?/EPA
United States
Environmental Protection
Agency
Health Effects Support
Documerit for Manganese
External Review Draft
-------
-------
Health Effects! Support Document
for Manganese
EXTERNAL REVIEW DRAFT
Contract Number: 68-C-01-002
Work Assignment Number: B-02
Prepared for:
U.S. Environmental Protection Agency
Office of Water
Health and Ecological Criteria Division
Washington, DC 20460
Prepared by:
Sciences International, Inc.
1800 Diagonal Road, Suite 500
Alexandria, VA 22314-2808
EPA-R-02-029
April 2002
Printed on Recycled Paper
-------
-------
TABLE OF CONTENTS
LIST OF TABLES • vi
FOREWORD - • • v"1
ACKNOWLEDGMENTS • • • x
1.0 EXECUTIVE SUMMARY - - • • - 1-1
2.0 IDENTITY: CHEMICAL AND PHYSICAL PROPERTIES 2-1
•
3.0 USES AND ENVIRONMENTAL FATE - - - 3-1
3.1 Production and Uses • 3-1
3.2 Sources and Environmental Fate 3-3
4.0 EXPOSURE FROM DRINKING WATER 4-1
4.1 Introduction 4-1
4.2 Ambient Occurrence : 4-1
4.3 Drinking Water Occurrence • 4-4
4.4 Results - • • • 4-9
4.5 Conclusion - 4-12
5.0 EXPOSURE FROM ENVIRONMENTAL MEDIA OTHER THAN WATER 5-1
5.1 Food • 5-!
5.1.1 Concentrations of Manganese in Food 5-1
5.1.2 Intake of Manganese From Food 5-1
5.2 Air • 5-5
5.2.1 Concentration of Manganese in Air 5-5
5.2.2 Intake of Manganese in Air 5-11
5.3 Soil • 5-12
5.3.1 Concentration of Manganese in Soil 5-12
5.3.2 Intake of Manganese in Soil 5-12
5.4 Other Media 5'12
5.5 Summary of Exposure to Manganese in Media Other Than Water 5-12
6.0 TOXICOKINETICS : 6-1
6.1 Absorption • • 6-1
6.2 Distribution 6~7
6.3 Metabolism ...» • • 6-H
6.4 Excretion 6'12
7.0 HAZARD IDENTIFICATION 7-1
7.1 HumanEffects - 7~l
External Review Draft—Manganese — April 2002
m
-------
7.1.1 Case Reports 7-1
7.1.2 Short-term Studies r 7-3
7.1.3 Long-Term and Epidemiological Studies r.. i. 7-4
7.1.4 Beneficial Effects 7-9
7.2 Animal Studies 7-10
7.2.1 Acute Toxicity 7-10
7.2.2 Short-Tenn Studies 7-12
7.2.3 Subchronic Studies 7-14
7.2.4 Neurotoxicity j 7-16
7.2.5 Developmental/Reproductive Toxicity 7-25
7.2.6 Chronic Toxicity 7-34
7.2.7 Carcinogenicity 7-34
7.3 Other Key Data 7-37
7.3.1 Mutagenicity/Genotoxicity 7-37
7.3.2 Immunotoxicity 7-42
7.3.3 Hormonal Disruption 7-42
7.3.4 Physiological or Mechanistic Studies ; 7-42
7.3.5 Structure-Activity Relationship 7-46
7.4 Hazard Characterization ; 7-46
7.4.1 Synthesis and Evaluation of Major Noncancer Effects i 7-46
7.4.2 Synthesis and Evaluation of Carcinogenic Effects ' 7-49
7.4.3 Mode of Action and Implications in Cancer Assessment 7-50
7.4.4 Weight of Evidence Evaluation for Carcinogenicity 7-51
7.4.5 Sensitive Populations j 7-51
7.4.6 Potential Childhood Sensitivity \ 7-51
7.4.7 Other Potentially Sensitive Populations 7-53
8.0 DOSE-RESPONSE ASSESSMENT .'. ;.... 8-1
8.1 Dose-Response for Noncancer Effects 8-1
8.1.1 RfD Determination 8-1
8.1.2 RfC Determination '< 8-3
8.2 Dose-Response for Cancer Effects '....'•• 8-4
9.0 RISK DETERMINATION AND CHARACTERIZATION OF RISK FROM!
DRINKING WATER ; 9-1
9.1 Regulatory Determination for Chemicals on the CCL ' . 9-1
9.1.1 Criteria for Regulatory Determination 9-1
9.1.2 National Drinking Water Advisory CouncilRecommendations 9-2
9.2 HealthEffects 9-2
9.2.1 Health Criterion Conclusion ; 9-3
9.2.2 Hazard Characterization and Mode of Action Implications 9-3
9.2.3 Dose-Response Characterization and Implications in Risk
Assessment 9-5
9.3 Occurrence in Public Water Systems ; 9-6
External Review Draft — Manganese—April 2002
IV
-------
9.3.1 Occurrence Criterion Conclusion • 9-6
9.3.2 Monitoring Data - 9-7
9.3.3 UseandFateData 9-8
9.4 Risk Reduction 9-9
9.4.1 Risk Reduction Criterion Conclusion 9-9
9.4.2 Exposed Population Estimates • 9-10
9.4.3 Relative Source Contribution 9-10
9.4.4 Sensitive Populations 9-11
9.5 Regulatory Determination Decision 9-11
10.0 REFERENCES
10-1
APPENDDC A: Abbreviations and Acronyms
APPENDDCB: Complete NIRS Data for Manganese
B-l
External Review Draft — Manganese — April 2002
-------
-------
LIST OF TABLES
Table 2-1. Chemical and Physical Properties of Manganese 2-2
Table 2-2. Chemical and Physical Properties of Manganese Compounds. 2-3
Table 3-1. Imports of Manganese and Ferromanganese to the United States
(thousand metric tons, gross weight) 3-1
Table 3-2. Manganese Manufacturers and Processors by State. 3-2
Table 3-3. Summary of Uses for Selected Manganese Compounds. 3-4
Table 3-4. ' Environmental Releases (in pounds) for Manganese in the United States,
1988-1998 - , - 3-7
Table 3-5. Environmental Releases (in pounds) for Manganese Compounds in the United
States, 1988-1998 3-7
Table 4-1. Manganese Detections and Concentrations in Streams and Ground Water. ... 4-5
Table 4-2. Manganese Detections and Concentrations in Bed Sediments and
