All of these animal studies represent important components of the
existing data base (and therefore of the criteria document). For that reason,
they have high priority for funding.
COLLABORATION WITH OTHER INVESTIGATORS
In addition to our in-house efforts, we participate in a considerable
amount of collaborative work with other investigators in the Research Triangle
area. Within EPA's Health Effects Research Laboratory, we have joined the
Experimental Biology Division (N. Chernoff) in a search for possible
teratogenic effects of NO2* A similar collaboration (with L. Reiter) seeks to
demonstrate behavioral effects of exposure to 03 and N02- With S. Sandhu
(Genetic Toxicology Division) we are searching for mutagenic effects of N©2
exposure.
There is also collaboration with the University of North Carolina, Duke
University, and North Carolina State University. Topics of these studies
include immunologic effects of 03 and N02» effects of oxidants on upper
respiratory tissue (scanning electron microscopy), and pharmacokinetic effects
of 03 on some xenobiotic agents.
275
-------
27. INHALATION TOXICOLOGY BRANCH Gardner
REFERENCES
Bartlett, D., Jr., C. S. Faulkner, and K. Cook. 1974. Effect of chronic
ozone exposure on lung elasticity in young rats. J. Appl. Physiol.,
37:92-96.
Menzel, D. B. 1980. Pharmacological mechanisms in the toxicity of nitrogen
dioxide and its relation to obstructive disease. In: Nitrogen Oxides
and Their Effects on Health (Lee, S. D., ed.). Ann Arbor Science, Ann
Arbor, Michigan.
Menzel, D. B., W. G. Anderson, and M. B. Abou-Donia. 1976. Ozone exposure
modifies prostaglandin biosynthesis in perfused lungs. Res. Comm. Chera.
Pathol. Pharmacol., 15:135-147.
Stokinger, H. E. 1957. Evaluation of the acute hazards of 03 and oxides of
nitrogen. Am. Med. Assoc. Arch. Ind. Health, 15:181.
276
-------
28. SOME SPECIFIC STUDIES PLANNED BY THE INHALATION TOXICOLOGY BRANCH
Judith A. Graham
Health Effects Research Laboratory, MD-82
Office of Research and Development
U.S. Environmental Protection Agency
Research Triangle Park, NC 27711
INTRODUCTION
This report discusses in more detail some of the projects overviewed by
Gardner (Chapter 27 of this volume). All are projects of the Inhalation
Toxicology Branch (Environmental Toxicology Division, Health Effects Research
Laboratory, Office of Research and Development, U.S. Environmental Protection
Agency, Research Triangle Park, North Carolina). It is not the intent of this
report to present data, but rather to indicate the focus and scope of these
studies.
CHRONIC NITROGEN DIOXIDE EXPOSURE USING MOUSE INFECTIVITY MODEL
One major planned study is a chronic nitrogen dioxide (NO2> exposure. We
recently built an in-house exposure facility suitable for NO2 (or any gas).
Currently, we are conducting some shorter-term studies to ensure that the
system will work when we actualy begin the chronic exposure.
277
-------
28. STUDIES PLANNED BY THE INHALATION TOXICOLOGY BRANCH Graham
The driving force behind this study is some previous work with the mouse
infectivity model. In this model system, a mouse is exposed to a pollutant
and then challenged with an aerosol of live Streptococcus pyogenes. Mortality
is followed for a holding period in clean air. This model system is one of
the most sensitive model systems available for ozone (03) and NC>2 animal
inhalation toxicology.
Previous NC>2 studies examined concentrations no smaller than 0.5 ppm. At
0.5 ppm, 90 d of exposure were required to observe effects. We are, of
course, interested in investigating concentrations that are closer to ambient
levels. Therefore, it will be necessary to extend the time of exposure.
Prior work also showed that, at least for NO2 and the infectivity model,
accurate interpretation of results requires that the chamber exposure patterns
match ambient exposure patterns. Therefore, we will perform more research
using a base-line NC>2 concentration upon which peak N(>2 concentrations are
superimposed. In the planned study, we will expose the animals to 0.2 ppm
continuously (except for short periods of cleaning the chamber). Then the
animals will receive a peak exposure of 0.8 ppm, 5 d/week, 1 h/d in the
morning and 1 h/d in the afternoon.
A number of parameters will be examined. Besides the Streptococcal
infectivity model, several mouse pulmonary function tests are now available
and will be applied.
278
-------
28. STUDIES PLANNED BY THE INHALATION TOXICOLOGY BRANCH Graham
PENTOBARBITAL-INDUCED SLEEPING TIME IN THE MOUSE
Although over the years it has been assumed that gases affect primarily
the lung, there is increasing interest in the extrapulmonary effects of gases
and other pollutants. There have been recent indications that extrapulmonary
effects occur upon exposure to oxidizing pollutants. One of the model systems
used in our laboratory is pentobarbital-induced sleeping time in the mouse.
In this particular model system, the animal is exposed to a pollutant,
pentobarbital is injected, and sleeping time is measured.
In our early studies, concentrations as low as 0.1 ppm 03 and 0.25 ppm
NO2 caused significant increases in pentobarbital-induced sleeping time.
Further evaluation of this model led to some interesting results. For
example, there appeared to be a difference between the toxicity of NO2 and O3.
As the concentration of 03 was decreased (e.g., from 1 ppm to 0.1 ppm), an
increasing number of days exposure (at 3 h/d) was required to observe
increased pentobarbital-induced sleeping time. A single 3-h exposure to 1 ppm
03 produced the effect, whereas at 0.1 ppm it was necessary to repeat the
exposure (3 h/d) for 16 d to observe about the same level of effect. However,
with NO2 the effect could be seen after the first day of exposure: a 3-h
exposure to 5 ppm NO2 or to 0.5 ppm N02 demonstrated the effect. Thus, in
this model system (at least for acute exposures) N02 appears to be more toxic
than 03. Of course, this conclusion differs from the vast body of
toxicological literature on the two gases. Our results suggest a different
279
-------
28. STUDIES PLANNED BY THE INHALATION TOXICOLOGY BRANCH Graham
mechanism of action between 03 and NC>2, and this is something that we want to
follow up.
To make sure that our results were not unique to pentobarbital, further
examinations employed a variety of drugs (hexobarbital, thiopental, and
zoxazolamine). We found that our results were indeed not unique to
pentobarbital.
We were also interested to determine the biological significance of these
effects. With regard to criteria documents, the concentration is important
but the effect is of prime importance. Knowledge of the significance of that
effect (e.g.: Are human beings at risk if this effect is observed in animals,
or even in humans?) is a very, very important question. This model system is
particularly useful because effects are observed at lower concentrations.
Disappearance curves for pentobarbital in the blood were performed to
determine (1) if there were effects on either phase 1 or phase 2 clearance,
and (2) if the effects were primarily on hepatic metabolism. Generally
speaking, changes in pentobarbital-induced sleeping time were attributed to
quantitative changes in xenobiotic metabolism.
Once we complete the kinetic studies described above, we plan to employ
another model system has already been developed. This model system,
antipyrine kinetics, is based on the fact that antipyrine is completely
metabolized by the cytochrome P-450 system. In humans, antipyrine is an
280
-------
28. STUDIES PLANNED BY THE INHALATION TOXICOLOGY BRANCH Graham
indicator of P-450 metabolism by the liver. We plan to test this model in
animals and perhaps use the results to provide some guidance to workers
engaged in human studies. Perhaps similar studies can eventually be performed
in humans and then related to the whole array of effects that can be observed
using the more invasive techniques with animals. Besides looking at some of
these whole-animal models, we want to examine in more detail some of the
events that may be occurring in the liver with the cytochrome P-450 system.
Thus, we will also run a rather typical evaluation of enzyme activities
(O-demethylase, N-demethylase, and hydroxylase).
Sex sensitivity studies were part of our initial evaluation of this model
system. (Given that EPA is charged with protecting the most sensitive segment
of the population, sex sensitivity is an appropriate focus of research.) We
began with female mice and the effect was observed. Then came the question:
Is there a sex sensitivity, or does this effect only occur in the mouse? (An
effect that occurs only in the mouse might not be extrapolatable to man.) We
found that the effect occurs in the rat and hamster as well as the mouse; in
all these species, the female was more susceptible. There are known sex
differences in P-450 metabolism in the rat, but these major differences do not
apply to other species. Thus, the sex difference we observed represents more
than classical differences in P-450 metabolism.
281
-------
28. STUDIES PLANNED BY THE INHALATION TOXICOLOGY BRANCH Graham
SENSITIVITY TO BRONCHOCONSTRICTING AGENTS
Our program includes & large battery of studies within the category of
lung metabolism and pharmacological regulation of the lung. The major
stimulus behind this body of research is the increase in airway resistance
observed in humans after exposure to O3 and NO2. This particular effect in
humans is one of the major reasons behind regulation of these pollutants, and
we want to better understand its mechanism of action.
Many current human clinical studies involve pharmacological perturbation
with carbachol, acetylcholine, or other parasympathomimetic agents to see if
the pollutant increases sensitivity to these bronchoconstrictors. Questions
have been raised as to the biological relevance of increased sensitivity to
bronchoconstrictors and as to how EPA should treat this information from a
regulatory standpoint. Our group is conducting some mechanistic studies that
are designed to yield an improved understanding of the toxicological events,
thereby giving a better understanding of the potential seriousness of these
events in man.
We are conducting a number of studies with 03 alone and with NO2 alone.
After we define the effects with these gases, we plan to expose the animals to
the same gases in combination. Early studies by Menzel (see Gardner, Chapter
27 of this volume) focused on HO2 effects on the metabolism of prostaglandins,
particularly PGE2« Concentrations of NO2 as low as 0.2 ppm (3 h) inhibited
metabolism of the bronchoconstrictor; the effect was maximal at 18 h post
282
-------
28. STUDIES PLANNED BY THE INHALATION TOXICOLOGY BRANCH Graham
exposure (i.e., there was a delayed effect). The study used an isolated
ventilated perfused lung (IVPL), so the effect observed was probably on the
epithelial cells of the capillaries in the IVPL preparation. We are
interested in expanding this study because the prostaglandins affect not only
airway tone but also vessel tone, and changes in these parameters might
influence not only airway resistance but also the relationship between
ventilation and perfusion in the lung. Also, there are a number of other
possible effects. Hopefully, expanding this study will lead to a better
understanding of the mechanism involved in airway resistance changes.
Other compounds (e.g., histamine and SRS-A) are also very involved in
bronchoconstriction, and we plan to measure those compounds as well. All are
related to arachidonic acid metabolism and products from the associated
reactions.
Also, the lung is the main site of the converting enzyme activity that
takes angiotensin 1 to angiotensin 2 (the most potent vasoconstrictor in the
body). We have accomplished some preliminary studies indicating possible
effects on these systems. We plan to extend these studies and see if changes
in the reflex system will be reflected in changes in cardiovascular function.
Planned studies that have not yet begun include an examination of cyclic
AMP and cyclic GNP; these are very important regulators of various aspects of
cell function. Also planned are studies of antioxidant metabolism. Most of
this work will be done by M. Mustafa (University of California - Los Angeles).
283
-------
28. STUDIES PLANNED BY THE INHALATION TOXICOLOGY BRANCH Graham
Many of Mustafa's studies will examine combinations of O3 and N02, something
in which we are particularly interested in light of his previous reports of
effects at concentrations as low as 0.1 ppm (and also at low concentrations of
NO2). We need to know more about combinations of effects and recovery from
those effects*
STUDIES TO IDENTIFY SUSCEPTIBLE POPULATIONS
With regard to what segments of the human population are more susceptible
and might need more protection, there are some obvious lines to be followed:
male vs. female, very young vs. young vs. older, and so on. These are very
large classifications. In spite of many years of toxicological studies with
03 and NO2, it still remains unclear as to whether the young are more or less
susceptible. Our group wants to improve the state of this knowledge; so far,
the only solid plans are for 03. However, there is a need for such work with
NO2, also.
The planned study will use rats. Animals will be exposed to 0.25 ppm,
0.12 ppm, and 0.08 ppm O3. A set of animals will be exposed from 0 to 6 weeks
of age (when alveolarization occurs). Also, a group of young adult animals
will be exposed. A control group of young adults will be exposed to 03 as
well as air.
Morphology and extensive morphometric procedures will be the prime
parameters used to evaluate age susceptibility. Pulmonary function will be
284
-------
28. STUDIES PLANNED BY THE INHALATION TOXICOLOGY BRANCH Graham
inserted into the system as soon as the animals are large enough to be
measured. It should be possible to do a reasonable battery of pulmonary
function studies when the animals reach ~20 g.
WORKSHOP COMMENTARY
R. K. Wolff; Did you mention that you were going to use 03, NO2» and some
inhaled particulate?
J. A. Graham; As part of our Branch's inhaled particle program, we're
interested in doing studies on the combined effects of particles plus gases.
We have not yet decided precisely what particles; the decision will be based
upon data from ongoing studies.
J. L. Whittenberger; The theme of this meeting is scientific research; a
secondary theme is relationship to standard-setting. Gardner [Chapter 27 of
this volume] gave as one of the major objectives of your Branch the
confirmation and extension of important findings. I agree that that's a very
important objective. It was mentioned that you'd like to confirm and extend
the Bartlett et al. (1974) study [see Gardner, Chapter 27 of this volume, for
details of reference]. The importance of that study was recognized in 1977,
if not earlier. Would you be willing to give a mini case history of why the
Bartlett study has not yet been confirmed and extended?
D. E. Gardner; My guess is that it has mainly been a matter of appropriate
funds and priority. In reviewing the documents, it becomes evident that this
is an important study when one considers the whole package of animal
toxicology programs. Now that we have some very sophisticated morphometric
techniques, we think we can not only reproduce the study but also extend it
and make it better.
J. L. Whittenberger; EPA didn't have to do that study with its own funds;
there's NIEHS "just down the road." Why didn't they repeat it?
J. A. Graham; Essentially it comes down to funds and also to the study's
relationship to criteria documents. When a criteria document is evaluated by
a large number of health scientists, the gaps really become immediate. That's
why it's a good exercise to read the criteria documents even if you're not
involved in the regulatory process: you see where the research is needed.
J. L. Whittenberger; When you say it's a matter of "funds," are you telling
me that somebody in Washington screwed up?
285
-------
28. STUDIES PLANNED BY THE INHALATION TOXICOLOGY BRANCH Graham
D. E. Gardner; We hope to correct this.
J. A. Graham; Dr. Gardner [Chapter 27 of this volume] discussed the
relatively low level of funding for oxidants research in fiscal year 1979 as
compared to fiscal year 1980. There is a major difference.
D. E. Gardner; If we had the money to fund the projects of every researcher
attending this workshop, all those projects would be very interesting. But
there is a limited budget, and with this limited budget we have to answer to
"the guy that's paying the bill." That's why we can only fund projects that
the customer says are most useful in the regulatory setting.
J. L. Whittenberger; So you're saying that the Program Office doesn't
consider this particular study to be very important?
J. A. Graham; No, indeed they do consider it to be important. There was
simply a financial limitation last year. I'm sure everybody in this room
could propose 10 excellent studies to add on to those outlined here by Dr.
Gardner and me. But we simply can't do it all at once. You're going to be
asking the same question about a different study four or five years from now.
D. E. Gardner; Part of the objective of this workshop plus the "straw man"
document [Miller et al. 1979; see Chapter 2 of this volume for details of
reference] is to bring some of these elements together. For example, Dr.
Dungworth's work [Chapter 20 of this volume] will fit in very nicely. Maybe
bringing all of these together will help solve some of our problems.
J. L. Whittenberger; Remember that 88% of environmental health research is
not funded by EPA.
Question; Can you provide a better explanation of how the decisions were
made? In answer to Dr. Whittenberger, you pointed out that one pressure on
you is the necessity to respond to your client. What client is interested in
the mechanistic work you were talking about; what client considers that top
priority?
J. A. Graham; The Oxidant Research Committee of the Office of Research and
Development.
D. E. Gardner; As described by Jones [Chapter 4 of this volume], all of our
work goes through research committees made up of clinicians, epidemiologists,
chemists, OAQPS staff, lawyers from the Office of General Counsel, and so on.
Every time we have a package that we want to "sell," we present it to these
committees. The committees then prioritize it. To the best of our ability,
we present what we think is appropriate.
Question; [inaudible]
286
-------
28. STUDIES PLANNED BY THE INHALATION TOXICOLOGY BRANCH Graham
J. A. Graham; Are you asking if a study done in rats can be extrapolated to
humans? I think Miller [Chapter 29 of this volume] addresses the quantitative
aspects of that question. As a generalization, we feel that if an effect can
be demonstrated in a number of animal species then it is quite possible that a
similar type of effect occurs in humans. We don't know yet at what
concentration that effect would occur in humans. But if young rats are more
sensitive than middle-aged rats, it's quite possible that a similar effect
occurs in humans. The results of such a study would provide guidance for
future epidemiologic studies or even human clinical studies (if they were
ethically possible).
Comment; How can one use cyclical exposures (especially with a base-line
exposure underneath that) as information in the standard-setting process?
J. A. Graham; EPA has two major information needs with respect to NO2? these
relate, respectively, to a short-term standard and a long-term standard.
Should each exist and, if so, at what concentration? The prime drive behind
this study is to answer questions relating to the short-term standard.
At present, EPA has an annual standard for NO2 which permits the
occurrence of cyclical peaks. The question is: Are cyclical peaks safe for
the public health, or do these commonly occurring peaks add an additional
burden for which control measures are needed? Our planned study will expose
animals to these patterns. One control will be the base-line concentration of
NO2 itself. Then we will have base line plus peak. Let's say that we add an
incremental 20-30% effect to the system. That might also apply to humans.
The ratio we have chosen for this particular chronic study (0.2:0.8) is the
typical urban ratio that occurs in the ambient environment.
D. E. Gardner; Those experimental concentrations may be slightly higher than
concentrations in the natural environment, but the ratio is consistent.
Question; When you go about setting a standard, then, you need some of the
information to help you set long-term standards and some for short-term
standards?
D. S. Gardner; Yes.
Comment; In several of the studies [described in this volume], the rat
appears to be an outlier, both with respect to other rodents and with respect
to other species. Why are you beginning another rat study?
J. A. Graham; The age susceptibility study will be done in rats because the
Bartlett et al. (1974) study was done in rats. We want to replicate and
extend that work. Also, a lot of base-line morphometric information is
available for rats, and it will probably be possible to share some controls
with other studies.
287
-------
28. STUDIES PLANNED BY THE INHALATION TOXICOLOGY BRANCH Graham
I think the whole question of species sensitivity to 03 and N©2 (or to
any pollutant, for that matter) is a very difficult issue. At this time, in
terms of pulmonary effects, I don't know which species is more or less
sensitive. At this workshop we heard an excellent presentation [Chapter 11 of
this volume] on some of the differences that exist, let's say, in circulating
anti-oxidant enzymes.
I think the rat has been used by Mustafa and Freeman. In terms of the
lung, Mustafa found effects at 0.1 ppm 03 for rats maintained on normal
Vitamin E levels. For rats with excess Vitamin E levels, the effect was not
seen until 0.2 ppm. So I think the rat is sensitive, at least in terms of
lung effect, and lung effects are primarily what we will examine in this
study.
Once again it's a question of limited funds. I agree that it would be
ideal to follow up these effects in several animal species.
288
-------
29. BIOMATHEMATICAL MODELING OF OXIDANT TOXICITY
Fred J- Miller
Health Effects Research Laboratory, MD-82
Office of Research and Development
U.S. Environmental Protection Agency
Research Triangle Park, NC 27711
INTRODUCTION
There is an acknowledged need to better assess target organ exposure
levels and the comparability of the animal toxicological data base for
standard-setting purposes. The current standards for oxidant gases are based
mainly upon human studies. All of the animal toxicological data (which are
95% of the data available) come in "through the back door": "These effects
are observed in humans and, by the way, some effects are also observed in
animals." In ethical consideration of what cannot be done experimentally with
humans, and in view of the need to understand mechanisms, researchers must
provide answers that will allow EPA to make better use of the animal
toxicological data base.
The problem is partially one of dosimetry: How do we correlate an effect
observed in an animal toxicological study with the exposure concentration
required to deliver the same amount of pollutant to the target site in man?
289
-------
29. BIOMATHBMATICAL MODELING OF OXIDANT TOXICITY Miller
We must improve our determinations of the quantities of pollutants that reach
given levels in the respiratory tract and that are associated with given
effects.
This report describes some preliminary attempts to biomathematically
model the deposition of ozone (03). Biomathematical modeling of deposition is
an expanding field, and one of increasing interest to EPA. At present, the
technique is not utilized to the greatest possible extent. Fuller utilization
will require more and better biochemical and physical data. Thus, another
goal of this report is to outline some areas in need of further research.
Many of the needed studies are not specific to 03, but would apply to modeling
the deposition of any gas.
NASAL PHARYNGEAL REMOVAL
It is exceedingly difficult if not impossible to measure the specific
amount of pollutant that reaches a given level in the lung. It is certainly
possible, however, to experimentally measure nasal pharyngeal removal. Nasal
pharyngeal removal of 03 in humans represents a major piece of missing data
that would enable us to put into perspective the animal studies and the
effects observed in human exposures. This information would permit various
manipulations to relate the likelihood of exposure relationships to
probability of effect.
290
-------
29. BIOMATHEMATICAL MODELING OF OXIDANT TOXICITY Miller
In one particular study (Miller et al. 1979) we looked at nasal
pharyngeal removal in rabbits. This was also done in guinea pigs. Both
species showed 50% removal of the pollutant over a concentration range of 0-2
ppm. These results for 03 satisfy such other models as that of Aharonson et
al. (1974) for penetration of soluble vapors into the nasal mucosa. In guinea
pigs, if the concentration exceeded 2 ppm, there was disproportionate removal
by the nasal pharyngeal cavity (an increase of 15%) and the similarity of
deposition was no longer maintained.
GAS TRANSPORT
The major components for analyzing gas transport are (1) convection,
which can be separated into advection and eddy dispersion, and (2) diffusion
in both the axial and radial directions, which is a function of the flows in
various areas in the lung. To accurately model the amount of pollutant
removed at each level of the lung also requires the incorporation of any
chemical reactions in addition to morphometric data. Fortunately, the
morphometric data are starting to become available.
MODELING PROCESS
These major components are brought together in a differential equation
that represents gas transport in the lung. The change in concentration per
unit time is a function of convection, axial and radial diffusion, and the
source terms. For a given pollutant, then, one needs to know these values in
291
-------
29. BIOMATHEMATICAL MODELING OF OXIDANT TOXICITY _ Miller
order to definitively solve the equation. As discussed below, we have
completed preliminary modeling of O3. However, much work remains to be done
to improve the data base.
MODELING ABSOLUTE DOSAGE: SOME PRELIMINARY RESULTS
In our initial 03 modeling, we used a reaction scheme that looks at the
attack of 03 on free fatty acid carbon-carbon double bonds. Also incorporated
were Mudd et al. (1969) data showing the reactions of 03 with the free amino
acids. Available data on lipids obtained from human tracheobronchial
secretions were also used.
After this initial work, we found data on the free amino acid
•*
concentrations in tracheobronchial secretions (Kohler et al. 1969). The
results discussed below were obtained in updated modeling which incorporated
these new data. This illustrates that we can always refine our estimates of
dose by incorporating additional biochemical data.
With respect to dose for each airway generation in the human lung, the
model predicts that the seventeenth generation—the first generation of
respiratory bronchioles—will receive the maximum dose of 03. For low
concentrations, the model predicts no penetration of 03 through the mucous
layer to conducting airway tissue; as the concentration increases, penetration
into these other generations occurs. However, for a given concentration it is
still the respiratory bronchioles that are hardest hit. Proceeding distally,
292
-------
29. BIOMATHEMATICAL MODELING OF OXIDANT TOXICITY Miller
the predicted dose decreases. In terms of the available histopathological
data for primates and rats (Dungworth et al. 1975; Stephens et al. 1973),
these results correlate well for main site of injury.
We also obtained modeling results for rabbits and guinea pigs. The
available morphometric data for these species are not exactly comparable to
humans, but there are similarities in terms of tracheobronchial bronchioles,
alveolar ducts, sacs, etc. The junction between the last bronchioles and the
alveolar ducts is again hardest hit in terms of 03 dose. Also, the conducting
airway dose again falls off as one proceeds distally. The shapes of the
curves are quite similar; there are differences in the absolute magnitude of
the predicted dose.
In guinea pigs, higher concentrations are required to predict dosages to
the conducting airway tissue. Once again, the terminal bronchiole, alveolar
duct, and respiratory bronchial areas are predicted to receive the maximum
dose.
We plotted maximal dose as a function of inhaled tracheal concentration
for man, rabbits, and guinea pigs. In our first plots, these curves started
to become linear above 100 yg/m^. When we incorporated the new data on
tracheobronchial free amino acids that are available to react with and deplete
O3 before it can strike conducting airway tissue, it became necessary to reach
~200 ug/m3 to observe a linear relationship between the respiratory
bronchiolar dose and the inhaled tracheal concentration.
293
-------
29. BIOMATHEMATICAL MODELING OF OXIDANT TOXICITY
By accounting for nasopharyngeal removal, this type of information
enables us to put into perspective the exposure concentrations and effects
observed in animal studies. It becomes apparent, for example, that a lower
tracheal concentration is required to deliver a given dose in man as compared
to the rabbit or guinea pig. Thus, our modeling data are useful in planning
experiments to fill gaps in the existing data base. (The 03 criteria document
probably reflects the most extensive toxicological data base for any
pollutant, yet many gaps remain. Very few studies have looked at a
commonality of end points in several species in terms of dose-response curves,
etc.)
