United States
Environmental Protection
Agency
Health Effects Research
Laboratory
Research Triangle Park NC 27711
EPA-600/1-79-038
September 1 979
Research and Development
A Biologic
Indicator for
Air Pollution
-------
RESEARCH REPORTING SERIES
Research reports of the Office of Research and Development, U.S. Environmental
Protection Agency, have been grouped into nine series. These nine broad cate-
gories were established to facilitate further development and application of en-
vironmental technology. Elimination of traditional grouping was consciously
planned to foster technology transfer and a maximum interface in related fields.
The nine series are:
1. Environmental Health Effects Research
2. Environmental Protection Technology
3. Ecological Research
4. Environmental Monitoring
5. Socioeconomic Environmental Studies
6. Scientific and Technical Assessment Reports (STAR)
7. Interagency Energy-Environment Research and Development
8. "Special" Reports
9. Miscellaneous Reports
This report has been assigned to the ENVIRONMENTAL HEALTH EFFECTS RE-
SEARCH series. This series describes projects and studies relating to the toler-
ances of man for unhealthful substances or conditions. This work is generally
assessed from a medical viewpoint, including physiological or psychological
studies. In addition to toxicology and other medical specialities, study areas in-
clude biomedical instrumentation and health research techniques utilizing ani-
mals — but always with intended application to human health measures.
This document is available to the public through the National Technical Informa-
tion Service, Springfield, Virginia 22161.
-------
EPA-600/1-79-038
September 1979
A BIOLOGIC INDICATOR FOR AIR POLLUTION
by
Russell P- Sherwin
University of Southern California Medical Center
Los Angeles, California 90033
Grants No. R800881 and R804545
Project Officer
Donald E. Gardner
Biomedical Research Branch
Health Effects Research Laboratory
Research Triangle Park, NC 27711
HEALTH EFFECTS RESEARCH LABORATORY
OFFICE OF RESEARCH AND DEVELOPMENT
U.S. ENVIRONMENTAL PROTECTION AGENCY
RESEARCH TRIANGLE PARK, NC 27711
-------
DISCLAIMER
This report has been reviewed by the Health Effects Research Laboratory,
U.S. Environmental Protection Agency, and approved for publication. Approval
does not signify that the contents neeessarily reflect the views and policies
of the U.S. Environmental Protection Agency, nor does mention of trade names
or commercial products constitute endorsement or recommendation for use.
11
-------
FOREWORD
The many benefits of our modern, developing, industrial society are
accompanied by certain hazards. Careful assessment of the relative risk of
existing and new man-made environmental hazards is necessary for the estab-
lishment of sound regulatory policy. These regulations serve to enhance the
quality of our environment in order to promote the public health and welfare
and the productive capacity of our Nation's population.
The Health Effects Research Laboratory, Research Triangle Park, conducts
a coordinated environmental health research program in toxicology, epidemio-
logy, and clinical studies using human volunteer subjects. These studies
address problems in air pollution, non-ionizing radiation, environmental
carcinogenesis and the toxicology of pesticides as well as other chemical
pollutants. The Laboratory participates in the development and revision
of air quality criteria documents on pollutants for which national ambient
air quality standards exist or are proposed, provides the data for registra-
tion of new pesticides or proposed suspension of those already in use,
conducts research on hazardous and toxic materials, and is primarily respon-
sible for providing the health basis for non-ionizing radiation standards.
Direct support to the regulatory function of the Agency is provided in the
form of expert testimony and preparation of affidavits as well as expert
advice to the Administrator to assure the adequacy of health care and
surveillance of persons having suffered imminent and substantial endanger-
ment of their health.
The report describes studies wherein guinea pigs and mice were exposed
to low levels of nitrogen dioxide either continuously or on sensitive
discriminants involving correlations of structural and functional altera-
tions. The two major test areas developed were the quantitative measure-
ments of Type 2 cell populations and protein leakage within the lung. The
increase in number and size of Type 2 pneumocytes with N02 exposure is
believed to represent a corresponding loss of Type 1 lung cells, and the
protein leakage within the lung is in accord with the concept that the
Type 1 cell is a critical barrier for fluid transport in the alveolar area.
The authors suggest that their research data supports the hypothesis that
to properly evaluate the possible health effects from air pollutants one
must give attention to the subclinical cellular alterations that deplete
the functional and structural reserves of organs which are the primary
target for air pollutants.
F. 6. Hueter, Ph.D.
Director
Health Effects Research Laboratory
iii
-------
ABSTRACT
Studies with nitrogen dioxide (NO.) were carried out at levels from 2 ppm
to a low of Q.k ppm, with either continuous or Intermittent exposures and with
the use of guinea pigs and mice. The major overall goal was to develop highly
sensitive discriminants involving correlations of structural and functional
alterations. Two major testing areas were developed, quantitative measurements
of Type 2 cell populations and protein leakage in the lung. A series of four
image analyzer studies showed that exposure of guinea pigs to continuous 2 ppm
NO- resulted in an increased number of Type 2 pneumocytes (p < .05), that there
was a very high correlation between hand and automated cell counts (r > 7.7;
p < .001), there was high correlation between hand counts of alveoli and
automated measurements of alveolar wall area (r > .056; p < .002), that duration
of exposure had a significant effect (F = 5.33; P < .05), and that the Type 2
cell increased in size (p < .025), with duration of the exposure also signifi-
cant (p < .05).
With respect to protein leakage, polyacrylamide gel electrophoresis (PAGE)
quantitative measurements showed a greater protein content in the lung lavage
fluid of guinea pigs that had been exposed to continuous 0.4 ppm NO- for 1 week
(p < .001) as opoosed to a control group. In two studies of the urine of guinea
pigs, the protein content was shown by PAGE and Lowry protein tests to be higher
for the exposed group of animals (p <.01 and p < .05 respectively for those
exposed to continuous and intermittent (4 hours per day) NO- - A newly inno-
vated test for protein content of lungs of mice, using a fluorescent protein
label (fluorescamine) showed higher levels of protein in the lungs of mice
exposed continuously to 0.47 ppm NO- for 12 ± 2 days (p < .025). The latter
studies were expanded through the use of a molecular probe, horseradish peroxi-
dase (HRP) as a means of quantitating plasma and other protein substances in
lung cells, tissues, air spaces, and lymphatic channels. In three independent
experiments, using continuous 0.6 ppm NO-, intermittent 0.8 ppm NO., and inter-
mittent 0.6 ppm N0_, an increased HRP content of the lungs of the exposed
animals was found ffirst two experiment) after three weeks of exposure p < .05
in each case. With respect to the third experiment, an elevation of HRP in
the lungs of exposed animals occurred at 6 weeks (p < .025). Of the kidney
HRP content studies, there was one significant difference (p < .05), again
for an exposed group of animals. Parallel enzyme kinetic assays of HRP were
also carried out for some of the studies and confirmed the findings. Ultra-
structural tracer studies of HRP distribution did not show overt differences
between the two groups of animals in the screening type of comparison carried
out.
In addition to image analysis and protein leakage findings the ultra-
structural studies were the first to show unidentified wall-free organisms in
the bronchi of guinea pigs; there was a remarkable attachment to microvilli
and an association with the basal portion of cilia. An extensive attempt to
isolate and culture these organisms by special L-phase and mycoplasma media
was unsuccessful, despite the cooperation of an expert pre-eminent in these
areas of culture work.
The diphosphoglycerate content of red blood cells was measured in a
colony of guinea pigs. Those exposed to continuous 0-36 ppm NO- for 1 week
were found to have a significant increase (p < .05).
IV
-------
Support in part was given to a few projects tangential to the research
direction of this program. The projects were opportunities to obtain new
kinds of measurements of lung cell vigor and viability, using material
supplied by an independent tissue culture program.
The overall findings from the image analysis and protein quantitat ion
studies imply that both structural and functional alterations of the lung
can be produced by supraambient and ambient levels of NO- exposure of
guinea pigs and mice. An increase In numbers and size or Type 2 pneumocytes
is believed to represent a corresponding loss of Type 1 lung cells, and a
leakage of protein is in accord with the concept that the Type 1 cell is a
critical barrier for fluid and protein leakage in the alveolar spaces.
Changes in diphosphoglycerate content of red blood cells and increased
content of protein in the urine indicate a systemic effect of NO-. Other
aspects of the work point to a role for N0« in placing additional burdens on
the host defense system, in particular possibly providing an opportunistic
circumstance for the pathologic emergence of organisms indigenous to the
lung.
-------
TABLE OF CONTENTS
Fo rewa rd if I
Abstract iv
Table of Contents. . . . '. vi
Figures viii
Tables Ix
Abbreviations xi
1. Introduction 1
2. Conclusions 3
3. Recommendations 6
A. teterials and Methods 7
Animals 7
Exposure Chambers 7
Pollutant 7
Lactage dehydrogenase identification of Type 2
pneumocytes for quantitative measurements 8
Lung lavage 8
Gel electrophoresis 9
Protein determinations 9
Electron microscopy 9
Horseradish peroxidase permeability studies 9
Enzyme kinetic assay for HRP 10
Oiphosphoglycerate determinations 11
Acid phosphatase 11
Ferr it in studies 11
Other procedures 11
Statistical analyses 12
Image analysis • 12
5- Results and Discussion 13
Three studies of continuous exposures to 10 ppm N02,
2 ppm N0£, and combined 3 ppm N0£ and 5 ppm S02 13
Wall-free organisms on bronchial mucosa of guinea pigs;
ultrastructural study 13
Bacterial, L-phase, and mycoplasma culture studies. . . 13
Continuous 2 ppm N02 exposure for 7, 1^, and 21 days ... 14
Semi-automated quantitat ion of area of Type 2
pneumocytes and alveolar wall cell area lA
VI
-------
TABLE OF CONTENTS (Cont.)
Exposure of guinea pigs to 2 ppm N0£ for 1, 2, and
3 weeks 20
Image analyzer quant I tat ion of numbers of Type 2
pneumocytes 20
Exposure of guinea pigs to continuous 0.4 ppm N0£
for one week. . . . 20
Polyacrylamide gel electrophoresis (PAGE) for
quantitation of protein content of lung lavage fluid. 20
0.36 ppm N0£ continuously for one week 28
Measurement of diphosphoglycerate content of red
blood cells of guinea pigs 28
Exposure of guinea pigs to 2 ppm N0£ for one week 30
An automated determination of ratios of Type 2
pneumocytes to alveolar wall area using an image
analyzer 30
0.4 ppm N0£ exposure, 4 hrs/day, for one week and
three weeks 34
Acid phosphatase in density equilibrium fractions
of lungs of guinea pigs 34
Ultrastructure 34
Continuous 0.5 ppm N02 for seven and 14 days, and
intermittent 0.4 ppm N02, four hours/day for eight, nine,
and ten days (proteinuria in guinea pigs) 34
Protein quantitation 34
Ultrastructure and histopathology 40
Continuous 0.47 ppm N02 for ten, 12 and 14 days 40
Fluorescami ne labeling studies 40
Ultraviolet microscopy 47
Continuous 2.0 ppm N02 (an experimental model and
automated methodology) 50
Experimental model 50
Automated methodology 50
Automated quantitation of hypertrophy 50
Horseradish peroxidase studies at three levels of N02
exposure: continuous 0.5 ppm N02, intermittent 0.8 ppm
N02, and intermittent 0.6 ppm N02 (clearance of
exogenous horseradish peroxidase in the lungs of mice) . 54
PAGE quantitation of lung homegenates 54
Enzyme kinetic assay 54
Electron microscopic studies 54
0.5 continuous and intermittent N02 exposures
(investigations supported in part by this project) ... 64
Liquid scintillation counting of labeled cells
in culture 64
Macrophage congregation in tissue cultures of mouse
lungs 65
Ultrastructural, histochemical aspects 66
Studies presently incomplete 66
References 68
vi i
-------
FIGURES
Number Page
1a Bronchus of a guinea pig exposed to a combination of
5 ppm 50,2 in combination with 3 ppm N02 ......... ^
1b Bronchus of a guinea pig exposed to 2 ppm NO- ....... 15
2 Average area-cell ratio, in arbitrary units, for each
of the animals ...................... 18
3 Phase contrast photomicrograph of linear alveolar
structures, as detected by the image analyzer ...... 21
k Plot of average hand cell count versus automated cell count
No. 1 .......................... 22
S Densitometric recordings of lavage fluid disc-gel
electrophoresis ..................... 23
6 Plot of hand count of alveoli versus automated
measurements of wall area for the 2^ slides . ...... 31
7 Comparisons of densitometric readings: disc-gel
electrophoretic determinations of urinary protein .... ^
8 Definition of the model .................. 51
9 Estimates of X and j ................... 52
VIII
-------
TABLES
Number Pag
1 Number of animals in which wall-free organisms were
observed l6
2 Absolute data ^
3 Means and standard deviations of the average area-LDH-
positive cell 19
k Comparison of averages ± SE of hand counts and automated
counts 2^
5 Linear regression analysis of relationship of hand counts
to automated counts 2**
6 Analysis of variance of ratios derived from automated
and hand counts 25
7a Analysis of variance of protein content 26
7b Protein content of lung lavage fluid 27
8 2,3'DPG phosphorus content of RBC (pmol/ml ) in guinea
pigs 29
9 Average values of measurements of alveolar cells and area
of alveolar wall for 2k animals 32
10 Coefficients of correlation between ratios of cells to
alveoli obtained by hand count and automated
determination of ratio of cells to wall area using
selected numerators and denominators 33
11 Band properties 35
12 Acid phosphatase readings (Experiment 1) 36
13 Analysis of variance of acid phosphatase data (Experiment 1) 37
Ik Acid phosphatase readings (Experiment 3) 38
15 Acid phosphatase readings (Experiment 3) 39
16 Data summary of urine analysis (Experiment 1) k2
17 Albumin .content of urine disc-gel electrophoresis
(Experiment 2) ^3
IX
-------
TABLES (Cont.)