Aquatic Biota Tissues (all sites) 4-5
Table 4-3. Manganese Occurrence in Ground Water PWS of NIRS Survey 4-8
Table 4-4. Occurrence Summary of Ground and Surface Water Systems by State for
Manganese 4-10
Table 5-1. Manganese Concentrations in Selected Foods 5-2
Table 5-2. Average Concentrations of Manganese in Ambient Air Sampled
from 1953-1982 - 5-6
Table 5-3. Manganese Levels in Air of Canadian Urban Locations as Determined
by Personal Exposure Monitoring 5-6
Table 5-4. Ambient Air Concentrations of Manganese in Relation to Traffic Density,
Montreal, Canada 1981-1994 5-8
Table 5-5. Estimated Atmospheric Mn Concentration in Relation to the Combustion
of MMT in Gasoline 5-9
Table 5-6. Mean Manganese Exposures from 3-day Indoor, Outdoor and Personal Air
Samples ; 5-11
Table 5-7. Summary of Human Exposure to Manganese in Media Other than Water — 5-14
Table 6-1. Normal Manganese Levels in Human and Animal Tissues 6-8
Table 7-1. Mean Neurological Scores of Residents in Three Areas of
Northwest Greece 7-5
External Review Draft — Manganese — April 2002
VI
-------
Table 7-2. Mean Neurological Scores of Residents in Germany Exposed to Different
Levels of Manganese in Well Water 7-8
Table 7-3. LD50 Values for Manganese Compounds 7-11
Table 7-4. Neurological Effects of Oral Exposure to Manganese , 7-18
Table 7-5. Developmental Effects of Exposure to Manganese . 7-27
Table 7-6. Reproductive Effects of Exposure to Manganese 7-33
Table 7-7. Follicular Cell Tumor Incidence in B6C3Fj Mice ' 7-35
Table 7-8. Summary of Carcinogenicity Studies Reporting Positive Findings for
Selected Manganese Compounds .., .'..-... 7-36
Table 7-9. Pulmonary Tumors in Strain A Mice Treated with Manganese Sulfate|a 7-37
Table 7-10. Genotoxiciry of Manganese In Vivo \ 7-38
Table 7-11. Genotoxicity of Manganese In Vitro 7-40
Table 9-1. Comparison of Average Daily Intake from Drinking Water and Other
Media" 9-11
External Review Draft — Manganese—April 2002
Vll
-------
FOREWORD
The Safe Drinking Water Act (SDWA), as amended in 1996, requires the Administrator of
the Environmental Protection Agency to establish a list of contaminants to aid the agency in
regulatory priority setting for the drinking water program. In addition, SDWA requires EPA to
make regulatory determinations for no fewer than five contaminants by August 2001. The criteria
used to determine whether or not to regulate a chemical on the CCL are the following:
The contaminant may have an adverse effect on the health of persons.
The contaminant is known to occur, or there is a substantial likelihood that the
contaminant will occur, in public water systems with a frequency and at levels of public
health concern.
In the sole judgment of the. administrator, regulation of such contaminant presents a
meaningful opportunity for health risk reduction for persons served by public water
systems.
The Agency's findings for all three criteria are used in making a determination to regulate
a contaminant. The Agency may determine that there is no need for regulation when a
contaminant fails to meet one of the criteria. The decision not to regulate is considered a final
agency action and is subject to judicial review.
This document provides the health effects basis for the regulatory determination for
manganese. In arriving at the regulatory determination, data on toxicokinetics, human exposure,
acute and chronic toxicity to .animals and humans, epidemiology, and mechanisms of toxicity were
evaluated. In order to avoid wasteful duplication of effort, information from the following risk
assessments by the EPA and other government agencies were used in development of this
document:
U.S. EPA 1994a. U.S. Environmental Protection Agency. Drinking Water Criteria
Document for Manganese. Office of Health and Environmental Assessment, Cincinnati,
OH CEAO-CIN-D008, prepared September, 1993, revised March 31,1994.
ATSDR. 2000. Agency for Toxic Substances and Disease Registry. Toxicological
Profile for Manganese (Update). Department of Health and Human Services. Atlanta,
GA. Available at http://www.atsdr.cdc.gov.
U.S. EPA 1996a. U.S. Environmental Protection Agency. Integrated Risk Information
System (IRIS): Manganese. Available at http://www.epa.gov/iris. Last revised December
1,1996.
In addition, primary references of studies published in peer-reviewed scientific journals
relevant to human risk assessment of manganese were also used in preparing this Drinking Water
External Review Draft — Manganese—April 2002 viii
-------
Support Document for Manganese. Recent studies of manganese were identified by literature
searches conducted in 1999 and 2000.