Many human data have been obtained in studies in which subjects exercised
while being exposed to 03 or nitrogen dioxide (NO2) (for example: Bates et
al. 1972; Hackney et al. 1975). Evaluating these studies with our model shows
that, as tidal volume increases, the model predicts a tremendous increase in
uptake by respiratory airway tissue. Due to the transit times involved, a
relatively constant amount is removed by conducting airway tissue and a
decreasing amount is depleted by the mucous. Assuming 750 and 200 yg/m-* as
the maximum and minimum concentrations, respectively, we can see the vast
importance of exercise level on the resulting dose. In many of the human
studies, exercise levels that approximately doubled the tidal volume (~1000
ml) were used. Assuming a concentration range of from 200 to 750 yg/m-*, the
model prediction ranges from removal of 45% of the 03 to delivery of ~80% to
respiratory bronchioles ( compared to the base line) . Such a prediction
294
-------
29. BIOMATHEMATICAL MODELING OF OXIDANT TOXICITY Miller
correlates with increased pulmonary function effects that are indicative of
decreasing airways.
Because EPA is charged with protecting individuals at any level of
activity and under any exposure conditions, it would be very helpful to know
the exposure level required to yield 400 yg/m3 at the human trachea.
Unfortunately, that information is not currently available. At any rate, if
we compare tidal volumes of 500 ml and 1750 ml while the subjects are resting
at an 800-yg/m3 trachea level and exercising at a 400-yg/m3 level,
respectively, the model predicts the same maximal dose to the respiratory
bronchioles. There are millions of adults jogging and children playing; what
are the effects and what are the safe exposure levels in view of such activity
levels? There is certainly justification, in the Los Angeles Air Basin, for
air pollution alerts and for restrictions on children's play during those
episodes. Mathematical modeling of dosage permits us to make meaningful
comparisons of experimental results and to better calculate safe exposure
levels, thereby helping to determine whether such restrictions are justified
for other locations and environmental conditions.
QUANTITATIVE BIOCHEMICAL DATA ON THE MUCOUS LAYER: A MAJOR RESEARCH NEED
Current modeling efforts focus on the sensitivity of dose to mucous
production. We are in the process of establishing bounds for
295
-------
29. BIOMATHEMATICAL MODELING OF OXIDANT TOXICITY Miller
The biochemical data required to model a gas greatly influence the amount
of useful information that can be obtained. Fortunately, information is
available for O3; O3 is so reactive that one can consider instantaneous
irreversible reactions with many biological substances. Unfortunately, this
is not the case for NO2. Reactions with NO2 are much slower. With sulfur
dioxide (SO2) there are different reactions to consider, again requiring a lot
of information.
To accurately predict how much gas will reach the respiratory bronchial
and alveolar regions requires quantitative biochemical data on what is
available in the mucous to react with the gas. Such data would enable us to
put species differences into perspective in terms of absolute dosage. Most of
the available animal data, however, are qualitative in nature. For example,
much of the work on mucous secretions has involved staining techniques: we
know that something is present bat not in what quantity. Some quantitative
•*
human data are available (Lewis 1971; Kohler et al. 1969).
The thickness of the mucous layer is one parameter that can impact upon
the probability of finding effects. Luchtel (1976) completed a study in
rabbits. There is controversy about the thickness of the mucous layer in
rats. We also need better data on regional production and transport of
mucous.
Effective axial diffusivity relates to the effective diffusion down the
airway. Past modeling has incorporated only molecular diffusion; the correct
296
-------
29. BIOMATHEMATICAL MODELING OF OXIDANT TOXICITY Miller
term is in fact ~2000 times greater in the first few generations of the
conducting airways. Some work on a human model has been completed by Scherer
et al. (1975). For animals, the needed data are (once again) lacking. Also
lacking are the diffusion coefficients in mucous and surfactant. Thus far in
our modeling, we have been forced to substitute the value for water.
REFERENCES
Aharonson, E. F., H. Menkes, G. Gurtner, D. L. Swift, and D. F. Proctor.
1974. Effect of respiratory airflow rate on removal of soluble vapors by
the nose. J. Appl. Physiol., 37:654-657.
Bates, D. V., G. M. Bell, C. D. Burnham, M. Hazucha, J. Mantha, L. D.
Pengelly, and F. Silverman. 1972. Short-term effects of ozone on the
lung. J. Appl. Physiol., 32:176-181.
Dungworth, D. L., W. L. Castleman, C. K. Chow, P- W. Mellick, M. G. Mustafa,
B. Tarkington, and W. S. Tyler. 1975. Effects of ambient levels of
ozone on monkeys. Fed. Proc., 34:1970-1974.
Hackney, J. D., W. S. Linn, D. C. Law, S. K. Karuza, H. Greenberg, R. D.
Buckley, and E. E. Pedersen. 1975. Experimental studies on human health
effects of air pollutants: III. Two-hour exposure to ozone alone and in
combination with other pollutant gases. Arch. Environ. Health,
30:385-390.
Kohler, H., P. Klossek, H. Aurich, and H. Strobel. 1969. Der gehalt des
menschlichen bronchialskretes an freien aminosauren. In: Folia
Bronchologica in Zeitschrift fur Erkrankungen der Atmungsorgane, pp.
259-265.
Lewis, R. W. 1971. Lipid composition of human bronchial mucus. Lipids,
6:859-861.
Luchtel, D. L. 1976. Ultrastructural observations on the mucous layer in
pulmonary airways (abstract). J. Cell. Biol., 70:350a.
Miller, F. J., C. A. McNeal, J. M. Kirtz, D. E. Gardner, D. L. Coffin, and D.
B. Menzel. 1979. Nasopharyngeal removal of ozone in rabbits and guinea
pigs. Toxicol., 14:273-281.
297
-------
29. BIOMATHEMATICAL MODELING OF OXIDANT TOXICITY Miller
Mudd, J. B., R. Leavitt, A. Ongun, and T. T. McManus. 1969. Reaction of
ozone with amino acids and proteins. Atmos. Environ., 3:669-682.
Scherer, P. W., L. H. Shendalman, N. M. Green, and A. Bouhuys. 1975.
Measurement of axial diffusivities in a model of the bronchial airways.
J. Appl. Physiol., 38:719-723.
Stephens, R. J., M. F. Sloan, M. J. Evans, and G. Freeman. 1973. Early
response of the lung to low levels of ozone. Am. J. Pathol., 74:31-58.
WORKSHOP COMMENTARY
Question; Do you know the competence of your predictions in terms of error
limits? Are the estimates really so precise that you can tell the difference
between humans and animals?
F. J. Miller; We plot our results on a log scale. Thus, the increments
represent two- and three-fold differences.
Question; You obtained these results from how many studies, for example, on
humans?
F. J. Miller; Reaction components are readily available from two papers:
Lewis (1971) and Kohler et al. (1969). Again, this is a function of the
sensitivity analysis: If the input data are off by a factor of 25, we're in
trouble; if they're off by a factor of 10, we're okay. As better data become
available, reiteration of the analysis builds confidence in the model's
predictions.
Question; Do you intend, in your future work, to give some indication of
confidence in the lines you produce? Do you have plots for that?
F. J. Miller; Solving the differential equation does not yield the confidence
limits. One can change the variables in the model: frequency, tidal volume,
breath-holding, and so on. One can look at all those things. I have just
given you a smattering of them.
Question; Do you have an indication, all along the way, of confidence in what
you put in?
F. J. Miller; Well, the differential equation applies to gas transport.
That's not a unique equation but, rather, the equation that would apply to any
coordinate system. We make simplifying assumptions for analyzing the
tracheobronchial airways and the rest of the lung. There's nothing "magical"
about the first equation.
298
-------
29. BIOMATHEMATICAL MODELING OF OXIDANT TOXICITY Miller
Comment; Some of the problem here is in going from small animals to large
animals. Recognizing that effects in humans are ultimately what we're after,
are there (or would you like to see) some large animal studies that would make
some of these measurements experimentally and allow you to confirm your models
as you go along?
F. J. Miller; We need that kind of information. The limiting factor is the
morphometric data. I understand that some data on monkeys are starting to
become available. The usefulness of this modeling approach is limited by the
availability of morphometric data; beyond that, of course, there is no limit
in terms of its application.
J. L. Mauderly; I have three comments. First, with respect to your last
statement, I think there are some fairly detailed morphometric data available
that range from human down through rodents.
Secondly, I am very glad to see the growing recognition that it1 s not the
air concentration but the inhaled material that causes the effect. We
physiologists have been saying this for years. The message is finally coming
through that it's not ventilation but alveolar ventilation that's important,
and that's a simple relationship that we've been working with for decades.
Finally, Graham [Chapter 28 of this volume] alluded to some studies by
age and said that you would be mentioning more about this...
J. A. Graham; The main questions are whether we should use rats or another
species, and whether data on rat age susceptibility can be related to humans.
J. L. Mauderly; Yes, that's what I was looking for. My comment is that we
are learning more and more with respect to lung structure and function with
age in these different experimental animals, but right now we know almost
nothing about the senescent lung in most of these species. We do know quite a
bit about the dog in terms of morphology, morphometry, and pulmonary function.
Very recently, quite a lot of data have been produced on rodents (rats and
hamsters; particularly rats). What we see is not very encouraging. We know
almost nothing about primates.
Now, certainly we know quite a bit about primate morphology and
morphometry and pulmonary function. A group at Davis has done a lot of work
in this area, but almost nothing on the aged primate and on the time course of
senescence. This is a "glass house" situation because we're doing and
planning a lot of studies in rats.
I've said on a number of occasions that the reason for this is that the
life span of the average beagle dog is longer than the political life span of
the average politician. This is probably about the only reason we're working
on rats. Statisticians encourage us to use rats, too.
299
-------
29. BIOMATHEMATICAL MODELING OF OXIDANT TOXICITY Miller
We know at the present time that the pattern of senescence in lung
structure and function in the dog is very similar to that in man, even though
the lung type is different. We know just from preliminary data that the
pattern is markedly different in rodents (certainly in the rat): there is a
growing lung throughout the life span, and this must impact on carcinogenesis
and the development of chronic lung disease. And we're skirting around this
issue and ignoring it and proceeding as though rats are men—which we know
they're not.
We also know, from some recent data, that lung function may peak at
mid-age in the rat (~18 mo) . Even one group at 27 mo—very close to the
expected life span for the Fischer rat—had lung function that was still very
good with respect to young adult normals. In contrast, human lung function
peaks in young adults and begins a slow decline that is progressive and almost
linear throughout the adult life span. That's not the case in rodents; in
fact, we don't even know yet if there is a terminal decline in rodent lung
structure and function.
This is a comment rather than a question. Actually, it1s a sermon. And
I think it's something we've got to be very much aware of.
F. J. Miller; I wouldn't deny the need to use other animal species.
J. L. Mauderly; The point is that we don't have this information about the
dog (and perhaps the primate) . We must continue to fund at least some studies
in animals known to have similar patterns of change with age.
C. A. Heckman; Aging studies at Oak Ridge National Laboratory are examining,
in a preliminary way, changes in the aging of the rat lung. So far, the
computer data indicate that there is probably more cellular septum and greater
septum thickness in the rat lung. This may be because the physiological and
functional measures are not quite as sensitive in these smaller animals.
There may be some more morphological changes in larger animals that can be
picked up by functional measurements.
Also, we've found some differences in the way the epithelium from the
trachea grows jln vitro in aging rats as compared to younger rats. Let me
emphasize that this is all preliminary work.
300
-------
30. OVERVIEW OF CURRENT AND PLANNED RESEARCH BY THE HUMAN STUDIES DIVISION
Robert E. Lee
Health Effects Research Laboratory, MD-51
Office of Research and Development
U.S. Environmental Protection Agency
Research Triangle Park, NC 27711
INTRODUCTION
The Human Studies Division (Health Effects Research Laboratory, Office of
Research and Development, U.S. Environmental Protection Agency, Research
Triangle Park, North Carolina) is a new entity. A reorganization late in 1978
combined the then existing Population Studies Division, which focused on
epidemiologic studies, with the Clinical Studies Division, which was centered
mainly in our Human Research Facility in Chapel Hill, North Carolina. These
two organizations were combined to create the new Human Studies Division.
The main intent of the reorganization was to enable better coordination
and information exchange between our epidemiologic and clinical programs.
Both programs are now under a single Director. Hopefully, the net result of
the reorganization will be better interaction between these two very different
research fields.
301
-------
30. HUMAN STUDIES DIVISION
EPIDEMIOLOGIC PROGRAM
Previous EPA-supported oxidant studies have focused mainly on the Los
Angeles area. Studies have involved the measurement of irritation symptoms in
various cohorts, including athletes and schoolchildren.
For the past several years we have funded a study with Copley
International to quantify the acute health responses measured by respiratory
disease from various patterns of exposure to nitrogen dioxide, ozone, and
suspended sulfates and nitrates. The investigators have followed daily
reporting of acute respiratory disease symptoms for a period of 26 weeks in
300 families in each of 4 communities. This study should be completed in June
1980.
For the past three years we have funded an effort at Loma Linda
University (Loma Linda, California) to replicate a study of cytogenetic
effects of photochemical oxidants in college freshmen in the Los Angeles
Basin. This study was designed: (1) to determine if there is an increased
risk of chromosomal aberration in peripheral lymphocytes among young adults in
an area of high ambient photochemical air pollution, (2) to determine if such
aberrations persist upon emigration from the polluted area, and (3) to
determine if new immigrants develop the chromosomal aberrations. We expect a
report on this study in February 1980.
302
-------
30. HUMAN STUDIES DIVISION Lee
Chapman (Chapter 34 of this volume) describes an ongoing air pollution
study in the Texas Gulf Coast area that is designed to assess the effects of
air pollution on local asthmatics1 symptoms and lung function, and on
performance symptoms and physiology in local residents who exercise vigorously
and repeatedly in a standardized way.
Dawson (Chapter 39 of this volume) describes an integrated set of human
and animal studies planned by the Air Resources Board of the State of
California.
CLINICAL PROGRAM
Our clinical research activities with photochemical oxidants are
described by Haak, Hazucha, and Ginsberg (Chapters 31, 32, and 33 of this
volume). These studies are conducted in our $8-million Human Exposure
Facility located in Chapel Hill, North Carolina.
WORKSHOP COMMENTARY
Question; We learned earlier [Chapter 28 of this volume] that certain animal
studies will be replicated by EPA. What key human clinical studies mentioned
in the 03 and W>2 criteria documents will be replicated?
R. S. Chapman; I would guess that the Orehek et al. (1976) study [see Chapter
33 of this volume for details of reference] will be replicated.
R. E. Lee; I think future studies will depend in some measure on the
discussions at this workshop. That's one of the reasons we asked you to come.
303
-------
31. HUMAN PULMONARY ADAPTATION TO OZONE
Edward D. Haak, Jr.
Health Effects Research Laboratory, MD-58
Office of Research and Development
U.S. Environmental Protection Agency
Research Triangle Park, NC 27711
INTRODUCTION
Animal and human studies have established that short-term exposure to low
levels of ozone (03) produces detrimental effects involving multiple organ
systems and cell types. In most cases, the health impact of these changes
(particularly the long-term consequences) is very unclear. We do not know
which of the observed detrimental effects has the greatest impact on health,
nor for that matter do we know which of the effects is the most sensitive
indicator of detrimental exposure. We do know that the detrimental effects
include decreased immuno-integrity of circulating lymphocytes, changes in
biochemical parameters, decreased pulmonary mucociliary clearance, and
decrements in standard pulmonary function tests.
This report describes a series of studies—the OZADAPT series—that was
designed to examine the effects of sequential 03 exposures on young healthy
nonsmoking men. In all, OZADAPT involved five studies and a total of 90 young
304
-------
31. HUMAN PULMONARY ADAPTATION TO OZONE Haak
healthy nonsmoking male subjects. These subjects were examined for pulmonary
function measures as well as biochemical and immunologic parameters.
A full presentation of the complex OZADAPT series cannot be given here.
This report presents some of the pulmonary function results and compares these
findings across the five separate studies, with particular attention to the
question of "adaptation" in terms of pulmonary function.
PROTOCOL
Identical screening procedures were used to select the subjects for all
five studies. Table 31-1 indicates the mean values and variation of height,
weight, and age for each group of subjects. Comparisons of demographic data
revealed no significant differences among the groups. Any subject with a
history of allergy or asthma was rejected, but no skin tests were performed.
As shown in Table 31-2, each OZADAPT study consisted of five consecutive
study days (Days 1 through 5 = Monday through Friday). Day 1 was always a
control day; exposures (4 h/d) were performed on Days 2 through 5. OZADAPT I
and IV were control studies in which subjects received only air; OZADAPT II,
III, and V involved exposures to 0.4 ppm 03. Pulmonary function data were
obtained at three times each day: GI (base line), C% (2 h), and 04 (4 h).
The pulmonary function data reported in this paper are for forced expiratory
volume
305
-------
31. HUMAN PULMONARY ADAPTATION TO OZONE
Haak
TABLE 31-1. HEIGHT, WEIGHT,
Parameter OZADAPT
height (cm) I
II
III
IV
V
weight (kg) I
II
III
IV
V
age (yr) I
II
III
IV
V
AND AGE OF
Mean
181.2
183.6
181.4
180.3
180.6
76.49
76.55
77.55
71.61
72.79
25.24
24.83
25.89
25.03
24.16
SUBJECTS
Standard
Deviation
6.30
6.04
5.84
8.99
7.08
7.05
9.72
11.41
9.64
8.88
2.58
2.34
1.96
2.83
2.63
RESULTS AND DISCUSSION
We observed significant differences in mean pulmonary function for
OZADAPT II (light exercise) versus OZADAPT III (heavy exercise). The mean
FEV-) data from OZADAPT II indicated a much lower decrement at the light (35
liters/min) exercise level. The mean response diminished to an insignificant
level by-Day 5, the fourth consecutive 03 exposure day. In contrast, the mean
FEV-i data from OZADAPT III showed a much larger decrement at the heavy
exercise (57 liters/min) level; subjects did not return to base-line values by
24 h after Day 2, the first 03 exposure day. The response to the second
exposure (Day 3) was similar in magnitude of mean decrease. By the fourth day
306
-------
31. HUMAN PULMONARY ADAPTATION TO OZONE
Haak
TABLE 31-2. EXPERIMENTAL PROTOCOL
OZADAPT
I
II
III
IV
V
Day
1
2
3
4
5
1
2
3
4
5
1
2
3
4
5
1
2
3
4
5
1
2
3
4
5
Exposure3
Air
Air
Air
Air
Air
Air
°3
03
03
03
Air
°3
03
03
°3
Air
Air
Air
Air
Air
Air
03
03
03
°3
Treadmill
Exercise'3
light
light
light
light
light
light
light
light
light
light
heavy
heavy
heavy
heavy
heavy
heavy
heavy
heavy
heavy
heavy
heavy
none
none
heavy
heavy
a4 h/d. 03 concentration =0.4 ppm.
bLight treadmill exercise = 4 mi/h, 0% grade, 15 min, 35 liters/min
ventilation. Heavy treadmill exercise = 4 mi/h, 10% grade, 15 min, 57
liters/min ventilation.
307
-------
31. HUMAN PULMONARY ADAPTATION TO OZONE
of consecutive 03 exposure (Day 5), however, the mean decrement had diminished
to little more than the variability in the control values.
Therefore, on the basis of mean decrement in FEVi across groups, one may
conclude that consecutive exposures are associated with a progressive
diminution in observed decrement. However, a consideration of individual
subject values across the week points up the great importance of individual
sensitivities, even among "normal" vigorous young men.
Figure 31-1 displays three individual patterns that were observed. (All
subjects received identical treatment, exercise, etc.) Subject #3 was barely
responsive across all exposure days. Subject #7, on the other hand, was very
responsive on the first and second exposure days, but his reactivity was
greatly diminished by the third and fourth exposure days. Subject #8 was
equally reactive on the first and second exposure days; more importantly, his
responsiveness persisted. This clinical observation of wide variance among
"normal" young men prompted us to undertake the additional study (OZADAPT V),
which was designed to further characterize the responsiveness of these
subjects.
In OZADAPT V, mean FEVi responses were minimal for Days 2 and 3 (rest).
There was a large response on Days 4 and 5 (heavy exercise). Decrements on
Days 4 and 5 were very similar to those observed on Days 2 and 3 in OZADAPT
III. Unfortunately, we could not continue the OZADAPT V exposures for a fifth
and sixth day to look for "adaptation."
308
-------
120.00r
U)
U>
o
VD
110.00*
100.00-
>" 90.00 •
80.00.
70.00
60.00
Day 1
Air
-- Subject #3
••• Subject #7
— Subject #8
Day 2
Day 3
Day 4
03
8 8
8
CO
o
q
CO
8 8
in co r«» oo <
EXPOSURE DAY CONDITION
8
o
Day 5
03
CN CO
o
o
q
in
I
I
i-3
8
H
i
SI
Figure 31-1. Response variability among three subjects in OZADAPT III.
-------
31. HUMAN PULMONARY ADAPTATION TO OZONE Haak
The results of OZADAPT V were not unexpected, but they are extremely
important. Whether an individual responds depends in part on his inherent
sensitivity and also on his exercise challenge or provocation level.
We decided to define "significant responsiveness" (in the clinical sense)
as a decrease of >20% in FEV^ or maximal midexpiratory flow rate (MMEF), even
though we could statistically define a meaningful decrement at a much lower
level. We designated the 20% level in order to permit a comparative
examination of all five studies. Our goal was to identify on an individual
basis (not a mean basis) subjects exhibiting evidence of significant effect
and the persistence of that effect across all four exposure days.
As shown in Table 31-3, none of the total 45 subjects in the light
exercise control study (OZADAPT I) and the heavy exercise control study
(OZADAPT IV) showed a 20% decrease in FEV-j or MMEF under any measurement
condition on any study Day- In view of the strictness of our criteria, this
observation is not surprising.
Table 31-4 indicates the number of responders (as defined by the same
criteria) for each of the four consecutive exposure days in each of the three
03 exposure studies (OZADAPT II, III, and V). Remarkably, 4 of 15 subjects in
OZADAPT II (light exercise) met these strict criteria on the first day of
exposure. Even though the responders decreased in number with repeated
exposure, 2 subjects persisted in responsiveness through the fourth exposure
day. Statistical comparison with the control group (OZADAPT I) was not
310
-------
31. HUMAN PULMONARY ADAPTATION TO OZONE Haak
TABLE 31-3. SIGNIFICANT RESPONSIVENESS* AMONG CONTROL SUBJECTS
OZADAPT
I
IV
Day
2
3
4
5
2
3
4
5
Number of
Significant Responders
0
0
0
0
0
0
0
0
aDefined as a decrease of >20% in FEV^ or MMEF.
significant because of the small number of subjects. In OZADAPT III, 7 out of
15 subjects were initial responders. Once again we observed a decrease in
number of responders with continued exposure, although 2 subjects were
responsive even after the fourth exposure day. In OZADAPT V, there were very
few responders on the first and second exposure days (rest). On the third and
fourth exposure days (heavy exercise), the numbers of responders were very
similar to those in the first and second exposure days of OZADAPT III. From
this, it is quite apparent that the ventilation level of exercise challenge is
an important provocative stress that may evoke individual sensitivity to 03
exposure. Note also that, even with this challenge, one third of the OZADAPT
V subjects did not respond in terms of our strict criteria.
311
-------
31. HUMAN PULMONARY ADAPTATION TO OZONE
TABLE 31-4. SIGNIFICANT RESPONSIVENESS3 AMONG SUBJECTS EXPOSED TO OZONE
OZADAPT Day
II 2
3
4
5
total
III 2
3
4
5
total
V 2
3
4
5
total
Number of
Significant Responders
4/15
2/15
1/15
2/15
4/15 (27%)
7/15
7/15
4/15
2/15
8/15 (53%)
1/15
3/15
9/15
7/15
10/15 (66%)
Probability
(I vs. II:)
0.010
0.111
0.341
0.111
0.010
(III vs. IV:)
0.003
0.003
0.050
0.241
0.001
(V vs. IV:)
0.500
0.112
<0.001
0.003
<0.001
aDefined as a decrease of >20% in FEV<| or MMEF.
CONCLUDING REMARKS
Investigators are only beginning to understand the many factors which
affect pulmonary function during short-term acute exposures to 03 and other
oxidants. Among these factors are: concentration of the gas, ventilation
(exercise) level, duration of exposure, proximity of pulmonary function
testing to exposure peaks, attitude, previous exposure experience, humidity,
temperature, and many others.
312
-------
31. HUMAN PULMONARY ADAPTATION TO OZONE Haak
Our OZADAPT studies stress that individual sensitivity is also very
important. The fact that the OZADAPT subjects were not "high-risk" subjects
underlines the need to investigate the presumably large population subsets
that are apparently intolerant of 03 exposure.
WORKSHOP COMMENTARY
R. S. Chapman; Has there been any subsequent follow-up on OZADAPT subjects
who seemed not to "adapt" like the group means, with respect to possible
atopic tendency not uncovered in the initial examination?
E. D. Haak; Are you asking how many of our responsive people were discovered
(say, by skin testing) to be truly allergic, and how many of the nonresponsive
were found to be not allergic?
R. S. Chapman; That would be a subquestion. Having discovered that some of
the subjects were indeed not "adapting," did you do any follow-up skin or
other testing to reveal what we might call an "occult allergic tendency"?
E. D. Haak; We have not gone back and done any skin testing on those
individuals who appeared to be persisting in their responses to 03 exposure.
As a matter of fact, we haven't gone back and done any follow-up testing of
any kind on any OZADAPT subjects. Rather, we have been involved in an
analytical appraisal of the study. We might be able to go back and do some
follow-up testing. Unfortunately, our subject population is a transient
student population; many of them have graduated and would not be available for
testing.