Number Page
18 Urine specific gravity (Experiment 2) 44
19 Total protein content (disc-gel) of urine 45
20 Protein content (mgA) with and without dialysis 46
21 Lung homogenate fluorescence 48
22 Serum fluorescence 49
23 The analysis of variance of the experimental data 53
24a Experiments and conditions, HRP studies 55
24b Mean animal weights, HRP studies 55
25 HRP content of mouse lung tissue (HRP/gm tissue) 56
(Experiment 1)
26 Two factor analysis of variance (Experiment 1) 57
27 HRP content of mouse lung tissue (HRP/gm tissue
(Experiment 2) 58
28 Two factor analysis of variance, serum and lung ...... 59
29 HRP content of mouse lung tissue (HRP/gm tissue)
(Experiment 3) 60
30 Two factor analysis of variance: lung HRP content
(Experiment 3) 61
31 Serum HRP J weeks 62
32 HRP content of mouse kidney tissue (HRP/gm tissue) 63
-------
ABBREVIATIONS
A — Angstrom
BSS — balanced (or basal) salt solution
C° — Centrigrade
cm — centimeter
cpm — counts per minute
DPG — 2,3-diphosphoglycerate
DAB — diaminobenzidine
EM — electron microscopy
Enzymes: AcPase — acid phosphatase
AtPase — adenosine triphosphatase
G6PD — glucose-6-phosphate dehydrogenase
HRP — horseradish peroxidase
LDH — lactate dehydrogenase
MOH — ma late dehydrogenase
g — gravity units
GTA -- glutaraldehyde
3H — tritium
hrs — hours
3H-TdR — tritiated thymidine
K — thousand
M — molar
ma — mi 11iampere
ml — mil 1i1iter
mV — millivolt
U -- micron
yd — microcurie (10"'6 Ci)
ymol — micromole (10~6 Moles)
nM — nanometer (10~9 meter)
N02 — nitrogen dioxide)
PAGE — polyacrylamide gel electrophoresis
PMI — postmortem interval (delayed explantation in vitro)
ppm — parts per mil Iron
RBC -- red blood eel Is
RH — relative humidity
rpm — revolutions per minute
Statistical data: Xi — number of Type 2 pneumocytes per microscopic field
Xz — number of alveoli in the same field
Y ~ Xi/Xa
T — treatment
D — duration
L — lung lobe
TDL — treatment-duration-.lobe
S — lung section
ATDL — animal-treatment-duration-lobe
X2 — area of each slide occupied by the alveolar wall
Y* — Xi/xf ratio of cells to baseline
Ta-Ts — from 3 to 6 congregating macrophages per target cell
UV -- ultraviolet 1ight
xi
-------
SECTION 1
INTRODUCTION
The lung and other organs of the body can sustain an enormous amount of
damage, and especially loss of tissue, without an overt effect. For example,
an entire kidney can be donated for transplantation purposes without any
clinical sign or patient awareness of impaired kidney function. Similarly,
it appears that over one half of the lung must be lost before the loss comes
to clinical or patient attention, and even before pulmonary function tests
first become positive. The reason such losses are inapparent is the fact that
the body in general has great compensatory mechanisms and an extraordinary
amount of cellular and tissue reserves. Thus, signs of clinical disease
(morbidity), or death itself (mortality), will generally be very crude dis-
criminants for use in establishing air quality standards. While attention to
exacerbation of illness can provide more sensitive and useful data, an evalu-
ation of air quality demands that attention also be given to subclinical
cellular alterations that deplete the functional and structural reserves of
organs which are the primary targets for air pollutants. We have termed an
abnormal loss of reserves "hypeinopenia." When this covert loss of reserve
reaches a point where a person becomes particularly susceptible or- vulnerable
to bodily insults and illness in general, that person can be said to suffer
subclinically from "morbility."
There can be no question that everyone, regardless of his or her en-
vironment, is to some extent losing cellular and tissue reserves of all organs.
The important question is what is the baseline in the well population in terms
of alveolar cells and alveoli that have been lost, and what correlation can be
made with the various ambient atmospheres across the country.
The projects we have proposed over the past several years have been
specific attempts to provide data on hypeinopenia as found in two animal
models, the guinea pig and the mouse, and with respect to one of the air
pollutants we considered to have high potential for an adverse health effect,
namely nitrogen dioxide (N0_). The principal objectives were to obtain
quantitative measurements that would reflect physiopathologic alterations that
could be related to the problem of air quality for human populations. Towards
this end, we applied our experience with examination of human lung biopsies
to a search for discriminants in an animal model that would be highly relevant
for the human experience. We noted, as have others, that a loss of Type 1
cells and a replacement by Type 2 pneumocytes (and/or Clara cells) is a common
denominator and early finding for a great variety of human lung disease. We
considered edema and protein loss to be additional early, common denominators.
Thus, the proposed projects have been based on animal models that permit
exploration of these specific alterations, i.e. a population shift in the
alveoli of the lungs and increased capillary permeability following air
pollutant exposure. Nitrogen dioxide (N0_) was singled out as one of the most
important of the air pollutants for study, both with respect to its high
potential for adverse health effects and as a prototype pollutant for the
development of new methodologies that can later be applied to other pollutants,
singly and in combination.
-------
The results of our studies have been submitted as annual progress
reports. They include a number of new discriminants we developed for the
measurement of early pathophysiologic alterations following supraambient and
ambient levels of NO , applied continuously -and intermittently over short term
periods of up to three months. More specifically, the efforts include quanti-?
tative measurements of numbers and size of Type 2 pneumocytes by image
analysis, measurements of protein leakage in lung lavage fluid by quantitative
electrophoresis and scintillation counts of tritiated albumin, measurements of
protein content of urine, ultrastructural studies, and other approaches as
presented elsewhere in this report. The results can also be found in Our
publications which we cite; the reprint sets were submitted earlier with the
the progress reports.
-------
SECTION 2
CONCLUSIONS
The purpose of these studies has been to develop highly sensitive dis-
criminants for evaluating the adverse health effects of air pollutants, be-
ginning with an investigation of the single pollutant, nitrogen dioxide (N02)
New methodologies were innovated as a means of detecting early structural and
functional changes in the lungs of guinea pigs and mice exposed to supra-
ambient and ambient levels of N02. The results from the use of these new
methodologies confirm our working hypothesis that the quantitation of cell
population shifts in the lung, in conjunction with quantitative measurements
of alterations in capillary permeability, are highly sensitive tests for
assessing adverse health effects.
The results of tests of cell population shifts with quantitative image
analysis (where guinea pigs were exposed to 2 ppm NO^ for 7 days, 14 days,
and 21 days) demonstrated that highly sensitive and versatile measurements
can be achieved. Specifica11y,_the initial study showed a high coefficient
of correlation (r > 0.8; p < 10~ ) with hand counts of the numbers of Type 2
pneumocytes expressed as a ratio to the numbers of alveoli in the same lung
sections. As with the. hand count, the lungs of the exposed animals showed a
greater number of Type 2 pneumocytes per alveolus (p < .05) than the control
animals. However, the relationship to duration of treatment shown by the
hand count (p < .025) was found by automated measurements to be of borderline
significance (p < .1), a finding subsequently brought to significance by new
measurements with an improved methodology (see below).
In a subsequent study of the image analyzer, a number of improvements
were made in the methodology to include an automated stepping stage, the use
of phase contrast images of the walls of the lung for quantitation, and a
fully automated determination of Type 2 pneumocyte/alveolar wall ratios.
High correlation coefficients were found in two specific areas: 1) between
hand and automated cell counts (r > 0.7; P < 0.001); and 2) between hand
counts of alveoli and automated wall area (r > 0.56; p < 0.002).
Final refinement of the image analysis methodology formed the foundation
for an experimental model use of this objective, quantitative approach. A
detailed analysis of the data obtained showed high correlation coefficients
for two independent analyses of hand count and automated data (p < 10 ),
and also demonstrated that there were two significant effects, NO., treatment
(F • 6.93, p < .02) and duration of NO,, exposure (F = 5-33, P < .vST-
The automated image analysis was expanded to include Type 2 pneumocyte
cell size as well as cell numbers. From this study, it was found that a
significant size increase for the exposed animals occurred (p < .025) as a
result of the 2 ppm NO- exposure, and that the duration of the exposure
C7, 1*», and 21 days) was also significant (p < .05). This study, and the
image analysis studies that preceded it, show that a study of the micro-
ecology of the lung is both a highly sensitive and meaningful means of
detecting early adverse health effects, as is well recognized for studies of
the ecology in general.
To achieve other means of detecting early adverse health effects, and
for the purpose of achieving correlations with image analysis studies, a
-------
number of physiologic methods were developed. The first of these pathophysio-
logic methods was a disc gel (^PAGE) measurement of the protein content of the
lung lavage fluid of guinea pigs exposed continuously to 0.4 ppm NO- for a one
week period. Higher levels of protein characterized the lavage fluids of the
exposed group of animals (p < .001). This methodology represents a relatively
simple means of detecting a very important pathophysiological alteration, one
that is a common denominator at the very early stages of a great many lung
diseases.
In a subsequent study, two experiments were carried out to compare the
protein content of NO. exposed guinea pigs to that of control animals: a)
continuous 0.5 ppm NO for one and two weeks, and b) intermittent 0.4 NO-
(4 hours per day) for 9 - 1 day showed an increase content of protein by
PAGE in the urine of the exposed animals (p < .01 and p< .05 respectively for
the two experiments). This preliminary study particularly warrants follow-up
investigations and ccnfirmation since it implies a systemic effect by exposure
to N02 atmosphere at an ambient level (0.4 ppm) of NO-.
The study of increased capillary permeability and protein leakage was
expanded through the development of a new methodology, namely the labeling of
plasma protein as it entered the lung, i.e. the use of a fluorescent labeling
compound for primary amines. The results of the study showed that continuous
0.47 ppm NO- for ± 2 days was responsible for an increase of labeled protein
in the lungs of the exposed animals (p < .025). This approach was used to
measure not only increased content of protein in alveolar fluid, as demon-
strated in the earlier studies, but also increased content of protein in
cells, tissues, and lymphatic channels of the lung. This method was.
innovated in response to a need for evaluating the bidirectional cell transport
in the lung of protein (plasma and cell breakdown origin) in the lung.
To expand the fluorescamine studies and also to achieve ultrastructural
correlations, the molecular probe, horseradish peroxidase (HRP), was used to
evaluate independently the influence of continuous 0.5 ppm NO-, intermittent
0.8 ppm NO., and intermittent 0.6 ppm NO-. In "the first of tne two experiments,
the HRP content of lung homogenates (following intravenous injection of HRP
and a testing delay for blood clearance) was found to be elevated (p < .05 in
both exposed groups of animals following three weeks of exposure. No differ-
ences were found after six weeks. In a third experiment, an exposed group of
animals again showed a higher HRP content (p < .025) but only after six weeks
of NO- exposure. In one of the experiments, a difference was noted (p < .05)
for just one of the experimental sets, the exposed groups again having a higher
HRP content than that of the controls. The studies require elucidation.
A number of other experiments were carried out during the investigation.
They include the following: a) the first ultrastructural demonstration that
an unidentified wall-free organism appeared to be a common inhabitant of the
bronchial mucosa of the guinea pig. This organism, with structural variations
from elementary bodies to budding rorms, was characteristically attached to
microvilli and often located in large numbers at the base of cilia. These
organisms are believed to represent a special life cycle for a bacterium
(L-phase), mycoplasma, or some other organism. Our working hypothesis is
-------
that NO- may provide an opportunistic circumstance for the organisms to
develop walls, or in some other way to become pathogens; however, the pre-
liminary work did not show a relationship between NO- exposure and these
organisms. An intensive effort to isolate and culture these organisms with
special L-phase and mycoplasma media was unsuccessful; b) the diphosphoglycer-
ate content of red blood cells of guinea pigs exposed to 0.36 ppm NO- for one
week was found to be higher than that of the control group (p < .05). This
again is a preliminary finding requiring confirmation, but suggesting that
N0_ may have adverse effects outside of the lung, and in this instance, on
the binding properties of red blood cells for oxygen; c) the project contri-
buted a relatively small part of support to other studies where credit for
the support was acknowledged in the report. These include studies with
macrophages in tissue culture and, in particular, the measurement of lung
cell outgrowth in tissue culture by a quantitative methodology involving the
uptake of tritiated fucose and leucine followed by liquid scintillation
counting. The published reports showed statistically significant differences
between the control and exposed groups of animals.
-------
SECTION I I I
RECOMMENDATIONS
The results of the studies strongly support the working hypothesis that
it is the depletion of cell reserves by air pollutants that is the key to
the establishment of air quality standards. A critical need is data on the
acceleration of the rate of cell reserve depletion caused by exposure to
ambient levels of pollutants. Only a beginning has been made in this direction.
Of great pertinence, the hallmark of the human emphysematous diseases is the
covert loss of over 80% of lung tissue before the loss comes to clinical attention.
We consider image analyzer quantitat ion an important advance in detecting very
early loss of Type 1 lung cells, and believe that the foundation we have
established(20,6) should be expanded into studies of the human well populations
as well as Into comprehensive animal model investigations. The very high sensitivity
of image analysis quantitation that can be expected with large volume quantitation,
and the demonstration of pathophysiologic alterations at levels of N02 as low as
O.^ppm, attest to the feasibility and realistic nature of the recommendation.
Some of the immediate questions raised are: 1) what is the lowest level of N02,
singly and in combination with other pollutants, that will cause detectable
structural and functional alterations? ; 2) are the alterations that result
reversible? ; 3) what are the health effects implications of the alterations?
In the latter respect, does protein leakage and Type 2 cell hyperplasia indicate
Type 1 cell damage and loss that represents an early stage of emphysema? Note
that early emphysema is in effect an "excessive" loss of lung cell reserves, and
that image analysis quantitation now makes possible this type of measurement in
both animal model and human studies.
There are other aspects of the microecologic evaluation deserving special
attention, notably in vitro stress testing of cells for vigor or reserve functional
loss. The functional reserves of cell populations will vary according to both
inherited and acquired factors, and clearly play important roles in creating
especially susceptible cell populations.One promising approach we have developed
for measuring individual cell population reserves is the postmortem interval stress
test,i.e. the capacity of the cells to withstand delayed explantation in culture.
This type of dynamic approach should be expanded to include biochemical and
ultrastructural correlations.
Lastly-, air pollution is but one factor in perturbations of the microecology.
The well recognized "constellation of events" from all factors demands some close
cooperation between investigators working in diverse areas. Special efforts are
needed to encourage more integration of work and the exchange of ideas. Our work
with a probable special life cycle of an opportunistic bacterium had the valuable
assistance of experts in mycoplasma/ L-phase cultures. Our work with molecular
tracers would have benefitted by parallel studies carried out by others using
organs other other than the lung that we supply . Note that our urine protein
measurements indicate that N02 very likely has systemic effects, and the DPG
studies as well. Human studies of protein excretion-pollution relationships
would seem to be a very important priority., and should be expanded to include
immunologic and other newer assays. Quantitative work on the bidirectional
transport of protein in the lung also deserves special support since a great
increase in workload no doubt precedes the occurrence of functional and structural
abnormalities which are themselves of a very subtle nature.
-------
SECTION 4
MATERIALS AND METHODS
ANIMALS
For the most part, the studies have used Swiss-Webster mice from various
vendors (Simonson, Charles River, and Hill Top Laboratories). For studies
using guinea pigs, purchases were made from Camm Laboratories, and the
animals were Hartley strain. On arrival, the animals were maintained in
quarantine by the Vivarium for 7-10 days and were then transferred to a
laboratory room maintained exclusively for our exposure chamber use. The mice
were housed in standard plastic cages with stainless steel grids, in a bedding
of wood shavings, and with a maximum of five mice per cage. The guinea pigs
were housed in galvanized iron guinea pig cage racks, with a maximum of five
guinea pigs per cage. For most of the experiments, a standard chow for the
respective animals was provided with food and water ad libitum. In some
instances, the guinea pigs were fed a rabbit chow to provide a low Vitamin
C content diet.
EXPOSURE CHAMBERS
The cages used for the study were two identical cube units, 27 cubic feet
in volume. The cages are wood construction, and the interiors have a complete
epoxy layer. Air flow was regulated to provide approximately one turnover
every two minutes, with a slight negative pressure from the exhaust fan used to
provide air flow. The air was filtered through two units in tandum, a parti-
culate filter and either a charcoal or Purafil filter. Both air intake and
air exhaust were common to both units; the only difference was a mixing
chamber to the air intake of the exposure unit. (In the final year of the
program, a completely stainless steel exposure chamber with laminar flow was
built to replace the epoxy unit). The cages were spaced within the chamber
to provide a correspondence between cage numbers of the control and exposed
animal groups. The bedding (or pans) was changed generally from two to three
times a week, depending upon the need.