Generally a Reference Dose (RfD) is provided as the assessment of long-term toxic effects
other than carcinogenicity. RfD determination assumes that thresholds exist for certain toxic
effects such as cellular necrosis. It is expressed in terms of milligrams per kilogram per day
(mg/kg-day). In general, the RfD is an estimate (with uncertainty spanning perhaps ah order of
magnitude) of a daily exposure to the human population (including sensitive subgroups) that is
likely to be without an appreciable risk of deleterious effects during a lifetime.
The carcinogenicity assessment for manganese includes a formal hazard identification.
Hazard identification is a weight-of-evidence judgment of the likelihood that the agent is a human
carcinogen via the oral route.
Guidelines that were used in the development of this assessment may include the
following: the Guidelines for Carcinogen Risk Assessment (U.S. EPA, 1986a), Guidelines for the
Health Risk Assessment of Chemical Mixtures (U.S. EPA, 1986b), Guidelines forMutagenicity
Risk Assessment (U.S. EPA, 1986c), Guidelines for Developmental Toxicity Risk Assessment
(U.S. EPA, 1991a), Proposed Guidelines for Carcinogen Risk Assessment (1996b), Guidelines
for Reproductive Toxicity Risk Assessment (U.S. EPA, 1996c), and Guidelines for Neurotoxicity
Risk Assessment (U.S. EPA, 1998a); Recommendations for and Documentation of Biological
Values for Use in Risk Assessment (U.S. EPA, 1988); and Health Effects Testing Guidelines
(OPTS series 870,1996 drafts; U.S. EPA 40 CAR Part 798,1997); Peer Review and Peer
Involvement at the U.S. Environmental Protection Agency (U.S. EPA, 1994b); Use of the
Benchmark Dose Approach in Health Risk Assessment (U.S. EPA, 1995). !
The chapter on occurrence and exposure to manganese through potable water'was
developed by the Office of Ground Water and Drinking Water. It is based primarily on
unregulated contaminant monitoring (UCM) data collected under SDWA. The UCM data are
supplemented with ambient water data as well as information on production, use, and discharge.
External Review Draft — Manganese—April 2002
IX
-------
ACKNOWLEDGMENTS
This document was prepared under the U.S. EPA contract No. 68-C-01-002. Lead
Scientist, Julie Du, Ph.D., Health and Ecological Criteria division, Office of Science and
Technology, Office of Water.
External Review Draft — Manganese—April 2002
-------
-------
1.0 EXECUTIVE SUMMARY
The U.S. Environmental Protection Agency (EPA) has prepared this Health Effects
Support Document to assist in determining whether to establish a National Primary Drinking
Water Regulation (NPDWR) for manganese. At high doses by inhalation, manganese is very
toxic, as seen by occupational exposure in miners. On the other hand, manganese is essential for
normal physiological function of animals and humans. The Food and Nutrition Board of the
National Academy of Science (NAS) sets an adequate intake for manganese at 2.3 mg/day for
men and 1.8 mg/day for women, and an upper limit for daily intake at 11 mg for adults (Food and
Nutrition Board, 2002). Manganese has a low aesthetic threshold in water. Based on staining
and taste, EPA has set a secondary level for manganese at 0.05 mg/L which is below the level that
may present a health concern. Available data suggest that regulation of manganese in public
water does not present a meaningful basis for health risk reduction. EPA will present a
determination and further analysis in the Federal Register Notice covering the Contaminant
Candidate List proposals.
Manganese (Chemical Abstracts Services Registry Number 7439-96-5) is an abundant
elemental metal that does not exist naturally in its pure form, but rather is found as a component
of over 100 minerals. It is also an essential nutrient, and a certain level of intake is necessary for
good health. The Food and Nutrition Board of the NAS has determined that the Adequate Intake
for manganese (AI) is 1.8 to 2.3 milligrams per day for an adult woman and man, respectively,
although others have argued that it may be higher. Manganese occurs naturally in soil, air, water,
and food at low levels.
Manganese and manganese compounds are used mostly in the production of manganese-
iron alloy through a smelting process. They are also used in fertilizer, fungicide, livestock feed,
and in unleaded gasoline as an anti-knock additive in the form of methylcyclopentadienyl
manganese tricarbonyl (MMT). Any of these uses may result in substantial releases of manganese
to the environment. Manganese is listed as a Toxic Release Inventory (TRI) chemical, with
releases to soil constituting most of the on-site releases, although air, surface water and ground
water are also important sinks for manganese release.
Human exposure to manganese occurs primarily through ingestion of foods containing
manganese. These include many nuts, grams, fruits, legumes, tea, leafy vegetables, infant
formulas, and some meat and fish. The relatively high levels of manganese in nuts, grains, and
many plant products and infant formulas are not well absorbed upon ingestion because these
foods also contain inhibitors of manganese absorption such as phytates, fiber, plant protein and
polyphenolic compounds (tannins). Manganese absorption is affected by other factors including
age (neonate compared to the adult), chemical species of manganese, dose, and route of exposure
in addition to the dietary factors mentioned above. Human exposure to manganese may also
occur through inhalation of manganese dust, intake of soil containing manganese compounds, or
drinking water contaminated with manganese.