313
-------
32. OZONE-INDUCED HYPERREACTIVITY AS MEASURED BY HISTAMINE CHALLENGE
IN NORMAL HEALTHY SUBJECTS
Milan J. Hazucha
Division of Pulmonary Diseases
Department of Medicine
School of Medicine
University of North Carolina
Chapel Hill, NC 27514
INTRODUCTION
Haak (Chapter 31 of this volume) has demonstrated that 0.4 ppm ozone (03)
significantly decreases pulmonary performance in human subjects who are
exercising. Even under resting conditions, 0.6 ppm 03 for 1 or 2 h results in
a significant impairment in group mean pulmonary function. However,
examination of each individual's performance under such conditions reveals
considerable variation in the degree of response to 03 exposure.
In previous studies, we observed that some subjects show little change
even at 03 concentrations as high as 0.8 ppm, while other, apparently more
sensitive subjects show decreases of up to 50-60% in maximal midexpiratory
flow rate following 2 h of exposure with exercise (Bates and Hazucha 1973).
Clearly, such a wide spectrum of individual variability is obscured if effects
are evaluated only as group mean responses. We also noted that the duration
314
-------
32. OZONE-INDUCED HYPSRREACTIVITY Hazucha
of effect is only a matter of hours: we were not able to detect by
spirometric means any residual effects at 24 h post-exposure (Bates and
Hazucha 1973). Recently, however, Golden et al. (1978) demonstrated that O3
exposure can lead to airway hyperreactivity that persists for several days
(and sometimes weeks) even in subjects who exhibit no continuing effects on
pulmonary mechanics.
Consequently, the study reported here attempted to answer two basic
questions: (1) Does 03 exposure alter airway reactivity as assessed by a.
standard bronchial challenge technique using histamine aerosol (Chai et al.
1975)7 (2) Is there any correlation between individual sensitivity to 03 and
degree of airway reactivity?
PROTOCOL
The subjects were 14 healthy individuals having no history of allergies
or asthma. A 3-d protocol was employed. On Day 1, subjects received a 2-h
air (control) exposure. On Day 2, subjects received a 2-h exposure to 0.6 ppm
03. Each exposure period included two 15-min periods of treadmill exercise (4
mi/h x 10% grade x 15 min = 1 mi walk). On Day 3, the subjects returned for
24-h follow-up testing.
A battery of pulmonary function tests (PFT's) was performed prior to each
exposure, 2 h post-exposure, and 24 h post-exposure. The PFT's included:
functional residual capacity (FRC) and airway resistance (RAW) acquired in a
315
-------
32. OZONE-INDUCED HYPERREACTIVITY Hazucha
plethysmograph, several measures of forced vital capacity (FVC), vital
capacity (VC), forced inspiratory volume (FIV), and maximal voluntary
ventilation (MW) measured by a dry seal spirometer. During exposure to air
or 03, FVC and RAW were measured at 1 and 2 h of exposure, 10 min after the
preceding exercise period. In the post-exposure 2-h recovery period, FVC was
monitored at 30-min intervals.
The PFT battery was always followed by a complete histamine challenge
test. Histamine aerosol was generated in a De Vilbiss #65 ultrasonic
nebulizer. The aerosol was passed through an impactor chamber in order to
narrow the particle size distribution to a mass median aerometric diameter of
2.0 ± 1.6 ym. Seven different concentrations of histamine were generated:
0.0 (saline), 0.3, 0.6, 1.25, 2.5, 5.0, and 10.0 mg histamine salt/ml aerosol.
Briefly, the challenge test consisted of inhaling 5 normal tidal breaths
(800-1000 ml/breath) of each concentration at 5-min intervals. One FVC and
one RAW measure were obtained 2 min after inhalation of each concentration.
RESULTS
Figure 32-1 shows the mean response and recovery of 14 subjects to 03 as
determined by changes in forced expiratory volume (FEV^). The broken line
shows the mean control (air exposure) data (± 1 S.D.) while the solid line
represents the mean 03 exposure data. Controls showed no statistically
significant changes over the exposure and recovery periods; the maximum
absolute difference between mean FEVi values was only 270 ml. In contrast, 03
316
-------
OJ
N)
M
I
5-
A
A
A
A
A
w
a
U)
2-
T
H
<
time (hours)
a
u
Figure 32-1.
Mean FEVj^ in 14 human subjects during and after exposure to air and O3. Solid line =
2-h exposure to 0.6 ppm 03; broken line = 2-h exposure to air (control).
-------
32. OZONE-INDUCED HYPERREACTIVITY Hazucha
inhalation progressively lowered mean FEV"i. After 2 h of breathing 03, mean
PEV-i had dropped by 23%, a significant decrease (p < 0.001). During the 2-h
recovery period, mean FEV^j returned slowly towards the pre-exposure level and
at 2 h was almost the same as the control.
The RAW response (Figure 32-2) was similar but opposite in direction.
After 2 h of O3 exposure, mean RAW increased by 45% (p < 0.001). At the end
of the recovery period, mean RAW was only 7% above the pre-exposure value (not
statistically different). These observations are in accord with previously
published results from our laboratory as well as other research centers.
Implicit in these dynamic lung function changes is the production by 03 of
significant upper airway bronchoconstriction. Although a 2-h recovery period
appeared sufficient for subsidence of the effects on a group mean basis, some
of the exposed subjects continued to exhibit minor residual effects.
Although effects on pulmonary mechanics (on a group mean basis) had
virtually disappeared at 2 h post-exposure, analysis of the histamine
dose-response curves reveals evidence of airway hyperreactivity. Figures 32-3
and 32-4 show the linear regressions obtained for FEV} and RAW histamine
dose-response data, respectively. For simplicity, only linear regression
lines for each condition (pre-exposure, 2 h post-exposure, or 24 h
post-exposure) are plotted; individual points and standard deviations are not
shown. The 2-h post-exposure regression is represented as a solid line. In
Figure 32-3, note the considerable (but not statistically significant)
differences in both intercept and slope of FEV-| at 2 h post-03 versus the
318
-------
u>
N)
2-
u
-------
UI
(O
Ul
(O
M
Z
D
M
!*>
fl
fl
0.0
468
HISTAMINE CONCENTRATION (mg/ml)
Figure 32-3. Mean FEV^ following histamine challenge in 13 human subjects before and after exposure
to air and O3. A = 2-h exposure to air (control); O3 = 2-h exposure to 0.6 ppm O3.
0 = pre-exposure; 2 = 2 h post-exposure; 24 = 24 h post-exposure.
a
0)
N
c
n
a-
-------
U)
w
O
25
§
S
M
468
HISTAMINE CONCENTRATION (mg/ml)
10
Figure 32-4. Mean RAW following histamine challenge in 13 human subjects before and after exposure
to air and CK. A = 2-h exposure to air (control)? O3 = 2-h exposure to 0.6 ppm O3.
0 = pre-exposure;
2-h exposure to air (control)? O3 = 2-h exposure to 0.6 ppm O3.
2 = 2 h post-exposure; 24 = 24 h post-exposure.
EC
D)
S-
-------
32. OZONE-INDUCED HYPERREACTIVITY Hazucha
other times and conditions. However, in the RAW regression lines (Figure
32-4), this post-C>3 slope is increased compared to the other conditions; thus,
the same histamine concentration elicited a greater response when administered
2 h after exposure to 03. Although the differences in intercept are not
statistically significant, the 2-h post-O3 slope is significantly (p < 0.05)
elevated, reflecting greater airway reactivity.
DISCUSSION
Recent studies by Lee et al. (1977) in dogs and Holtzman et al. (1979) in
human nonsmokers have demonstrated increased bronchial sensitivity immediately
after 03 exposure. These authors state that such hypersensitization results
from exposure of bronchial irritant receptors caused by damage to the
epithelium. In both studies, pretreatment with atropine and vagal cooling
blocked this hyperirritability; therefore, the investigators concluded that
the bronchomotor response is mediated via vagal cholinergic pathways.
Unfortunately, the histamine challenges employed in these two studies were
done at the end of exposure, not allowing for a complete recovery from 03.
For this reason, the airway "sensitization" observed in these studies is not
necessarily indicative of true hyperirritability. It is possible that the
increase in sensitivity (bronchoconstriction) was related to a persisting
O3~induced bronchoconstriction to which the histamine effects were added.
That is, the decrease in the airway radius induced by histamine was
superimposed on the (>3-induced bronchoconstriction, resulting in a predictably
greater effect on airway resistance than if the 03 effect were not present.
322
-------
32. OZONE-INDUCED HYPERREACTIVITY Hazucha
However, Habib et al. (1979) explored the variability of airway response to
histamine and concluded that base-line bronchomotor tone is not a major
contribution to the increase of resistance. In our own studies, we allowed
pulmonary function to return to the pre-exposure level, in order to avoid this
confounding of measurements with double intervention (03 and histamine). Even
though FVC, FEV-j, and RAW values reached control levels, however, the airways
remained hypersensitive. Such increased sensitivity may last up to several
weeks in some subjects (Golden et al. 1978).
Closer examination of individual responses in our study showed a wide
spectrum of sensitivity following 03. Some subjects did not appear to respond
either to histamine or to 03, while a few subjects reacted strongly to both
challenges. On reviewing the individual responses to both 03 and histamine
(considered separately), the subjects appear to fall into two distinct
categories based on type of response: reactors (decrease in FEV-j of >20% and
increase in RAW of >100% compared to control values) and nonreactors (the
remaining subjects). When the same separation criteria are applied to the
histamine dose-response data, subjects who responded to O3 seem to have been
more reactive to histamine. This reactivity was further enchanced by O3
exposure. Despite incomplete data analysis, the correlation between response
to O3 and enhanced response to histamine is striking.
Several plausible explanations for this apparent interdependence of
histamine reactivity and 03 response have been explored by other investigators
(Golden et al. 1978). It is possible that 03-induced peripheral airway
323
-------
32. OZONE-INDUCED HYPERREACTIVITY Hazucha
obstruction causes maldistribution of the aerosol, with a greater proportion
going to the larger airways and consequently eliciting increased response.
Ozone damage of epithelial cells with consequent exposure of vagal sensory
receptors is another possible mechanism (Holtzman et al. 1979). Such denuded
receptors may be more sensitive to inhaled histamine. Accordingly, some
investigators (Orehek et al. 1977) have proposed the analysis of data in terms
of sensitivity and reactivity to histamine. Our data have not yet been
analyzed in this manner; therefore, it is difficult to make an in-depth
evaluation of various mechanisms contributing to the magnification of
response.
Regardless of mechanism, these studies demonstrate that 0.6 ppm 03 will
not only impair dynamic lung function but also increase at least transiently
the sensitivity of the tracheobronchial tree. Such hypersensitivity may last
well beyond restoration of lung function as determined by spirometry. The
high correlation between 03 reactivity and magnitude of histamine response
leads us to conclude that the histamine bronchial challenge can be used as a
screening test for 03 reactivity, and might also predict reactivity to other
oxidants.
REFERENCES
Bates, D. V., and M. Hazucha. 1973. The short-term effects of ozone on the
human lung. In: Proceedings of the Conference on Health Effects of Air
Pollutants (Assembly of Life Sciences, National Research Council,
National Academy of Sciences). Report prepared for the Committee on
324
-------
32. OZONE-INDUCED HYPERREACTIVITY _ __ Hazucha
Public Works, U.S. Senate. S. Res. 135, Serial No. 93-15. U.S.
Government Printing Office, Washington, D.C., pp. 507-540.
Chai, H., R. S. Farr, L. A. Froehlich, D. A. Mathison, J. A. McLean, R. R.
Rosenthal, M. D. Sheffer, II, S. L. Spector, ana R. G. Townley. 1975.
Standardization of bronchial inhalation challenge procedures. J. Allergy
Clin. Immunol., 56:323-327.
Golden, J. A., J. A. Nadel, and H. A. Boushey. 1978. Bronchial
hyperirritability in healthy subjects after exposure to ozone. Am. Rev.
Resp. Dis., 118:287-294.
Habib, M. P., P. O. Pare, and L. A. Engel. 1979. Variability of airway
responses to inhaled histamine in normal subjects. J. Appl. Physiol.:
Resp. Environ. Exercise Physiol., 47:51-58.
Holtzman, H. J. , J. H. Cunningham, J. R. Sheller, G. B. Irsigler, J. A. Nadel,
and H. A. Boushey. 1979. Effect of ozone on bronchial reactivity in
atopic and non-atopic subjects. Am. Rev. Resp. Dis., 120:1059-1067.
Lee, L.-Y., E. R. Bleecker, and J. A. Nadel. 1977. Effect of ozone on
bronchomotor response to inhaled histamine aerosol in dogs. J. Appl.
Physiol.: Resp. Environ. Exercise Physiol., 43:626-631.
Orehek, J., P. Gayrard, A. P. Smith, C. Grimaud, and J. Charpin. 1977.
Airway response to carbachol in normal and asthmatic subjects. Am. Rev.
Resp. Dis., 115:937-943.
WORKSHOP COMMENTARY
J. D. Hackney; Were the subjects exercising or resting?
M. J. Hazucha; The subjects were exercising twice: during the last 15 min of
the first hour and during the last 15 min of the second hour of exposure.
J. D. Hackney; To follow up on your dose-response findings, did you test the
subjects' responsiveness to histamine 24 or 48 h later?
M. J. Hazucha; Yes, we tested responsiveness at 24 h post-exposure. There
still appeared to be increased responsiveness to histamine, but the
correlation between 03 response and response to histamine was smaller.
J. D. Hackney; And at 48 h?
M. J. Hazucha; We didn't do any testing at 48 h.
325
-------
32. OZONE-INDUCED HYPERREACTIVITY
Hazucha
R. K. Wolff; Did you make any measurement of the actual inhaled dose of
histamine as opposed to the concentration in the nebulizer?
M. J. Hazucha; Although we know how much of the aerosol was inhaled, we
really don't know how much of it was retained. We believe that if the subject
inhaled approximately the same amount of aerosol at each dose, most likely the
same amount of aerosol was retained and thus the doses were comparable. I
wish we could do such measurements. It is possible, but we lack the technical
means at present.
R. K. Wolff; Did you compare the histamine and the methacholine bronchial
challenges?
M. J. Hazucha; We did. In another study (using exactly the same protocol) we
challenged subjects with methacholine. The magnitude of response seemed to be
about the same as with histamine, and we were again able to identify
responders and nonresponders. The data are not yet completely analyzed so I
cannot give you exact figures.
Question; Do you plan to use 0.25 ppm 03 in your system?
M. J. Hazucha; Yes, in our forthcoming 03 threshold study.
Comment: Are you referring to 03~plus-N02 studies? We have plans to look at
a study similar to the Orehek study, to try to confirm or deny those
observations. We will also look at 0.1 ppm 03 to confirm or deny the report
of De Lucia and Adams. This should follow on the heels of the study discussed
by Ginsberg [Chapter 33 of this volume] . We expect to start with a low-level
03 exposure which should feature the same points made by Haak [Chapter 31 of
this volume].
S. V. Dawson; Did you see an apparent increase in the variance of the results
under 03 exposure? This is something that I see recurring fairly often. It's
not the mean but actually the variance that's changing, suggesting possibly
that some kind of a control system is becoming unstable. A random effects
model—a standard statistical model which isn't used much in biology—examines
whether variance is significantly changed from one regime to another. Have
you looked at that in your 03 studies?
M. J. Hazucha; No, we have not yet statistically analyzed these variances.
We are indeed planning to do in-depth statistical analyses of the variances of
single and group data. Specifically, we will look at correlations between
responders and nonresponders; we will evaluate the two group responses within
and between days.
The increased variance which appeared during exposure and diminished
during recovery was caused, we believe, by the spread of responses induced by
03. Some of the subjects reacted considerably, while some didn't respond at
all. Your suggestion that we analyze these variances is very well taken.
326
-------
32. OZONE-INDUCED HYPERREACTIVITY Hazucha
M. Goldman; In relation to Dr. Dawson's question on variability, [Figures
32-1 and 32-2] indicate the number of subjects to be ~14 in the beginning but
only ~9 during recovery. This may be an explanation for the variability
change. Were the same number of individuals measured throughout exposure and
recovery?
M. J. Hazucha; Yes. Fourteen subjects were studied from the very beginning
to the very end. In some conditions, however, we were unable for technical
reasons to recover data. That's why [the figures] show only 9 of the 14
subjects included in the recovery period.
327
-------
33. RESPONSE OF NORMALS AND ASTHMATICS TO LOW-LEVEL NITROGEN DIOXIDE
Joel F. Ginsberg
Health Effects Research Laboratory, MD-58
Office of Research and Development
U.S. Environmental Protection Agency
Research Triangle Park, NC 27711
INTRODUCTION
The oxides of nitrogen are important constituents of air pollution;
nitrogen dioxide (NO2> is particularly important. Our interest in defining
the adverse effects of low-level NO2 exposure was stimulated by the increased
use of diesel fuel in transportation and power plants.
High levels of NO2 are known to cause inflammatory lung disease in humans
and are implicated as carcinogens in animals. Lower levels, such as those in
the ambient air, obviously need further evaluation in man, particularly in
sensitive individuals (i.e., those most likely to be at risk). A recent
French study (Orehek et al. 1976) suggested that 0.1 ppm NO2 may be harmful to
asthmatics. This finding is at variance with previous work showing no harmful
effects at levels of up to 1 ppm in normal human subjects. In the Orehek
study, 20 mild to moderately severe asthmatics, aged 17 to 44, were exposed
for 1 h to 0-1 ppm NO2 and air at weekly intervals. The protocol also
328
-------
33. RESPONSE OF NORMALS AND ASTHMATICS TO NITROGEN DIOXIDE Ginsberg
included nonspecific bronchial challenge with carbacholine, a synthetic analog
of the naturally occurring neurotransmitter acetylcholine. Two bronchial
challenges were performed in each subject. In terms of specific airway
resistance (SRAW), N(>2 exposure caused the dose-response curve to shift to the
left, and lower threshold doses of carbacholine were required to cause
increases in SRAW. From this it was concluded that very low levels of NO2 can
adversely affect some asthmatics. This conclusion suggests that stringent
regulatory standards, particularly a 1-h standard, deserve consideration.
PROTOCOL
In an effort to corroborate these findings, we designed a similar study
that.is now in progress. Eventually, 15 normals and 15 asthmatics will have
been exposed to 0.1 ppm N02 and to air (each for 1 h) in a randomized,
double-blind, crossover design.
Our asthmatics range from asymptomatic to mildly severe. Where they are
asymptomatic, the diagnosis is based on a history of reversible wheezing or
dyspnea that is clearly seasonal and not associated with upper respiratory
tract infection. All of our asthmatic subjects have at least one positive
cutireaction to a battery of allergens common to our area. Our normals have
no symptoms or family history of atopy, and all have negative skin tests to
the same battery of allergens. All subjects are nonsmokers.
329
-------
33. RESPONSE OF NORMALS AND ASTHMATICS TO NITROGEN DIOXIDE Ginsberg
Subjects abstain from all medications for at least 48 h, and all
theophylline-containing foods are withheld for 20 h. Asthmatics are not
studied within 4 weeks of bronchodilator use, and none are taking steroids.
No subject is studied within 4 weeks of symptoms of an upper respiratory tract
infection.
Bronchial challenges are performed as recommended by Chai et al. (1975)
modified for an ultrasonic aerosol generator (De Vilbiss #65). We use
twofold-increasing concentrations of acetyl methacholine (hereafter referred
to as "methacholine") for seven doses. The normals receive a first dose of
0.31 mg/ml and a maximum dose of 20 mg/ml. The asthmatics receive a first
dose of 0.07 mg/ml and a maximum dose of 5 mg/ml. Five tidal-sized
inhalations with a 4-s breath hold are used to ensure a high percentage of
particle retention. Within 5 to 6 min of each dose, airway resistance (RAW)
and thoracic gas volume (TGV) are measured plethysmographically by the
technique of DuBois et al. (1956).
Our major departures, then, from the Orehek study are: (1) inclusion of
normals; (2) skin testing to exclude occult atopy in the normal group and to
ensure that all the asthmatics share a reagenic mechanism; (3) less clinically
severe asthma in the asthmatic subjects; (4) bronchial challenge with
methacholine (which has a shorter duration of activity due to its affinity for
the enzyme acetylcholinesterase) instead of carbacholine; and (5) more
bronchial challenges (including pre-exposure and 24 h post-exposure) at
varying concentrations.
330
-------
33. RESPONSE OF NORMALS AND ASTHMATICS TO NITROGEN DIOXIDE Ginsberg
Bronchial challenge produces dose-response curves which can be analyzed
for the threshold dose of methacholine (defined as the dose causing a 100%
increase in SRAW) and for the slope of the line connecting responses at doses
above the threshold. While the mechanisms are not clear, changes in these
parameters are gaining favor with investigators as a reflection of the adverse
effects of oxidant pollutants. Of course, we must remember that
methacholine-induced bronchospasm does not reproduce an asthmatic attack
(which is certainly more complex than simple smooth muscle contraction).
However, methacholine-induced bronchospasm may simulate asthma closely for a
very short period of time (e.g., 15-30 min). It is also important to remember
that a methacholine-induced increase in SRAW does not necessarily relate to
increased morbidity among the asthmatic population.
PRELIMINARY RESULTS
We have obtained and plotted some preliminary data for four asthmatics
and four normal subjects. In Figures 33-1 through 33-5, specific airway
resistance (RAW x TGV/1000) (SRAW) is plotted linearly on the ordinate, and
the methacholine dose is plotted logarithmically on the abcissa. The
logarithmic dose plot allows us to stretch the early part of the curve that
would otherwise (for asthmatic subjects) go up very rapidly and be compressed
at the left-hand side of the graph. The term "PDAR" appears in some figures
and is used to describe bronchial challenge. The PDAR base line represents
the first challenge given to each subject. The PDAR 4 line indicates another
bronchial challenge performed in every subject at least 5 d after the
331
-------
33. RESPONSE OF NORMALS AND ASTHMATICS TO NITROGEN DIOXIDE Ginsberg
base-line bronchial challenge. (A minimum of 5 d was chosen because of
compatibility with previous work and good reproducibility of bronchial
challenge at 7-d intervals.) The PDAR 5 line represents a challenge performed
24 h after PDAR 4; PDAR 6 is a challenge performed 24 h after PDAR 5.
Exposure to either air or NC>2 occurred immediately before the second of these
three PDAR's. To summarize, the PDAR base line represents a dose response at
some time distant from the other three, which are dose responses taken at 24-h
intervals with an exposure to either air or NC>2 occurring just before the
second of the three.
Figure 33-1 represents an asthmatic subject exposed to air and challenged
with 2.5 and 5 mg/ml methacholine. Note that the base line appears shifted to
the right compared to the pre-exposure PDAR 4 curve. The PDAR 5 curve is
similar in appearance. Arrows mark each threshold dose. The threshold doses
and the slopes of the lines beyond the thresholds are similar for PDAR 4 and
5. The PDAR 6 curve, however, has a somewhat different appearance, with a
higher threshold and a reduced slope—in short, a shift to the right. This
trend toward a depression of reactivity on the third of three consecutive days
seems to be real and may reflect a repeat treatment effect of the methacholine
challenge. To assume that bronchial responsiveness is similar from week to
week, then, may not be valid.
Figure 33-2 (two normal subjects) displays only the dose responses
obtained immediately after air and NC>2, 1 week apart (i.e,., two of the total
seven bronchial challenges done in each subject). Note that both subjects
332
-------
U)
30
CO
25 -
20 -
Figure 33-1.
15 -
QC
CO
10 -
5 -
BASELINE
PDAR FOUR
PDAR FIVE
PDAR SIX
10-1
1QO. 5 100
LOG (DOSE LEVEL)
10° -5
101
Specific airway resistance following three bronchial challenges in an asthmatic subject
exposed to air.
I
to
w
§
CO
o
o
CO
§
2
H
n
i
i
-------
33. RESPONSE OF NORMALS AND ASTHMATICS TO NITROGEN DIOXIDE
Ginsberg
30
25 •
20
< 15 H
cc
in
10 -
5 •
AIR
1Q-1 1Q-0.5 1QO 1Q0.5
LOG (DOSE LEVEL)
30
25 -
20 -
CO
10 •
5 -
— AIR
NO,
10-1 1Q-0-5 i00 1Q0.5 1Q1
LOG (DOSE LEVEL)
Figure 33-2. Specific airway resistance following two post-exposure
bronchial challenges in two normal subjects exposed to air
and NO2*
334
-------
33. RESPONSE OF NORMALS AND ASTHMATICS TO NITROGEN DIOXIDE Ginsberg
have thresholds of >5 mg/ml. However, the slopes, of the lines beyond the
thresholds differ: the lower subject obviously shows a much increased slope.
Other workers have suggested that these parameters (dose threshold and slope)
characterize a given subject but do not necessarily correlate with one
another; i.e., a high threshold is not always associated with a reduced slope,
as demonstrated by the subject in the lower panel of Figure 33-2.
Another important aspect of Figure 33-2 is the apparent lack of pollutant
effect on base-line SRAW.
Figure 33-3 (two asthmatic subjects) again depicts only the dose
responses obtained immediately after air and NO2» 1 week apart. These
subjects display lower thresholds than the normals in Figure 33-2; in each
case, the threshold is <1 mg/ml. Also, both have steep slopes.
The plots in Figure 33-3 (at least those in the lower panel) again
suggest no significant pollutant effect on airway responsivity. There is some
suggestion of a pollutant-associated shift to the right in the upper panel.