The temperature in both chambers was maintained at 23 C - 2 C and at
ambient humidity (generally 50-65% RH). The humidity indicators used in the
chambers were Abbeon AB—1-67.
POLLUTANT
In the study, a single pollutant nitrogen dioxide was used at various
concentrations and at intermittent as well as continuous periods of exposure.
The nitrogen dioxide was obtained from Matheson Chemical Company and it was
added to a volume of silicone fluid (medical grade 360, 500 centistokes, Dow
Chemical Co.). The methodology for this NO- delivery system was developed
earlier in our laboratory and the details can be found in the published report
(19)- In brief, silicone fluid, an inert substance, will absorb N02 inter-
molecularly in an unusually large quantity (the precise amount is now the
subject of an ongoing investigation). Once the N07 gas phase has equilibrated
in the separatory funnel, and the concentration has been found to be
satisfactory through monitoring, the N02 laden silicone is allowed to drip
-------
into the mixing chamber where the NCL is released by Venturi forces of the
mixing chamber. The completely clear silicone fluid is collected in a glass
container at the bottom of the mixing chamber, and is discarded. This delivery
system is highly advantageous. , particularly since the height of the silicone
and air flow can be readily controlled, and thus the level of N0_ established
very easily maintained. The exhaust fan is connected to a constant voltage
unit and the air flow periodically measured by a flow meter. Monitoring of
the NCL was achieved by Saltzman fritted bubbler analysis with routine
standardization and bubble meter monitoring of the fritted bubbler airflow.
Periodic comparison of the NO- levels was achieved by loan use of chemilumin-
escent NO. detectors,and grab sample spot checks.
Lactate dehydrogenase identification of Type 2 pneumocytes for quantitative
measurements
The methodology was first reported in a guinea pig study (20)• *n brief,
the lungs are removed from the animals immediately after death by a lethal
intraperitoneal pentobarbital injection. With the lung placed on a cold
surface, 6% gelatin is instilled via the trachea to achieve inflation of the
lung corresponding to the thoracic volume, and with slow, gentle pressure on
the hypodermic syringe used for injection. Loss of the sharp angulation of
the parenchymal markings was the immediate endpoint for cessation of
injection. The gelatin was allowed to solidfy at 4 C and then the lungs were
placed in a deep freeze for cryostat sectioning (-20 C). Cryostat sectioning
was done at 15 micra, with guinea pig lungs in the coronal plane and mouse
lungs in the horizontal plane. The frozen sections on glass slides were
lyophylized and processed for the lactate dehydrogenase (LDH) reaction. The
details can be found in the earlier report (20). In brief, a tetrazolium
salt is used as the electron acceptor and is reduced to an insoluble formazan
in the presence of the LDH enzyme, with lactate as the substrate. The formazan
reaction product is most heavily deposited in the cytoplasma of Type 2
pneumocytes with a variable and entirely different, type of deposition in
macrophages and with negligible deposition in Type 1 and endothalial cells
under the conditions of the reaction used.
Lung lavage
Immediately after death of the animal by intraperitoneal pentobarbital
injection, the thoracic and abdominal cavities were opened and a portion of
the anterior rib cage resected to provide a wide exposure of the lungs. The
trachea was then prepared f or cannula,tion by removing fatty tissue and free
blood. A short horizontal slit was made in the trachea approximately 5 mm
below the larynx to permit the insertion of a 14 gauge 4 inch long cannula
(for the guinea pig) with attached 20 ml syringe, the cannula and syringe
having been previously filled with Gey's balanced salt solution. For the
mice, a 1 ml syringe was used with an attached 22 gauge cannula. The cannula
was securely tied in place by two encompassing sutures. The lungs were
gently lavaged over four complete cycles of injection and retrieval, with the
lungs being inflated to a point where they occupied nearly completely the
thoracic space. For the guinea pigs, 10 ml of BSS were introduced and 7 ml
recovered. For the mice, 1 ml of BSS was introduced and 0.7 ml recovered.
At the completion of lavage, the cannula was removed from the syringe and the
8
-------
contents of the syringe entered into a chemically clean screw top vial. The
specimens were immediately placed in a refrigerated centrifuge (4 C) and the
supernate stored at 4°C before subsequent testing (k).
Gel electrophoresis
For the gel electrophoresis procedure, a 125 Lambda sample of each
lavage fluid was introduced into the loading gel portion of the preparation.
The method of Ortistein (21) was followed with the exception that sucrose was
omitted from the loading gel, the tubes were rinsed in a detergent solution,
and the order of preparing the vaious gels was reversed, i.e. separatory gel
first, followed by stacking and loading gels. A bromophenyl blue dye was
used in the buffer of the electrophoresis apparatus (BioRad) and a current
of 4 ma was applied for 45 minutes or until the tracking dye extended 1.5
inches beyond the stacking-separating gel interface. The proteins were
stained in 1% amino black for a minimum of one hour. The gel columns were
destained -for four hours and washed in a solution of 7% acetic acid with a
current of 10 ma per tube. A gel scanner recording was made of each tube at
a wave length of 559 mm, at 0.26 mm slit width, and with constant scan and
chart recording speeds. Each gel electrophoresis run involved 12 specimens
representing two matched animal pairs with triplicate samples from each
animal. Additional runs were employed as needed. The scan recordings were
quantitated by either of two methods: a) initially, the use of heavy weight,
semitransparen't drafting paper for tracing the scans, followed by weighing
of the cut-out scans; b) planimetry; c) the use of an integrated chart re-
cording; The gel measurements included those for LDH, ferritin, horseradish
peroxidase (KRP), protein (Amido black and Coomassie blue), and other sub-
strates (16,18).
Protein determinations
Protein analysis on lavage fluid, lung homogenates, and urine were carried
out using a Lowry procedure (22,23) and polyacrylamide gel electrophoresis
(PAGE).
Electron microscopy
Our ultrastructural methodology has been covered in earlier reports
( 1 ,24). In brief, portions of bronchus and lung were cut into 1 mm cubes,
fixed in cacodylate buffer 2% glutaraldehyde, postfixed in 1% osmium, dehy-
drated in ethyl alcohol, embedded in araldite , -epon, or other resins (aral-
dite-opon; Vestopal W), double stained after thin sectioning (500 X) with
uranyl acetate and lead citrate. The sections were examined with a Philips
301 electron microscope, or an Hitachi HU7.
Horseradish peroxidase permeability studies
Following each exposure period, the mice were heated for 30 minutes under
an infrared lamp (37°C) in order to dilate the tail veins. They were then
given an intravenous injection of low to high dose HRP, depending on the
experiment. The high dose was 0.5 mg of HEP (Sigma type II, 50 mg/ml of BSS)
per gram body weight, using a 1 ml syringe and a 27 gauge needle. The low
-------
and other doses are covered in the Results. Each animal was killed at
specified postinjection periods for the permeability studies. Immediately
after death, blood was collected from the jugular veins and spun down in a
microfuge for five minutes. The serum was placed in a deepfreeze at -85°C.
The right ventricle of the heart was perfused with 20 ml BSS (22°C) until
maximal blanching of the lung was achieved, generally a perfusion period of
one minute. The lungs were removed en bloc, placed on a BSS moistened gauze
sponge, compressed gently two to three times by fingertip pressure to express
fluid from the vascular bed and air spaces, placed in a tared vial, weighed and
then frozen in a deepfreeze at -85°C. A similar perfusion and handling
procedure was carried out for the right kidney.
For the HBP assay, frozen lung tissue was homogenized by means of a
Polytron in 2 ml of BSS (5°C) containing 10% isopropanol and 10~"* phenomethyl-
sulfonylfluoride. The homogenate was spun down in duplicate 25 micro liters
and 50yL samples of supernate were loaded on gels for PAGE analysis. A portion
of the supernate was also used to determine protein content by Lowry's method,
and also for comparative studies with the enryme kinetic assay for HRP (see
below). PAGE was run at 2-3 mA per gel for three hours.
The gels were incubated in a benzidine-guaiacol stain for 30 minutes in
the dark at 25 C, and then were fixed in 7% acetic acid. Immediate scanning
was done on a Beckman Acta 3 spectrophotometer.at 525 nM with a scanning speed
of 1.5 cm/min. The tracings were measured through the use of an integrated
recorder with settings at 10 mV at 12,000 cpm, with a chart speed of 4 ons/min.
and a span of 2. Measurements were obtained of HRP content of lung, kidney,
and serum (16,18).
Enzyme kinetic assay for HRP
Assays were -carried out for both lung homogenates and for portions of
gels containing the HRP band (as indicated by a corresponding stained gel
sample). Also included in the enzyme kinetic assay was one of the triplicate
samples of 'serum and kidney homogenate, two processed for PAGE and one for
the assay. For the gels, 1.5 cm portion of the gel containing the HRP band
was homogenized in 1 ml distilled water using a Polytron, and the homogenates
spun at 1,000 rpm for 10 minutes in a refrigerated centrifuge. The samples
were allowed to warm to room temperature and were assayed by a modification
of the 0-diansidine method. A supernate sample was placed in a 1.4 ml glass
cuvette (path length of 1 cm) and 0.9 ml of substrate added to the 0.1 ml of
sample. Following immediate mixing, the rate of color development at 460 nM
was recorded using a Gilford spectrophotometer with a recorded chart speed
of 2 inches per minute, a span of 1, a ratio of 1.0. A temperature of 25 C
was maintained by a Lauder Model K2-R circulating water bath (25).
For the electronmicroscopic studies of HRP, several procedures were used.
These included: 1) perfusion of the trachea with 2% glutaraldehyde, in an
amount sufficient to expand the lungs to fill the thoracic cavity, followed
by tying of the trachea and suspension of the lung in a 2% GTA solution for
two hours. The lungs were then thinly sliced and the sections processed
according to the DBA reaction, with control sections for endogenous
peroxidase; 2) perfusion of the left ventricle with 3 ml of BSS followed by
3 ml of 2% GTA, and after five minutes of perfusion fixation in GTA, perfusion
with 5 ml of DAB solution with suspension in DAB for 30 minutes.
10
-------
Diphosphoglycerate determinations
The method for a simplified assay of 2,3-DPG was developed earlier (26).
In brief, 10 ml of guinea pig heart blood, with potassium oxylate as an
anticoagulant, was centrifuged at 4 C and the packed RBCs resuspended and
washed in isotonic saline. The RBCs were hvimolyzed with cold distilled water
and the protein component precipitated with trichloroacetic acid. The
filtrate was absorbed with charcoal to remove all but two forms of phosphorus,
inorganic phosphorus and that bound to 2,3-DPG. The filtrate was then tested
independently for the latter two compounds by the means of a colorometric
assay based on a. molybdenum compound. Duplicate samples of 0.25 ml and 0.5
ml aliquots were tested (5).
Acid phosphatase
Lung tissue was rapidly sliced into 1 mm cubes in a drop of 0.4 M sucrose
on a waxed plate and then placed into an homogenizer tube containing 1.4 ml
of cold 0.4 M sucrose for each 100 mg of lung tissue. After 10 complete
passes of a glass pestle at 600 rpm, the homogenate was centrifuged for 10
minutes at 140 g and the supernate obtained diluted to its original volume
with cold 0.4 M sucrose at pH 7.2. Two aliquotes of the supernate were
overlayed onto fresh gradient tubes, and each contained 26 ml of a linear
gradient ranging from 0.149 to 2.255 M sucrose (5% to 60% sucrose), and this
in turn was overlayed with 0.1 M sucrose to within 1/8 inch of the top of the
gradient 2. Using a SW 27 rotor, the preparation was centrifuged for 150
minutes at 22K rpm (70Kg) in a Beckman L-350 ultracentri£uge prechilled to
1 C. One of the two gradients was fractionated using a peristaltic pump and
fractionater. The fractioning pipe was inserted from the top of the gradient
tube and lowered by 64.5 turns of the control handle. This placed the tip of
the pipe approximately 1 nM below the lowest observed band. At slow pump
speed, 20 drops (1 ml) were collected for each fraction and 20 fractions were
collected from each gradient tube. Acid phosphatase (27) and protein (28).
Assays were done on all 20 fractions from each animal specimen. For the
enzyme assay, 0.2 ml of each fraction was added to 0.5 ml of substrate, and
the mixture was incubated at 37 C for 30 minutes, following the addition of
5 ml of 0.1 NaOH, a readout at 405 nM was obtained (7,17).
Ferritin studies
Ferritin tracing was used for both PAGE studies and electron microscopy.
All animals were given an intravenous injection of horse spleen ferritin
(Wbrthington) via femoral or tail veins (1 Gm./Kg and.5mg/Gm body weight of
guinea pigs and mice respectively). The methodology followed was essentially
that of Farquhar et al (28) . Both the EM and PAGE procedures were essentially
identical to those described above (29) .
Other procedures
A number of other procedures were carried out during the period of
support. Since the data have not been published only brief mention is made
of the methodology. The main reason for incompletion of the studies was inter-
ruption of the work when the postdoctoral fellows (Drs. Yuen and Hanson)
carrying out the investigations could not be assured of continuing support and
accepted positions elsewhere. Dr. Charles Hanson took on the assignment of
specifying the species of proteins being found in lavage fluid and urine as
11
-------
as well as other biochemical aspects. His work involved PAGE, column separa-
tion of proteins, lung membrane preparations, and SDS-PAGE analyses. Other^
incompleted methodologies include fluor^ietric assays of fluorescamine (carried
out by Lester Layfield), FITC labeled plasma proteins in lung homogenates,
and ferritin tracing studies for kidneys and lungs (Dr. Yuen).
Statistical analyses
The analyses generally use a Student's t test for independent variables
and a two-factor analysis of variance. The details of the analyses are
provided in the published reports. The statistical approach was under the
direction of Stanley P. Azen (a co-author of the book "Statistical Analysis,
a Computer Oriented Approach," Academic Press, New York, 1972) and the problems
of developing and correlating the various methodological quantitative
analyses have been covered in our collaborative report (12).
Image analysis
With the cooperation of Mr. Brian Partridge of the Cambridge Instrument
Company (Imanco) and Mr. Gus Faulhaber of Leitz, the Quantimet 720 and the
Leitz Clasimet respectively, were used for a part of our work. The details
can be found in the published report. In brief, a very practical and highly
reproducible means of obtaining large volume quantitative measurements of
Type 2 cells, alveolar walls, internal surface area, and other parameters have
been demonstrated with the use of these instruments.
12
-------
SECTION 5
RESULTS AND DISCUSSION
Three studies of continuous exposures to 10 ppm NO , 2 ppm NO , and combined
3 ppm NO^ and 5 ppm S09.