External Review Draft — Manganese—April 2002
1-1
-------
The primary target of manganese toxicity is the nervous system, and common symptoms
of toxic exposure include ataxia, dementia, anxiety, a "mask-like" face, and manganism, a
syndrome similar to Parkinson's disease. These effects, when observed, are generally the result of
very high exposures via inhalation, as might occur in an industrial setting, and are riot seen among
the general population exposed to low or moderate manganese levels. Manganese has very low
toxicity by oral ingestion and reports of adverse effects by this route are rare. Because manganese
is an essential nutrient, concern for toxic over-exposure must be balanced against the potentially
negative effects of nutritional deficiency resulting from under-exposure.
An epidemiological study performed in Peloponnesus, Greece (Kondakis et al., 1989)
showed that lifetime consumption of drinking water containing naturally high concentrations of
manganese oxides may lead to neurological symptoms and increased manganese retention
(through the concentration of manganese in hair) for people over 50 years old. For the group
consuming the highest concentration (around 2 mg/L) for more than ten years, the authors
suggested that some neurologic impairment may be apparent. The study raises concerns about
possible adverse neurological effects following chronic ingestion from drinking water at doses
within ranges deemed essential. However, the study did not examine manganese intake data from
other routes/sources (i.e., dietary intake, inhalation from air, etc.), precluding its use as a basis for
theRfD.
Another long-term drinking water study in Germany (Vieregge et ai, 1995) found no
neurological effects in people older than 50 years of age who drink water containing 0.3 to 2.16
mg/L of manganese for more than ten years. However, this study also lacks exposure data from
other routes and sources, and the manganese concentration range in water is very wide. Tims, the
study cannot be used for quantitative assessment.
A small Japanese community (total 25 individuals) ingested high levels of manganese in
contaminated well water (that leaked from dry cell batteries buried near the wells) over a three-
month period (Kawamura et al., 1941). Manganese intake was not determined at the time of
intoxication, but when assayed months later, it was estimated to be close to 29 mg/li (i.e., 58
mg/day or approximately 1 mg/kg-day assuming a body weight of 60 kg). Symptoms included
lethargy, increased muscle tonus, tremor, mental disturbances, and even death. Autopsies
revealed macroscopic and microscopic changes in the brain tissue. In contrast, six children (1- to
10-yr-old) were not intoxicated as were the adults by this exposure. The elderly were more
severely affected. Some effects may have resulted from factors other than manganese exposure.
There is no information available on the carcinogenic effects of manganese in humans, and
animal studies have reported mixed results. EPA considers manganese to be not classifiable with
respect to carcinogeniciry, Group D, according to the!986 Guidelines for Carcinogen Risk
Assessment. Data from oral exposure suggest that manganese has a low developmental toxicity.
In various surveys, manganese intakes of adults eating western-type and vegetarian diets
ranged from 0.7 to 10.9 mg per day (Freeland-Graves, 1994; Gibson, 1994 as cited by Food and
Nutrition Board, 2002). Depending on individual diets, a normal intake may be well over 10 mg
External Review Draft—Manganese — April 2002
1-2
-------
per day, especially from a vegetarian diet (Schroder et al. 1966). Thus, from the dietary surveys
taken together, EPA concludes that an appropriate reference dose (RfD) for manganese is 10
mg/day (0.14 mg/kg-day; IRIS, 1996). This RfD is unique, with an uncertainty factor (UF) of 1
applied to a human chronic NOAEL of 0.14 mg/kg-day. The UF of 1 is used because the
NOAEL (with no apparent LOAEL) is based on chronic human dietary intake surveys, not the
typical toxicity studies, and because of the essentiality of the trace element.
EPA derived a health-related benchmark for evaluating the occurrence data, called the
health reference level (HRL), of 0.30 mg/L. The HRL is six times the s-MCL of 0.05 mg/L. The
HRL is based on the dietary RfD and application of a modifying factor (MF) of three for drinking
water as recommended by IRIS (IRIS, 1996), and on an allocation of an assumed 20% relative
source contribution'from water ingestion as opposed to total manganese exposure. The
modifying factor accounts for concerns raised by the Kondakis study (1989), the potential for
higher absorption of manganese in water compared to food, consideration of fasting individuals,
the concern for infants with potentially higher absorption and lower excretion rates of manganese,
and the potential for increased susceptibility to neurotoxic effects of ingested manganese as
compared to adults. For example, Dorman et al. (2000) reported that rat pups dosed for 21 days
posmatally with 11 or 22 mg Mn/kg-day (by mouth in drinking water) exhibited significant
increases in the startle response compared to controls. Significant increases in striatal DA
(dopamine) and DOPAC (dihydroxyphenylacetic acid) concentrations, in the absence of
pathological lesions, were also observed in the high-dose treated neonates. Because manganese is
an essential nutrient in developing infants, the potential adverse effects from manganese deficiency
may be of greater concern than potential toxicity from over-exposure. Potentially sensitive sub-
populations include children, the elderly, pregnant women, iron-deficient individuals, and
individuals with impaired liver function.
Exposure to manganese in drinking water is ubiquitous in the United States. Data from
the National Inorganics and Radionuclide Survey (NIRS), conducted between 1984 and 1986 by
EPA, were used to characterize manganese occurrence in public water systems (PWSs).
Although somewhat out of date, these data indicate that occurrence estimates are relatively high,
with approximately 68% of ground water PWSs (an estimate of approximately 40,000 systems
nationally) showing detections of manganese, affecting about 55% of the ground water PWS
population served (approximately 47.5 million people nationally). The median levels for detects
and the 99th percentile concentration for all samples were 0.01 milligram per liter (mg/L) and 0.63
mg/L, respectively. Based on this survey information (which consisted only of ground water and
not surface water sampling), and using supplemental surface water levels from Safe Drinking
Water Act (SDWA) compliance monitoring data from five States, EPA concluded that population
exposure to manganese in PWSs is potentially high.