Figure 33-4 presents the full spectrum of seven dose responses obtained
in the asthmatic subject of Figure 33-1. This subject's dose responses should
probably be examined week by week by comparing, in this case, PDAR 4 to PDAR
1, both of which are pre-exposure dose-response curves. For the week of air
exposures (upper panel), this subject's dose-response curves are for some
reason shifted to the left of the base line. Therefore, the effect of
335
-------
33. RESPONSE OF NORMALS AND ASTHMATICS TO NITROGEN DIOXIDE
Ginsberg
10-0-5 100 100.5
LOG (DOSE LEVEL)
101
10°
LOG (DOSE LEVEL)
Figure 33-3. Specific airway resistance following two post-exposure
bronchial challenges in two asthmatic subjects exposed to
air and NO2«
336
-------
33. RESPONSE OF NORMALS AND ASTHMATICS TO NITROGEN DIOXIDE
Ginsberg
30-
cc
25-
20-
15
10
5
—— BASELINE
—••PDAR FOUR
-•— .PDAR FIVE
. — .PDAR SIX
„./
10-1 1Q-0.5 10° 100-5
LOG (DOSE LEVEL)
30
25-
20
|l5-
DC
CO
10-
-BASELINE
.PDAR ONE
PDAR TWO
.PDAR IHREE
™ ••••^^^•^
10° 10°5
LOG (DOSE LEVEL)
101
101
Figure 33-4. Specific airway resistance following seven bronchial challenges
in an asthmatic subject exposed to air and N02.
337
-------
33. RESPONSE OF NORMALS AND ASTHMATICS TO NITROGEN DIOXIDE Ginsberg
pollutant is best examined by comparing bronchial challenges done 24 h before
and immediately after exposure (in this case, PDAR 1 and PDAR 2 in the lower
panel). Unfortunately, this subject is not a particularly good example.
Still, comparing pre- and post-exposure curves for a given week probably gives
a better reflection of the effect of pollutant than comparing only
post-exposure curves (as in the Orehek study).
Figure 33-5 shows the mean responses of the four asthmatics and four
normals. Again, this depicts only post-exposure curves which, as we have just
discovered, may not be the best way to examine the data. Nevertheless, these
mean values for both groups show large between-group differences in both
threshold and slope; the asthmatics obviously have lower thresholds and
increased slopes.
Although these data would have to be considered quite preliminary, the
mean pre- and post-exposure values suggest no effect of N(>2 in either normals
or asthmatics. The apparent NO2~associated rightward shift in asthmatics may
represent a repeat treatment effect of methacholine.
PRELIMINARY CONCLUSIONS
We conclude that there may be week-to-week variability in parameters of
response to bronchial challenge in a group of mildly symptomatic asthmatics.
Comparisons of responses after air and pollutant should probably be made
between pre- and post-exposure curves rather than on the basis of
338
-------
u>
w
w
VO
35
30-
25-
NORMALS-AIR
NORMALS-NO2
ASTHMATICS-AIR
ASTHMATICS - NO2
20-
I
cc
15-1
10-
10-1
10-0.5 1QO
LOG (DOSE LEVEL)
1QO-5
.0
S= 0.73
01
I
0}
M
z
o
I
n
a
Figure 33-5. Mean specific airway resistance following two post-exposure bronchial challenges in
four normal and four asthmatic subjects exposed to air and NO».
3
i
i
-------
33. RESPONSE OF NORMALS AND ASTHMATICS TO NITROGEN DIOXIDE Ginsberg
post-exposure curves only. A confounding variable in this type of analysis
may be a repeat treatment effect of methacholine. Our preliminary data
suggest that a 1-h exposure to 0.1 ppm 1K>2 has no demonstrable effect on
airway responsiveness in either normals or mildly symptomatic asthmatics.
REFERENCES
Chai, H., R. S. Farr, L. A. Froehlich, D. A. Mathison, J. A. McLean, R. R.
Rosenthal, M. D. Sheffer, II, S. L. Spector, and R. G. Townley. 1975.
Standardization of bronchial inhalation challenge procedures. J. Allergy
Clin. Immunol., 56:323-327.
DuBois, A. B., S. Y. Botelho, and J. H. Comroe, Jr. 1956. A new method for
measuring airway resistance in man using a body plethysmograph: Values
in normals and in patients with respiratory disease. J. Clin. Invest.,
35:327-335.
Orehek, J., J. P. Massari, P. Gayrard, C. Grimaud, and J. Charpin. 1976.
Effect of short-term, low-level nitrogen dioxide exposure on bronchial
sensitivity of asthmatic patients. J. Clin. Invest., 57:301-307.
WORKSHOP COMMENTARY
P. E. Morrow; It's important to do the bronchial challenge before and after
exposure; at least that has been our experience.
How might you explain a residual effect of methacholine that persists
from day to day? With carbachol, which is established to be a longer-acting
drug, we certainly see no effect from a previous administration even over a
matter of hours. I'm surprised that there might be a residual effect from
methacholine on a day-to-day basis. Do you have any thoughts about this?
J. F. Ginsberg; That's a good question. We really don't have an answer; we
were equally surprised. We are currently looking at methacholine shelf life;
the fact that we make up methacholine fresh daily may have something to do
with the decreased responsiveness that we see. If there is some type of
adaptation to methacholine, it may possibly be due to a longer persistence on
receptors by more active metabolites.
340
-------
34. HEALTH EFFECTS OF AIR POLLUTANTS IN THE TEXAS GULF COAST AREA
Robert S. Chapman
Health Effects Research Laboratory, MD-54
Office of Research and Development
U.S. Environmental Protection Agency
Research Triangle Park, NC 27711
INTRODUCTION
The 1977 Clean Air Act Amendments mandate that EPA perform a study of the
nature, transport, and effects of pollutants in the Gulf Coast area. EPA's
Health Effects Research Laboratory at Research Triangle Park, North Carolina
is sponsoring the biomedical portion of this project; a companion effort by
the Environmental Sciences Research Laboratory (Research Triangle Park) will
examine pollutant sources and transport.
This report describes the planned epidemiologic studies, emphasizing the
background work that led to their design. Also described are some of the
various scientific "shoals" that we hope to avoid. No data are reported, for
this project remains in the planning and coordination stages.
341
-------
34. TEXAS GULF COAST STUDY Chapman
BACKGROUND
The specific congressional charge for this work occurs in Section 403(d)
of the Clean Air Act Amendments of 1977:
(d) The Administrator of the Environmental Protection Agency
shall conduct a study of air quality in various areas throughout
the country including the gulf coast region. Such study shall
include analysis of liquid and solid aerosols and other fine
particulate matter and the contribution of such substances to
visibility and public health problems in such areas. For the
purposes of this study, the Administrator shall use environmental
health experts from the National Institutes of Health and other
outside agencies and organizations.
Various legislative support documents prepared by relevant congressional
committee staff broaden the scope of this effort to oxidants as well as
aerosols. Background material also suggests that this work focus on the city
of Houston, Texas.
We perceive a dominant attitude in the Houston area that it is
unreasonable to develop national regulations from data that have been
collected primarily in one geographic area. (With respect to oxidants, of
course, this area is Southern California.) Although this attitude is
sometimes expressed in a more heated than enlightening fashion, there is
underlying validity in the complaint. As more studies are conducted and more
data become available, it becomes very apparent that a high degree of
circumspection is required whenever one generalizes results gathered in one
342
-------
34. TEXAS GULF COAST STUDY Chapman
area to other areas. Therefore, we believe that is is scientifically
reasonable to undertake studies in the Texas Gulf Coast.
In view of these factors, we have resolved not to second-guess which
pollutants to assess. Rather, we intend to determine the range of pollution
exposure in the Gulf Coast, the concentrations, the temporal distributions,
etc. before we irreversibly commit ourselves to embark on any specific study.
Since Congress appears to want a rather well defined package of studies,
we feel that the project demands a certain balance between short- and
long-term ambient pollution exposure. Similarly, we consider ourselves well
advised to deal with a spectrum of health in the population (insofar as
resources permit); specifically, we will examine individuals who are
considered healthy as well as those who are sick. Finally, although the
studies will be tailored to the local situation in the Texas Gulf Coast, we
have selected study parameters similar to those already studied in Southern
California. We thus hope to enable maximum comparability of results from the
two geographic areas.
PLANNING STUDY
Our first effort towards fulfillment of the congressional charge was to
enter into a contract with Radian Corporation (Texas) assisted by the
Southwest Research Institute (Texas) for an assessment of the kinds of studies
(in both the air quality and biomedical areas) that would be most useful. The
343
-------
34. TEXAS GULF COAST STUDY _.__ _. Chapman
biomedical portion of the resultant planning document recommended five or six
different kinds of studies as "highest priority-"
First, the planning document recommended that we assess the influence of
local air pollution on the development of lung cancer in the area. We
consider this to be a quite reasonable recommendation: over the past 15 years
or so, lung cancer rates (at least for white males) have practically doubled
in Harris County, the county in which Houston is located. The current annual
incidence in white males is estimated at ~70 per 100,000 population, compared
to a national incidence of ~40.
The planning document also recommended that we assess whether the local
ambient air has any mutagenic potential. Because no actual human effects
would be assessed, such a study does not seem particularly appropriate for the
biomedical arm of this rather circumscribed congressional charge. Perhaps in
connection with a study of cancer some mutagenicity assessment would be
appropriate; opinion is divided on that point.
Thirdly, the planning document recommended that EPA perform a study of
maximally exercising individuals who exercise repeatedly and in a standardized
fashion. This seems to be a quite reasonable recommendation, and we have
planned such a study. The background for this work comes from both
experimental and epidemiologic studies. There is now a great deal of
experimental evidence that oxidant (at least, ozone) effects become more and
more discernible as the level of exercise increases. On the epidemiologic
344
-------
34. TEXAS GULF COAST STUDY Chapman
side, the very interesting study almost 15 years ago of high school
cross-country runners in San Marino, California (Wayne et al. 1967) remains
one of the most convincing of all air pollution epidemiologic studies. The
Wayne et al. study has certainly been stressed in the preparation of oxidant
criteria documents and standards. Therefore, a study of maximally exercising
people fulfills several of our preliminary criteria.
Another recommendation of the planning document was that EPA undertake an
ongoing health surveillance system, i.e., longitudinal tracking over several
years of the occurrence of acute respiratory illness in children. We do not
feel that this type of effort fits comfortably into the program that Congress
intended: (1) Such a study is rather open-ended? (2) a careful study to
relate disease incidence to local air pollution would of necessity be very
time-consuming and very, very expensive (consuming all the resources that
Congress appropriated for the entire effort).
Finally, the planning document recommended that EPA undertake serial
studies of asthmatics, whose symptoms, physiology, and/or daily fluctuation in
use of medication might be correlated to short-term ambient pollution
exposure. We consider this to be an interesting recommendation, largely
because there are again both experimental and epidemiologic underpinnings for
the work. A great deal of the experimental underpinning comes from Dr. Jack
Hackney, whose work with reactive and unreactive normal subjects as well as
chronic asthmatics suggests that persons with reactive airways or wheezing
345
-------
34. TEXAS GULF COAST STUDY Chapman
tendency may be unusually sensitive to experimental and/or ambient oxidant
exposures (Hackney et al. 1975a, 1975b; Linn et al. 1978). For reasons
discussed below, we propose to do the asthma assessment not in children but in
adult-onset asthmatics.
To summarize, the three recommended studies that EPA will most likely
pursue are: (1) a study of lung cancer in the area; (2) a serial assessment
of asthmatic symptoms, simple pulmonary physiology, and medication use; and
(3) a serial assessment of the performance, physiology, and (probably)
symptoms of people exercising vigorously in the field in a standardized and
repeated fashion.
We originally contemplated a fourth study that was not specifically
recommended by the planning document: a cross-sectional study in which
questionnaires concerning persistent respiratory symptoms and relevant
covariates would be distributed to adults in areas of differing pollution
exposure. In EPA's hands, this general strategy has proven workable as well
as useful from a regulatory standpoint. Previous cross-sectional studies on
sulfur oxides and particulates have demonstrated an internal consistency of
results (at least, qualitative consistency) that tends to reinforce the
findings of other EPA studies. For the time being, limitations of financial
resources and manpower preclude the inclusion of a cross-sectional
questionnaire study in the Gulf Coast project. Should one of the three
planned studies prove unfeasible, however, we would certainly reconsider this
type of effort.
346
-------
34. TEXAS GULF COAST STUDY Chapman
FEASIBILITY ASSESSMENT
Prior to collecting any data, we intend to complete a thorough
feasibility ascertainment for each of the three planned studies. Toward this
end (and toward scientific assistance throughout the remainder of the
project), we are assembling an extramural advisory group that will formally
convene from time to time, and with whose members the investigators will
maintain frequent informal contact. The disciplines to be represented in the
group are epidemiology, biostatistics, air quality measurement, historical
estimation of air quality, pulmonary medicine and physiology, and clinical
allergy. Hopefully, the advisors will help us avoid any severe methodologic
or interpretive problems.
Lung Cancer Study
The proposed study of lung cancer will receive extensive feasibility
assessment. The venue for the proposed effort is Harris County, Texas, the
county in which Houston is located. Currently, 700 to 750 new cases of lung
cancer are estimated to occur in Harris County each year, the majority in
white males. Female rates are much lower (~15 per 100,000 population).
However, we may be well advised to examine females, since they may be less
likely to undergo occupational exposure in the petroleum and petrochemical
industries.
347
-------
34. TEXAS GULF COAST STUDY Chapman
A few years ago, MacDonald (1976) geographically charted the incidence of
lung cancer in Harris County. She found a correspondence between
concentration of incidence and either (1) proximity to the Houston Ship
Channel (a petroleum and petrochemical industrial hub) or (2) location in a
downwind path from the Channel. However, Marmor (1978) pointed out that
MacDonald did not have the opportunity to measure and assess socioeconomic and
occupational variables. She did not quantitate smoking differences, nor did
she specifically consider population mobility (certainly a crucial factor in
rapidly growing Harris County).
Henderson and associates recently completed a case-control study of lung
cancer in Los Angeles (Pike et al. 1979). Preliminary observations (Menck et
al. 1974) resembled those by MacDonald: lung cancer cases appeared to cluster
in an industrial area of the Los Angeles region. Following a more careful
case control study, however, all of the discernible effects in this geographic
area were observed to result from differential occupational exposure.
Henderson was unable to turn up any effects resulting from ambient exposure
alone.
Our feasibility effort for the cancer study will be complex, particularly
with respect to air quality estimation extending back over the past few
decades. We hope to be able to geographically characterize the county in
terms of high, intermediate, and low pollution classifications for individual
pollutants and for combinations of pollutants. We plan to assort cases and
comparison subjects into these zones and then, after adjusting for appropriate
348
-------
34. TEXAS GULF COAST STUDY Chapman
covariates, determine whether the proportion of cases in high pollution areas
differs from the proportion in the control group. Obviously, the success of
this case-control study will directly depend on the confidence with which we
can construct past exposure estimates for Harris County. Should
reconstruction prove unsuccessful, we will have to choose between adapting the
study design or abandoning the study altogether.
Another important feature of the lung cancer feasibility study will be to
ascertain the potential cooperation of the local medical community. The study
cannot succeed in the absence of very good relations with the pulmonary
clinicians in the area.
Our feasibility study will also seek to evaluate the existing tumor
registry capacity. How quickly will we be informed when a new lung cancer
case occurs? A lack of rapid turnaround would require us to build our own
surveillance system into the study. Some well established tumor registries
take, at best, ~12 to 18 mo to inform an investigator of a specific new case.
For lung cancer, this is far too slow: the 1-yr mortality for newly diagnosed
cases of lung cancer is ~85%.
In summary, successful completion of the lung cancer feasibility
study—particularly the estimation of air quality for previous years—will be
a considerable task both scientifically and politically. In our opinion,
however, the rise in lung cancer rates as well as their current absolute
magnitude justify this effort.
349
-------
34. TEXAS GULF COAST STUDY Chapman
TENTATIVE PROTOCOL
Performance of the feasibility and full-scale studies will proceed under
a cooperative agreement executed in late 1979 between EPA and the University
of Texas School of Public Health at Houston. Principal investigators from the
University of Texas will be Dr. Patricia Buffler and Dr. Reuel Stallones, Dean
of the School of Public Health. This author will serve as principal
investigator (biomedical) for EPA. All parties anticipate the effort to be a
cooperative agreement in more than name only.
Lung Cancer Study
Should our feasibility efforts predict favorably for success of the lung
cancer study, we plan to collect data from cases and comparison subjects over
the entire three-year period of 1981-1983. If the whole project succeeds, a
final report will be issued by mid-1984.
Asthma Study
Our ultimate goal in the asthma study will be to follow ~60 adult-onset
asthmatics on a daily basis for 6 mo. We plan to distribute to these subjects
some device by which each can easily measure one or more parameters of lung
function each study day. The device used may be a Mini Wright Peak Flow Meter
or perhaps an instrument such as the Vitalor, which yields a permanent tracing
of the forced vital capacity maneuver.
350
-------
34. TEXAS GULF COAST STUDY Chapman
Our reason for concentrating on adult-onset asthmatics is that we want to
consider the important covariate of fluctuation in medication use. To date,
this covariate has not been formally assessed in serial studies of asthmatics;
therefore, available studies are open to considerable questions of
interpretation. Warnings to doctors about the dangers of nebulizer use in
children have appeared in the pediatric literature. Some sobering stories of
nebulizer overdose and paradoxical nebulizer effects in children have
convinced us that childhood asthma, despite its theoretical advantages, may
not be an appropriate field of study in this project. We welcome further
comment on the relative appropriateness of childhood and adult asthma.
The "nebulizer chronolog"—an instrument to measure nebulizer use—is not
fully tested but may be useable by the time our study starts (probably in the
spring of 1981). The nebulizer chronolog records the time of day at which
each squirt of the nebulizer occurs; thus, we will have at least the daily
number of squirts (although it's unlikely that we'll be able to tell whether
one squirt is as big as another).
Assuming the feasibility study forecasts success for the asthma project,
we expect to issue a report by mid-1982.
Exercise Study
With respect to the exercise study, it is not possible in this report to
outline a proposed schedule. The schedule will depend integrally on the
351
-------
34. TEXAS GULF COAST STUDY Chapman
number of subjects that are ultimately selected, on their frequency of
standardized exercise, on the range of pollution exposure at the study site,
and on the complexity of that exposure.
In this author's opinion, the most desirable subjects would be "dedicated
adult joggers" who strive to exercise vigorously each and every day, who run a
standard distance, and who perhaps even try to improve their time each day.
It is not certain that we can obtain such a group, largely because of the
absolute requirement for standardized exercise on every occasion. Perhaps no
informal group of adult joggers, regardless of personal dedication, would
fulfill this requirement. For this reason, we will canvass the area not only
for adult subject groups but also for possible high school and college groups.
We have already determined (to our dismay) that the high school cross-country
schedule will not permit replication of the Wayne et al. (1967) study
mentioned earlier. This is because high school cross-country meets tend to
occur at different locations within cities and even within tri-county areas.
Such constant changes of exercise location might exert important unmeasured
effects on the results.
We hope to produce our report on the exercise study shortly after the
report on asthma (i.e., mid-1982 or perhaps a little later).
352
-------
34. TEXAS GULF COAST STUDY Chapman
REFERENCES
Clean Air Act Amendments of 1977. Public Law 95-95, August 7, 1977.
Hackney, J. D., W. S. Linn, J. G. Mohler, E. E. Pedersen, P. Breisacher, and
A. Russo. 1975a. Experimental studies on human health effects of air
pollutants: II. Four-hour exposure to ozone alone and in combination
with other pollutant gases. Arch. Environ. Health, 30:379-384.
Hackney, J. D., W. S. Linn, D. C. Law, S. K. Karuza, H. Greenberg, R. D.
Buckley, and E. E. Pedersen. 1975b. Experimental studies on human
health effects of air pollutants: III. Two-hour exposure to ozone alone
and in combination with other pollutant gases. Arch. Environ. Health,
30:385-390.
Linn, W. S., R. D. Buckley, C. E. Spier, R. L. Blessey, M. P. Jones, D. A.
Fischer, and J. D. Hackney. 1978. Health effects of ozone exposure in
asthmatics. Am. Rev. Resp. Dis., 117:835-843.
HacDonald, E. J. 1976. Air pollution, demography, cancer: Houston, Texas.
J. Am. Med. Women's Assoc., 31:379-395.
Marmor, M. 1978. Air pollution and cancer in Houston, Texas: A causal
relationship? J. Am. Med. Women's Assoc., 33:275-277.
Menck, H. R., J. T. Casagrande, and B. E. Henderson. 1974. Industrial air
pollution: Possible effect on lung cancer. Science, 183:210-212.
Pike, M. C., J. S. Jing, I. P. Rosario, B. E. Henderson, and H. R. Menck.
1979. Occupation—"Explanation" of an apparent localized excess of lung
cancer in Los Angeles County. In: Energy and Health (SIAM-SIMS
Conference Series, No. 6). Society for Industrial and Applied
Mathematics, Philadelphia, Pennsylvania, pp. 3-16.
Wayne, W. S., P. F. Wehrle, and R. E. Carroll. 1967. Oxidant air pollution
and athletic performance. J. Am. Med. Assoc., 199:901-904.
WORKSHOP COMMENTARY
Comment; We have heard a lot [at the Research Planning Workshop on Health
Effects of Oxidants] about the animal infectivity model. I haven't heard you
describe any planned studies to investigate human susceptibility to infection.
353
-------
34. TEXAS GULP COAST STUDY Chapman
Are you planning any, or is that something that will be available only from
animal toxicology?
R. S. Chapman; I'd love to do a study like that on a community level. But,
frankly, I'm not sure that it's doable right now. An epidemiologic study
dealing with pollution and respiratory infection would require careful
assessment of specific etiologic agents over time; this would be a very
difficult prospect in itself. It would require a group of unusually willing
subjects, because (in my opinion) one would have to do serologic testing at a
fairly short interval. For instance, there's evidence now that the immune
period for respiratory syncytial virus in children may be as short as 3 mo.
This means at least a couple of bleedings per year—a rather ticklish prospect
on an epidemiologic plane. I'm disheartened to say, though, that a bigger
deterrent is the fact that such a study would need to be long-term (a decade
of data collection, at least). Frankly, I don't think the present EPA funding
structure could realistically absorb such a long-term study.
Comment; We see so much of these infectivity models in terms of the criteria
document; I would think it would be important to~
R. S. Chapman; I agree with you one hundred percent. There have been some
indirect assessments along this line in self-administered cross-sectional
questionnaire studies in which mothers were asked whether children had certain
specific syndromes diagnosed by a doctor over a given period of time. Because
they rely on the mother's recall, such studies aren't (theoretically, anyway)
ideal. More prospective assessments of acute respiratory illness in all
family members have been and are being performed by EPA; Lee [Chapter 30 of
this volume] mentioned one such study. These studies allow one to compare
communities with respect to gross incidence of upper and/or lower respiratory
illness, perhaps with an allergic overtone. I don't think they allow the
analysis to become any more specific than that. Since these studies are
performed on a daily basis, time series statistics might conceivably be
developed to allow some shorter-term inference. However, I feel that
including some assessment of community experience with specific etiologic
agents would be a highly desirable adjunct to this kind of work.
Question; With respect to your asthmatic pulmonary function study, will you
draw upon the data that were collected in the Houston Area Oxidants Study?
R. S. Chapman; Those data may be helpful in locating potential subjects.
Before we start, we hope to know where virtually every adult-onset asthmatic
in the area lives so that we can focus on a couple of residential clusters
(ideally, just one cluster), in order to minimize our aerometric requirements
for ambient measurement. Other than that and some of the initial recruitment
methods, I'm not sure that the Houston Area Oxidants Study data will be
particularly useful to us. If those data are carefully reanalyzed, their
usefulness may increase.
354
-------
34. TEXAS GULF COAST STUDY Chapman
Comment; Everything you've described has involved human pulmonary effects or
animal experimentation that includes extrapulmonary effects. Do you plan to
look at any human extrapulmonary effects?
R. S. Chapman; The only possible nonpulmonary work would be some simple
cardiovascular physiological measurements (perhaps heart rate, blood pressure,
and/or electrocardiogram) on subjects in the exercise study.
Question; What about blood chemistry?
R. S. Chapman; Probably not, although the door isn't entirely closed. We
must make certain decisions with respect to our financial resources. For one
thing, we won't be able to rely on ambient measurements alone; we will
probably have to do some indoor measurements before we're through. (Whether
those are performed in the homes of subjects or in similar homes nearby has
not been decided. We're asking a lot of the subjects already.) Blood
chemistries, desirable as they might be, probably won't be possible just
because of resource limitations.
Question; Was Houston involved in the third National Cancer Survey?
Comment; Yes.
Question; Have you compared Houston with other cities with respect to lung
cancer? Was it always higher?
R. S. Chapman; No; of the six-county area, Harris County has the second
highest incidence rate. Galveston has a somewhat higher rate.
Comment; I was more interested in a comparison to the other cities examined
in the National Cancer Survey.
[Comments inaudible]
R. S. Chapman; One of the reasons we have to find some relatively low
exposure areas within Harris County is to provide grounds on which to assess
pollution effects. If there is homogeneous exposure throughout the county,
there's no point in doing a case-control study.
Some of the available data (admittedly, incomplete) suggest that there is
some kind of exposure gradient over the years. Whether we can estimate that
gradient is another question entirely.
Question; Do I understand correctly that you intend, in the exercise study,
to obtain hourly values for sulfate and nitrate? Can that be done?
R. S. Chapman; If we gave you that impression, we either weren't thinking
very clearly or didn't express ourselves very clearly in writing. I don't see
any theoretical deterrent to gathering hourly filter samples for those
355
-------
34. TEXAS GULF COAST STUDY Chapman
substances. I don't see much point in going finer than that; hourly values
themselves may be difficult to obtain. I'd appreciate comment on this point.