Wall free organisms on bronchial mucosa of guinea pigs; Ultrastructural study
The three studies, all with guinea pigs, were as follows: three pairs
control and exposed (continuous 10 ppm NO for three weeks); six pairs control
and exposed (continuous 2 ppm N09 for three weeks); and eleven control and
exposed (3 ppm N0« and 5 ppm S0? continuously for three weeks). All the
animals were studied for the presence or absence of the wall-free organisms
discovered on an earlier screening examination of the electron microscopic
sections. The midportion of the bronchus of the right middle lobe from each
animal was excised under aseptic conditions and portions were taken in a
specific sequential order for the following studies: electron microscopy,
bacteriologic culture with trypticase soy broth, and cultures for L-phase
organisms and mycoplasma using special media applied with and without thallium
acetate by Dr.Leonard Hayflick of Stanford University.
t
Of approximately 500 organisms found in the bronchi of ten guinea pigs,
347 were wall-free organisms (Fig. 1) and the others mature bacterii. With
respect to the occurrence (Fig. 1) of the organisms in the animals, no
differences were found between control and exposed animals, i.e. six of 19
control and four of 19 exposed (Table 1). Further, five of ten animals having
wall-free organisms in their bronchi also showed mature bacteria.. Of 30
animals where wall-free organisms were not found in the bronchi,five did show
intrabronchial mature bacteria. The wall-free organisms were located
primarily at the base of cilia and in between microvilli. Particularly note-
worthy was the apparent attachment in some instances of the wall-free organisms
to both cilia and microvilli. Elementary bodies were also associated with
wall-free organisms. The specific ultrastructural details have been reported
(1 ). In brief, the wall-free organisms ranged in size from 281 nm to 950 nm,
with most of the organisms (92%) less than 500 nm in diameter. The plasma
membranes (trilaminar membrane) had the following measurements, from external
to internal surfaces, 27 S, 20 X, and 2? &. A floccular, irregular thickening
of the outer surface generally measured less than 150 A in thickness, the so-
called "fuzz" layer of wall-free organisms.
Bacterial, L-phase, and mycoplasma culture studies
Of 160 L-phase cultures of lungs, 27 of tracheas, and 48 of bronchi, no
conclusive L-phase organism was identified. In a few instances, cultures
suggestive of L-phase organisms were studied by electron microscopy but a
heavy growth of bacteria around the suspected colonies made selection difficult
and obviated any attempt at confirmation.
13
-------
Figure la. Bronchus of a guinea pig exposed to a combination of 5 ppm SC^ in
combination with 3 ppm N02- Several microvilli appear to be attached to this
wall-free organism, with one microvillus exhibiting an intimate juxtaposition
to 25% of the circumference of the organism. The two, small, spherical,
structures at one pole of the organism are believed to represent elementary
bodies. The cytoplasm of the two cells underlying the organism had focal
areas of presumable degeneration, with loosening of the cytoplasmic matrix as
shown here. The bar is 0.5 u- in length; the organism, one of the largest
found, is 500 nm. x 775 nm. in cross diameters in diameter.
-------
Figure Ib. Bronchus of a guinea pig exposed to 2 ppm N02- Two of the
organisms show bud-Like processes, with one process containing a fairly large
dense body (arrow). The cytoplasm of the underlying cell shows a diffuse
looseness of the cytoplasmic matrix. There is also a suggestive degeneration
of mitochondria. The bar is 0.5 p. in length.
15
-------
Table 1 i
Number of Animals in which Wall-free Organisms were Observed
Experiment Control Exposed
1 0/3 1/3
2 1/6 2/6
3 5/10 1/10
Total 6/19 4 of 19
Numerator represents number ,of animals in which wall-free
organisms were observed; denominator represents number of animals in study,
16
-------
Summary
This study resulted in the first published report on the spontaneous
occurrence of wall-free organisms in the lungs of guinea pigs and their
association with microvilli and cilia (l ). The demonstration of an intimate
relationship between the wall-free organisms and microvilli suggests both a
means of extraordinary resistance to bronchial clearance and a special event
in the life cycle of this so far unidentified organism. A relationship of
exposure of the animals to air pollutants was not established; the numbers
of animals included in the study was insufficient for a definitive evaluation
of air pollutant influence, i.e. increased incidence or conversion.
•
Continuous 2 ppm N02 exposure for 7, 14, and 21 days
Hypertrophy of alveolar wall cells in guinea pigs
Semi-automated quantitation of area of Type 2 pneumocytes and alveolar wall
cell area
Twenty-four male guinea pigs were equally divided into control and NO.
exposed animals and the right lower lobes of four pairs each were quantitated
at each of three exposure periods. The automated stepping stage of the micro-
scope of the image analyzer provided seven microscopic fields (0.38 x 0.47 nm)
for Type 2 and alveolar wall measurements. The measurements included: 1) the
sum of areas of the lactate dehydrogenase(LDH) positive alveolar wall cells;
2) number of alveolar wall cells; 3"6) numbers of wall cells remaining after
step-wise increases in the sizing threshold, i.e. the number of cells remaining
when the analyzer sequentially excluded cells with linear dimensions of less
than 5 micra, 10 micra, 15 micra, and 20 micra respectively. The data were
printed out by a desk calculator interfaced to the image analyzer (2).
The data were based on ratios of the sums of alveolar wall cell areas
divided by the sum of the number of cells counted in that field. This was done
for each of the seven fields. The data have been summarized in Tables 2
and 3- In brief, an analysis of variance showed significant differences
between the NO- and control groups of animals on the basis of the NO-
exposure (p < .025) and also for the duration of the experiment, i.e. greater
change after longer exposure (p < .05). The specific values for each of the
animals has been presented in Figure 2.
Summary
In a study of the lungs of 24 guinea pigs,.half of which were exposed to
2 ppm NO- by image analyzer quantitation of Type 2 pneumocytes, an increase
in the size (hypertrophy) of the Type 2 pneumocyte occurred in the NO-
exposed animals as opposed to the controls (p < .025). The duration of
exposure to the NO- also appeared to influence the size of these cells
(p < .05). The results were based on an image analyzer quantitation of 9,824
lactate dehydrogenase positive alveolar wall cells.
The precise meaning of the hypertrophy is not clear at this time. How-
ever, it is well recognized that both an increase in the numbers and size of
the Type 2 pneumocyte follows a great variety of insults to the lung tissues,
and is considered one of the early signs of damage to the Type 1 pneumocyte.
The findings implicate some impairment of gas exchange secondary to both the
17
-------
220-1
200-
CO O* O •—
R R Sfi
CN CM CM CN CN CN
7 DAYS GAS EXTOL'JRE
14 DAYS GAS EXPOSURE
d'CN CMCN
21 DAYS GAS EXPOSURE
Figure 2. Average area-cell ratio, in arbitrary units, for each of
the 24 animals; 2773 and 2782 appear as outliers, but high value of
latter may be due to experimental treatment. Treatment and duration
both significant at p < .05 level. Treatment was significant for
seven-day group and, with animal 2782 included, for 21-day group
also (solid bars, exposed animals; striped bars, control animals).
18
-------
Table 2 -Absolute Data*
C«U Counts •
Control
Exposed
Total
Mean Anas
&
Diameters,
With Standard Deviations
Control
4.986
4.838
9.S24
C 2.773
S 2.773
Area (jq M)
56.50 = 27.9
50.18 = 18.2
Diameter (p.)
8.27 = 2.0
7.86 = 1.5
Exposed
£2.782
71.31 = 24.7
9.42 = 1.5
42.782 65.41 = 14.6 9.08 = 1.0
'Number of cells counted in 12 control and 12 exposed slides: aosolute values (in microns of the mar-
ginal means of the average area-cell ratio and diameter-ceil ratio for control and exposed animals. Relative
areas are given in Table 2 and displayed in the Figure; conversion from area to diameter assumed arcu-
lanty of cells. As possible outliers, animals 2773 and 2782 were both included and excluded in the calcu-
lations.
Table 3 -Means and Standard Deviations of the Average
Area-LDH-Positive Cell*
Duration of Experiment (Days)
7
C 58.8=24.2
Tr.»trr,~i« n - 4
X 94.2 = 29.9
n-4
14
68.4 = 26.9
n-3
90.9 = 14.1
n-4
21
97.3 = 19.0
n-4
113.3 = 16.7
n-3T
"Relative values in arbitrary units of the means of the average area-cell ratio for the six different treat-
ment-duration groups. The two outliers .ve'e omitted; the dagger indicates the substituted value wnen the
outlier in the 21-day exposed group is included in the analysis.
When n - 4. this value is 136.1 = 47.7.
19
-------
presumed alteration of Type 1 lung cells and the barrier effect to gas
exchange of the thickened Type 2 pneumocyte.
Exposure of guinea pigs to 2 ppm NO. for 1, 2, and 3 weeks
image analyzer quantitation of numbers of Type 2 pneumocytes
In an earlier study (30), randomly selected fields of 141 lung sections
from 2k guinea pigs (half of which had been exposed to 2 ppm NO.) were
examined and a ratio determined of the numbers of lactate dehydrogenase
positive Type 2 pneumocytes to the numbers of alveoli. The methodology
involved projecting 141 35 mm slides on a paper screen and having a techni-
cian count by hand, with pencilled marks for checking off structures counted,
all cells and alveoli found in the 25X microscopic field photographed. In
the present study, the same photographic slides were quantitated by image
analysis, using a Quantimet 720 image analyzer for the detection of LDH
positive pneumocytes and the alveolar counts as previously determined by
hand counts. The findings are presented in Tables 4 to 6, and Figure 3«
The correlation analysis (Fig. 4) demonstrated a highly significant linear
relationship between the average hand count and each of the two automated
cell counts that were carried out (p < 10 ). The correlation coefficient
between the two automated cell counts was 0.945 with a 95% confidence
interval of 0.92 to 0.97- The calculated regression equations for each set
of data are shown in Table 5.
Summary
A prior hand count of 56,587 Type 2 pneumocytes and 27,720 alveoli was
repeated with the use of an automated image analyzer. Type 2 pneumocytes
(LDH positive) were again shown to be increased in a study of 141 photographic
slides of 25X microscopic fields for lungs of 24 control and exposed (2 ppm
NOj) guinea pigs. A high coefficient of correlation was obtained (R > 0.8;
p < 10 ), and the time required for counting was reduced by a great order of
magnitude. Further details can be found in the published report (3 ).
Exposure of guinea pigs to continuous 0.4 ppm NO. for one week.
Polyacrylamide gel electrophoresis (PAGE) for quant^itation of protein content
of lung lavage fluid
The study involved lung lavage of 18 white male, short hair Hartley
strain guinea pigs, equally divided into control and experimental (0.4 ppm
NO.) groups. The results have been presented in a published report (4 ).
In brief, a nested 2-factor analysis of variance (TableTa) showed a high level
of statistical significance (p < .001 for the greater amount of protein found
in the lavage fluid of the NO- exposed animals (Table 7b).. Of further interest,
the densitometric recordings of the lavage fluid PAGE findings showed a
prominent peak (albumin) followed by two minor peaks believed to be a mixture
of a and 6 globulins and X globulins respectively (Fig. 5). The increases found
were believed to be the result of a general rise in protein content rather than
a selected increase in lower molecular weight proteins.
20
-------
Figure 3. Phase contrast photomicrograph of linear alveolar
structures, as detected by the image analyzer.
21
-------
420-
360-
300-
8 240-j
UJ
u
2 180-
120-
60-
fir
90 150 210
AUTOMATED CELL COUNT
r
270
I
330
Figure 4. Plot of average hand cell count versus automated
cell count No. 1.
22
-------
CS-72-4206
i i
— -&r
" • ; : 1 i
,E i | i
•f \ i -: i
' • i • i
. I
• ' •
i
i
1 !
r— iA-! - - -
- n
V ! i: } . t f \
i ; i - 1 ' ! 1 '
• !
!
CS-72-^
i T
H-
122]
Figure 5. Densitometric recordings of lavage fluid disc-gel
electrophoresis. A and B, Typical findings for control animal
specimens. There is one prominent peak and two small following
peaks, which are believed to represent albumins, a mixture of
«t- and JJ-globulins and y-globulins, respectively. C and D, All
of these peaks are accentuated, and the globulin peak is much
better defined. E depicts exposed animal specimen which was an
exception to consistently high values found. Total protein levels
can be found in Table 1, i.e., animals 62, 03, X-^, Xy, and Xg,
respectively. The levels are expressed in arbitrary units based
on the weights of paper cutouts of tracings. Dominance of exposed
animal values, with the exception of animal Xg is evident.
23
-------
Table
COMPARISON OF AVERAGES ± SE OF HAND COUNTS AND AUTOMATED COUNTS
Animal
Group
Control
NO? Expand
Type of
Count
Hand*
Automated
No. 1T
Automated
No. 2"
Hand*
Automated
No. 1T
Automated
No. 2**
Duration of Experiment /days)
7 14
209.92 ±7.57 1 98.S6 ± 8.00
178.1313.70 178.13 ±9.84
205.94 ±12.35 17&28±5.S6
184.46 ±12.48 158.29 ±5.61
21
187.40 ±8.06
160.88 ±10,21
128.79 ±8.76
200.33 ± 7.51
180.29 ± 8.74
158.71 ±9.63
Each number reprev*nts the average of 24 slides from 4 animals (14 day exposed represents 21 slides from
4 animals).
••average ±SE for hand count.
* -aver age ± SE for first automated count.
• '"average ± SE tor-second automated count.
Table 5
LINSAfl REGRESSION ANALYSIS OF RELATIONSHIP OF HAND COUNTS
TO AUTOMATED COUNTS
Automated
Count No. 1
(n-141)
Automated
Count No. 2
(n-60)
Regression of Hand
On Automated Cell
Correlation
Coefficient
0.333- y
(0.80. 0.891
0-3O7* y
(0.69, 0.88)
Call Counts (y)
Counts (x)
Regression
Equation
• 63 .8 + 0.767x
- 94.6 + 0.663*
Regression of
On Automat*
Hand Alv
Correlation
Coefficient
0.891*
(0.88, a92)
0.887*
(0.82, 033}
Hand Call Count
Hand Alveolar Count '
id Cell Count (jtj
•olar Count
Regression
Equation
y - 0.6O7 + 0.803X
y - 0.960 + 0.743x
•P <10~*: numbers in parentheses are 95 per cent confidence intervals.
24
-------
Table 6
ANALYSIS OF VARIANCE OF RATIOS DERIVED FROM
AUTOMATED AND HAND COUNTS
Soure*
Treatment
(T)
Duration
(0)
Lung Sit*
(L)
T x O
Animal
A (TOD
Section
S (ATDL)
Oegreas
of
Fr««dom
1
2
1
2
41
96
Type
of
Count
Auto'
Hand
Auto
Hand
Auto
Hand
Auto
Hand
Auto
Hand
Auto
Hand
Mean
Squares
2.38
2.04
1.83
2.28
0.006
0.191
1.86
1.87
0.686
0.511
0.081
0.064
F
Ratio
4.20
3.99
2.67
4.46
0.007
0.372
2.71
3.66
9.30
7.98
P
Valua
<0.05
= 0.05
<0.10
< 0.025
NS
NS "
<0.10
-------
TABLE 7a. ANALYSIS OF VARIANCE OF PROTEIN CONTENT
Source
Treatment (T)
Animal A (T)
Replicate R (AT)
Degrees of
Freedom
1
16
34
Mean
Square
.02630
.00161
.0000388
F
Ratio
P
Valve
16.34
41.82
< .001
< .001
26
-------
TABLE 7b. PROTEIN CONTENT OF LUNfi LAVACE FLUID*
Control Animals
Headings
Ni
Keudlnga
(1)
(2)
(3)
H
4204
(C,)
.08660
.08615
.08645
.08640
(«. 00022)
4206
(C2)
.06497
.05997
.05989
.06161
(«. 00291)
4208
(Cs)
.07914
.07583
.07177
.07558
(». 00369)
4210
.10456
.09700
.08512
.09556
(*. 00979)
4212
(C5)
.11352
.11411
.11347
.11370
(•.00035) (»
4214
.10000
.10252
.11065
.10439
.00556)
4216
(C7)
.09055
.09635
.09234
.09308
(».00296)
4218
(CB)
.10750
.11752
.10408
.10970
(». 00698)
4220
.08203
.07420
.08014
.07879
(«. 00408)
Exposed Animals
(1)
(2)
ff
4205
.16227
.151154
.18065
.16982
4207
(X2)
.13744
.14200
.14521
.14155
4209
(Xa)
.11299
.11866
.11752
.11639
4211
(X.,)
.12578
.13818
.11968
.12788
4213
(xs)
,14814
.14325
.14751
.14630
4215
(Xfi)
.15142
.15620
.15531
.15431
4217
(X7)
.15214
.15011
.15108
.15111
4219
.15558
.14276
**
.14917
4221
(Xq)
JD7160
.06917
**
.07038
(».01641) (*.00390) (*.00299) (».00942) (».00266) (*.00254) (».00101) (>.00906) ('.00171)
(1) * Expraoued In arbitrary unite according to weights of gel scan cut-oute (Cm)
(2)** Damaged gal
-------
Studies were also carried out in an attempt to correlate proteinuria
and dIsphosphoglycerate content of red blood cells with the protein content
of the lavage fluid. However, the findings were not technically satisfactory
and the data were not published.