When the detected concentrations are evaluated at a draft health reference level (HRL) of
0.3 mg/L, approximately 6.2% of the NIRS PWSs have detections > J/2 HRL (> 0.15 mg/L),
consisting of about 3,700 ground water PWSs nationally, and affecting approximately 4.6% of the
population served (estimated at four million people nationally). The percentage of NIRS PWSs
with detections > HRL of 0.3 mg/L is approximately 3.6% (about 2,200 ground water PWSs
External Review Draft — Manganese—April 2002 1 -3
-------
nationally), affecting 2.7% of the population served (estimated at approximately 2.3 million
people nationally). It is important to note, however, that when average daily drinking water
intakes for manganese are compared with intakes from a normal diet, drinking water accounts for
a relatively small proportion of total manganese intake. \
External Review Draft—Manganese—April 2002
1-4
-------
2.0 IDEISfTITY: CHEMICAL AND PHYSICAL PROPERTIES
Manganese is an abundant element which makes up about 0.1% of the earth's crust
(ATSDR, 2000), Although the elemental (metal) form of manganese does not occur naturally in
the environment, manganese is a component of over 100 minerals (ATSDR, 2000).. The most
common mineral forms include manganese dioxide, manganese carbonate, and manganese silicate
(ATSDR, 2000). Manganese exists in 11 oxidative states, with the most common valences being
2+, 4+, and 7+ (U.S. EPA, 1994a). Although there is no recommended daily allowance (RDA)
for manganese, it is essential for the proper function of several enzymes and is necessary for
normal bone structure and brain function (U.S. EPA, 1994a). The chemical and physical
properties of elemental manganese are presented in Table 2-1. Chemical and physical properties
for manganese compounds are summarized in Table 2-2.
External Review Draft — Manganese—April 2002
2-1
-------
Table 2-1. Chemical and Physical Properties of Manganese.
Property
Chemical Abstracts Services
(CAS) Registry No.
Chemical Formula
Atomic Number
Molecular Weight
Synonyms
NIOSH Registry of Toxic
Effects of Chemical Substances
(RTECS)No.
Hazardous Substances Data
Bank(HSDB)No.
Boiling Point
Melting Point
Vapor Pressure (at 1,292°C)
Density (at 20°C)
Water Solubility
Acid Solubility
Information
1439-96-5
Mn
25
54.94
Elemental manganese; Collodial manganese;
Cutaval; Magnacat; Tronamang ;
009275000
00550
1,962°C
1,244°C
1 mmHg ;
7.21-7.44 g/cm3
Decomposes 1
e i
Dissolves in dilute mineral acids ;
Sources: ATSDR (2000); U.S. EPA (1994a); ChemlDp/tw (2000)
External Review Draft — Manganese—April 2002
2-2
-------
en
•SJ
O
a
S
o
(J
9)
I
I
e
PM
T3
S
I
G"
If
£5
*
U
£
0 «
§
f
s s
f t
lfffl
* "'
a .
onate;
te;
onate;
ite
S
M
Na
§
§
I
3s
a?
de;
de;
ride
ese c
ese di
bichl
(Il) c
as,
R
•a
s
en
"*hk
I
I
i
I
-------
e
0>
|
|
en
|
O
I-
O
U
O)
09
o>
a
C3
CUD
i
cu
p-
I
•a
.a
i
PH
TS
a
a
"3
u
JO
U
s
03
H
I
IG1
pa s^
o O
1 1
el
i
H
Q
CO
S
it
01
s
'Tl
g>J i
Itl
T
9
wi
I
."3
es
S
O
S
sS
I
•s
82
P §
II
» H
.
«N
»*»
I
1
I
-------
s
•-c
i
09
T3
9
O
I
u
ta
I
OS
I
O)
"S
es
l
u
•s
H
I
**•»
If f
Sf& S a
1*
.
U
1
I
•S3 o
a .s B?
tn
a
1,50
55 .S B?
a
•43
•3
anese dinit
acid, man
s
.s a?
1
T
•8
1
CO
9
ingnaic acid;
iurn salt;
jleon mineral
-------
-------
3.0 USES AND ENVIRONMENTAL, FATE
The uses and environmental fate of manganese in air, water, and soil have been extensively
reviewed by ATSDR (2000) and U.S. EPA (1994a). Information from these documents and
other sources is summarized below.
3.1 Production and Uses
Manganese is a naturally occurring element that constitutes approximately 0.1% of the
earth's crust. It does not occur in the environment in its pure metal form, but is ubiquitous as a
component of over 100 minerals, including many silicates, carbonates, sulfides, oxides,
phosphates, and borates (ATSDR, 2000). Manganese occurs naturally at low levels in soil, water,
air, and food. Of the heavy metals, manganese is surpassed in abundance only by iron (ATSDR,
2000).
In the United States, most manganese ore is smelted to produce ferromanganese, which is
a manganese-iron alloy (ATSDR, 2000). The latter is used primarily in the production of steel to
improve stiffness, hardness, and strength. The ore is mined in open pit or shallow underground
mines, though little has been mined in the U.S. since 1978 (ATSDP^, 2000; USGS, 2000). Almost.
all of the manganese ore used in steel production in the United States is imported (see Table 3-1;
ATSDR, 2000). Large quantities of ferromanganese are imported as well (USGS, 2000). Table
3-2 provides further information by State of the widespread manufacture and processing of
manganese.