Do hourly measurements of nitrate make any sense? I think they're doable, but
do they make any sense? Is the distortion-to-real-information ratio so high
that the hourly measurement is going to be "garbage"? Or does the ratio stay
constant regardless of the sampling interval?
Comment; It's a variable measurement demanding care in technique.
J. D. Hackney; Do I understand correctly that you intend to study adult-onset
asthmatics?
R. S. Chapman; Yes, but I can be talked out of it! I have a nebulous
suspicion that there may be something different about adult-onset asthmatics
in comparison to childhood-onset asthmatics who remain asthmatic as adults. I
have no concrete evidence. Do you think that my concern holds any water at
all?
J. D. Hackney; I just wanted clarification.
Question; Are you speculating that the adult-onset asthmatic is less likely
to be an atopic asthmatic and more likely to be a perennial nonseasonal
asthmatic?
R. S. Chapman; In all candor, my thinking wasn't that refined. It is highly
unlikely that we will have more than 60 subjects stay with us throughout the
course of this work. Therefore, we might be well advised to impose a certain
homogeneity on the group. I could certainly be convinced that it would be
best to choose adult asthmatics with childhood onset rather than adult-onset
asthmatics. On the other hand, if there's absolutely no reason for that, we
could certainly widen the field of acceptance.
Question; In the exercise study, how do you plan to perform all the pulmonary
function assessments "at once" (when they all "pile up on you")?
R. S. Chapman; That's a good question—one that we haven't really worked out
yet. The number of testing technicians ultimately hired will probably depend
largely on the number of subjects. My own feeling is that no single group
will contain more than ~15 subjects. It seems to me that three testing
technicians could test this number of people spirometrically in a rather short
period of time. I would not expect a single subject (especially a trained
one) to stay with the testing technician for more than 3 min.
But another question arises: Would we be well advised to forget about
testing them immediately post-exercise? Would we be smarter to test them,
say, 20 min post-exercise after they've had a shower? That would give us a
standardized interval for testing; testing immediately post-exercise,
especially after an exhausting routine, might not be very informative.
356
-------
34. TEXAS GULF COAST STUDY Chapman
M. Goldman; With respect to the lung cancer study, isn't change in population
makeup a crucial facet? There is probably a 20- to 30-year latent period for
lung cancer. Therefore, if there ^s an environmental association, those at
risk may have received it at another location. Secondly, it's very difficult
to remove the overwhelming effect of cigarette smoking both in women and in
men. If we do this, there may be very little left. This is especially true
when we try to determine what the "petrochemical" environment was 20 years ago
and what it is at this time. In sum, are the problems of an intracounty
assessment surmountable?
R. S. Chapman; That is a succinct statement of our own concern.
357
-------
35. OVERVIEW OF RESEARCH AND REGULATORY ACTIVITIES
OF THE CALIFORNIA AIR RESOURCES BOARD
John R. Holmes
California Air Resources Board
1102 Q Street
Post Office Box 2815
Sacramento, CA 95812
INTRODUCTION
This report surveys the research and regulatory functions of the Air
Resources Board (ARB) of the State of California. Chapters 36 through 39 of
this volume present details of some of the oxidant health effects studies.
BACKGROUND
The California State Government has sponsored air pollution research on a
substantial scale for the past 10 years or so. This author heads a program
which actually began in the University of California as Project Clean Air.
About 1970, that program was shifted from the University to ARB and expanded
to include investigators from outside the University of California system.
Our charge is to do applied research on problems that are unique to and
critical in California. Over the years, we have tried to adhere fairly
358
-------
35. CALIFORNIA AIR RESOURCES BOARD Holmes
closely to that mandate. The program was developed in much the same way that
EPA's program is now being developed: client relationships were established
with all the various ARB divisions and branches. Input from these clients
allows us to propose, on a yearly basis, a research program that responds to
their needs.
Year-to-year variability in the funding level sometimes makes it
difficult to plan with certainty. Funding has ranged from ~$1.5 to ~$3.5
million a year. We operate on a zero-base arrangement: each year ARB puts
together a set of projects and takes it through the budget process from the
Governor's Office to the Legislature. Appropriation is on a
project-by-project basis.
For the budget year beginning in July 1980, we hope to have ~$4 million
for research within the State of California. The Governor has proposed a
fairly healthy increase for us this year? what the Legislature will do with
this proposal remains to be seen.
RESEARCH PROGRAM
The ARB research program is divided into seven different categories; each
is discussed below. Most of the reports originating from this program are
available through the National Technical Information Service in Washington,
D.C. Available from our office in Sacramento is a short report containing
one-page summaries of all ARB projects conducted over the past two years.
359
-------
35. CALIFORNIA AIR RESOURCES BOARD Holmes
Effects of Air Pollution
Research into the effects of air pollution represents a major part of our
program. Health effects research has accounted for ~20% to ~25% of the total
research budget over the past five years; the fraction allocated to health
effects research in the proposed 1980 budget is somewhat larger. A number of
problems remain to be resolved, and ARE will need better and better
information to establish California air quality standards in the years ahead.
The other important area is research into effects on vegetation; ~10% of
our budget is devoted to this category. Agriculture is the State's largest
industry; the value of crops produced in 1978 was ~$12.5 billion, with another
$2 billion generated by the forest products industry. Thus, the State of
California is very concerned about limiting damage to crops and forests.
Economic Impacts of Air Pollution Control
In the last couple of years, ARB has become very involved in estimating
the economic impacts of control options. One of our contractors is working to
incorporate air pollution control in a statewide input-output economic model.
This model is designed to consider economics at the stage when control
measures are first proposed. When the model is up and running, we'll be able
to test various proposed control measures for economic impacts on directly
affected industries. Beyond that, the model will permit us to determine how
the costs of air pollution control filter down through our State's economy.
360
-------
35. CALIFORNIA AIR RESOURCES BOARD Holmes
Emissions Inventory and Control Technology
The third category of our research program—emissions inventory and
control technology—receives the largest fraction of funding. About 40% of
our budget is devoted to this research, which includes studies to improve
emissions inventories for stationary sources and cars, feasibility studies on
the kinds of control equipment that might be adopted for use in California
(e.g., scrubbers and ammonia injection as a means to control nitrogen oxides
emissions from large sources), and so on.
Atmospheric Processes
About 15% of our budget is committed to a catchall category called
"atmospheric processes." We've sponsored a great deal of smog chamber work by
James Pitts' laboratory at the University of California - Riverside. These
studies are designed to sort out some of the intricacies of photochemical smog
formation and to determine the fates of various organic molecules present in
the air (solvent hydrocarbons, pesticide hydrocarbons, etc.).
Also funded in this category is a great deal of work in the field. For
example, ARB sponsors tracer studies and airborne measurements to establish
source-receptor relationships and track pollutant movements between the
State's various air basins or air quality control regions. Results from this
work are calling into question a fundamental assumption of our State law—the
assumption that the State can be divided into separate and independent air
361
-------
35. CALIFORNIA AIR RESOURCES BOARD Holmes
basins or sheds. We now know that a great deal of pollution is transported
from such locales as the San Francisco Bay Area to the Great Central Valley,
and from the Los Angeles area down to San Diego and out into desert areas.
So, what happens in the way of controls upwind is a very important determinant
of what happens in air basins downwind. Our field studies establish these
relationships.
Air Quality Modeling
The fifth category, air quality modeling, entails the development of
models that simulate emissions and such atmospheric processes as chemical
transformation and dispersion. There are regional models in various stages of
development for San Diego, Los Angeles, Sacramento, and Fresno. The Bay Area
model was developed independently of our support.
Although air quality models seem to be the wave of the future, our
experience shows them to be a mixed blessing. In principle, such models can
provide very precise information on the benefits of various control measures.
But air quality models are also "data hungry": To exercise a model properly
and with confidence requires a great deal of air quality data, meteorological
data, emissions data, and so forth. Thus, although ARB is making progress in
this area, we have also discovered that it takes longer than expected to
reduce theory to practice.
362
-------
35. CALIFORNIA AIR RESOURCES BOARD Holmes
Meteorological Forecasting
Because air pollution can reach very high levels in such areas as Los
Angeles and the Central Valley, meteorological forecasting is an important
part of our research program. We need to be able to forecast the occurrence
of severe episodes of photochemical smog in Los Angeles and of the trapping of
smoke from agricultural burning in the Valley during winter months. When
employed by research meteorologists, advanced statistical techniques can be
used to very handily forecast the occurrence of such emergencies at least 24
and possibly as long as 36 hours in advance.
Measurement and Measurement Methods
Finally, ARB sponsors a fair amount of work on air pollution measurement
and measurement methods. These efforts focus on ambient levels. For example,
we've worked with the California Institute of Technology and with the
University of California - Berkeley to develop methods to survey acid
precipitation. A report on acid rain in California will be issued within the
next two months.
The laboratory of James Pitts at the University of California - Riverside
has developed sophisticated measurement techniques for vapor-phase ammonia, a
compound that is very important in the formation of photochemical aerosols due
to its neutralization of sulfuric and nitric acids. Understanding ammonia is
important to understanding the nature of the particulate burden in many areas
363
-------
35. CALIFORNIA AIR RESOURCES BOARD
Holmes
of the State as well as for dealing with the problem of visibility
degradation.
CALIFORNIA AIR POLLUTION STANDARDS
For more than a decade, the State of California has promulgated its own
ambient air quality standards and other standards to limit community exposure
to various pollutants and combinations of pollutants. These California
standards fall into three distinct categories: ambient air quality standards,
alert levels, and standards for hazardous substances.
Ambient air quality standards are, of course, levels that are considered
to be safe over the long term for the general population.
Alert levels are specific ambient levels at which extraordinary measures
are taken. For example, the State has established alert levels for ozone that
require special action on the part of industry and, in some cases, the
population in general. The best known is the school alert level for ozone:
when ozone rises to 0.20 ppm and that concentration is expected to persist for
an hour or more, schools and colleges are notified. Most of our medical
advisors believe that children, especially, should curtail their outdoor
physical activities when ozone reaches this level.
Recently the State has considered standards for hazardous substances.
This is a relatively new area for the Federal Government as well as for us.
364
-------
35. CALIFORNIA AIR RESOURCES BOARD Holmes
These hazardous substance standards are somewhat different from our ambient
standards or alert levels in that they are targeted specifically at the
vicinities of sources (factories, etc.). For example, about two years ago ARE
adopted a hazardous substance standard for vinyl chloride. In California,
vinyl chloride is not a ubiquitous pollutant—it's localized to three or four
individual sources in the Los Angeles Basin. Our standard provides that vinyl
chloride levels will be monitored in the vicinity of each plant (i.e., at the
property line or thereabouts). The rationale for this approach is that most
of the emissions from such facilities do not emanate from a stack or any
particular large source; rather, they are "fugitive emissions" (principally,
leaks and accidental releases). It is very difficult to write an emissions
regulation that addresses fugitive emissions in an effective way, so ARB
adopted an air quality standard for the vicinity of each vinyl chloride
facility. Thus, each facility is free to figure out the best way to reduce
its own emissions to a level that is safe for the surrounding community. This
makes the job of ARB easier, too: it's simpler to write this type of
regulation as compared to a detailed engineering specification type of
regulation.
The ARB procedure for adopting standards is somewhat different (and in
some ways simpler) than EPA's procedure. Responsibility is divided between
ARB, which actually adopts the standard, and the State Health Department,
which is required under State law to advise ARB on health-related standards.
The State Health Department, in turn, is advised by a committee of physicians
appointed by the California Medical Association.
365
-------
35. CALIFORNIA AIR RESOURCES BOARD Holmes
In contrast to the private or closed-door method of EPA (in which the
Administrator proposes standards on the basis of staff recommendations),
California standards are adopted via a public hearing process. In this
author's opinion, there are virtues to both systems. In the California
system, the ARB staff does a great deal of the groundwork prior to the
hearings, and the hearings can be lengthy. However, the Board itself—the
group of individuals who must ultimately make the decisions—is exposed to a
much wider range of viewpoints than in the Federal system.
Our system also differs from the Federal system in that there are no
rigid timetables for attainment of standards. The ARB views standards more as
ultimate objectives of the air pollution control program. The State statutes
do require "reasonable progress" toward attainment of standards, but there are
no rigid timetables. As technology develops and as economic conditions
permit, ARB and the local air pollution control officials implement emission
controls to move ahead toward our standards. This gives us a fair degree of
flexibility that is not present in the Federal system.
In California law there are no distinctions between primary or
health-related standards and secondary or welfare-related standards.
Standards adopted by ARB are assumed to be protective of both human health and
welfare. There are only a couple of standards that are strictly
welfare-related: the standard for the hydrocarbon ethylene (a specific
toxicant for certain kinds of plants), and the State visibility standard,
366
-------
35. CALIFORNIA AIR RESOURCES BOARD Holmes
which has turned out to be our most stringent standard, the toughest to meet,
and the one that we've made least progress toward attaining.
WORKSHOP COMMENTARY
Question; Do the budgetary numbers that you quoted represent the budget of
your research program, or is that the total budget of ARE (which, I presume,
involves a great deal of expenditure for administrative costs and all the
noninvestigative functions)?
J. R. Holmes; Those figures are for actual extramural research projects.
Those are the dollars that we "shovel out the door," so to speak.
Question; How does the yearly uncertainty in funding levels affect your
ability to get other people to do work related to ARE goals?
J. R. Holmes; Well, it's certainly a problem. Some of the projects that we
would like to do almost require a guarantee of continued funding (for example,
prospective epidemiologic work). We've talked to the Legislature staff and
tried to convince them that we ought to have this sort of funding capability,
but so far we haven't met with success. The Legislators simply don't want to
commit funds over that period of time.
367
-------
36. LUNG INJURY AND DEPLETION IN THE MOUSE
FOLLOWING EXPOSURE TO NITROGEN DIOXIDE
Russell P. Sherwin
Department of Pathology
School of Medicine
University of Southern California
2025 Zonal Avenue
Los Angeles, CA 90033
INTRODUCTION
This report summarizes some preliminary animal experimental data which we
consider highly relevant to the question of human adverse health effects
resulting from exposure to oxidants. Lung tissue of mice exposed to nitrogen
dioxide (NO2) was analyzed for hyperplasia of Type II cells and protein
leakage.
BACKGROUND
Preliminary studies of human lungs from the Los Angeles Medical
Examiner-Coroner's Office and the Los Angeles County-University of Southern
California Medical Center suggest that a progressive depletion of lung tissue
occurs in everyone. The rate at which this occurs and the extent of the
depletion in the well population are presently unknown. However, the crude
368
-------
36. PULMONARY EFFECTS OF NITROGEN DIOXIDE IN THE MOUSE Sherwin
indicators available (in particular, "naked eye" measurements of emphysema,
and the frequency of bronchiolitis and structural disruption by microscopic
study) point to a serious underestimation of the extent of destructive lung
disease in the general population.
The underestimation of lung disease is largely due to the tremendous
amount of lung reserves (e.g., ~50 billion Type II cells and ~30 billion Type
I cells in ~300 million alveoli). Substantial numbers of these lung cells can
be lost in the absence of overt clinical symptoms or signs, and routine
pulmonary function tests may not become clearly positive until as much as half
of the lung has been irreversibly altered or destroyed. Routine autopsies are
not much more accurate: criteria for recognition and quantitation of the
various kinds of emphysema fall far short of a uniform definition and
application, and this affects most of the autopsies carried out in major
universities and hospitals.
In essence, the major problem of air pollution is a disturbance in the
microecology of the human body, i.e., the diverse cell societies not only of
the lung but of other organs as well. As with the visible ecology, vanishing
species of life is an early and serious part of the adverse effect. With
respect to the lung, the Type I cell is of special concern: the ultrathin
lining it provides for the alveoli is critical to gas exchange. Thus, the
first target of our research effort was a quantitation or inventory of the
Type I cell population.
369
-------
36. PULMONARY EFFECTS OF NITROGEN DIOXIDE IN THE MOUSE Sherwin
METHODS AND RESULTS
Inventory of Type I Cell Population
The methodology we used to count Type I cells is an indirect one, based
on the now well established fact that loss of Type I cells is manifested by a
replacement or hyperplasia of Type II cells. First, we developed a means to
single out the Type II cell for quantitation (Sherwin et al. 1967).
Subsequently, mouse whole lung sections were processed for lactate
dehydrogenase (LDH) responsiveness. We found LDH positive granules of the
Type II cell to be densely packed around the nucleus, whereas LDH positive
granules of macrophages and other lung cells were widely dispersed throughout
the cytoplasm and relatively weakly stained. It should be noted that
intra-alveolar macrophages are, for the most part, removed by frozen section
processing procedure (i.e., intra-alveolar gelatin from perfusion-inflation is
washed out during processing and carries macrophages with it). Furthermore,
interstitial macrophages are not detected by the gray value setting since they
have extremely thin and elongated cytoplasmic processes within the
interstitium, and this minimizes the LDH response.
An image analyzer with a highly sophisticated detection system and field
editor provided a very practical means to obtain accurate large-volume
measurements not only of the number of Type II cells but of their sizes as
well (Margolick et al. 1973; Sherwin et al. 1973). The number of Type II
cells was adjusted to a base line of number of alveoli as reflected in the
370
-------
36. PULMONARY EFFECTS OF NITROGEN DIOXIDE IN THE MOUSE Sherwin
amount of alveolar wall area detected in the same field by the image analyzer.
Additional measurements obtained at the same time were: numbers of Type II
pneumocytes according to size distributions (i.e., by diameters of >8 ym, >10
ym, and >12 ym) , internal surface area of the alveoli, and linear intercepts
of the alveolar walls. Computer processing of the data permitted large-scale
statistical analysis and diverse kinds of data analysis (Sherwin et al.
1979a). Macrophage quantitation and total alveolar counts are currently in
progress.
Most recently, we studied the effects of 0.34 ± 0.02 ppm NO2 on
Swiss-Webster adult male mice. A total of 120 mice were equally divided into
control and exposed groups. Exposed animals received NO2 for 7 h/d, 5 d/week,
for 6 weeks. The lungs of these animals were inflated with gelatin to
approximate the volume of full lung expansion, and frozen sections were
processed for the LDH reaction. Eight whole lung sections were obtained from
the left lung of each animal and numbered according to location in a sagittal
plane running medially (from the lung hilum) to the lateral lung peripheral
area. Four fields were quantitated for each lung section, for a total of 32
fields per animal. The data obtained are summarized in the tables that
follow.
Table 36-1 displays the data from each field treated independently and
analyzed. The findings are:
371
-------
36. PULMONARY EFFECTS OF NITROGEN DIOXIDE IN THE MOUSE
Sherwin
TABLE 36-1. LUNG TISSUE DATA (FIELDS ANALYZED SEPARATELY)
FOR MICE EXPOSED TO NITROGEN DIOXIDE VS. AIR
Standard
Exposure Mean Deviation
Number of Type II Cells (>8 pm)b
N02 314.35 145.46
air 288.05 130.04
Area of Type II Cells (>10 ym)
N02 3129.09 2885.15
air 2722.58 2265.38
Alveolar Wall Area (no sizing) Minus
N02 39166.40 21265.98
air 38044.79 21759.30
Alveolar Internal Surface Area
N02 2747.92 1231.38
air 2622.36 1180.57
Linear Intercepts of Alveolar Walls
NO2 8229.40 3749.41
air 7786.04 3526.56
Alveolar Wall Area (>10 ym) * Number
NO2 82.84 28.82
air 88.61 34.43
Type II Cell Area * Number of Type II
N02 10.91 4.94
air 10.64 3.99
Internal Surface Area
N02 6.88 0.96
air 6.86 1.04
Standard T
Error Value
3 . 58 5 . 39
3.30
71.01 4.42
57.34
Area of Type II Cellsc
523.37 1.48
550.91
30.30 2.95
29.88
92.25 3.45
89.29
of Type II Cells (>10 ym)<3
0.71 -5.16
2.87
Cells
0.12 1.70
0.10
0.02 0.54
0.03
Freauency
Distribution
3211
3210
3209
3211
3210
3207
3206
3210
Probability
(two-tailed) a
p + 0
p -»• 0
0.14(NS)
0.003
0.001
p + 0
0.09
0.59
aNS » not significant.
bper lung field.
cFor wall area with sizing of >10 vm minus area of Type II cells (>10 urn), p = 0.441 (US).
^Alveolar wall area * number of Type II cells = inverse of number of Type II cells per lunq field
adjusted to amount of lung tissue (or number of alveoli); 1 unit «• 3.2 ym2. AH combinations of two
wall areas divided by three Type II cell measurements (>8 ym, >10 ym, >12 urn) provided p values + 0.
All Type II cell measurements (>8 ym, >10 ym, >12 ym) provided p values + 0.
372
-------
36. PULMONARY EFFECTS OF NITROGEN DIOXIDE IN THE MOUSE Sherwin
1. The numbers of Type II cells (>8 pm, >10 ym, and >12 yra in
diameter) were increased for the exposed group (p * 0).
2. The mean areas per field of the Type II cells were greater for
the exposed group (p > 0).
3. The value, alveolar wall area T number of Type II cells per
field, was greater for the control animals (p •»• 0), and
indicated a a greater number of Type II cells for the exposed
animals since this is an inverse of the number function. Also,
the wall area base line was not significantly different between
groups, indicating that the ratio difference is a function of
the number of Type II cells.
Table 36-2 displays the results of a two-way nested analysis based on 60
pairs of animals using a mean field value for each animal. We found an
increase in Type II cells for the exposed group, but this increase was not
statistically significant, apparently because of the great variability between
animals. Also, the analysis assumed a normal distribution (e.g., no high
responders).
Again using animal rather than field comparisons, we ranked the animals
according to upper quartile distribution and subjected the ranking to
chi-square analysis (Table 36-3). A statistically significant (p < 0.025)
difference in the number of Type II cells was noted when the number of Type II
cells was adjusted for alveolar wall area. (This adjustment compensates for
variations of inflation of the lung during processing.) Without the
adjustment, the difference was still apparent but only at the borderline of
statistical significance (p < 0.1).
373
-------
w
BLE 36-2. LUNG TISSUE
Measure
Number of
Type II Cells
(>8 vim)
Number of
Type II Cells
(>12 pm)
Number of
Type II Cells
(>10 um)
Type II Cell
Area
Alveolar Wall Area
Alveolar Wall Area
Type II Cell
Area
Alveolar
Internal
Surface Area
DATA (TWO-WAY NESTED ANALYSIS) FOR MICE EXPOSED
A(G)
G
Residual
A(G)
G
Residual
A(G)
G
Residual
A(G)
G
Residual
A(G)
G
Residual
A(G)
G
Residual
A(G)
G
Residual
aAnimal nested within group; slide
^NS = not significant.
Distribution
Frequency
112
1
3067
112
1
3067
112
1
3067
112
1
3067
112
1
3067
112
1
3067
112
1
3067
and field are
Mean
Square
327,573
323,856
9,939
169,835
122,024
5,148
230,859
217,392
7,101
107,837,392
91,516,928
3,242,773
10,420,183,040
58,261,504
202,329,168
4,300,115,968
26,804,224
89,261,632
3,270,255,616
675,282,944
69,220,512
considered replicates.
TO NITROGEN DIOXIDE VS. AIR
Frequency
Ratio Probability13
32.96 p * 0
0 . 99 NS
32.99 p * 0
0.72 NS
32.51 p + 0
0 . 94 NS
33.25 p -»• 0
0.85 NS
51.50 p -»• 0
0.06 NS
48.17 p -»• 0
0.01 NS
47.24 p -v 0
0.21 NS
W
ffl
V
1
5
Q
t>
a
a
**
"fl
w
•>
i
o
m
3
d
I
o
•H
H
4
H
z
1
§
•J
"A
ft
M
01
tr
(D
|
-------
36. PULMONARY EFFECTS OF NITROGEN DIOXIDE IN THE MOUSE
Sherwin
TABLE 36-3. LUNG TISSUE DATA (CHI-SQUARE ANALYSIS OF CONTINGENCY TABLES)
FOR MICE EXPOSED TO NITROGEN DIOXIDE VS. AIRa
Number of Type II Cells (<8 pm)
L U Total
N02 38
air 47
Total
35
19
10
29
57
57
114
X = 3.75 (0.05 < p < 0.1)b
Area of Type II Cells
NO2
air
Total
2
L
37
46
83
U
18
11
29
Total
57
57
112
2.63 (NS)
Alveolar Wall Area
Alveolar Internal Surface Area
N02
air
Total
2
L
44
41
85
U
13
16
29
Total
57
57
114
NO2
air
Total
2
L
41
44
85
U
16
13
29
Total
57
57
114
0.42 (NS)
0.42 (NS)
Linear Intercepts of Alveolar Walls
L U Total
N02
air
Total
2
X = 1.16
1
40 17
45 12
85 29
(NS)
57
57
114
Alveolar Wall Area *
Number of Type II Cells (>10
0
NO2 48
air 37
Total
85
9
20
29
Total
57
57
114
X = 5.60 (0.01 < p < 0.025)c
1
particular group and membership in the
(p > 0.1).
bSame for Type II cells of >10 ym and >12 ym.
GFor Type II cells of >8 ym: 0.05 < p < 0.1.
375
-------
36. PULMONARY EFFECTS OF NITROGEN DIOXIDE IN THE MOUSE Sherwin
Protein Content of Lung Tissue Following Exposure to Nitrogen Dioxide
In other studies, we used a molecular probe, horseradish peroxidase
(HRP), to measure the protein content of lung tissue from N02~exposed mice.