Summary
An analysis of protein content in lung lavage fluid by PAGE and densito-
metric measurements has shown a greater amount of protein in the lavage fluid
from guinea pigs exposed to an ambient level of NO., namely 0.4 ppm continu-
ously for one week (p < .001). The loss of protein was considered to reflect
an altered capillary permeability from the NO- exposure, and the possibility
exists that some of the protein may represent an increased turnover of lung
cells. The PAGE method developed for this study is a very practical means for
detecting this important pathophysiologic alteration. In terms of meaning for
air quality standards, it is well recognized that increased protein content of
the urine is an important sign of renal impairment. We believe the same
principle applies to the lung although we do not presently have such informa-
tion in respect to human populations and air quality. Further, the animal
model system appears to be an excellent means of providing this new kind of
discriminant for assistance with the establishment of air quality standards.
0.36 ppm NO. continuously for one week
Measurements of diphosphoglycerate content of red blood cells of guinea pigs
There were nine pairs of animals in the experiment, but only four pairs,
or eight animals, were investigated for 2, 3~DPG content of their RBC. The
specimens of the other animals were unsuitable for analysis due to storage in
the refrigerator beyond 48 hours. Quadruplicate blood samples for each
animal were tested and showed consistently high levels for those specimens
from the NO. exposed animals (p < .05) (Table 8). The mean values for the
control animals ranged from 16.5 ymol to 20.4 jimol, and the standard devia-
tions were less than 1 ymol with one exception. For the exposed animals, the
mean values were 21 umols to 32.2 ymols, with standard deviations from 0.7
Umold to 3*7 mols.
Summary
The 2,3~OPG content of red blood cells of eight guinea pigs, half of
which were exposed to 0.4 ppm continuous NO. for one week, showed signifi-
cant differences, with the exposed animals Raving an increased content
(p < .05). While the findings are clearly preMminary, the study did demon-
strate the feasibility of this newly developed, simplified method for the
determination of OPG, and it also showed good reproducibility. There has
been no prior report dealing with air pollutant effects on DPG other than
mention of its dissociation from hemoglobin under the influence of carbon
monoxide in tissue culture preparations. The significance of the finding is
that the DPG level essentially reflects poor cell oxygenation? The
implication suggested is that an ambient level of NO^ exposure may affect
oxygen binding by the red blood cell. (* "dysoxia")
28
-------
Table 8
2,3-OPG Phosphorus Content of RBC (^mol/ml*) in Guinea
Group
Control
SO
Mean
Exposed
SO
Mean
1
21.30
19.30
19.80
20.05
0.3508
20.1125
24.650
25.450
17.950
25.750
3.6959
23.4500
2
23.20
24.40
17.00
17.00
3.9564
20.4000
21.175
28.775
22.775
24.575
3.2756
24.3250
3
18.625
17.025
16.625
17.325
0.8655
17.4000
30.925
34.625
32.925
30-325
1.9619
32.2000
Pigs
4
17.650
16.650
15.550
16.250
0.8770
16.5250
20.600
22.000
20.700
20.700
0.6683
21.0000
*OPG jumol values are eaual to one half of the phosphorus values.
29
-------
Exposure of guinea pigs to 2 ppm NO. for one week
An automated determination of ratios of Type 2 pneumocytes to alveolar wall
area using an image analyzer
In two earlier studies Type 2 pneumocytes were counted by hand and by
an image analyzer, with the latter using one of the hand count measurements
as a denominator for a ratio determination. In the present study, the lungs
of 2k animals, equally divided into control and exposed groups, were
analyzed using microscopic slides from an earlier study (30) and substitu-
ting one of the serial sections for the original one used. The purpose of
the study was to achieve a count of Type 2 cell pneumocytes based entirely
on image analysis rather than be dependent upon a hand count for the
denominator, i.e. numbers of alveoli. The denominator selected was the wall
area as measured by the Quantimet and as displayed by a phase contrast
image for image analyzer detection. The details of the study have been pre-
viously reported (6 ). In brief, a stepping stage of the image analyzer
automatically selected six fields in addition to an initial one randomly
selected by the investigator. Two sets of five scans each were performed in
each field, in accordance with five sizing thresholds, i.e. at threshold 1,
all detected LDH positive pneumocytes were counted, including those repre-
sented by a small amount of cytoplasm. The other thresholds (2-5) detected
cells with diameters greater than 5, 10, 15, and 20 mi era respectively. The
LOH positive Type 2 pneumocytes were quantitated with light transmission
microscopy; the alveolar walls by phase contrast without changing the field.
Data analysis was carried out for five values of Type 2 pneumocytes and nine
for wall areas, for the lungs of each animal. The 45 sets of ratios were
correlated with the ratios of cell/alveoli obtained from the hand count
analysis of corresponding photographic slides. The results are presented in
Table 9, which shows that of the nine correlations between numbers of alveoli
counted by hand and those evaluated by image analyzer, the correlation with
Wall.-Wall, was the most significant (R= 0.563; P < 0.002). Figure 6 shows
the 2k data pairs as a graphic plot. Of the five correlations between the
numbers of alveolar wall cells counted by hand and those by image analyzer,
the correlation of cell, was the most significant (R - 0.77; P < 0.001).
However, all other correlations were almost as significant. Table 10 presents
the correlations between the ratios of cells/alveoli from the hand count and
those from the cell/wall area; these gave the highest correlation coefficient.
The conclusion was made that the tedious hand count methodology of cells and
alveoli could be abandoned completely in favor of cells and wall area in view
of the strong correlation (R > 0.7; P < 0.001).. Further the methodology has
now opened the way for a greatly expanded study not only of numbers of Type 2
pneumocytes but increases in size as well. In addition, a mathematical
analysis for determining numbers of alveoli and internal surface area from the
image analysis methodology now appears to be feasible.
Summary
A method has been developed which permits a totally automated quantita-
tion of numbers and size of alveolar wall cells and amount of wall area, for
the purpose of quantitat ing hyperplasia and hypertrophy of Type 2 pneumocytes
following air pollutant exposure at near ambient and/or ambient levels. The
30
-------
160-
150-
14O-
130-
120-
110-
100-
90-
»
7O-
60-
3000
4500
6000
7500
9000
WALL AREA (WALL —WALL )
Figure 6. Plot of hand count of alveoli versus automated
measurements of wall area for the 24 slides. The correlation
coefficient is 0.563.
31
-------
Table 3
AVERAGE VALUES OF MEASUREMENTS OF ALVEOLAR CELLS AND
AREA OF ALVEOLAR WALL FOR 24 ANIMALS
Calh
Call?
Calls
CalU
Calls
Sizing Factor
> 0
> 5
>10
> 15
>20
No. of
Maan
58.3
30.6
13.3
8.4
3.0
Calls
SE
4.4
3.8
Z2
U
0^
Sizing Factoi
Wall (tun)
Want
Wall}
Wall3
walU
wails
> 0
> 5
>10
>15
>20
Wall A
Maan
15910.3
9879.4
5298 J)
3034.4
16S8.6
r«a*
SE
1049.5
983.0
785.5
S8Q.7
415.0
Calculated Quantities
Call]
Call!
Calll
Caill
+ Cailj > 5*
+ Cell3 >10f
+ CalU >15f
+ Cells >20r
89.0
71.6
64.7
61.3
73
6.4
5.5
5.0
Walli - Wall?
wall] - Walla
Walli - WalU
Walli - Walls
< 5
<10
<15
<20
6O3O.9
10614.3
12875.8
14251.7
249.3
451.2
624.6
751.7
•Artrtrary Units.
f Calls graatw man sizing factor ara counted as two calls.
32
-------
Table 10
COEFFICIENTS OF CORRELATION BETWEEN RATIOS OF CELLS TO
ALVEOLI OBTAINED BY HAND COUNT AND AUTOMATED
DETERMINATION OF RATIO OF CELLS TO WALL AREA USING
SELECTED NUMERATORS AND DENOMINATORS
O •nominator
W»Mi -
Numerator Wall] • Wall; W«ll3 W»IU Walls
C*li
Cttti + C»ll2
Ctfli + Ca«3
C*Ui + C*il4
C«i + C*«s
0.42
0.63
0.64
OJ6
OJ1
0,79
0^0
0^0
0.79
0.79
0.73
0.79
0.79
0.7T
0.76
0.66
0.77
0.77
0,73
0.71
058
0.73
a73
068
0.64.
33
-------
study has demonstrated a high correlation between hand and automated cell
counts (R >0.7; p < 0.001) and between hand counts of alveoli and automated
wall area measurements (R >0.56; p < 0.002). A great usefulness of this
methodology for assistance in the establishment of air quality standards is
indicated by these findings. There is also a great potential for other
applications, including assistance in the definition of early pathologic
lesions.
0,4 ppm NO- exposure, 4 hrs/day, for one week and three weeks
Acid phosphatase in density equilibrium fractions of lungs of guinea pigs
Three independent experiments, each including eight guinea pigs, half
exposed to 0.4 ppm N02 for four hours/day for seven days, were carried out
for acid phosphatase content of lung homogenates centrifuged on a 5%~6% linear
sucrose density gradient. Five distinct bands, from six pooled fractions,
were observed in each gradient (Table 11) and these were isolated, pooled, and
tested spectrophometrically for acid phosphatase activity. Two differences
were found with respect to the N0_ exposed animals in all of the three studies,
a greater overall (all fractions) amount of acid phosphatase activity
(p <.025) and a density gradient region containing primarily membrane frag-
ments which appeared to contribute most heavily to the overall difference in
acid phosphatase activity (p < .001). The details are provided in Tables
12-15- The findings are considered to be of a preliminary nature in view of
the small number (24) of animals involved in the study, but there is the
implication that this ambient level of N0« applied intermittently for one
week may have resulted in subtle alterations in parenchymal cells and
macrophages with a resultant increase of lysosomal enzymes.
Ultrastructure
The studies showed the pooled fraction with the highest content of acid
phosphatase to be characterized by membrane fragments and cytoplasmic debris;
no specific structural relationship was established, but other fractions had
much less of this material.
Summary
In three independent experiments, 24 guinea pigs were exposed to 0.4
ppm N0~» four hours/day, for seven days (Experiments 1 and 2) and 21 days
OExperiment 3). Density gradient centrifugation was carried out with the
homogenized lung tissue and acid phosphatase determined in five pooled
fractions of each gradient. The acid phosphatase content of all five frac-
tions was higher in the exposed group of animals than in the controls (p < .025)
There were also significant differences between the pooled fractions (p < .001).
The acid phosphatase content, as determined by assay of the supernate from
whole lung homogenates, did not show differences due to NO. exposure.
Continuous 0.5 ppm N02 for seven days and 14 days, and intermittent 0.4 ppm
NO., four hours/day for eight, nine and ten days.
Proteinuria rn guinea pigs
Protein quantitation
34
-------
Tabla 11. BAMl) PROPERTIES
Pooled
Fractions
I
II
III
IV
V
VI
Bands
Fluffy white
Light brown
(No band
observed)
Opaque white
Opaque white
Translucent
light brown
Band
Position*
1.0
2.2
3.9
4,8
5.2
Mean Density
of Band
2.174
2.077
1.939
1.866
1.834
Band
Width dm)
2.0
1.0
3.5
2.0
1.5
Fractions
Pooled
1, 2
3, 4, 5
6, 7, 13,
14
8, 9, 10.
11, 12
15, 16, 17
18
19, 20
Band
Margins
Vague
Sharp
Vague
Vague
Sharp
*Dlstance (on) from bottom of gradient tube
-------
Table 12. ACID PHCSPKATASS READINGS*
(Experiment 1)
Pooled Fraction
Control
Ani.im.1s
Mean
SD
Eroosed
Aninals
Significance
of Treatment
C 1
C 2
C 3
C 1*
X 1
X 2
X 3
X U
I
.0159
.0101
.0109
.0161
.013250
.003193
.02W*
.0259
.0093
.0132
.018206
. 00820U
P < .05
II
.021U
.0080
.0213
.0196
.017575
.0061*36
.0217
.0259
.0229
.0236
.023525
.001767
P < .05
III
.0191*
.0078
.0210
.0201*
.017150
.006268
.0187
.0232
.021*8
.0188
.021375
.003100
p < .05
IV
.0189
.0106
.0225
.0199
.017975
.00511*5
.0191*
.02U8
.021*9
.0131
.020550
.005591
3S»»
V
.010U
.011*2
.0113
.0122
.012025
.001625
.0102
.015l»
.OH9
.0171*
.013725
.003269
NS»»
VI
.0085
.0129
.0103
.0121
.010950
.001962
.0078
.0099
.0118
.0178
.011825
.001*205
rrs»*
* Adjusted tar proreln content of each pooled fraction
»* NS = Sot significant - ? > .05
Overall significance: p < .025
36
-------
Table 13- ANALYSIS OF VARIANCE OF ACID PHOSPHATASE DATA
(Experiment l)
Treatment
(T)
Fraction
CD
T x F
Animal
A (TF)
Degrees of
Freedom
5
5
36
Mean
Squares
.000137
.OOOllU
.000008
.000022
F
Ratio
6.22
5.19
0.35
P
Value
< .025
< .001
HS
37
-------
Table 14. ACID PHOSPSATASE HEADINGS*
(Sxnerlrent
Pooled Fraction
Control C 1
Animals C 2
C 3
C 1*
Mean
SD
Mean**
SD»*
Exposed X 1
Am'mnls X 2
X 3
X 1*
Mean
SD
Significance
of Treatment
I
.0030
.001*3
.0138
.0013
.005600
.005602
.002933
.001601
.0035
.001*1
.0069
.OOUl
.001*650
.001526
3S
II
.0072
.0089
.Oll»2
.001*2
.008625
.OOU193
.006766
.002379
.0092
.0127
.0107
.0117
.011075
. 001U93
p < .05**
III
.0081
.0108
.0161
.0068
.010U50
.001*113
.008566
.00201*0
.0110
.0127
.0106
.0087
.010750
. 0016" U2
!IS
IV
.0126
.0158
.01U3
.0089
.012900
.002969
.0121*33
.003U53
.0157
.0169
.0135
.0113
.011*350
.0021*73
35
Y
.0033
.0038
.0029
.0021*
.003100
.00059U
.003166
.000709
.0031
.0031
.0031*
.0025
.003025
.000377
JIS
VI
.0019
.0030
.0022
.0018
.002225
.00051*3
.002233
.000665
.0019
.0022
.0027
.001*7
.002875
.001260
as
* Adjusted for protein content of each pooled fraction
** Data based on exclusion of one aninal (C 3)
^S = Hot significant - p > .05
Overall significance: H.S. (p < .005 without animal C3)
38
-------
Table '5- ACID PHOSPHATASE HEADINGS*
(Experiment 3)
Pooled Fraction
Control
Aniaals
Mean
3D
Exposed
Qfrf tfif^ g
Mean
SD
Significance
of Treatment
C 1
C 2
C 3
C k
X 1
X 2
X 3
X U
I
.161
.fc55
.657
.061
.333500
.272902
.301
.1*03
.315
.190
.302250
.087397
SS«»
II
.U96
= 566
1.121
.221
.601000
.377292
.655
.887
.731
1.226
.87U750
.253298
!JS»*
III
.777
.681
l.W»0
.566
. 866000
.392267
.956
1.107
.978
.986
1.006750
.006750
!IS*»
17
l.lUo
1.088
.909
.676
.953250
.2096U9
1.1»35
1.577
1.163
.862
1.259250
.315660
HS*»
V
.8U8
.718
.6U6
.520
. 683000
.137098
.897
.903
.905
.588
. 823250
.156870
2TS»«
VI
.576
.596
.l»2l*
.1*21
. U89250
.083719
.523
.585
.630
.851
. 6^7250
.1U27'*0
jfS»»
* Adjusted for protein content of supemate (vhole homogenate)
** HS * Hot significant - p > .05
Overall significance: p < .025
39
-------
This study consists of two independent experiments, one involving a
colony of 14 guinea pigs, half of which were exposed to continuous NO-,
with testing of three of the animals at seven days and four at 14 days, 2)
in the second experiment, there were 24 animals, half of which were exposed
to 0.4 ppm N02 for four hours per day, with testing at 8, 9, and 10 days
after exposure. The urine was collected from the bladder with a sterile
hypodermic needle and syringe and processed for polyacrylamide gel electro-
phoresis (Fig. 7\ Lowry protein determinations, and refractometer measure-
ments of specific gravity. The results have been reported previously (8).