Table 3-1. Imports of Manganese and Ferromanganese to the United States (thousand
metric tons, gross weight).
Compound
manganese ore
ferromanganese
1984
308
330
1988
499
449
1995
394
310
1996
478
374
1997
355
304
1998
332
339
1999 f
535
325
years 1984 and 1988: ATSDR, (1997)
years 1995 to 1999: USGS, (2000)
^estimated
Manganese compounds are produced through reactions of various elements and
compounds with either manganese ores or ferromanganese (ATSDR, 2000). Some common
manganese compounds include manganese chloride, manganese sulfate, manganese (II, III) oxide,
manganese dioxide, and potassium permanganate (ATSDR, 2000). Uses of these compounds are
varied, implying widespread environmental release. Significantly, approximately 80% of the
potassium permanganate used in the United States is expended in wastewater and drinking water
treatment (U.S. EPA, 1984). Manganese dioxide is used in the production of matches, dry-cell
batteries, fireworks, and as a precursor for other manganese compounds. Manganese chloride is
also used as a precursor for manganese compounds. A large
External Review Draft — Manganese—April 2002
3-1
-------
Table 3-2. Manganese Manufacturers and Processors by State.
Sttte'
AL
AR
AZ
CA
CO
CT
DE
FL
QA
HI
IA
ID
IL
IN
KS
KY
LA
MA
MD
ME
MI
MN
MO
MS
MT
NC
ND
ME
NH
NJ
MM
NV
NY
OH
OK
OR
PA
PR
RI
SC
SD
TO
TX
UT
VA
VT
WA
Wl
WV
WY
Number of facilities
60
29
8
55
14
16
1
26
42
1
51
3
121
161
30
63
17
26
17
8
128
28
49
23
1
57
5
18
4
27
1
2
63
231
48
17
179
5
5
57
7
54
85
23
23
1
27
126
15
2
Range of maximum amounts
on-site in thousands of
0-50,000
0-50,000
1-1,000
0-500,000
1-10,000
1-1,000
10-100
0-10,,000
0-10,000
10-100
0-10,000
1-1,000
0-50,000
0-50,000
1-500,000
0-500,000 ;
0-10,000 !
0-1,000
1-50,000
1-100
0-10,000
0-10,000
0-10,000
0-50,000
100,000-500,000
0-10,000
1-100
0-10,000
1-1,000
1-10,000
10-100
100-50,000
0-10,000
0-500,000
0-10,000
1-10,000
0-100,000
0-1,000
1-1,000
0-10,000
1-100
0-50,000
0-10,000
1-100,000
0-1,000
10-100
0-1,000
0-50,000 ;
1-500,000
0-1,000 1
Activities and uses' :
1,23,6,7,8,9,12
1,23,5,7,8,9,12,13
1,4,5,7,8,9,10,12
1,23,4,5,6,7,8,9,10,1,1,12,13
23,4,9,12
23,9,10
1A8
8,9,10,13 ;
1,23,5,7,8,9,10,12,13
9
1,23,5,7,8,9,10,12
9 i
1,23,4,5,8,9,10,11,12,13 i
1,23,4,5,6,7,8,9,10,11,12,13
13,4,5,8,9,12,13
1,2,3,4,5,6,7,8,9,10,12,13
1,23,5,7,8,9,10,12,13 >
1,23,4,5,9,10 ;
2,4,9,10,13
13,9
1,23,4,5,6,7,8,9,10,12,13
8,9,10,12
1,5,8,9,12 i
8,9,13
1,23,4,5,6,7
1,23,5,8,9,10,1 1,12,13
23,9
1,23,8,9,12,13
8,9
1,23,4,7,8,9,10 ,
9
23,7
1,23,4,5,7,8,9,10,12,13
1,23,4,5,6,7,8,9,10,12,13 i
1,23,4,5,6,8,9 :
23,9,12,13 (
1 ,23,4,5,7,8,9,10,1 1,12,13
9
23,9,10
1,23,5,7,8,9,10,13
9,13
1,23,4,5,6,7,8,9,10,12,13
1,23,4,5,6,8,9,10,12,13 ;
23,7,9,12,13
13^,7,8,9
9
U3,6,8,9 j
U3A6,7,8,9,10,12,13 1
8A10,13 , 1
1,5
•Post office State abbreviations used
*Dibi ta TR1 ire maximum amounts on-site at each facility
•Activities/Uses
1. Produce 8. Formulation component
2. Import 9. Article component
3. OB-site use/processing 10. Repackaging
4. Sale/distribution 11. Chemical processing aid
5. Byproduct 12. Mannfhntining aid
6. Impurity 13. Ancillary/other uses
7. Readmit
source: ATSDR (2000) compilation of1996 TRIdata
External Review Draft — Manganese—April 2002
3-2
-------
proportion (60%) of U.S. manganese sulfate is used as a fertilizer, while the remainder is used in
varnish, fungicides, and as a livestock feed supplement. An organic manganese compound,
methylcyclopentadienyl manganese tricarbonyl (MMT), was used as an anti-knock additive in
unleaded gasoline before it was banned in 1977. However, a 1995 court decision required EPA
to reregister MMT and its use is ongoing (ATSDR, 2000).
The uses of manganese compounds vary widely depending on the chemical form. Table 3-
3 summarizes key uses of selected manganese compounds.
3.2 Sources and Environmental Fate
Manganese compounds are widely distributed in air, soil, and water. Sources of
atmospheric manganese include industrial emissions, fossil fuel combustion, and erosion of
manganese-containing soils. Volcanic eruptions can also contribute to levels of manganese in air.