In brief, intermittent exposures of Swiss-Webster adult male mice to NC>2 at
levels as low as 0.4 ppm for periods of 3 and 6 weeks resulted in a greater
content of HRP in the lungs of exposed animals as quantitated by
polyacrylamide gel electrophoresis (Sherwin et al. 1979b). In three
independent experiments, with two test periods per experiment, exposed animals
showed greater mean HRP values for five of six periods (upper quartile
analysis). Using a two-factor analysis of variance alone, statistically
significant differences were found in three of six periods. Ultrastructural
studies of animals injected with the HRP probe showed HRP to have free access
to the basal lamina of the lungs of both control and exposed animals; studies
of the electron micrographs (incomplete at this writing) have shown no
differences in distribution. Pinocytotic vesicles of both endothelium and
epithelium appear to play major roles in the transport of HRP from the
capillary to the alveolar lumen as part of the bidirectional protein transport
mechanism.
DISCUSSION
While the findings reported here are preliminary, they are in line with
earlier reports showing both Type II cell hyperplasia and protein leakage at
higher levels of NO2 exposure. Other studies are in progress to confirm and
376
-------
36. PULMONARY EFFECTS OF NITROGEN DIOXIDE IN THE MOUSE Sherwii
elaborate upon these preliminary results (see Addendum, below). The question
of reversibility is of special concern. It should be noted that Type II cell
hyperplasia and protein leakage are very early findings and common
denominators in diverse kinds of human lung disease.
REFERENCES
Margolick, J. B., S. P. Azen, and R. P. Sherwin. 1973. An image analyzer
quantitation of Type II pneumocytes. Am. Rev. Resp. Dis., 108:704-707.
Sherwin, R. P., S. Winnick, and R. D. Buckley. 1967. Response of lactic acid
dehydrogenase-positive alveolar cells in the lungs of guinea pigs exposed
to nitric acid. Am. Rev. Resp. Dis., 96:319-323.
Sherwin, R. P., J. B. Margolick, and S. P. Azen. 1973. Hypertrophy of
alveolar wall cells secondary to an air pollutant. A semi-automated
quantitation. Arch. Environ. Health, 26:297-299.
Sherwin, R. P., K. Kuraitis, and V. Richters. 1979a. Type II pneumocyte
hyperplasia and hypertrophy in response to 0.34 ppm nitrogen dioxide: An
image analyzer computer quantitation (abstract). Fed. Proc., 38:1352.
Sherwin, R. P., D. Okimoto, D. Mundy, and J. Bernett. 1979b. Clearance of
horseradish peroxidase in the lungs of mice exposed to an ambient level
of nitrogen dioxide (abstract). Lab. Invest., 40:280.
ADDENDUM
At the time of review of the draft Proceedings, data from another study
had again shown hyperplasia and hypertrophy of Type II cells. In this study,
mice were exposed to 0.3 ppm NO2 for 6 h/d, 5 d/week, for 6 weeks, with tests
at 4 and 10 weeks post-exposure. Also, comparisons of the three test periods
showed impaired mitochondrial growth (size) . The mitochondrial and Type II
cell alterations persisted to the 10-week post-exposure period.
377
-------
37. PULMONARY AND PSYCHOPHYSIOLOGICAL EFFECTS OF OZONE
Steven M. Horvath
Institute of Environmental Stress
University of California
Santa Barbara, CA 93106
INTRODUCTION
This report summarizes some preliminary results of human clinical
exposures to ozone (03) performed at the Institute of Environmental Stress at
the University of California - Santa Barbara. Results of some of the
individual studies have been published elsewhere (Folinsbee et al. 1978, 1980;
Gliner et al. 1980; Horvath et al. 1979).
A special focus of our work is the pulmonary functional response to 03 in
combination with exercise. Another focus is the habituation response to 03
(and individual differences in that response). Finally, recent studies
examine some effects of 03 on the central nervous system.
PULMONARY EFFECTS OF OZONE IN COMBINATION WITH EXERCISE
Exercise at a fixed level of ventilation had dramatic effects on
pulmonary function following 03 exposure. At 0.5 ppm 03, we observed marked
378
-------
37. PULMONARY AND PSYCHOPHYSIOLOGICAL EFFECTS OF OZONE Horvath
pulmonary function decrements immediately following 15 min of exercise
(regardless of the time at which the exercise occurred). Variations in the
amount of decrement reflected individual sensitivities to 03. Prolongation of
the exercise to 30 or 60 min resulted in higher decrements. In addition to
exercise, environmental temperature was implicated in the degree of pulmonary
function decrement.
Other studies examined the capacity for exercise following 03 exposure.
We observed a 10% decrement in maximal exercise capacity following a 2-h
exposure to 0.75 ppm 03 during which subjects exercised intermittently. A
slight reduction in the 03 concentration combined with omission of the
intermittent exercise resulted in no decrement, however.
In all such studies, it is very important to directly measure and control
the subject's level of ventilation. Attempting to infer ventilation from
heart rate, etc. (as in some studies) is not adequate. We performed various
03 exposures (0, 0.1, 0.3, and 0.5 ppm) at various subject ventilation levels
(10, 30, 50, and 70 liters/min) and confirmed that forced vital capacity
(FVC), forced expiratory volume (FEV^, and other parameters decrease markedly
and consistently with the level of ventilation, and more so with the level of
inspired 03.
Knowing the subject's ventilation level during periods of exercise and
rest allowed us to calculate the effective dose in each experimental trial.
Knowing the effective dose permitted us, in turn, to construct straightforward
379
-------
37. PULMONARY AND PSYCHOPHYSIOLOGICAL EFFECTS OF OZONE Horvath
equations that predict pulmonary function decrement as a function of three
variables: ventilation, concentration, and exposure. We are confident of the
validity of these equations: all are significant at p < 0.01.
PULMONARY HABITUATION TO OZONE IN COMBINATION WITH EXERCISE
Habituation exposures were performed under a protocol similar to that
described by Haak (Chapter 31 of this volume). However, our subjects
practiced for 2 weeks prior to exposure, allowing us to obtain base-line
pulmonary function measures. Exposures to filtered air and 03 (0.2, 0.35, and
0.5 ppm) extended for 3 or 5 d. Relative humidity was maintained at 45%, to
simulate the environment of a hot summer day in Los Angeles.
Evaluations of several pulmonary function parameters (including FEV<| and
FVC) revealed a general pattern: a maximum decrement on the first or second
day of exposure followed by a return to near-normal on the third or fourth
day. Despite this general pattern, however, individual subjects displayed
considerable variability in habituation. For example, one subject showed a
dramatic fall in FEV^ on the first day of exposure but had fully recovered by
the second day; no further change was seen. A less sensitive subject showed a
slight change in FEVj on the first day of exposure but no decrement
thereafter.
At present, we are following the highly sensitive subjects by having them
return to the laboratory at various time intervals. Our goals are to
380
-------
37. PULMONARY AND PSYCHOPHYSIOLOGICAL EFFECTS OF OZONE Horvath
determine the length of time for which habituation persists and to identify
the mechanism by which habituation occurs.
PSYCHOPHYSIOLOGICAL STUDIES
Studies of subjective,symptoms in O3-exposed subjects showed a marked
correlation between the number of symptoms and the environmental temperature.
In studies of vigilance task performance, O3~exposed subjects showing
significant pulmonary function decrement in the first hour of exposure showed
a gradual decline in the ability to detect signals during the second hour of
exposure. The most recent studies suggest a correlation between vigilance
performance decrement and 03 concentration.
Preliminary studies of 03 effects on the electroencephalogram (EEC)
suggest a decrement in the usual 8-10 Hz range.
REFERENCES
Folinsbee, L. J., B. L. Drinkwater, J. F. Bedi, and S. M. Horvath. 1978. The
influence of exercise on the pulmonary function changes due to exposure
to low concentrations of ozone. In: Environmental Stress: Individual
Human Adaptations (L. J. Folinsbee, J. A. Wagner, J. F. Borgia, B. L.
Drinkwater, J. A. Gliner, and J. F. Bedi, eds.). Academic Press, New
York, pp. 125-145.
Folinsbee, L. J., J. F. Bedi, and S. M. Horvath. 1980. Respiratory responses
in humans repeatedly exposed to low concentrations of ozone. Am. Rev.
Resp. Dis., 121:431-439.
381
-------
37. PULMONARY AND PSYCHOPHYSIOLOGICAL EFFECTS OF OZONE Horvath
Gliner, J. A., S. M. Horvath, R. A. Sorich, and J. Hanley. 1980. Psychomotor
performance during ozone exposure: Spectral and discriminant function
analysis of EEG. Aviat. Space Environ. Med., 51:344-351.
Horvath, S. M., J. A. Gliner, and J. A. Matsen-Twisdale. 1979. Pulmonary
function and maximum exercise response following acute ozone exposure.
Aviat. Space Environ. Med., 50:901-905.
WORKSHOP COMMENTARY
R. K. Wolff; Regarding the temperature effects, does the subject's
ventilation go up at a higher temperature? Why do you think temperature has
that effect?
S. M. Horvath; Body temperature goes up. At a higher temperature ventilation
increases, so we lower the work rate in order to maintain the ventilation
constant. In our view, to look at the effects of 03 on the pulmonary
functions of man requires at least a constant ventilation. If you don't have
a constant ventilation, all you're looking at is variability. As a
consequence, there is less variability in our obtained data simply because we
spend so much time looking at and measuring the ventilation. We know the
maximum work capacity of each subject and can predict within 5% what level of
work will maintain a certain ventilation. We continually monitor the
ventilation and lower or raise the work rate to maintain the ventilation.
Question; Does your "effective dose" refer to absorbed 03 or inspired 03?
S. M. Horvath; Inspired 03. We were originally interested in measuring how
much 03 is really taken up. The exposure level does not tell you what amount
actually gets into the lungs. If the Air Resources Board can provide the
funding, we will obtain that information. In the meantime, we still don't
know how much actual 03 is taken up by the individual. We do know how much is
presented, and the prediction equations are based entirely on that.
382
-------
38. EPIDBMIOLOGIC STUDIES
OF OXIDANT HEALTH EFFECTS IN THE LOS ANGELES AREA
Roger Detels
School of Public Health
University of California
Los Angeles, CA 90024
INTRODUCTION
Since 1972, the University of California - Los Angeles (UCLA) and the Los
Angeles County Lung Association have collaborated on studies of the
relationship between chronic exposure to various types and levels of air
pollution and lung function test performance. The primary objective of these
studies is to determine if changes in lung function test performance over time
can be correlated to residence in areas exposed to high and moderate levels of
photochemical oxidants. No previously reported studies have attempted a
similar correlation in a large, free-living population; thus, our project may
be definitive in establishing the magnitude of such correlation (if any).
Coinvestigators for the project include this author as well as Dr. Stanley
Rokaw, Dr. Frank Massey, Dr. Donald Tashkin, and Mrs. Anne Colson.
To date, we have completed base-line testing in four areas and retesting
in one of those four areas. This report presents some preliminary results of
383
-------
38. EPIDEMIOLOGIC STUDIES IN LOS ANGELES Detels
a cross-sectional study completed for the years 1972-1977. There are many
problems associated with the cross-sectional study design; these can be
resolved, in part, by completing longitudinal follow-up of the cohorts already
established.
STUDY SITES
Four study sites were selected to provide a range of pollution exposures.
Glendora, located in the East San Gabriel Valley, is exposed to high levels of
photochemical oxidants. Burbank, in the East San Fernando Valley, is exposed
to moderate levels of photochemical oxidants. Lancaster, located two mountain
ranges north of downtown Los Angeles, is exposed to low levels of all ambient
air pollutants. The Long Beach site is exposed to low levels of photochemical
oxidants but relatively high levels of sulfur dioxide, particulates, and
(presumably) hydrocarbons.
These four sites were also selected for contiguity to monitoring stations
of the Southern California Air Quality Management District, which can provide
continuous measurements of levels of selected air pollutants. Monitoring
stations are located immediately adjacent to or within the four study sites
with the exception of Glendora, where the monitoring station at Azusa (~2-3 km
upwind of the study site) is used to estimate exposures.
The monitoring data we compared consisted of average annual means of
daily maximum hourly average concentrations of selected air pollutants over
384
-------
38. EPIDEMIOLOGIC STUDIES IN LOS ANGELES Detels
the six study years. The Azusa station had the highest levels of
photochemical oxidants. Lancaster showed oxidant levels that were higher than
originally expected but lower than the levels in Burbank and Azusa. Long
Beach had the lowest levels of oxidants but the highest levels of sulfur
dioxide. Nitrogen dioxide levels were highest for Long Beach and Burbank.
Particulates were high in Azusa. Particulates were not measured in Long
Beach, but isopleths indicate that Long Beach had the highest particulate
levels. Hydrocarbon measurements were a problem because they were not made in
Long Beach until 1977. We had hoped that one result of the planned joint
ARB/EPA studies would be better measurements of hydrocarbons and particulates.
PULMONARY FUNCTION TESTING
Subject Recruitment
At the beginning of each test period, articles concerning the research
program appeared in the local newspapers, and letters of introduction were
sent to residents of the target areas sequentially so as to permit lung
function testing within one week of notification. Telephone contacts and
construction of household rosters were done by individuals recruited from the
communities in which the testing was being carried out.
The percentage of households enumerated ranged from 84% in Lancaster to
98% in Glendora. The proportion of residents completing lung function testing
ranged from 70% in Burbank to 79% in Lancaster; the actual number completing
385
-------
38. EPIDEMIOLOGIC STUDIES IN LOS ANGELES Petals
lung function testing ranged from 3403 in Glendora to 4509 in Lancaster. An
additional 1 to 8% of residents of the four areas completed a respiratory
questionnaire only. The demographic characteristics of the nonrespondents
were similar among the four study areas.
The prevalence of residents with a history of asthma, bronchitis, or
emphysema ranged from 10 to 13% in the four communities and was highest in
Lancaster, the study area exposed to the lowest levels of pollutants.
Lung function testing was completed only on residents who were seven
years of age and older. We felt that children younger than seven years of age
would not be able to follow the procedures. Also, it was necessary that each
subject be able to walk up the steps to the mobile lung research laboratory.
Individuals who were so infirm as to not be able to walk up to the laboratory
could not be tested, although we were very willing to assist them if possible.
Testing Protocol
Prior to pulmonary function testing, subjects underwent an interview
schedule administered in a separate trailer. Lung function tests were
administered in a mobile lung research laboratory. These tests included:
forced expiratory spirometry with computer recording of the entire flow volume
curve; the single breath nitrogen test, including the change in level of
nitrogen between 750 and 1250 cm3 of expired air as well as the closing volume
fraction; and body plethysmography, including determination of thoracic gas
386
-------
38. EPIDEMIOLOGIC STUDIES IN LOS ANGELES Detels
volume, functional residual capacity, and specific airway resistance. (We
were quite surprised at the very low refusal rate for plethysmography: <1%.)
Each spirometric flow volume curve was recorded directly onto tape so that we
could select best breath, best two breaths, etc. using computer algorithms.
The results of the other two tests were recorded by hand.
Quality Assurance
Several strategies were used to determine the reliability of the lung
function testing: (1) every tenth participant was immediately retested; (2)
100 residents of each area were retested three times over the course of a
year; (3) a 3% probability sample of individuals completing lung function
testing was reexamined using more intensive techniques at the UCLA pulmonary
function laboratories. In addition, instruments were calibrated and
recalibrated several times during each test day. Reliability for the
spirometric indices was excellent; for the plethysmographic indices—fair; and
for the single breath nitrogen indices—poor.
PRELIMINARY RESULTS
We have completed a preliminary analysis of the following parameters:
cough and sputum, any symptom, forced expiratory volume (FEV,), peak flow, and
closing volume fraction. In order to minimize the number of variables between
study groups, the results presented here are limited to white residents who
did not have Spanish surnames and who had never changed residence or
387
-------
38. EPIDEMIOLOGIC STUDIES IN LOS ANGELES Detels
occupation because of a respiratory problem. Further, these results are only
for individuals who were either current smokers or who had never smoked, and
includes only individuals who were 25 to 59 years of age. Analyses of results
for younger subjects, older subjects, and former smokers are ongoing.
Among residents who had never smoked, the worst test values were most
frequent in Glendora, the area exposed to the highest levels of photochemical
oxidants. The frequency of worst test values among participants from Burbank
(the area exposed to moderate levels of oxidants) was low and similar to the
frequency among residents of Lancaster (the area exposed to low levels of all
pollutants). The second highest frequency of worst test values occurred among
residents of Long Beach (the area exposed to relatively high levels of sulfur
dioxide, particulates, and possibly hydrocarbons).
Among current smokers, the distribution of worst test values was similar
to the distribution among never-smokers. Two of the tests that presumably
measure predominantly small airway function (AN27E-n_i9C-n and the maximum flow
at low lung volumes) correlated with smoking but not with residence.
Differences in results (any test) between the best and the worst study areas
were greater than differences (same test) between never-smokers and current
smokers.
The results of the spirometric tests in the Burbank participants were
closer to those in the Lancaster participants than in the Glendora
participants, suggesting the possibility of a nonlinear dose-response curve or
388
-------
38. EPIDEMIOLOGIC STUDIES IN LOS ANGELES Detels
a threshold value for oxidants lying somewhere between the levels occurring in
Burbank and the levels occurring in Glendora. Another possible explanation is
that more affected residents of the Burbank study area had moved away before
the program started. This possibility will be resolved by longitudinal
follow-up of the established cohorts.
In summary, the highest frequency of worst test values occurred among
participants in the area exposed to the highest levels of photochemical
oxidants, and the lowest frequency of worst test values occurred among
participants in the area exposed to the lowest levels of all ambient air
pollutants.
DISCUSSION
These results should not be viewed as conclusive evidence of a
relationship between chronic exposure to photochemical oxidants and impairment
of lung function. Although most of the biases we have been able to identify
would tend to decrease observed differences in comparison to true differences
between the communities, cross-sectional studies cannot account for all of the
variables which may confound these comparisons. Thus, it is essential that we
pursue the original objective of these studies: to observe whether changes in
lung function test performance are commensurate with the levels of the various
pollutants measured concurrently over the five-year test interval.
389
-------
38. EPIDEMIOLOGIC STUDIES IN LOS ANGELES Detels
WORKSHOP COMMENTARY
Comment; Case-study epidemiology is always a tough thing, and you left
yourself a good caveat. Still, with regard to oxidants, should you be
measuring people where they live or where they work? Shouldn't you be looking
at people who work outdoors—telephone linemen, ditch diggers, and so on—to
get some idea of what exposure to those ambient concentrations means? Spizer
and Ferris and others have found very large differences between ambient levels
of photochemical oxidants and levels of photochemical oxidants indoors where,
unfortunately, most of us spend our busy days.
And how about the correlation with income that is seen in the
epidemiology of cancer in large cities (high income, low cancer)? Does this
reflect the fact that people who don't have much money always live down by the
bayou, down in the valley, next to the railroad tracks, close to the factory?
It's never clear to me which is cause and which is effect, but this is
something for you to consider here. I suppose that you have considered it.
R. Detels; Yes, we have. First, with regard to socioeconomic status, I'm
afraid I didn't make it clear that we matched these communities on mean
income, on cost of housing, and on racial distribution. That's why they are
predominantly white. If we had taken any other kind of a mix it would have
been impossible to match them and have different air pollution exposures.
With regard to place of residence vs. place of employment, we took down
histories of where they live and where they work, how much time is spent in
commuting, etc., and we'll be looking at that. However, Lancaster is ~60-70
mi from downtown Los Angeles. Therefore, the majority of residents do not
make that commute. The same is not true for the people from Long Beach,
Burbank, and Glendora, which are the three polluted areas. People from
Glendora only have "one way to go," and that's to reduce their pollutant
exposure by commuting. Therefore, it is our impression that the biases that
would be introduced by commuting to an area of different pollutant exposure
would tend to diminish the observed differences from the true differences.
However, we will be looking at subgroups on the basis of commuting patterns.
We also take occupation into consideration. We know that the
distribution of occupations is similar in these four communities.
Question; What do ambient values have to do with the inside of a Lockheed or
Douglas plant?
R. Detels; I can't answer that. But I think that if one sees differences in
breathing parameters between communities, then one has to hypothesize
differences—occupational or other exposures—if one is unwilling to accept
residence as the important variable. We found no differences in occupations
that could account for these differences.
390
-------
38. EPIDEMIOLOGIC STUDIES IN LOS ANGELES Detels
We haven't looked at the difference between .indoor and outdoor exposures.
Frankly, that's difficult in a residential type of study. The object of this
study is to look at whether breathing ability is associated with air pollution
levels in the area of residence. We realize that this objective has some
shortcomings; if we see no difference then we can't say much. However, if we
see a difference, the argument against a residential effect is a little harder
to make.
Question; Would you explain the kinds of exposure data that were available to
you?
R. Detels; Data were available for a seven-year period from continuously
operating stations of the Southern California Air Quality Management District.
Question; And you had access to the exposure data from that time up to the
current time? At what frequency?
R. Detels; The Southern California Air Quality Management District's
monitoring is continuous. The Air Resources Board and the Southern California
Air Quality Management District have kindly provided us with their measurement
ti»>es so that we can select any parameter we want. We picked the mean of the
daily maximum. However, one can select other parameters. Dr. Yuji Horie of
the Technical Services Division has looked at exposure in terms of the number
of days over the air quality standards for these areas. That is another way
of trying to document pollution differences. Using that technique, the
differences are even more dramatic. We have data dating back to the early
1960's.
Question; Have you done any analysis in terms of length of residence?
R. Detels; We have that information. We deliberately picked communities
having a high proportion of individuals who had lived there five years
previously. We did not divide it into individual years of exposure in that
community.
When you stop to think about it, if somebody has lived in Burbank for 5
years but spent the previous 20 years in downtown New York City, how does he
compare to somebody who has spent 5 years in Burbank and the previous 20 years
in Apple Valley, California? I don't know. We have the information, but we
don't know how to get at that.
Question; For your follow-up, do you plan to repeat all the same
measurements? We have a concern, for example, in terms of the single breath
nitrogen test. You indicated that there was very poor reliability.
R. Detels: That's correct. First of all, it's a gratuitous measurement. The
instrumentation is on the test unit. To take it requires an extra 3-4 min of
the participant's time and little else. Second of all, we did see •"
association with closing volume in the four areas and we dxd see a correlation
391
-------
38. EPIDEMIOLOGIC STUDIES IN LOS ANGELES Detels
with cigarette smoking. If we're going to look at these tests to see whether
they're reasonable predictors (one of the subobjectives of this study), it's
important to see that the test itself is consistent in an individual over a
five-year course. If we find somebody with a low closing volume (which we
think may be an early sign of respiratory disease) in 1973 and then we find
him with a decreased FEV-| but a good closing volume five years later, I would
question whether the closing volume is a very good predictor. I would think
that the closing volume measurement must at least be consistent with what we
saw in 1973.
If we had to go out and buy all of that instrumentation from scratch in
order to do the retesting, there would be a very reasonable question raised.
But since it's gratuitous and we can get that consistency estimate, I think we
would be somewhat remiss not to do it.
Comment; That's my point. It may be consistent but it may not be reliable.
R. Detels; In my opinion, in order to be consistent it must be reliable. It
may not be sensitive, it may not be specific, but it must be reliable.
F. J. Miller; What percent of response did you get in your follow-up?
R. Detels; This is one of our urgent concerns. In the follow-up for Burbank,
we can account for ~84% of the cohorts. For ~16% we have demographic
information and base-line test results but no retest information. Our problem
is that people moved out of Burbank. Unfortunately, we can't bring the mobile
laboratory to these people because they are scattered over a wide area.
Therefore, we have to ask them to drive several hours, if they live in
Southern California, to be retested. Retesting of individuals who have moved
further away is completely unfeasible. Most of our refusals were, in fact,
from people who had moved.
To date, we have completed ~75% of the testing in Lancaster. At this
point, it appears that the response rate will be better than in Burbank.
The response rate in the remaining two areas will be better because we
initiated the annual follow-up about a year and a half after the original
Lancaster cohort was done. Unfortunately, for two years we had no follow-up
in Burbank; this may be part of our problem. I want to emphasize that, the
longer we delay, the more people will move out and the lower the response rate
will be.
F. J. Miller; Are they moving from the area or are they just not responding,
to avoid the bother of testing? Also, you can minimize your differences by
the fact that they choose to move out of the polluted area.
R. Detels; We send a letter to each individual who has moved away. If there
is no response, we send a second letter. Finally, we telephone the
individual. One of the major questions we ask is: "Why did you move?" So,
392
-------
38. EPIDEMIC-LOGIC STUDIES IN LOS ANGELES ______^_ Detels
if they moved for respiratory reasons, we have that information. In addition,
we will compare the lung function test parameters at base line for those who
moved out vs. those who remained. We want to determine if there are
differences between these two groups. We did that using FEV1? there were no
large differences. We are also looking at the demographic characteristics,
since we have complete questionnaires on all subjects.
Thus, we have better documentation of the characteristics of our retest
nonrespondents than most studies do. We know a lot about their respiratory
physiology at base line, and we know a lot about their demographic
characteristics.
Comment; Are there sufficient numbers of people who move from an area because
of respiratory problems to make it worthwhile to do a correlation of their
response? In other words, let's assume that certain people can adapt to
photochemicals in the area and others cannot. Those who cannot adapt, move;
they would be a subgroup that might be very interesting to study.
R. Detels; That's true. As long as they move within the general Southern
California area, we have that potential. We could either bring the laboratory
to them or bring them to the laboratory.
W. Frietsch; Could you briefly describe this study's origin? What were the
major objectives?
R. Detels; There were two reasons the study came about. First, I moved into
the Los Angeles area about 1971 and decided that if I was going to breathe it,
I might as well study it. About that time, the National Heart Lung Blood
Institute issued a request for proposal for a population study of areas
exposed to air pollution. We responded with a proposal to look for the
changes in lung function parameters in groups of people exposed to different
levels of photochemical oxidants and other pollutants over a five-year period.