In brief,the two-factor analysis of variance statistical study showed
significant effects for NO. exposure (p < .01 for the first experiment and
p < .05 for the second experiment but without an influence by the duration
of exposure, i.e. one week vs. two weeks. The same statistical significance
was found in both of the analyses carried out, i.e. PAGE and Lowry protein
determinations (Tables 16-20X In a study of water consumption by 162 control
and 116 experimental animals in four independent experiments, no statisti-
cally significant differences were found.
Ultrastructure and histopathology
Representative sections of the cortex and medulla of the kidney were
processed for ultrastructural examination as well as routine histologic
study. No kidney abnormality was noted on routine study. The ultrastruct-
ural studies are incomplete (loss of Dr. Ted Yuen to the project).
Summary
Urinary proteins were quantftated by PAGE, Lowry protein tests, and
refractometry. There were two experiments, with 14 animals in the first
with continuous 0.5 ppm exposure for one and two weeks, and 24 in the second
with four-hour per day 0.4 ppm NO- exposure for eight to ten days. Both
PAGE and Lowry protein measurements showed a greater amount of urinary
protein for the exposed animals (p < .01 and p < .05 respectively). While
this study is clearly of a preliminary nature, there is an implication that
exposure to NO- may have an adverse systemic as well as local lung effects.
An increase in protein content of the urine is often an early indication of
subclinical renal disease in humans and this often occurs in the absence of
structural abnormality by routine histological examination.
Continuous 0.47 ppm NO. for ten* 12 and 14 days.
Protein leakage in the mouse lung
Fluorescamine labeling studies
Forty-four Swiss Webster male mice were divided equally into control and
experimental groups, matched according to weight. A fluorescamine solution
was injected immediately after its constitution (2 mg of 'fluorescamine powder
in 0.15 ml of 100% ethyl alcohol, followed by a mixing of 0.05 ml of saturated
supernate to basal salt solution for a final volume of 0.4 ml + The solution
was injected into a tail vein over a period of three minutes - 30 seconds. At
precisely 40 minutes after injection, and following a lethal dose of intra-
peritoneal pentobarbital sodium, the chest cavity was opened and blood
collected from the right ventricle. Fifty yl of serum was placed in 0.5 ml
40
-------
/v.
•\ Jl
Figure 7. Comparisons of densitome trie readings: disc-gel
electrophoretic determinations of urinary protein. A represen-
tative number of gel scan tracings are displayed, control group
on top and exposed on bottom. As a rule, three major peaks
presumably albumins, combined
-------
Table 16. (Experiment 1)
DATA SUMMARY OF URINE ANALYSIS
7 Days Exposure 14 Days Exposure
Protein
S.G.
(mg/Z)
CTL 1.007 5.2
XL "1.035 17.1
C2 1.013 5.4
X2 >1.035 14.8
C3 1.009 5.7
X3 1.027 14.8
Gel Protein Gel
S.G.
Units* (mg/%) Units*
.13134 C4 1.008 5.4 .22477
.74471 X4 1.018 ' 23.7 1.09729
.22255 C5 1.013 5.4 .23359
.62972 X5 1.035 19.5 .89827
.34600 C6 1.013 4.9 .11765
.62227 Xg 1.020 5.7 .27630
C7 1.013 5.2 .14577
X7 1.012 5.3 .20001
*Weight of paper cut-outs for entire protein fraction
42
-------
Table 17. (Kitpurlmuiit 2)
ALBUMIN CONTENT OP URINE DISfCEL ELECTROPHONESIS
8 Days Exposure
'Sample* Sample
1
Cj 71
•Xi 125
i C? 151
X2 211
C3 56
X3 -
C4 44
X4 108
2
67
140
157
223
54
~
44
96
FT
69
132.5
154
217
55
—
44
107
S.D.
2.83
10.61
4.24
8.49
1-.41
--
0
8.48
9 Days Exposqre
Sample* Sample
1
C5 34
X5 221
C6 52
X6 31
C7 39
X7- 274
Cfl 154
Xfl 347
2
40
209
60
28
39
260
160 .
331
FT
37
215
56
29.5
39
267
157
339
S.D.
4.24
8.48
5.66
2,12
0
9.90
4.24
11.31
10
Days Exposure
Sample* Sample
1
C9 301
X9 32
C10 47
XlO 52
GU 400
xll 532
Ci2 306
Xi2 1500
2
330
32
48
50
390
500
321
1400
315.5
32
4.75
51
395
516
313.5
1450
S.D.
20.51
0
.71
1.00
7.07
22.63
10.61
70.71
* Sample unite are bauod on weights of paper cut-outa from albumin fraction.
-------
Table |8
(Experiment 2)
Cl
*1
C2
X2
C3
X3
C4
X4
8 Daytt Exposure
Urine Vol. Urine
ml S.G.
4.4 1.023
1.4 1.015
0.5 1.028
0.9 1.017
3.8 1.035
__
1.0 1.035
1.0 1.028
9 Days
Urine Vol.
ml
C5 2.0
XS 2.4
C6 2.0
X6 2.2
€7 0.4
X7 1.0
CB 2.0
X8 5.0
Exposure
Urino
S.G.
> 1.035
1.021
> 1.035
1.029
1.028
1.022.
1.035
1.027
\
10 Days
Urine Vol.
ml
Cg 1.0
x9 3.6
CIQ 2 .0
X10 0.8
GH 2 .0
*U 2.0
C12 1.6
X12 3.2
Exposure
Urine
S.G.
1.035
1.021
1.031
1.028
> 1.035
1.021
> 1.035
1.033
-------
Table 13 (Experiment 2)
TOTAL PROTEIN CONTENT (DISC-GEL) OF URINE
Control
Mean
S.D.
Exposed
Mean
S. D.
8 days 9 days 10 days
..09094 .14423 .32136
.04135 .08915 .19410
.169113 .28282 .43248
.09048 .20680 .41568
* n *» 4 for each group, except for the 8 day exposed group
where n « 3
45
-------
Table 20
Protein
Cl
C2
Xl
x,
Content (aig/%)
-
2.000
1.020
1.085
1.754
With and Without
+
.456
.471
.703
.828
Dialysis
Gels
.1296
.1291
.1975
.2719
46
-------
basal salt solution at 4 C for fluormetric assay. The lungs were washed
briefly in cold basal salt solution, and pressed between two gauze sponges
to remove surface blood and capillary blood content, and homogenized. The
fluorescence of each sample was recorded between 410 nm and 700 nm, with the
excitation wave length set at 390 nm (emission peak was found to be 480 nm) .
For each of the various samples, a standard was prepared with a fluorescein
stock solution that was diluted as needed to 5x10 . The peak reading
obtained from the standard was used to standardize subsequent readings, both
at the beginning and the end of each animal pair unit experiments.
^Fluormetric readings were obtained of lung homogenates and serum. The
findings for the lung are listed in Table 21, and the details have been
presented in a published report (9). In brief, there were consistently
greater mean values for the exposed animals for each of the three time
periods compared with the control counterparts. Also, the mean value of the
combined exposed groups (i.e. for all three periods of exposure) was greater
than that of the control counterparts (p< .025). The results of the serum
tests are given in Table 22; no significant differences were found. Further,
a linear regression was carried out comparing serum and lung levels of bound
fluorescamine. The correlation coefficient was determined to be 0.18547 and
thus not significant.
Ultraviolet microscopy
An ultraviolet microscopy study (FITC excitation filter and Zeiss OGI
barrier filter, with epi1uminescence) showed small aggregates of amorphous
material randomly distributed throughout the lung, plus a generalized back-
ground fluorescence of the lung tissue which was greater than that seen with
the autofluorescence of lungs from mice not receiving the f1uorescamine in-
jection. The overall impression was that labeled protein was distributed
throughout all tissues of the lung, with no aggregate localization that was
consistent in terms of distribution and amount. It was not possible to dis-
tinguish control from exposed animals by microscopy. The findings are
consistent with others who have shown a bidirectional transport of protein in
the lung, which includes intracel1ular protein aggregates and lymphatic
clearance. This is the first study using the fluorescamine label in an
attempt to demonstrate a greater loading of the protein transport system of
the lung following air pollutant exposure, and also appears to be the first
to attempt this type of quantitation regardless of methodology. Hypotheti-
cal ly, a measure of protein transport in the lung may be much more sensitive
than lavage fluid measurements in view of the great capacity of the cells of
the lung for protein reabsorption.
Summary
The protein content of homogenized lung tissue from 44 mice, half of
which were exposed continuously to 0.47 ppm NO^ for 10, 12, and 14 days, was
assayed fluormetrically following intravenous injection of fluorescamine, a
new protein labeling reagent for quantitative assay use. The mean protein
values of all exposed animals were higher than those of the control animals
(p < .025). No correlation was noted between serum fluorphor levels and
those in the lung homogenate. Ultraviolet microscopy studies support the
working hypothesis that protein leakage is associated with an increased load
of protein in the protein transport system of the lung.
47
-------
LUNG
TABLE 21
COSTBDL
DAY 10
DAY 12
DAY 14
79
87
73
92
69
82
48
95
=ean: 73.75
S.D.; 14.32
72
87
83
107
79
108
119
140
mean: 100.00
S.D.: 22.71
85
79
130
56
102
52
-
-
mean: 34.00
S.D.; 29.23
106
81
128
114
120
90
-
-
aean: 106.50
S.D.: 18.01
39
71
39
97
146
150
73
89
aean: 94.37
S.D.; 37.23
79
80
102
67
96
205
300
500
=ean: 173.62
S.D.: 152.33
~ Control mean: 86.0
S.D.: 23.1
Exposed sean: 130.4
S.D.; 97.1
p < .025
48
-------
DAY 10
TA3LE 22
SEREM FLUORESCStCS
DAI 12
* Concrol aiean: 123.5
S.D.: 39.9
Exposed mean: 139.3
S.D.: 47.6
p < .25
DAT 14
143
169
95
111
132
150
141
mean: 134. 42
S.D.: 24.71
152
83
91
86
150
108
75
aaan:106.42
S.D.: 32,06 -
179
160
87
115
171
146
-
mean: 143. 00
S.D.: 35.49
141
132
201
242
209
114
-
mean:173.16
S.D.: 51.04
76
40
119
72
163
73
-
swan: 91.33
S.D.; 43.18
119
104
125
147
200
178
—
mean: 145. 50
S.D.: 37.07
49
-------
Continuous 2.0 ppm N02
An experimental model and automated methodology
Experimental model
The data from an experiment involving 2k white, all male guinea pigs,
half of which were exposed to NO- for 1, 2, and 3 weeks, were analyzed by a
partially nested analysis of the variance model. There were three crossed
factors: 1) treatment (T) with two levels (control or N02 exposed); 2) dura-
tion (D). with three levels (1, 2, or 3 weeks); 3) and loBe (L), with two
levels (upper and lower). The other factors in the design were the error
factors given for the four animals (A) for each treatment-duration lobe (TDL)
combination; the section (S) factor (three levels) nested within treatment-
duration lobe animal (ATDL); and count factor (Error) with two levels nested
within all factors. The expression for this model system is given in Figure 8.
Table 23 presents the results of the analysis of variance. The results_show
that exposure to NO. after 2 ppm level resulted in an increase in Y from
1.99 to 2.23 cells/alveolus (p = .06). _Thus, this experimental model was
shown to detect a 12% increase in Y using only small numbers of animals
and a slightly supra-ambient concentration of NO..
Automated methodology
Using the data from the earlier studies on automated counting of cell
size and cell numbers, various estimates of XI and X2 were considered as
partially shown in Figure 9- The quantities cell, and wall, served as correc-
tion factors in the estimation. Since cell, is a count of cells larger than
5(i-l) yM, and since these cells have already been counted in cell,, the
cell, plus eellj effectively counts these cells twice. Hence in this way
cell, plus eellj corrects for the counting of two contiguous cells as one
large cell. Similarly, the subtraction of wall, from wall, has the effect of
excluding walls thicker than 5 (j-1) pM. This Is desirable since alveolar
walls are very thin. By correlating quantities X. and X-* (averaged over the
seven microscopic fields). With X. and X- respectively, it was demonstrated
that the best estimator of^X] was X. =• cell, (r =» 0.77; P < .001), and the
best estimator of X2 was X'' = wall, - wall- (r = 0.56; p < .002). Since the
35 mm slides could not be used for a phase contrast study, the area scanned
was slightly different, and hence a less pronounced correlation of wall area
with alveoli was anticipated. The correlation of Y*= X,/X * with Y was
r = 0.79 (p < .001)- For a fuller explanation, see the report (12).
Automated quant?tation of hypertrophy
The analysis of variance of the data (using averages for each of the 2k
animals) revealed no significant treatment-duration-interaction effect, but a
significant treatment effect (F = 6.93, p < .02), and a significant duration
effect (F = 5.33, p < .05). Although no direct correlation with a hand quan-
titation was possible, these results closely parallel results of the analysis
of the variance of the hand hyperplasia analysis of variance (treatment
p < .06, duration p < .025; Table 23.
The present study adds strong support to the feasibility of the automated
approach to lung cell and lung tissue quantitation. In particular, the image
50
-------
Figure 8
Definition of the Model
Y..7, . . . .. . ., , . - -
ijklim i j T,J k ijkl ijklm
-------
Figure 9
Estimates of Xj and X2
for i=2,.. .5, and
*
Estimates of Z and
52
wa111-wa11 .