Almost 80% of industrial emissions of manganese are attributable to iron and steel production
facilities. Power plant and coke oven emissions contribute about 20%. Although soil erosion is
considered an important source of atmospheric manganese, quantitative data for contributions
from this source are not available. Due to generally low vapor pressure, manganese compounds
in air exist primarily as suspended particulate matter. Because particle size is small, atmospheric
manganese distribution can be widespread. These particles will eventually settle out via the
process of dry deposition into surface waters or onto soils. Little information is available on the
chemical reactions of atmospheric manganese, but it is expected to react with sulfur and nitrogen
dioxide. The half-life of manganese in air is only a few days (ATSDR, 2000).
The fuel additive methylcyclopentadienyl manganese tricarbonyl (MMT) is expected to
contribute to urban air concentrations of manganese compounds. The fuel-enhancing properties
of MMT were first discovered in the 1950s, and the compound has been used as an additive in
leaded and unleaded gasoline since the 1970s in the United States and Canada (Lynam et al.,
1999). MMT was banned for use in unleaded gasoline in the United States in 1977 in accordance
with provisions in the Clean Air Act, which stated that all gasoline additives that were not
"substantially similar" to gasoline were required to obtain a waiver proving that the additive did
not "cause or contribute to the failure of emission control systems" (Lynam et al., 1999). The
U.S. EPA lifted this ban under court order in 1995, and MMT has been used freely since that
time.
Gasoline without MMT contains virtually no manganese (Lynam etal., 1999). The
currently allowed maximum level of MMT in unleaded fuel is 0.03125 gram of manganese per
U.S. gallon of gasoline (0.0083 g/L or 10.4 ppm). The amount of manganese emitted from the
tailpipe of an automobile using MMT-containing fuel depends upon the type of engine, driving
cycle, and age of the vehicle. Estimates for manganese in vehicular exhaust vary between 4% and
41% of the manganese consumed (Ardeleanu et al., 1999). The remaining fraction apparently
remains in the vehicle (Ardeleanu et al., 1999). Early analysis of emissions suggested
External Review Draft — Manganese — April 2002
3-3
-------
Table 3-3. Summary of Uses for Selected Manganese Compounds.
Compound
Methylcyclopentadienyl manganese tricarbonyl
(MMT)
Manganous carbonate
Manganese chloride
Manganous acetate
Manganese etfaylenebisdithiocarbamate ,
Manganese oxide
Manganese phosphate
Manganese sulfate
Manganous trifluoride
Manganese borate
Manganous nitrate
Manganese dioxide (electrolytic manganese,
pyrolusite)
Potassium permanganate
Use
i
Fuel additive ;
Ferrites; animal feeds; ceramics; acid soluble
manganese source
i
Catalyst in organic compound chlorination; trace
mineral supply for animal feed; brick colorant; dye;
dry-cell batteries; linseed oil drier; disinfecting;
purifying natural gas \
Mordant in dyeing; drying agent for paint and varnish;
bister ;
Agricultural fungicide
Ferrites; ceramics; fertilizer; livestock feed additive
Ingredient of proprietary solutions for phosphating iron
and steel !
Livestock feed additive; fertilizer; glazes; varnishes;
ceramics; fungicides
Fluorinating agent in organic chemistry
Drying agent for varnish and oil; linseed oil drier;
leather industry
Porcelain colorants; manufacture of reagent grade
manganese dioxide
Dry-cell batteries; matches; fireworks; porcelain; glass
bonding materials; amethyst glass; manufacturing
manganese steel; oxidizer
Oxidizing agent; water and air disinfectant; antialgal
agent; metal cleaning, tanning, and bleaching agent;
fresh flower and fruit preservative !
Sources: U.S. EPA (1994a); ATSDR (2000); Merck (1983).
External Review Draft—Manganese — April 2002
3-4
-------
that manganese from combustion of MMT is emitted primarily as manganese tetroxide (Mn3O4)
(der Haar et al., 1975d as cited in Lynam et al., 1995). However, more recent testing suggests
that when very low levels of MMT are combusted (i.e., concentrations comparable to the
currently allowed levels), manganese is emitted primarily as manganese phosphate and sulfate.
The reported valence of the emitted manganese is +2.2, with a mass median aerodynamic
.diameter of 1 to 2 microns (Ethyl Corporation, 1997; Ressler et al., 1999; Wong et al., 1998; all
as cited in Lynam et al., 1999). Uncombusted MMT rapidly decomposes to manganese oxide,
carbon dioxide, and organic compounds in the atmosphere and has a half-life of only a few
seconds in the presence of sunlight (Lyman et al., 1999; Zayed et al., 1999a). Data on the
occurrence of manganese in air resulting from combustion of MMT and other sources are
presented in Section 4.2.
Manganese is listed as a Toxic Release Inventory (TRI) chemical. In 1986, the
Emergency Planning and Community Right-to-Know Act (EPCRA) established the TRI of
hazardous chemicals. Created under the Superfund Amendments and Reauthorization Act
(SARA) of 1986, EPCRA is also sometimes known as SARA Title IE. The EPCRA mandates
that larger facilities publicly report when TRI chemicals are released into the environment. This
public reporting is required for facilities with at least 10 full-time employees that annually
manufacture or process more than 25,000 pounds, or use more than 10,000 pounds, of TRI
chemical (U.S. EPA, 1996e, 2000a).
Under these conditions, facilities are required to report the pounds per year of manganese
released into the environment both on- and off-site. The on-site quantity is subdivided into air
emissions, surface water discharges, underground injections, and releases to land (see Table 3-4).