That remains the major objective. Now, we have a. whole host of subobjectives.
But I want to emphasize that it was not designed as a cross-sectional study.
We were well aware of the pitfalls of the cross-sectional study. Still, that
remains the best way to form a population-based cohort.
F. J. Miller; From all the caveats you have mentioned (in terms of analyses
that could be done, etc.), my general reaction is that there are a couple of
years of analyses to perform before you can really consider an expanded study
or something on the order of "goals." It appears that you have a tremendous
amount of data that still need to be interpreted and put in perspective for an
effective study.
R. Detels; You're absolutely right. Given the best of all possible worlds, I
would have tested all these communities at the same time. Then I would have
"put them in the icebox" and said, "Don't anybody move; it's going to take me
two years to go through the analyses I'd like to do." Such analyses would not
be associated with the ultimate objective, and none would overcome the
393
-------
38. EPIDEMIOLOGIC STUDIES IN LOS ANGELES Petals
problems of a cross-sectional study. But such analyses would involve very
interesting cross-sectional studies: comparisons of different areas,
comparisons of different tests, comparisons of different age groups,
comparisons of different analytical strategies, and so on. The problem is
that subjects continue to move. The longer we delay the fewer people will be
available for retesting.
The second problem is that two measurements taken over five years and
used to estimate mean annual change cannot be compared to a mean estimated
from two measurements taken over seven years. The actual rate of change in
lung function parameters may not be constant over time. Such a comparison
assumes that the rate of change is linear. I'm almost sure that it is in fact
not linear. For instance, one of the groups in which we're most interested
are those who are progressing from young adulthood to adulthood. Their
changes in lung function are certainly not linear. And I'm sure there are
other groups in which the changes are not linear.
Limited funding has forced us to make compromises. When cuts were
necessary, those cuts were made in analysis. You cannot cut the collection of
field data. Without collection of data there is no study.
We have an opportunity to do all the analyses, but if we do only the
analyses we'll lose the cohort. If we lose the cohort, we're stuck with
another cross-sectional study and all of its attendant problems. So, although
I would prefer to stop and do two or three years of thorough analyses, I don't
think we have that option. We did not have that option in the past because we
had to focus on the collection of quality data in which we could have
confidence. Had we diverted the funding from collection of data to analyses,
I'm not sure that we would have obtained anything worthwhile to analyze.
394
-------
39. PROPOSED SCIENTIFIC PROGRAM OF THE COOPERATIVE STUDY
OF OXIDANT HEALTH EFFECTS IN THE LOS ANGELES AREA
Stanley V. Dawson
Research Division
California Air Resources Board
1102 Q Street
Post Office Box 2815
Sacramento, CA 95812
INTRODUCTION
This report describes a program to further our knowledge of oxidant
effects on human health in the Los Angeles area. The primary aim is to reduce
uncertainty about health implications of ambient air quality standards.
Funded jointly by the California Air Resources Board (ARB) and EPA, this
program consists of an integrated set of human and animal studies to be
conducted cooperatively over the next five years. Table 39-1 summarizes these
various studies. Most of the initial work will use established methods, but
innovative approaches will be considered in the program's evolution.
EPIDEMIOLOGIC STUDIES
The key question of how variously polluted atmospheres relate to measured
health effects in populations is being addressed in a large population study
395
-------
TABLE 39-1. COMPONENT INVESTIGATIONS
U)
vo
Atmosphere
Group Studied
End Point(s)
Investigator,
Institution
8
ambient
0.3 ppm 03
pollutant mixtures
(oxidant + sulfurous)
0.5 ppm 03
2.0 ppm 03
0.2-1.2 ppm 03
0.2-0.8 ppm 03
census tracts (4)
emergency room walk-in's
asthmatics
human sample
exercising rats
exercising dogs
young dogs
young rats
young monkeys
monkeys
lung function, cruestionnaire
chest complaints
medication, lung function
lung function
clearance + morphometry
lung function + morphometry
terminal lung function
+ morphometry
extensive morphometrv
(ongoing function)
macrophage motility
(ongoing structure)
R. Detels
UCLA
J. Hackney
Rancho Los Amigo Hospital
T. Crocker
UCI
R. Phelen
UCI
W. Tyler
UCD
L. Schwartz
UCD
M
S
H
z
-------
39. OXIDANT HEALTH EFFECTS IN LOS ANGELES
by R. Detels and associates at the University of California - Los Angeles
(UCLA) (see Chapter 38 of this volume). In this study, one full round of
pulmonary function testing and questionnaires has been completed in four
census tracts (4000-6000 population each, aged 7 and up) located in variously
polluted regions in the Los Angeles area. Follow-up studies of individuals
originally tested have been completed (84% retention rate) in Burbank (high
oxidant, medium sulfur dioxide (802)) and nearly completed in Lancaster (low
oxidant, lowest 802). Follow-up studies will begin in Long Beach (lowest
oxidant, highest 802) in spring 1980 and in Glendora (highest oxidant, medium
SO2) in 1981.
The initial part of this study, a cross section of the four census
tracts, has already yielded some useful results. Pulmonary function test
results were more below normal in the more "smoggy" areas, but the
relationship between test results and pollutants was not simple. Because of
the potential for different population conditions in different areas, such
cross-sectional studies are thought to be of limited validity in establishing
the relationship between mixed pollutants and health effects. The serial or
follow-up study of pulmonary function decrement in given individuals over time
will permit a more direct assessment of pollutant effects on populations in
the four tracts. Preliminary results of the follow-up study show greater
pulmonary function decrement in Burbank than in Lancaster, an area of
substantially less air pollution of every measured species.
397
-------
39. OXIDANT HEALTH EFFECTS IN LOS ANGELES Dawson
The UCLA group plans two additional smaller-scale epidemiologic studies.
Tentatively, the first will involve a reporting network for chest complaints
by emergency room "walk-in's." The other will be a comparative asthma panel
study in districts that have distinctly different oxidant levels but are
matched in other ways. The use of medication dispensers that can provide
assured counts and the use of quantitative self-testing will be explored.
THE INTEGRATED PROGRAM
In developing an integrated program that starts with these epidemiologic
studies, we sought projects that might answer questions not addressed in
purely epidemiologic studies, and we sought to add new exposure information to
sharpen the epidemiologic studies themselves. One limitation of epidemiologic
studies is the lack of confidence that one can place on the effect of a
particular level of a particular pollutant, while a strength of the Betels
cohort study is the determination of actual rate of lung function loss. An
exposure chamber study, on the other hand, can provide a clear response to a
particular level of pollutant, but the long-term significance of that
immediate response is open to question. Our integrated program proposes to
directly link these two types of studies by drawing samples for an exposure
study from Detels1 well characterized epidemiologic cohorts subsequent to
follow-up tests of pulmonary function.
As a further aid in interpreting the epidemiologic studies, four animal
studies are planned. Using rats and dogs under carefully controlled exposure
398
-------
39. OXIDANT HEALTH EFFECTS IN LOS ANGELES Dawson
conditions, the first study will examine the lung damage and impairment
attributable to two important environmental influences: exercise and
pollutant mixtures (both key factors in Los Angeles). The second and third
animal studies will address questions for which the Detels data are not yet
analyzed: namely, effects on the very young. Experiments on dogs, rats, and
monkeys will examine the extent of lung damage due to ozone (03) exposure
during the development of the lung, with special reference to the stage of
alveoli formation. Lung damage will be assessed morphometrically in all three
of the animal studies—a difficult approach in humans, although a comparative
post-mortem study on human adults is a future possibility. The fourth animal
study will explore a mechanism that is considered to be a likely key to the
accelerated loss of lung function that may occur in polluted regions of Los
Angeles as well as to the damage observed in terminal bronchioles of animals
exposed to as little as 0.2 ppm O3. The hypothesis is that low levels of O3
impair alveolar macrophage motility and that stagnant macrophages release
enzymes which attack the unciliated bronchiolar epithelium. The approach of
the study is to lavage (wash) and then measure the motility of macrophages
from certain lung segments of O3-exposed monkeys. The lavage technique
carries future potential for assessment of the human lung condition relative
to pollution exposure.
399
-------
39. OXIDANT HEALTH EFFECTS IN LOS ANGELES Dawson
IMPROVEMENT OF POPULATION EXPOSURE ESTIMATES
Some program resources have been set aside for exploring ways to improve
exposure estimates in the Detels study. This work will be conducted early in
the overall program.
The first aspect of this project is to ensure that the basic data from
the monitoring stations are as comprehensive and complete as possible. A few
standard pollutants were not actually measured during the study, though all
standard pollutants are now scheduled for routine monitoring in the two
remaining tracts. In addition, there is an important opportunity to measure
fine particles in all four stations near the tracts.
This effort involves two further aspects. One is the question of how the
actual exposures of individual subjects might have differed from the values
reported at the monitoring stations. Measurements and/or modeling of local
sources and indoor variations may be helpful in answering this common
epidemiologic question. The other question is of how individuals may have
been temporarily affected by pollutants at the time of the pulmonary function
measurements.
HUMAN LABORATORY EXPOSURES
The primary exposure of subjects from epidemiologic cohorts will be
performed by J. D. Hackney and associates at Rancho Los Amigos Hospital in
400
-------
39. OXIDANT HEALTH EFFECTS IN LOS ANGELES
Downey, California. Fifty subjects from the Burbank cohort will be exposed to
0.3 ppm 03 for ~2 h. The preliminary design calls for half of these subjects
to be drawn from the group that experienced the most pulmonary function loss
and for the other half to be drawn from the group that experienced the least
loss. Testing of the immediate response to 03 will determine the significance
of any correlation with loss of function over five years.
Should a significant positive correlation between long-term pulmonary
function loss and short-term O3 sensitivity be found, two contrasting
explanations could be offered. One explanation would be that lungs which are
a priori most responsive to O3 insult in the short term are also most
susceptible to long-term function loss. The other would be that existing
pulmonary function loss predisposes the lung to short-term O3 sensitivity.
Either explanation would carry importance in standard-setting, and future
studies (e.g., similar studies of groups with less O3 exposure or longitudinal
studies of individuals with high and low short-term 03 sensitivity) might be
able to distinguish between them. Related studies by Hackney and others
support the possibility of detecting relatively small but significant O3
responses and intergroup differences in response. However, the likelihood of
successfully detecting differences between groups showing different long-term
pulmonary function loss is difficult to predict. Even if the attempt to
distinguish between groups is not successful, considerable additional
information on responses to "worst case ambient" 03 levels will have been
obtained on a group of Southern California residents who, though still
401
-------
39. OXIDANT HEALTH EFFECTS IN LOS ANGELES Dawson
self-selected, bear a well defined relationship to a typical middle-income
community population.
Because of the phenomenon of desensitization or adaptation to 03
exposure, it seems likely that the clearest test of immediate 03
responsiveness of these subjects would be performed relative to an external
atmosphere that is as free of pollutant as possible. Thus, special
precautions will be taken to reduce effects of environmental background: we
intend to work with our subjects during seasons of low smog and to estimate
their environmental pollutant exposures. In subsequent years, it may be
possible to arrange for a portable exposure chamber for the two most distant
census tracts. Ultimately, it may still be worth transporting subjects to a
cleaner environment (e.g., Santa Barbara) for several days of tests. General
animal studies of the phenomenon of desensitization and of the correlation
between immediate 03 responsiveness and rate of pulmonary function loss during
chronic exposure would provide useful correlates to these human studies.
Subjects who participate in the human laboratory exposures could be
measured in various new ways with no additional inconvenience to them. A
proportion of program resources support pilot studies of innovative ideas that
appear likely to yield results. With A. Richters at the University of
Southern California we are considering exploratory studies on the blood
samples that will (in any case) be obtained. Leukocyte population counts and
leukocyte function tests of both T cells and B cells might reveal differences
between groups. In addition, with J. F. Mead at UCLA we are considering
402
-------
39. OXIDANT HEALTH EFFECTS IN LOS ANGELES __ Dawson
exploratory studies of exhalate to measure pentane concentration as a marker
of peroxidation in the lung lining.
ANIMAL LABORATORY STUDIES
A study to examine the effect of exercise on dogs and rats exposed to
pollutant mixtures for short periods is planned by T. Crocker and associates
at the University of California - Irvine (UCI). That laboratory has developed
exposure chambers which permit careful control of the aging of atmospheric
pollutant mixtures. In the initial tests, the atmospheres will consist of:
1. Toxic hydrocarbons produced by photochemical reactions,
including formaldehyde, acrolein, or peroxyacetyl nitrate
alone at 0.1-2 ppm
2. 03 alone at 0.2-0.6 ppm
3. Particulate ammonium sulfate at 1 mg/m3 plus SO2 at 0.5-5 ppm
4. Combinations of (1.) and (2.) and of (2.) and (3.)
The hydrocarbons and SO2 are water soluble; hence, combined
hydrocarbon-particulate or SO2-particulate atmospheres will be studied at low
and high humidities. Beagle dogs will be exercised on a refrigerated
treadmill; pulmonary function will be monitored by face mask, esophageal
balloon, expired gas, and carotid blood gases. Rats will also be exercised;
the post-exposure ability to clear inhaled particles from the lung will be
measured. Detailed lung morphometric measurements will include
characterization of lesions and of cell replication.
403
-------
39. OXIDANT HEALTH EFFECTS IN LOS ANGELES Dawson
Studies of lung development are planned initially for 03 exposures only.
R. Phelan at UCI will expose beagle puppies to 2.0, 0.5, and 0 ppm O3 for 4
h/d for 5 d at an age of active development, based on a protocol successfully
developed for 1.0 ppm. The facility for housing the dogs maintains an
exceptionally pure environment. After development is complete, the lungs will
be subjected to pulmonary function tests and detailed morphometry using
special methods for preparing the microscopic sections. Mean linear intercept
will be measured. W. Tyler at the University of California - Davis (UCD)
plans to augment and continue rat and monkey studies that are in progress.
The rat studies are both episodic (0.8 ppm 03 for 3 d during various stages of
development) and chronic (1.2, 0.8, and 0.2 ppm). Measurements by the most
advanced morphoraetric methods will include internal surface area, capillary
volume and surface area, and mean thickness of the air/blood barrier. The
monkey studies are chronic, involve an exposure level of 0.8 ppm 03, and start
at 6 mo of age? measures of pulmonary function and extensive morphometry will
be completed (as in the rats).
L. Schwartz of UCD will perform lavage of alveolar macrophages on healthy
monkeys exposed to 0.2-0.8 ppm 03. The macrophages produced by lavage will be
examined in a number of ways; the primary approach will be to measure
motility- The response to lung lining material and chemotactants will receive
attention. Data on macrophage structural alteration (obtained in ongoing
studies) will be compared to the observed functional changes.
404
-------
39. OXIDANT HEALTH EFFECTS IN LOS ANGELES
Dawson
OVERALL ASPECTS
An unusual feature of this program is the inclusion of a statistical
working group both as an aid to investigators and as a management mechanism.
Nine distinguished statisticians from Stanford University, the University of
California - Berkeley, and UCLA have agreed to join a Design and Evaluation
Committee coordinated by L. Breiman, an independent statistical consultant
based in Santa Monica. Operating under very specific guidelines, the
Committee will be responsible for the following functions:
1. To critically review all experimental design for statistical
integrity and power, including design modification at the
proposal stage and as necessary thereafter to ensure
statistical integrity and power;
2. To critically review the statistical analysis of all
experimental data;
3. To actively assist in the analysis of all data;
4. To develop a continuing evaluation of the statistical
implications of all results bearing on oxidant
standard-setting.
In addition, the ARB program management staff will be responsible not
only for coordinating the investigations but also for synthesizing the
results. Some of this synthesis—especially that involving extrapolation of
animal results to man—may employ mathematical modeling approaches as well as
biological concepts. The most useful overall contribution of the synthesis
may be to obtain, in connection with inputs from the statisticians, a clearer
405
-------
39. OXIDANT HEALTH EFFECTS IN LOS ANGELES v Dawson
definition of the degree of confidence that can be attached to statements of
oxidant effects at ambient levels.
WORKSHOP COMMENTARY
Question; Could you outline the approximate resources that would go to each
of the studies? Also, how did you arrive at that emphasis in terms of meeting
study objectives?
S. V. Dawson; To answer the last part of your question, we looked at the
merits of each study and the relevance of each in combination with the others.
To answer the first part, with respect to [Table 39-1], the Schwartz, Tyler,
and Phelan studies are each funded in the range of $70,000 to $90,000 a year;
the Crocker study is funded at $200,000; the Hackney study is funded at
$150,000; and the Detels study is funded at $530,000.
Comment; In my understanding, the original question to be answered by the
study concerned the health effects that occur in Los Angeles, since in Los
Angeles the standards (particularly for 03) are fairly routinely exceeded.
After all the negotiations, is that still the major objective of the study?
Does the design really focus in on that question?
S. V. Dawson; I would certainly claim that we're looking for health effects
caused particularly by levels of 03 and oxidant in Los Angeles. All of these
projects were screened for relevance to that question. It's true that, in
terms of pollutant, four of the studies address only 03. In another exposure
experiment, 03 is added on to other things (the sulfurous compounds and the
rest of the total oxidants). We have had trouble, perhaps, distinguishing
between 03 effects and oxidant effects. This is at least a step in that
direction.
Question; Were the studies assembled in sequential order? Did you really
look at the major objective and design a study to achieve that, or did you
more or less pick up pieces of different projects and assemble them?
S. V. Dawson; From the statements of interest we received, we assembled those
pieces that were most relevant to our objective and that were of sufficient
scientific merit. Several of the submissions were substantially modified; we
went pretty far in that direction to actually shape the thing but, as you
know, investigators have certain interests and you can only move them a bit.
To a certain extent, one has to take what one can get in the real world.
Question; What documentation for your presentation exists now?
406
-------
39. OXIDANT HEALTH EFFECTS IN LOS ANGELES
Dawson
S. V. Dawson; The basic scientific program is outlined just as presented
here.
J. A. Graham; I have three or four general questions. First, I don't
understand the rationale for doing two studies on the young until results are
available from at least one of them. Obviously, it's nice to work with a
number of animal species, but I don't see the rationale for doing all of this
work with O3 and no work with NO2, for example. I don't see the rationale for
duplication without prior data.
S. V. Dawson; Your point is well taken, but the investigators do have some
interesting results. Phelan has some very interesting results with that 4%
shift. Tyler has some very significant results on the alveolar-volume-to-
body-weight ratio, and his proposal is to do detailed morphometry to look at
that. So, as far as rats and dogs go, the investigators are producing some
very interesting results on 03. But the Committee is also very intrigued with
the monkey study. The monkey lung is more like the human lung and, if the
hypotheses about damage are right, the monkey should be very important for
looking beyond the rat and dog.
J. A. Graham; I guess that should be included in the more detailed proposal,
for the purpose of comparing results with those from extensive studies in this
area by Dr. Freeman.
My second comment concerns the 03 concentrations. I don't see how
studies at 2 ppm or even 1.2 ppm are going to be highly relatable to the rest
of the studies: those are very high concentrations compared to ambient
levels.
S. V. Dawson; First of all, Phelan is the only one that has suggested 2 ppm.
I argued that you really only need a couple of pairs of animals at 2 ppm to
get sufficient power in your test. This would indicate the direction in which
the line is going, thereby giving you a better calibration for the very
difficult lower 03 level.
J. A. Graham; That assumes that the mechanism is the same at the high
concentration. In the modeling by Miller [Chapter 29 of this volume], the
tissue dose of O3 was found to be dependent upon concentration; at higher
doses the O3 molecule may hit the pulmonary tissue; at lower doses, the
molecule may not hit the tissue.
S. V. Dawson; I absolutely agree with you.
J. A. Graham; My last question is: When you refer to macrophage motility,
are you talking about chemotaxis?
S. V. Dawson; Yes. This is a standard migration test that Les Schwartz has
published.
407
-------
39. OXIDANT HEALTH EFFECTS IN LOS ANGELES Dawson
J. A. Graham; For 03 in the monkey, why should that parameter—chemotaxis—
be any more sensitive than all the parameters already studied and reported
relative to the macrophages?
S. V. Dawson; Well, I don't think it's likely to be a more sensitive
parameter, but it ties to the mechanism of damage. It tests the hypothesis
that the respiratory bronchiole becomes tied up by all of this. One could
argue, perhaps, that finding an effect in that area at 0.2 ppm 03 supports
that hypothesis.
J. A. Graham; Those studies aren't going to be done on humans. In terms of
trying to correlate the epidemiologic, human clinical, and animal studies, I
don't see how macrophage chemotaxis directly relates to humans.
S. V. Dawson; Well, the monkey is as close as we can get. It's hard to do
that morphometry on the human.
P. E. Morrow; In what species do you plan to do the pentane exhalation study?
S. V. Dawson: Humans.
P. E. Morrow; How do you propose to do that, inasmuch as the dietary
component (lipid diet) and also the level of oxidant are so significant in
what you would apparently measure? The active level of oxidant breathed and
dietary component of lipid affect the breakdown. How do you deal with that in
the human population?
S. V. Dawson; When a subject is exposed to 03, there are really two different
aspects. First, if the subject comes from a polluted area, does he produce
more pentane, reflecting more oxidant damage to his lungs? Secondly, when
given 03, does he produce more pentane before or after the experiment?
Comment; As I understand it, the use of pentane and isopentane and the
relationship between them can help overcome some of the issues of metabolic
production in the background. Ethane has also been used.
Comment; With pentane, the method of detection is much more precise. The
ongoing levels have to be active? there1s no "memory" of what the individual
breathed yesterday.
Question; There is some published work on pentane and ethane exploration with
presumably oxidizing cytotoxins but not respiratory pollutants. Wouldn't it
be useful to have some animal work on this as well?
S. V. Dawson; Yes, that's absolutely valid and we will explore it. One of
the suggestions is that we might do such work at UCD.
408
-------
39. OXIDANT HEALTH EFFECTS IN LOS ANGELES ^^ Dawson
Question; In the walk-in chest complaints study, how do you plan to correct
for the self-selection bias? Secondly, what kinds of objective measurements
do you plan to analyze for? What are the end points?
Comment; Of necessity, we will use the somewhat gross end point of a
self-selected complaint. The frequency of such complaints in areas of widely
differing pollution levels and yet subject to the same barrage of publicity
about the episodes was thought to be a way to look at real effect. We have
already done some feasibility studies on emergency room visits; the study
could probably be more finely tuned by the use of data on respiratory therapy
services or the medications administered to people coining there. (We exclude
nonrespiratory complaints.) The concept is to set up a network and have it in
place so that, when acute health advisory levels are reached, the frequency of
visits during those periods can be compared to other seasons or years.
Question; Where [in Table 39-1] is the leukocyte function test to which you
referred?
S. V. Dawson; I don't know. That's tentative; we're just exploring it as a
pilot study that might possibly be funded.
A. P. Altshuller; Could you or Dr. Hackney expand somewhat on the directions
the human experimental work will take beyond what has been done to date?
Comment; As far as we know, all previous work has been on very highly
self-selected individuals—laboratory technicians, college students,
etc.—that were nonrepresentative of general populations. In our case, we
certainly don1t have any means of forcing people from the general population
to go into the study, but at least we are sending out letters asking the
general population, individual by individual, to join our study. That, as far
as I know, is completely new. Also, I think that the 0.3 ppm level is well
worth getting additional data on.
A. P. Altshuller; The study will be conducted at that level only?
Comment; Yes, that's the plan at least for the first year.
J. D. Hackney; The point I would make is the following: The epidemiologists
will present us with some subjects who are asserted to be sensitive to
photochemical pollutants in view of a more rapid decrement in forced
expiratory volume over five years. These subjects will be compared with
subjects showing almost no decrement. It may be that the subjects showing a
very large decrement are also sensitive to O3 over a short-term 2-h exposure;
such results would be exciting. There would be at least two possible
explanations; (1) the O3 somehow caused the rapid decrement; (2) having a
decrement is itself a predisposing factor for sensitivity to short-term 03
exposure. Either one of those explanations would be of interest. That s the
positive side.
409
-------
39. OXIDANT HEALTH EFFECTS IN LOS ANGELES Dawson
On the negative side, being able to do the study will depend on Dr.
Detels1 identification of these people. He already has them but the question
is whether they will volunteer in sufficient numbers.
Dr. Detels would do all the matching. He would send us the subjects and
we would test them. At the end, we'd try to sort it all out and see if there
was some correlation between epidemiologic decrement and short-term clinical
response.
410
-------
40. PANEL DISCUSSION
Timothy Crocker, Chairman
Aubrey P. Altshuller
Donald C. Borg
Robert S. Chapman
William Frietsch, III
Marvin Goldman
John R. Holmes
Fred J. Miller
James L. Whittenberger
T. Crocker; Of the possible items of agenda for this Panel Discussion, it is
my impression that most of those present favor some sort of scientist-
administrator exchange. In my view, such an exchange is potentially very
fruitful in that we might receive some information about the character of
future scientific review and program selection activities. In a certain
sense, we stand at the beginning of a phase of investigative effort. As
researchers, we all want this effort to be relevant and meritorious, but we
need to understand the intended mechanism for judging the quality and
relevance of our work.
To begin, Dr. John Holmes, Director of the Research Section of the
California Air Resources Board (ARE), will describe the cooperative agreement
with EPA under which ARB will conduct scientific review.
411
-------
40. PANEL DISCUSSION
J. R. Holmes; I certainly concur with the point made at the outset of this
meeting: It's very important, in selecting and managing research projects of
this kind, to bear in mind that we need scientific excellence and regulatory
relevance. In developing a program, we must consider both aspects.