-------
Table 23
THE ANALYSIS OF VARIANCE OF THE EXPERIMENTAL DATA
Source
T
D
L
TD
A(TDL)
S(ATDL)
Error
df
1
2
1
2
41
96
144
MS
4.02
4.69
0.42
3.62
1.02
0.13
0.04
F
3.94
4.60
0.41
3.54
7.85
3.25
_
P
.06
<.025
NS
NS
<.001
<.001
-_
53
-------
analysis methodology has the potential of contributing to the definition of
early pathologic lesions in the lung, especially with respect to the types of
lung damage that result in cell population shifts and the covert loss of the
cell populations themselves.
Summary
A series of studies designed to develop and automate a method for the
quantitation of Type 2 pneumocyte populations of the lung and their changes
in response to 2 ppm NO- exposure has been presented and discussed. The
validity and sensitivity of the automated approach was strongly supported by
high correlations between manual and automated cells counts (r> 0.8,
p < 10 6) and between their respected baselines (numbers of alveoli and
alveolar wall area; r = 0.56, p < .002) as well as by the automated detection
of significant hypertrophy of Type 2 cells following NO- exposure (p < .025).
Horseradish peroxidase studies at three levels of N02 exposure: continuous
0.5 ppm NO., intermittent 0.8 ppm NO-, and Intermittent 0.6 ppm NO^
Clearance of exogenous horseradish peroxidase in the lungs of mice
PAGE quantitation of lung homogenates
The combined studies resulted in data from 264 mice, with 48 pairs of
mice per each experiment except for Experiment 1 (Table 2k). The results of
this interrelated study of three independently carried out investigations are
as follow: 1) the total HRP content (combined a and 8 bands) of homogen-
ized mouse lung tissue was greater for the exposed animals of Experiments 1
and 2 after three weeks of N02 exposure, but not at six weeks of exposure
(Tables 25-28); p < .05 in both studies by a Student's t test of the three
week exposure period alone; 2) in the third experiment, the HRP content of
the lung homogenates was again greater for the exposed animals (Tables 29+30)
but at six weeks and not at three weeks (p < .025); 3) differences in HRP
content were also found with respect to the individual a and 3 bands of HRP,
but the findings did not show a consistent pattern for the increases; 4) the
serum content of HRP did not appear to be influenced by N02 exposure (Table
31); 5) the HRP content of the kidney was increased (p < .05) for the
exposed group of animals of one of the three experiments (Table 32).
Enzyme kinetic assay v
In Experiment 1, the mean HRP lung content at three weeks for exposed
animals was again great (p < .025; Table 25). For Experiment 2, the three
week levels for the exposed animals were greater than those of controls but not
significant in view of high variation in the exposed group (6.4 ± 5 control
vs. 9.4 ± 11 exposed). In Experiment 3, kidney HRP at six weeks was again
greater for exposed animals (p < .05; Table 32).
Electron microscopic studies
These were incomplete at the time the project was terminated. A search
was made for differences between the control and exposed groups of animals in
terms of the presence of HRP in the various types of cell junctions. At the
time of termination of the project, no overt differences could be identified.
54
-------
Table 24a
EXPERIMENTS AND CONDITIONS
Experiment No.
1
2
3
A 48
B 24
A 48
B 48
A 48
B 48
Duration
(Wks)
3
6
3
6
3
6
NO^
ppm
0
0
0
.5
.8
.6
Exposure
Period
Contin.
6Hr/day
5 d/wk
6Hr/day
5 d/wk
Table" 24b
MEAN ANIMAL WEIGHTS (GMJ
- Ml 08
0
3
6
C:
X:
C 22.6
Weeks
X 22,9
C 31.3
Weeks
X 32.4
C 38.7
Weeks
X 38.8
±
±
±
±
±
±
2.
1.
2.
3.
3.
3.
7
7
8
3
7
5
26.
26.
33.
34.
34.
Ml 09
8 ±
5 ±
4 ±
0 ±
2*±
33-3*±
2
4
3
4
4
6
.9
.4
.7
.2
.5
.3
29
29
35
36
38
39
Mil;
.0 ±
.0 ±
.7 ±
.3 ±
.4 ±
.3 ±
1
2.0
1.9
2.2
2.6
3-1
3.3
Control Animals
Exposed (NO,
,)
Animals
y» mean weight loss in 3 weeks for exposed group and
3% gain for controls vs. }}% •+ 32% weight gain for
Ml08 and Mill animals at S weeks.
55
-------
TABLE 25. EXPERIMENT 1
HRP CONTENT OF MOUSE LUNG TISSUE (HRP/GM TISSUE)
Peak Group
a
8
0*8
a/B
HRP/ek*
C
X
C.
X
C
X
C
X
C
X
3 Weeks
M p Value
30.53
38.09
8.49
11.13
39.02
49.22
3.9
3.8
10
15
± 11. 1
< 0.025
± 12.9
± 4.0
< .05
± 5.6 ( ".!)
± 14.1
< .05
± 17-8
± 1.2
N.S.
± 1.2
± 4
< .025
± 9
6 Weeks
M
44.0 ±
•-
48.4 ±
9.9 ±
12.9 ±
53.8 ±
61.3 ±
2.7 ±
2.3 ±
8 ±
12 ±
p Value
11
N.S.
15
4
N.S.
7
13
N.S.
22
1.0
< .1
0.8
7
N.S.
15
Values in parentheses are protein compensated and shown only when
different
—ft
*Enzyme kinetic assay: 10 ° Gm HRP/Gm tissue
i"HRP values by integrated gel scan areas (Peroxidase stained gels)
56
-------
TABLE 26. EXPERIMENT 1
TWO FACTOR ANALYSIS OF VARIANCE
A. Serum HRP
Peak
a
3
cH-6
a/6
B. Lung HRP
a
6
a+6
o/B
Content*
NO, Factor
N.S.
N.S.
N.S.
< .025
Content*
» . •
N.S. (<0.025)
<0.05 (N.S.)
<0.05 (<.025)
N.S.
Time Factor
*\» 0
«v. 0
»\. 0
•\i 0
<\t 0
N.S. (~ 0)
<.005 (~ 0)
< .025
N02 Time
Interaction
N.S.
N.S.
N.S.
N.S.
N.S.
N.S.
N.S.
N.S.
*HRP/Gm Tissue
Values in parentheses are protein compensated and shown only when
different
57
-------
TABLE 27. EXPERIMENT 2
HRP CONTENT OF MOUSE LUNG TISSUE (HRP/GM TISSUE)
Peak Group
a
3
a+0
a/B
HRP/ek*
C
X
C
X
C
X
C
X
C
X
M
48.4
57.4
20.9
31.6
69.3
89.0
2.7
2.3
6.4
9.4
3 Weeks
p Value
± 16
< .1
± 22 (N.S.)
± 12
< .05
± 32 (N.S.)
± 26
< .05
± 49 (N.S.)
± 1
< .1
± 1
± 5
N.S.
± 11
M
48.6 ±
. 45.6 ±
10.9 ±
12.0 ±
59.5 ±
57.7 ±
5.0 ±
4.3 ±
5.0 ±
4.9 ±
6 Weeks
p Value
13
N.S.
10
5
N.S.
8
16
N.S.
16
2
< .1
1
2
N.S.
2
Values in parentheses are protein compensated and shown only when
different
-8
*Enzyme kinetic assay : 10 Gm HRP/Gm tissue.
58
-------
TABLE 28. EXPERIMENT 2
TWO FACTOR ANALYSIS OF VARIANCE
A. Serum HRP Content*
Peak N02
a
3 <
a+£
a/3
B. Lung HRP Content*
a
3
a+3
a/3
Factor _. _
Time Factor
N.S. N.S.
.05 < .001
N.S. < .025
N.S. -v 0
N.S. N.S. (N.S.)
N.S. <.001 ( % 0)
N.S. <.005
N.S. ~ 0
N02 Time
Interaction
•v .05
< .05
< .05
N.S.
N.S.
N.S.
N.S.
N.S.
Values In parentheses are protein compensated and shown only when
different
*HRP/6m Tissue
-------
TABLE 29. EXPERIMENT 3
HRP CONTENT OF MOUSE LUNG TISSUE (HRP/GM TISSUE)
3 Weeks 6 Weeks
Peak Group M p Value M p Value
C 51.2 ± 16.8 43.3 ± 13.0
N.S. .005
X 51.6 ±21.6 59.7 ± 23.7 (< .025)
C 12.5 ± 10.9 10.8 ± 6.4
N.S. .005
X 18.2 ± 23.0 23-9 ± 20.0 (< .01)
C 63.7 ± 25.6 54.1 ± 17.3
a+3 N.S. .0025
X 69.8 ± 43.3 83.7 ± 39.8 (<.005)
C 5.55± 2.64 4.56± 1.37
N.S. < .05
X 5.71± 3.75 3.67± 1.85
Values In parentheses are protein compensated and shown only when
different
60
-------
TABLE 30: EXPERIMENT 3
TWO FACTOR ANALYSIS OF VARIANCE
Lung HRP Content
Peak
a
8
**
a/6
N02 Factor
< 0.05
(N.S.)
. 0.01
(< 0.025)
< 0.025
(< 0.05)
N.S.
T. .. „ NO, Time
T.me Factor |ntc?actlon
N.S. < 0.05
(N.S.)
N.S. N.S.
N.S. N.S.
(< 0.05)
< 0.01 N.S.
Values In parentheses are protein compensated and shown only when
different
61
-------
TABLE 31: SERUM 3 WEEKS
Experiment 1
Peak
a
6
a+fi
a/B
Group
C
X
C
A
C
X
C
X
ii
2
3
1
1
3
4
1
1
.3
.2
.3
f
• O
.3
.8
-3
.8
2
3
1
1
3
4
0
0
P Value
.1
< .05
.0
.0
N.S.
.2
.1
< .1
.1
.3
<.0025
.7
3
6
3
5
6
11
1
1
Experiment 2
M p Value
.*>> 3
< .01
-3 5
.0 2
< .005
.5 *»
.5 6
< .cos
.8 3
.1 .4
N.S.
.0 .3
experiment 3
M t
2
2
1
1
4
k
1
2
.8
.8
.5
.5
.3
.3
.9
.0
2.7
3.3
1.4-
1.7
4.1
l».7
0.8
0.6
> Value
N.S.
N.S.
N.S.
N.S.
6 weeks data: No significance for all items of the three experiments.
62
-------
TABLE 32. HRP CONTENT OF MOUSE KIDNEY TISSUE (HRP/GM TISSUE)
Peak
a
3
a/B
CMS
HRP/ek*
Group
C
X
C
A
C
X
C
X
C
X
Experiment 1
« p' Value
29.0*6.0
N.S.
28.0*5.0
16.0*3.0
< .025
14.0*3-0 (<.05)
1.8*0.3
< .05
2,1±0.5
45.0*8.0
N.S.
42.0*6.0
6.4*1.0
ti.S.
6.2M.G
•Experiment
2
H p Value
27.0* 6.0
<
31.1*11.0 (<
17.0* 7.0
<
25.0*15-0
1.7* 0.4
•<
1.4± 0.3
44.0*12.0
<
57.0*26.0 (<
5.0* 1.0
<
8.0* 7-0
.1
.05)
.025
.0025
.05
.025)
.05
Experiment
H p
19.5* 6.0
17.4+ 4.0
14. 8± 5.0
13. 8± 5.0
1.3* 0-3
1.3* 0.2
35.2±10.0
31-3* 9-0
3
Value
N.S.
N.S.
N.S.
N.S.
Values in parentheses are protein compensated, and shown only when
different.
-8
*Enzyme kinetic assay : 10 Gm HRP/Gm tissue.
63
-------
An evaluation was also started on the relative distribution of HRP in the
various cell amd tissue compartments of the lung parenchyma,
Summary
In three independent experiments, involving a total of 264 mice, the
exogenous HRP content of lungs, kidneys, and serum, were determined with and
without the influence of three different NCL exposures (continuous 0.5 ppm,
intermittent 0.6 ppm, and intermittent 0.8 ppm). The purpose of the study was
to evaluate possible alterations In capillary permeability and clearance of
leaked protein from the lung tissues and alveolar spaces following ambient
levels of NO^ exposure. Both PAGE with gel scanning densitometric measurements
and enzyme kinetic assay were carried out to quantitate HRP content in the serum
and tissues. In all three experiments, the HRP lung content was greater for
the exposed animal groups: 49 ± 18 vs. 39 ± 14 (p < .05; 89 ± 4g vs. 69 ± 26
(p < .05) and 84 ± 40 vs. 54 ± 17 (p < .025). The increase in HRP was found
at three weeks for the first two experiments and at six weeks for the third.
The findings are believed to represent increased capillary permeability and
decreased clearance of HRP from cells, tissues, and spaces of the lung. One of
the kidney studies showed an increase (p < .05) in HRP content for the exposed
animals. No significant differences were noted in the serum content of HRP
between the control and exposed groups. Ultrastructural correlations are
incomplete.
A potential use of HRP assay for assistance in establishing air quality
standards is suggested by the findings. Protein leakage is an important patho-
physiologic abnormality and is a common denominator for a great many pathologic
lesions at their early stage of development. Delayed HRP clearance tests
supplement those for measuring protein leakage into alveolar spaces, and It
has the potential for greater sensitivity since the total protein content of
the cellular and tissue protein transport system should be several magnitudes
greater than the protein content of the alveolar spaces in the early stages of
protein leakage.
0.5 continuous and intermittent N0_ exposures
Investigations supported In part by this project
Liquid scintillation counting of labeled cells in culture
One of the exploratory investigations, using tritium labeling methodolo-
gies, was applied to material derived from an independently support grant
project. Supplemental tissue culture preparations were set up for this specific
application. The experiment consisted, of a study of 30 mice, half exposed to
NO- for 36 days, and explantation of lung tissues into culture immediately
(0 hours) and after specified periods (5 days and 7 days) of storage in the
refrigerator at 4°C ("postmortem interval stress" testing PMI), which we
innovated as a means of evaluating vigor (survival time in-culture;a separate
study) as well as viability of lung cells following in vivo NO,, exposure. For
the 0 hour group, 150 cultures with five explants each were pufsed at seven
days in vitro with 20 uCi/ml 3H-fucose for 24 hours. The methanol fixed mono-
layers were evaluated by liquid scintillation counting. The same procedure was
carried out for 108 cultures (five-day group) and for 72 cultures (seven-day
groups), the decreasing numbers of cultures available reflecting PMI attrition.
64
-------
The numbers of large, medium, and small colonies found in those cultures
which were explanted at 0 hours were as follows: for the control group, 63,
172, and 103 respectively; for the exposed group, 128, 142, and 70 respec-
tively. With respect to the PMI studies, the colonies were measured by a
micrometer and separated into six groups, beginning with 1 mm diameters and
extending to the largest group of 6 mm in diameter by 1 mm increments. The
findings were as follows: at the five day PMI period, the control cultures
had 36, 76, 62, 58, 27, and 2 colonies in the respective small to large
groups; the exposed had 30, 42, 53, 71, 38, and 12 for the comparable groups.
Fo the seven-day PMI period, the findings were: for the controls 18, 31, 30,
58, 11, and 4; for the exposed 12, 26, 52, 53, 2k, and 4.
For the 3H-fucose label scintillation counts, there were the following
findings: of 150 0 hour cultures, the mean values for the controls and exposed
groups were 1039 ± 590 and 1018 ± 462 respectively; these differences were not
statistically significant. For the 108 cultures of the five-day PMI group,
the control and exposed animals had mean values of 72k ± 400 and 1057 ± 519
respectively, a difference statistically significant (p < .001) indicating an
increase in numbers of cells, possibly with increased fucose incorporation.
For the seven-day groups, there were the following control vs. exposed mean
values: 800 ± 401 vs. 779 * 465. The difference was not statistically signi-
ficant, but the numbers of cultures were considered to be suboptimal for an
appropriate quantitation. In effect, this approach can be considered a
stress test of different lung cell populations in a search for discriminants
that can quantitate loss of functional and structural reserves. These findings
have been presented at the 1976 Federation Meetings (10). ^H-Leucine labeling
studies were also done.
Macrophage congregation in tissue cultures of mouse lung
A relatively small part of the study was supported by funds under EPA
R 80081-05, primarily assistance in the N0« monitoring and other aspects of
the in vivo part of the experiment. This work has been published with
acknowledgment of R80081 support (Hl,.15).ln view of the relatively small role
played by this grant support, only a summary of the findings is presented here.
The lungs of 12 mice, half of which were exposed to continuous 0.5 ppm
nitrogen dioxide for 3 weeks, were explanted in culture, and the instances of
macrophage congregation were quantitated according to numbers of target cells
involved, categories of congregation from three to 11 or more, numbers of
macrophages participating in each category for the total cultures, and the
influence of delaying explantation for 24 and 96 hours. A total of 9042
macrophages and 2140 epithelial and spindle target cells were counted in the
outgrowths from 306 explants. The incidence of macrophage congregation (or
numbers of target cells) was greater for the cultures from the N02 exposed
animals, both with respect to total incidences between groups (p •* 0 ), and
the 0 hour (p < 0.001) and 24 hours Cp < 0.01) culture sub-groups. In addi-
tion, the values for Ts to T6 macrophage congregation were individually and
consistently greater for the exposed animal group. Postmortem interval stress
at 96 hours appeared to result in large colonies, but they were reduced
greatly in number. Also the incidence of macrophage congregation fell by
28% as compared to 0 hours and 24 hour subgroups.
65
-------
Ultrastructural, histochemical aspects
The macrophage congregation study was subsequently extended into ultra-
structural, histochemical aspects where, again, some support was provided
by EPA-R-80081. A summary of the findings is as follows;
The study is based on mouse lung in tissue cultures, with 85 colonies
from control and 104 from exposed animal groups, randomly selected from 626
culture preparations. A quantitative study was made of tritiated thymidine
uptake by macrophages in culture with respect to: a) macrophages unassoci-
ated with lung cells; b) macrophages associated with lung cells; and c)
spindle and epithelial cells. A semi-quantitative study measured the degree
of positivity of macrophages for LDH, G6PD, MOH, and ATPase. Those were
correlated with post-mortem intervals (PMl), i.e., the explantation after
storage at 4 C for 2k hours to 5 days, and with cytostructure, including
ultrastructural histochemistry (13,14).
The preliminary findings are: l) a special macrophage type has been
identified. The cell shape is generally round as is the nucleus. The
nucleus shows a scalloped chromatin margination, mitochondria are peripher-
ally distributed, dense bodies rare, and the Golgi large; 2) ATPase is
strongly positive and LDH weak, while the reverse is true for other macrophages.
MDH is also strongly positive whereas only 10% of other macrophages are
positive. All are weak with G6PD, but 25% of others do show positivity. The
numbers of LDH positive macrophages and the degree of response are greater
for the NO exposed mice, but only in the 5 day PMl group (p < .005). MDH is
also greater at 0 hours (p < .05). ATPase is greater for the exposed group
for 0 hours, 2k hours, and 96 hours (p < .01). H-TdR shows no differences
between groups (1495 of 4962 exposed vs. 745/2777 for controls, 30% vs. 27%).
The uptake for lung cells is greater for the exposed group, M = k$% vs. M =
31% (p < .05); 3) at five days PMl, the numbers of macrophages in exposed
animal cultures are greater than that of the controls (p < .025), based on
colony area evaluations.
A definitive publication covering the ongoing (804545) ul
-------
andLowry protein quantitation; 4) 30 guinea pigs, half of which were exposed
to intermittent (4 hours per day) .35 ppm NO- for protein and AcPase content
of lung lavage fluid and plasma, plus diphosphoglycerate content of red blood
cells; 5) 36 guinea pigs, half of which were exposed to continuous 0.42 ppm
N02 for 7 days, 9 days, 13 days, 15 days, and 4 weeks for a study of blood
diphosphoglycerate content, AcPase of lung, and a study of membrane protein
of lung and kidney. Dr. Hanson's studies with lung lavage protein determina-
tions confirmed our earlier findings, i.e. an approximately 20-30% increase
in protein in the lung lavage fluid from the exposed animals (p < .05). His
findings with diphosphoglycerate of red blood cells are inconclusive. His
work also indicated that density gradient separation appeared to provide
positive correlations between exposure and acid phosphatase increase when
determinations using whole lung homogenates failed to do so, and the lung
homogenate pellet from ultracentrifugation was shown to contain dense aggre-
gates of cell membranes. He was able to demonstrate 18 to 20 bands in his
attempt to obtain more specific AcPase fractions? the the results of this
work presently remain inconclusive; 6) some studies were carried out by Dr.
Hanson with mice exposed to continuous 0.44 ppm. These studies were also ex-
ploratory and no conclusive findings were derived from them. The pilot
studies were carried out with mouse tissues (lung, kidney, heart, and adrenal
gland) as well as with guinea pig tissues, and these studies included investi-
gations of catecho1 amines.
In an attempt to complete earlier studies carried out by Dr. Ted Yuen,
Dr. Yuen was hired on a part-time basis, but could only contribute a rela-
tively small number of hours to the project. Ultrastructural investigations
of the transport of ferritin and horseradish peroxidase in the lungs and
kidney was studied, in conjunction with the pathobiologic investigation of
the main project. An attempt to achieve qualitative and/or quantitative
distinctions between tracer molecule distributions in the lungs of control
and NO. exposed animals at ambient levels of exposure are incomplete at this
time. Our working arrangement with Dr. Yuen ended as of December 26, 1976.
Dr. Richters carried out a study involving electronmicroscopy of ferritin
deposition in the lungs of 38 animals, half of which had been exposed to 0.5
N0_. No conclusive differences were found in the distribution between control
ana exposed animals, but note was made of five exposed animals with unusual
sites of deposition as opposed to two control animals.
67
-------
REFERENCES
1. Sherwin, R.P., Yuen, T.G.H., and Richters, V.: The spontaneous occurrence
of wall-free organisms in the lungs of guinea pigs and their association
with microvilli and cilia, Am. J. Clin. Path. 60:268-280, 1973
2. Sherwin, R.P., Margol.ick, J.B., and Azen, S.P.: Hypertrophy of alveolar
wall cells secondary to an air pollutant. A semi-automated quantitat ion.
Arch. Environ Health 26:297-299, 1973
3. Margolick, J.B., Azen, S.P., and Sherwin, R.P.: An image analyzer quanti-
tation of type 2 pneumocytes. Am. Rev. Respir. Dis. 108:704-707, 1973
C\
4.j Sherwin, R.P., and Carlson, D.A.: Protein content of lung lavage fluid of
guinea pigs exposed to 0.4 ppm nitrogen dioxide. Disc-gel electrophor-
esis for amount and types. Arch. Environ. Health 27:90-94, 1973
5. Mersch, J., Dyce, B.J., Haverback, B.J., and Sherwin, R.P.: Diphosphogly-
cerate content of red blood cells. Measurements in guinea pigs exposed
to 0.4 ppm nitrogen dioxide. Arch. Environ. Health 27:94-95, 1973
6. Sherwin, R.P., Margolick, J.B., and Azen, S.P.:An automated determination
of ratios of type 2 pneumocytes to alveolar wall area using an image
analyzer Am. Rev. Respir. Disease 108:1015-1018, 1973
7. Sherwin, R.P., Margolick, J.B., and Aguilar, E.A.: Acid phosphatase in
density equilibrium fractions of the lungs of guinea pigs exposed inter-
mittently to 0.4 ppm nitrogen dioxide. Fed. Proc. 33:633, 1974
8. Sherwin, R.P., and Layfield, L.J.: Proteinuria in guinea pigs exposed to
0.5 ppm nitrogen dioxide. Arch. Environ. Health 28:335-341, 1974
9. Sherwin, R.P., and Layfield, L.J.: Protein leakage in the lungs of mice
exposed to 0.5 ppm nitrogen dioxide. A fluorescence assay for protein.
Arch. Environ.. Health 31:50-52, 1976
10. Sherwin, R.P., and Richters, V.: The influence of 0.4 ppm nitrogen
dioxide in vivo on the outgrowth of lung cells in tissue culture.
Fed. Proc. 35:632, 1976
11. Richters, V., Elliott, G., and Sherwin, R.P.: Macrophage congregation in
tissue cultures of mouse lungs: Influence of 0.5 ppm nitrogen dioxide
in vivo exposure. Fed. Proc. 35:717, 1976
12. Azen, S., Margolick, J.B., and Sherwin, R.P.: An experimental model and
automated methodology for the analysis of the effects of ambient levels
of air pollution on the lung. Applied Math. Computation 3:95-102, 1977
13. Sherwin, R.P., Richters, V., Bernett, J., and Paller, M.: Macrophage
congregation and influence of nitrogen dioxide: Ultrastructural and
histochemical aspects. XI Cong. Int. Acad. Path, (accepted for publi-
cation)
68
-------
14. Richters, V., Sherwin, R.P., Paller, M., and Bernett, J.: A multidisci-
pline postmortem study of the mouse lung macrophage: Influence of 0.6
ppm nitrogen dioxide in vivo. (Presented at the 16th Annual Hanford
Biology Symposium, Richland, Washington, September 27, 1976)
15. Richters, V., Elliott, G., and Sherwin, R.P.: Influence of 0.5 ppm
nitrogen dioxide exposure of mice on macrophage congregation in the
lungs. In Vitro 14:458-464, 1978
16. Sherwin, R.P., Okimoto, D., and Mundy, D.: Sequestration of exogenous
peroxldase in the lungs of animals exposed to 0.5 ppm nitrogen dioxide.
Fed. Proc. 36:1091, 1977
17. Sherwin, R.P., Margolick, J.B., Aguilar, E.A., and Hanson, C.W.:
Intermittent 0.4 ppm nitrogen dioxide exposure of guinea pigs: increase
in acid phosphatase in density equilibrium fractions of lung homogen-
ates. (submitted for publication)
18. Sherwin, R.P., Okimoto, D.T., and Mundy, D.!.: Serum and lung clearance
of exogenous horseradish peroxidase: Influence of low levels of nitrogen
dioxide. (Submitted for publication)
19. Sherwin, R.P., and Yuen, T.G.H: Silicone fluid for the metering and
monitoring of nitrogen dioxide. Arch. Environ. Health 24:331-336, 1972
20. Sherwin, R.P., Winnick, S., and Buckley, R.D.: Response of lactic acid
dehydrogenase positive alveolar cells in the lungs of guinea pigs exposed
to N02. Am. Rev. Respir. Dis. 96:319-323, 1967
21. Ornstein, L.: Disc electrophoresis: 1. Background and theory. Ann. NY
Acad. Sci. 121:321-349, 1964
22. Lowry, O.H., et al.: Protein measurement with the folin phenol reagent.
J. Biol. Chem. 193:265, 1951
23- cf. Pesce,A.J. Methods Used for the Analysis of Proteins in the Urine
Nephron 13: 93-104, 1974,for general review.
24. Yuen, T.G.H., and Sherwin, R.P.: Hyperplasia of type 2 pneumocytes and
nitrogen dioxide (10 ppm) exposure: A quantitation based on electron
photomicrographs. Arch. Environ. Health 22:178-188, 1971
25. Steinman, R.M., and Conn, Z.A.: The interaction of soluble horseradish
peroxidase with mouse peritoneal macrophages in vitro.
J. Cell Biol. 55:186, 1972
26. Dyce, B.J.: A simple sensitive rabbit assay procedure for 2,3-diphospho-
glycerate in multiple specimens of blood. Thesis. University of
Southern California, Los Angeles, 1971
69
-------
27. The Colorimetric Determination of Acid Phosphatase. In Sigma Technical
Bull. #104, Aug. 1971
28. Farquhar, M.G., Wissig, S.L., and Palade, G.E.: Glomerular permeability.
1. Ferritin transfer across the normal glomerular capillary wall.
J. Exp. Med. 113:47-66, 1961
29. Yuen, T.G.H., Richters, A., and Sherwin, R.P.: Unpublished data
30. Sherwin, R.P., Dibble, J., Weiner, J.: Alveolar wall cells of the guinea
pig: Increase in response to 2 ppm nitrogen dioxide.
Arch. Environ. Health 24:43-^7, 1972
70
-------
TECHNICAL REPORT DATA
(Please read Instructions on -the reverse before completing)
1. REPORT NO.
EPA-600/1-79-038
|3. RECIPIENT'S ACCESSION NO.
4. TITLE AND SUBTITLE
A BIOLOGIC INDICATOR FOR AIR POLLUTION
5. REPORT DATE
September 1979
6. PERFORMING ORGANIZATION CODE
7. AUTHOR(S)
Russell P. Sherwin
8. PERFORMING ORGANIZATION REPORT NO.
9. PERFORMING ORGANIZATION NAME AND ADDRESS
University of Southern California
Medical Center
Los Angeles, CA 90033
10. PROGRAM ELEMENT NO.
1AA816
11. CONTRACT/GRANT NO.
R800881 and R804545
12. SPONSORING AGENCY NAME AND ADDRESS
Health Effects Research Laboratory
Office of Research and Development
U.S. Environmental Protection Agency
Research Triangle Park, NC 27711
RTP, NC
13. TYPE OF REPORT AND PERIOD COVERED
14. SPONSORING AGENCY CODE
EPA 600/11
15. SUPPLEMENTARY NOTES
16. ABSTRACT
This report describes studies wherein guinea pigs and mice were exposed to low
levels of nitrogen dioxide either continuously or on an intermittent schedule. The
major objective was to develop highly sensitive discriminants involving correlations
of structural and functional alterations. The two major test areas developed were
the quantitative measurements of Type 2 cell populations and protein leakage within
the lung. The increase in number and size of Type 2 pneumocytes with N0£ exposure
is believed to represent a corresponding loss of Type 1 lung cells, and the protein
leakage within the lung is in accord with the concept that the Type 1 cell is a
critical barrier for fluid transport in the alveolar area. The authors suggest that
their research data supports the hypothesis that to properly evaluate the possible
health effects from air pollutants one must give attention to the subclinical cellular
alterations that deplete the functional and structural reserves of organs which are
the primary target for air pollutants.
17.
KEY WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
b.lDENTIFIERS/OPEN ENDED TERMS
c. COS AT I Field/Group
Bronchial
Kinetics
Electrophoresis
Fluorescamine
Ultraviolet
06F.P
18. DISTRIBUTION STATEMENT
iRELEASE TO PUBLIC
19. SECURITY CLASS (This Report)
UNCLASSIFIED
21. NO. OF PAGES
82
20. SECURITY CLASS I This page I
UNCLASSIFIED
22. PRICE
EPA Form 2220-1 (Rev. 4-77) PREVIOUS EDITION is OBSOLETE
71
------- |