In ARB, we have developed a joint committee consisting of members from
the ARB Research Advisory Committee as well as from an Air Quality Advisory
Committee of the California State Department of Health. These individuals are
scientists who are also involved (either peripherally or directly) in making
recommendations that serve as a basis for air quality standards. In my
opinion, the panel did an outstanding job in assembling a package of studies
that are both scientifically excellent and relevant to what we view as the
problems in setting an oxidant standard both nationally and in California. At
this point, none of the individual studies are "cast in stone"; there is still
time to hone and polish and make them even more relevant and scientifically
sound. The investigators and certainly ARB are very interested in any
comments from this group; we've received a number already.
In my opinion, before seeing the project through to completion, we need
to set up a system of scientific guidance that differs from the systems
currently used by EPA and ARB. In the management portion of our proposal to
EPA, we outlined plans for a joint technical committee to consist of
representatives from the State of California (2), EPA (2), the National
Laboratories (1), and the scientific community at large (2). This committee
is large enough to incorporate a number of different disciplines and yet small
412
-------
40. PANEL DISCUSSION
enough to permit efficient discussions and limited travel cost. I think this
approach will work out well. But I would appreciate any comments or
additional guidance on how technical review of the project should be carried
out.
W. Frietsch; We in EPA also have a review group that is looking at this
project (as well as others) strictly from a planning viewpoint [see Chapters 2
and 3 of this volume].
T. Crocker; My understanding is that Dr. Paul Altshuller is involved with a
separate Oxidant Research Committee within EPA's general approach to the
criteria air pollutants. This Committee develops strategy plans and documents
and will be responsible, as I understand it, for a mission statement. I may
be overstating the level of supervision that the Committee will undertake, but
I know that Dr. Altshuller has given some thought to potential developments
and to the level of guidance that may be forthcoming from this Committee.
Would you care to speak to this point, Dr. Altshuller?
A. P. Altshuller; Well, perhaps a bit of explanation is in order. During
congressional deliberations several years ago, the problem of the relevancy
and responsiveness of EPA in-house research to the regulatory needs of the
Agency came to a head. A fairly drastic change in the distribution of
resources was under consideration; the Deputy Administrator asked that we be
given a year to look into alternatives. We proposed a pilot Research
Committee program; our proposal was accepted. The pilot program was gradually
413
-------
40. PANEL DISCUSSION
expanded and today includes essentially all of EPA's research; the Oxidant
Research Committee is just 1 of 16 Committees covering air, water, radiation,
pesticides, etc.
The individual Committees have levels of responsibility for resources
ranging from $10 million to $60 million. The Oxidant Research Committee's
responsibility is in the ~$30 million range. Obviously, the Committee cannot
(and was not designed to) perform in-depth peer review for each small segment
of a $30-million program. Rather, the purpose of each Research Committee is
to interface between those responsible for research and those responsible for
regulatory development and enforcement. Thus, each Committee is largely
composed of individuals with technical backgrounds: personnel from EPA's
regulatory divisions, from the Office of Enforcement, from regional offices,
and from the Office of Research and Development (ORD). Each Committee is
co-chaired by a representative from ORD (in the case of the Oxidant Research
Committee, Ken Berry) and a representative from the lead program office. For
oxidant research, the lead program office is the Office of Air Quality
Planning and Standards.
The Committees serve as a mechanism for expression of concerns about
scientific relevance, responsiveness, and the timeliness with which outputs
from various research programs become available. The obvious question is:
How can we assure scientific quality? In part, this depends on the competence
and the dedication of the technical personnel within EPA. It also depends on
a peer review mechanism. Dr. Steven Gage, Assistant Administrator for
414
-------
40. PANEL DISCUSSION
Research and Development, has been very concerned about the peer review
mechanism for both in-house and external research. We are revising the
research grant program to return to a peer review procedure resembling that of
the National Institutes of Health (NIH) but which will also include a
relevancy review. At present, we don't have all the answers on how to
integrate these two aspects, but both are vital. Regardless of whether a
research program involves pass-through money in another agency, agency
agreements, contracts, or what have you, the same general question—how to
achieve a scientific product that will be accepted by the scientific community
and also usable in criteria and standards development—remains a vital issue.
Question; Has any significant peer review mechanism been set up?
A. P. Altshuller; A specific mechanism for research grant peer review is
being set up. It will consist of a group of four committees that will
periodically review grant proposals for both scientific acceptability and
relevancy. Because of certain congressional and other limitations on the
movement of funds, we have a number of "pigeonholes" where we have to "put the
money," so to speak, and we will fill these up with proposals until the
funding limit is reached. To do this, we'll work down a list of projects
scored for merit as well as relevancy. In a sense, it's a tripartite system:
scientific merit, relevancy, and the amount of money available for the
subcategory.
Comment; You still haven't answered the question.
415
-------
40. PANEL DISCUSSION
A. P. Altshuller; Would you care to enlarge on your question?
Question; Aside from research grants, is any peer review system under
consideration?
T. Crocker; Dr. Altshuller, I believe the question refers to the oxidant
program that is the subject of this workshop. You're talking about an
extramural grant review program that may not be related to our particular
focus.
A. P. Altshuller; Well, it would be related to all grants, including those
concerned with oxidant air pollutants. We're just working on possible
mechanisms for review. A question which does come up is: To what extent can
the other body of expertise available to EPA on a continuing basis—namely,
the Science Advisory Board (SAB) and its subcommittees—participate? Just
last week, we had some discussions with the Director of SAB, Richard Dowd; we
discussed some preliminary proposals to involve the Clean Air Scientific
Advisory Committee in more in-depth review of criteria pollutant research,
with particular emphasis on oxidants. The question is: Where do we go from
there, when we get down to the question of reviewing program components that
represent a million, half a million, or a quarter of a million dollars? This
would demand large numbers of people who are willing to commit some
significant fraction of their time. Very likely we would end up with well
over 100 individuals just in the area of health. These 100 individuals would
have to be willing to commit a certain fraction of their time to interact in
416
-------
40. PANEL DISCUSSION
peer review of a wide variety of projects in health, modeling, ecosystems,
control technology, etc. ftnd this has not been worked out.
Comment: The details of the review system have all been set down by Dr. Gage.
If you're interested in obtaining those details, I suggest that you contact
Dr. Marlin at EPA Headquarters.
A. P. Altshuller; Having talked quite a bit with Dr. Marlin, I'm not sure
that the array of detailed information is really there. The mechanics may be
there for the research grants and for the cooperative agreements. As for the
various other categories, it is probably a little premature to say that we
have all the answers.
F. J. Miller; SAB will review the EPA in-house projects on March 20. SAB
will also perform scientific review of the Department of Energy (DOE)
transferred projects. So mechanisms have been set up. If your question was
whether the DOE-EPA projects will be reviewed, the answer is "Yes."
D. C. Borg; With regard to Dr. Altshuller's point, it does take a lot of
people to carry out peer review (a la NIH, in any case). What's more, as
thorough and reassuring as that sort of review may be, there are certain
disadvantages. It grinds with extreme slowness, to say the very least. Other
agencies (perhaps less well known to this audience) carry out peer review in a
somewhat different fashion. For years, the National Science Foundation (NSF)
has proceeded with peer review (in the physical sciences) much as do editors
417
-------
40. PANEL DISCUSSION
for peer review journals: namely, asking for outside written opinions and
then presenting a consensus of that opinion to an internal review board. In
the life sciences area, there are also panels. Those who have served on the
NSP panels know that they don1t have as much time to spend on an individual
grant application as does an NIH study section. The NSF panels perform
something of an intermediary function: they are much more advisory than the
NIH study sections. In the physical sciences, however, NSF has managed to do
a fairly creditable job without all of that. Thus, I would remind EPA that
the NIH way is not the "only" way to achieve peer review.
A. P. Altshuller; Yours is a rather interesting comment, for this reason:
The EPA research grant/cooperative agreement review system that was in place
from 1973 until 1979 was very similar to the NSF system. Extramural reviewers
were asked to individually review each project and to submit written comments.
A single in-house scientific reviewer also assessed the project. A judgment
was made by technical management within the individual Laboratory. (Later,
that judgment added up to whether or not the project was funded.) It's
precisely this sort of system that was criticized, resulting in a return to an
NIH-type system.
One of the principle criticisms was that we were involving too few
investigators. Critics complained that we tended to be satisfied—once we had
found a group of investigators who could do the job in a specific
discipline—to "ride along" with those investigators in other disciplines.
I'm not sure that was a valid criticism; in all honesty, I don't know where
418
-------
40. PANEL DISCUSSION
the additional investigators would have been. As. far as I could discover, we
had sufficient resources to fund essentially all of the competent
investigators. In some areas it may have been a specious criticism; in
others, it may have been a relevant criticism.
M. Goldman; As someone who is not overly familiar with the EPA system of
review, I've still not received an answer for a question I jotted down on my
way to this workshop: How do you equitably separate the problem of relevancy
(which I assume to be an intra-agency one) from that of scientific merit? At
this workshop, I hear what seems to be a plea for an equitable peer review
system that will take care of both the intra- and extramural programs
regardless of how they are handled administratively. In other words, the
science—regardless of who" s doing it—should be equally good, and a mechanism
for assuring this must exist. This relates, perhaps, to a mechanism to
provide continuity of support for longer-term programs. If high-quality
science is to be attracted to these programs, the issues of peer review and of
how information will be delivered to the various clients are important parts
of the overall spectrum. From what I've heard here, progress is being made
but the system is not yet complete. -Phis business of "extramural versus
intramural program methods" leaves me and others with some confusion.
T. Crocker: Your comments fairly accurately summarize the situation in terms
of the structure and plan for Agency supervision. It's moving but it hasn't
jelled.
419
-------
40. PANEL DISCUSSION
With respect to extramural grant review, Dr. Whittenberger chairs a
special SAB Subcommittee that has been involved in [the redirection of Theme 1
of the Energy Health Effects program; see Chapter 3 of this volume]. That
body, of course, serves at the request of EPA; still, in the sense that it
consists of extramural scientists, the Subcommittee provides something of an
extramural peer review function. Perhaps Dr. Whittenberger would comment on
possible expansion of the Subcommittee1s mission to meet some of the issues
that have been raised.
J. L. Whittenberger; Our Subcommittee was not asked to consider scientific
merit. We were asked to assist in the planning process in cooperation with
EPA scientists and administrators. We have raised questions about scientific
merit but do not consider it our responsibility to exercise the kind of
judgment that we would perhaps like to exercise. We're not constituted to do
that. In our last meeting, we did discuss the need for peer review for
scientific merit as well as for relevance, and suggested that SAB might take
responsibility for setting up subcommittees that would be more properly
constituted to carry out such peer review. This discussion occurred at the
end of our meeting; we have taken no action to implement it.
I think it would be worthwhile to respond to Dr. Alpen's comments
[Chapter 6 of this volume] and to some of the comments by Dr. Goldman
illustrating the need for a much better understanding, on the part of
scientists outside EPA, of the constraints under which EPA operates as a
regulatory agency. Dr. Altshuller mentioned some of the items on my
420
-------
40. PANEL DISCUSSION
checklist; I'll touch on some that he didn't mention. These comments will
express my point of view as an individual, not as an SAB representative.
With regard to Dr. Alpen's comments [Chapter 6 of this volume], when I
wear my "Harvard hat" I agree with everything he says about toxicology and the
relevance of different kinds of studies in respect to toxicologic analysis of
any obnoxious compound. When I put on my "EPA hat," however, I am forced to
adopt a very different concept of what is "relevant." One must never forget
that research conducted within a regulatory agency is very different from
research conducted outside a regulatory agency. EPA's research program is
driven by program offices: primarily, the Office of Air, Noise and Radiation;
the Office of Water and Waste Management; and the Office of Toxic Substances.
When EPA was established, research and development were put in one office that
was supposed to respond to the program needs of other parts of the Agency.
This arrangement has continued to present certain problems.
We must remember that the program offices are driven by the specific
requirements of legislation. Sometimes EPA is undeservedly blamed for
situations for which it is not responsible. The fact that scrubbers are
required in power plants that can use low-sulfur coal does not reflect
perversity on the part of EPA but, rather, legislation that was written so as
to keep people employed in Appalachia. Another example is the Toxic
Substances Control Act, an act that many consider almost impossible to
administer.
421
-------
40. PANEL DISCUSSION
Several years ago, the program offices complained about ORD research:
they said that the scientists wanted to "do their thing" without responding to
the needs of the program offices. Congress became upset and, as Dr.
Altshuller said, demanded that something be done. Dr. Gage set up a task
force to examine how other agencies handle this problem. The end result was
the creation of the Research Committees.
The Research Committees have certain advantages and certain limitations.
One problem is the need to support long-term basic research as well as more
immediate regulatory-related research (i.e., research to provide certain
"answers" within six months or a year). Somehow, the long-term research needs
seem to "get lost" before they are ever funded. This is due not only to
limitations of the Research Committee approach but also to limitations
inherent in the legislative process. Congress recognizes EPA's need for basic
research; the Authorization Subcommittee encouraged Dr. Gage to plan a
so-called Anticipatory Research Program. But Dr. Gage has had trouble
obtaining funding for that program because the Appropriations Committee
doesn't look at things the way the Authorization Subcommittee does.
Dr. Gage has been able to set up so-called Centers of Excellence in
universities or consortia of universities. In designating a Center of
Excellence, EPA adds some relatively long-term funding to orient an already
strong program to EPA needs.
422
-------
40. PANEL DISCUSSION
On the subject that is paramount in this Panel Discussion—peer
review—I'd like to supplement some of the previous comments on EPA's
"credibility problem." From Mr. Costle on down, there is a great deal of
concern within EPA about peer review of all research activities (both
intra- and extramural). As Dr. Altshuller said, EPA originally had a peer
review mechanism which worked well in many parts of the Agency. It did not
work as well in other parts of the Agency, and those places may have had more
influence on Dr. Gage. Thus, he made the decision to set up the centrally
administered Grants Program. This program, which is spelled out in the
Federal Register, is now getting underway. The first committee meeting is
scheduled for next month and the notifications will be made, I believe, in
April.
That takes care of a certain part of EPA-funded research, but what about
the cooperative agreements, contracts, and so on? I don't believe those
mechanisms have really been developed yet, although I know that Dr. Gage is
very interested in putting such systems in place.
Some of the problems that I've referred to very briefly were the subject
of a study by the National Academy of Sciences reported in 1977. In 1978-9, I
participated in a study, mandated by Congress, in which we looked at EPA
health effects research from the point of view of relevance to program needs
as well as quality. After receiving our report, Mr. Costle asked the
Environmental Health Advisory Committee (an SAB committee) to assume an
ongoing responsibility for peer review of EPA's major health effects
423
-------
40. PANEL DISCUSSION
laboratories. This will encompass the intramural and extramural research that
is not covered by the centrally administered Grants Program.
I'd be glad to answer any questions about peer review.
R. S. Chapman; I'm concerned, because the essential focus of workshops such
as this, and of peer review generally, tends to be the review of scientists
within EPA by scientists outside EPA. For two reasons, I think that such a
focus is not very effective in resolving limitations in EPA's scientific
credibility. First, there is already substantial agreement between intramural
and extramural scientists on the standards required to ensure scientific merit
in research projects. Second, instead of this constant droning about review
of scientists by scientists, what we need is an educative function by
scientists both within and without EPA to all levels of the administrative and
legislative structure under which we operate. Such a function, if given
"teeth" and if perpetuated, would be far more useful than this constant
reshuffling of various ways to have scientists critique the work of other
scientists, which just isn't "getting the baby washed," to be perfectly honest
with you. In my opinion, we are today under the same administrative and
legislative pressures to do work that1s against our better judgment as we have
been ever since EPA was formed. And so I believe that education "up the line"
will be a lot more useful and fruitful than yet another mechanism by which
scientists can review scientists. Within the Agency, we're quite inclined to
agree with scientists who criticize the merits of our work; we tend to say,
"Yes, we agree with you." When our critiques are carried into the
424
-------
40. PANEL DISCUSSION
administrative area, however, our better judgment tends to be paired against
the conflicting pressure of "The mandate is here now, the money is here now,
the money won't be here if you don't think the work can be done now." The
former pressure always loses out to the latter. Until the former pressure has
more authority, I think that this kind of exercise is, to a considerable
extent, rather futile.
J. L. Whittenberger; I understand your feelings and I agree that the program
of which you are a part was subjected to a very critical and in many ways
unfair review. Personally, I felt that the CHESS program (one of the programs
that took place over a number of years in the unit to which Dr. Chapman
belongs) was a very important program; it should have been constructively
criticized but continued. One of the reasons we don't today know more about
the effects of oxidants on the health of people in Los Angeles is the
dismantling of that program.
On the other hand, I agree strongly with Dr. Goldman that scientists
anywhere ought to be subject to critical review by their peers. In visiting
many EPA laboratories, I have encountered an almost uniform desire to be
critically reviewed. Many scientists feel that their work has to stand the
test of peer review not only for publication in scientific journals but also
for introduction as evidence in court. They want that evidence to stand up.
So, I hope you'll be patient with the attempts to extend consistently applied
peer review within EPA.
425
-------
40. PANEL DISCUSSION
R. S. Chapman; I agree completely. I'm certainly not in favor of less peer
review for EPA scientists. I'm in favor of widening the perspective of
discussions like this (which, I might say, always come at the end of a meeting
instead of at the beginning or middle) to consider perhaps two more aspects:
(1) the entry of peer review at a higher administrative level, and (2) giving
the working scientists some tools for meaningful implementation of their ideas
coupled with the ideas of the peer reviewers. I'm certainly not arguing for
less peer review and scientist-to-scientist contact; rather, I'm arguing for
more effective use of it.
J. L. Whittenberger; I certainly agree. I encountered that same feeling in
many of the EPA laboratories. The scientists desire far more feedback, both
from SAB and from the scientists who reviewed their work.
T. Crocker; I think what Dr. Chapman is advocating might go even to the level
of peer review at upper administrative levels of the agencies and of the
Office of Management and Budget. In other words, who tells those gentlemen
how much to assign to a specific project in a specific year?
J. L. Whittenberger; I would extend your comment to Congress as well.
T. Crocker; In other words, it doesn't do us "bench people" a lot of good to
chew at each other* s quality when the decisions about relevance and resources
are made far beyond our reach. I think Dr. Chapman is asking how we might
reach that level. Possibly, increased communication between administrators
426
-------
40. PANEL DISCUSSION
and scientists would help. But high-level administrators are not present at
this workshop.
427
-------
MAILING ADDRESSES:
AUTHORS AND DISCUSSANTS
Edward L. Alpen
Lawrence Berkeley Laboratory
University of California
Berkeley, CA 94720
Aubrey P. Altshuller
Environmental Sciences Research Laboratory, MD-59
Office of Research and Development
U.S. Environmental Protection Agency
Research Triangle Park, NO 27711
Donald C. Borg
Medical Department
Brookhaven National Laboratory
Associated Universities, Inc.
Upton, NY 11973
Robert S. Chapman
Health Effects Research Laboratory, MD-54
Office of Research and Development
U.S. Environmental Protection Agency
Research Triangle Park, NC 27711
David L. Coffin
Health Effects Research Laboratory, MD-70
Office of Research and Development
U.S. Environmental Protection Agency
Research Triangle Park, NC 27711
Daniel L. Costa
Medical Department
Brookhaven National Laboratory
Associated Universities, Inc.
Upton, NY 11973
Timothy Crocker
University of California
Irvine, CA 92717
428
-------
AUTHORS AND DISCUSSANTS
Walden E. Dalbey
Respiratory Carcinogenesis
Oak Ridge National Laboratory
Post Office Box Y
Oak Ridge, TN 37830
Stanley V. Dawson
Research Division
California Air Resources Board
1102 Q Street
Post Office Box 2815
Sacramento, CA 95812
Roger Detels
School of Public Health
University of California
Los Angeles, CA 90024
Robert T. Drew
Medical Department
Brookhaven National Laboratory
Associated Universities, Inc.
Upton, NY 11973
Donald L. Dungworth
Department of Pathology
School of Veterinary Medicine
University of California
Davis, CA 95616
William Prietsch, III
Energy Effects Division, RD682
Office of Research and Development
U.S. Environmental Protection Agency
Washington, DC 20460
Donald E. Gardner
Health Effects Research Laboratory, MD-82
Office of Research and Development
U.S. Environmental Protection Agency
Research Triangle Park, NC 27711
Joel F. Ginsberg
59 Whispering Pines Drive
Sugar Mountain, TN 37377
429
-------
AUTHORS AND DISCUSSANTS
Marvin Goldman
Laboratory for Energy Related Health Research
University of California
Davis, CA 95616
Judith A. Graham
Health Effects Research Laboratory, MD-82
Office of Research and Development
U.S. Environmental Protection Agency
Research Triangle Park, NC 27711
Edward D. Haak, Jr.
Health Effects Research Laboratory, MD-58
Office of Research and Development
U.S. Environmental Protection Agency
Research Triangle Park, NC 27711
Sonja Haber
Medical Department
Brookhaven National Laboratory
Associated Universities, Inc.
Upton, NY 11973
Jack D. Hackney
Environmental Health Services
Rancho Los Amigos Hospital
University of Southern California
7601 East Imperial Highway
Downey, CA 90282
Wanda M. Haschek
Biology Division
Oak Ridge National Laboratory
Post Office Box Y
Oak Ridge, TN 37830
Milan J. Hazucha
Division of Pulmonary Diseases
Department of Medicine
School of Medicine
University of North Carolina
Chapel Hill, NC 27514
Carol A. Heckman
Biology Division
Oak Ridge National Laboratory
Post Office Box Y
Oak Ridge, TN 37830
430
-------
AUTHORS AND DISCUSSANTS
Rogene P. Henderson
Inhalation Toxicology Research Institute
Lovelace Bioiaedical and Environmental Research Institute
Post Office Box 5890
Albuquerque, NM 87115
John R. Holmes
California Air Resources Board
1102 Q Street
Post Office Box 2815
Sacramento, CA 95812
Steven M. Horvath
Institute of Environmental Stress
University of California
Santa Barbara, CA 93106
Edward Hu
Health Effects Research Laboratory, MD-82
Office of Research and Development
U.S. Environmental Protection Agency
Research Triangle Park, NC 27711
Junichi Iwai
Medical Department
Brookhaven National Laboratory
Associated Universities, Inc.
Upton, NY 11973
Michael H. Jones
Strategies and Air Standards Division, MD-12
Office of Air Quality Planning and Standards
U.S. Environmental Protection Agency
Research Triangle Park, NC 27711
Robert E. Lee
Health Effects Research Laboratory, MD-51
Office of Research and Development
U.S. Environmental Protection Agency
Research Triangle Park, NC 27711
T. Major
Biology Division
Oak Ridge National Laboratory
Post Office Box Y
Oak Ridge, TN 37830
431
-------
AUTHORS AND DISCUSSANTS
Ann C. Marchok
Respiratory Carcinogenesis
Oak Ridge National Laboratory
Post Office Box Y
Oak Ridge, TN 37830
Joe L. Mauderly
Inhalation Toxicology Research Institute
Lovelace Biomedical and Environmental Research Institute
Post Office Box 5890
Albuquerque, NM 87115
Fred J. Miller
Health Effects Research Laboratory, MD-82
Office of Research and Development
U.S. Environmental Protection Agency
Research Triangle Park, NC 27711
Samuel C. Morris
Biomedical and Environmental Assessment Division
Brookhaven National Laboratory
Associated Universities, Inc.
Upton, NY 11973
Paul E. Morrow
Radiation Biology and Biophysics
School of Medicine and Dentistry
University of Rochester
Rochester, NY 14642
A. C. Olson
Biology Division
Oak Ridge National Laboratory
Post Office Box Y
Oak Ridge, TN 37830
John A. Pickrell
Inhalation Toxicology Research Institute
Lovelace Biomedical and Environmental Research Institute
Post Office Box 5890
Albuquerque, NM 87115
Otto Raabe
Laboratory for Energy Related Health Research
University of California
Davis, CA 95616
432
-------
AUTHORS AND DISCUSSANTS
Gerald Rausa
RD681
Office of Research and Development
O.S. Environmental Protection Agency
Washington, DC 20460
Nathaniel W. Revis
Biology Division
Oak Ridge National Laboratory
Post Office Box Y
Oak Ridge, TN 37830
Karen M. Schaich
Medical Department
Brookhaven National Laboratory
Associated Universities, Inc.
Upton, NY 11973
C. C. Scott
Department of Nutrition
Harvard University School of Public Health
665 Huntington Avenue
Boston, HA 02115
Russell P. Sherwin
Department of Pathology
School of Medicine
University of Southern California
2025 Zonal Avenue
Los Angeles, CA 90033
Steven A. Silbaugh
Institute of Environmental Stress
University of California
Santa Barbara, CA 93106
F. Snyder
Oak Ridge Associated Universities
Oak Ridge, TN 37830
Beverly E. Tilton
Environmental Criteria and Assessment Office, MD-52
Office of Research and Development
U.S. Environmental Protection Agency
Research Triangle Park, NC 27711
433
-------
AUTHORS AND DISCUSSANTS
James L. Whittenberger
Department of Physiology
School of Public Health
Harvard University
665 Huntington Avenue
Boston, MA 02115
M. Jean Wiester
Health Effects Research Laboratory, MD-82
Office of Research and Development
U.S. Environmental Protection Agency
Research Triangle Park, NC 27711
Hanspeter Witschi
Biology Division
Oak Ridge National Laboratory
Post Office Box Y
Oak Ridge, TN 37830
Ronald K. Wolff
Inhalation Toxicology Research Institute
Lovelace Biomedical and Environmental Research Institute
Post Office Box 5890
Albuquerque, NM 87115
434
t US. GOVERNMENT PRINTING OFFICE: 1M1 -757-064/0221
------- |