For manganese, releases to land constitute most of the on-site releases, with an abrupt decrease
occurring in 1989. It is unclear whether this sharp decrease is real or a function of changes in TRI
reporting requirements in the late 1980s and early 1990s (see discussion below). Land releases
have fluctuated modestly since that year with no trend evident. Air emissions are also an
important mode of on-site release. Though the first four years of record for air emissions are
markedly higher, no trend is apparent for the remainder. Surface water discharges and
underground injections are less significant on-site releases, with underground injections sharply
decreasing in 1994. Low levels of underground injection have continued to the present. Off-site
releases of manganese are considerable. Though in 1990 there was a large drop when compared
to previous years, the late 1990s showed a steady increase in pounds released. These TRI data
for manganese were reported from 49 States, excluding Alaska (U.S. EPA, 2000b).
Only 1% of environmental manganese is released to water (Table 3-4). The primary
sources for surface and ground water releases are industrial facility effluent discharge, landfill and
soil leaching, and underground injection. Manganese, in the form of potassium permanganate,
may be used in drinking water treatment to oxidize and remove iron, manganese, and other
contaminants (ANSI/NSF, 2000), in addition to its use in industrial wastewater purification and
odor abatement (ATSDR, 2000; U.S. EPA, 1984). Transport and partitioning of manganese in
water is dependent on the solubility of the manganese form. The chemical form is controlled by
factors such as pH, oxidation-reduction potential (Eh), and the available anions. Often, manganese
External Review Draft — Manganese—April 2002
3-5
-------
in water will settle into suspended sediments. Little information is available on the biodegradation
of manganese-containing compounds in water, but factors such as pH and temperature are
important for microbial activities. Data for occurrence of manganese in drinking water are
presented in Section 4.3. :
Approximately 91% of environmental manganese is released to soil. The main source of
this release is land disposal of manganese-containing wastes. The ability of manganese compounds
to adsorb to soils and sediments is contingent upon the cation exchange capacity and organic
content of the soil or sediment. Adsorption can vary widely based on differences in these two
factors. Oxidative microbial activity may increase the precipitation of manganese minerals and
increase the dissolution of manganese in subsurface environments. Occurrence data for
manganese in soils are presented in Section 5.3.
TRI data are also available for the release of manganese compounds (Table 3-5). Releases
to land again constitute the largest proportion of on-site releases. With the exception of 1997 and
1998, releases to land have generally decreased over the period of record. Air emissions are also
an important mode of release and no trends are evident in the data. Significantly, surface water
discharges and underground injections are much more substantial for the compounds than for
elemental manganese, and have been generally increasing (dramatically in some years) since the
early 1990s. These data must be interpreted with caution, however, as they reflect changes in the
requirements for reporting releases. In 1998, only releases of 75,000 Ibs/yr were required to be
reported; this value is now 25,000 Ibs/yr. Therefore, although the values may seem to be
increasing, they are likely comparable to past releases that were previously unreported. Further,
the TRI data are meant to reflect releases and should not be used to estimate general exposure to
a chemical (U.S. EPA, 2000c, d). |
I
Increases in surface water discharges and underground injections of manganese
compounds have contributed to increases in total on- and off-site releases in recent years. The
latter have returned to, or exceeded, the higher levels seen in the late 1980s and early 1990s. Off-
site releases, a large component of total releases, are also at their highest levels since reporting
began in 1988. These TRI data for manganese compounds were reported from all 50 States (U.S.
EPA, 2000b).
External Review Draft — Manganese — April 2002
3-6
-------
Table 3-4. Environmental Releases (in pounds) for Manganese in the United States,
1988-1998.
Year
1998
1997
1996
1995
1994
1993
1992
1991
1990
1989
1988
On-Site Releases
Air
Emissions
970,658
751,743
816,733
699,897
818,600
901,827
721,047
1,113,160
1,1'68,809
2,444,211
1.586.675
Surface Water
Discharges
260,403
146,364
117,571
117,277
89,332
243,999
235,307
143,105
139,358
150,965
321.993
Underground
Injection
3
7
8
17
10
504
304
272
881
556
255
Releases
to Land
9,995,895
9,920,481
10,111,563
8,279,054
8,452,582
7,530,152
6,543,600
9,906,511
9,031,215
7,984,172
20.229.826
Off-Site
Releases
15,967,545
16,209,483
15,191,636
12,753,204
14,076,682
12,150,694
11,997,270
14,590,589
11,364,721
20,559,164
20.087.660
Total On- &
Off-Site
Releases
27,194,504
27,028,078
26,237,511
21,849,449
23,437,206
20,827,176
19,497,528
25,753,637
21,704,984
31,139,068
42.226.409
source: U.S. EPA (2000b)
Table 3-5. Environmental Releases (in pounds) for Manganese Compounds in the
United States, 1988-1998.
Year
1998
1997
1996
1995
1994
1993
1992
1991
1990
1989
1988
On-Site Releases
Air
Emissions
1,566,352
1,549,505
1,828,684
2,928,644
3,060,424
2,324,442
2,079,044
1,531,832
2,276,084
1,847,528
1,801,463
Surface Water
Discharges
4,471,582
4,202,876
2,119,241
1,627,184
857,825
685,737
733,728
709,557
721,787
907,866
681,469
Underground
Injection
7,755,610
14,412,830
15,630
3,590
5,930
8,740
22,569
15,327
2,842
1,005,518
6.816,070
Releases
to Land
52,820,578
50,141,026
40,334,426
41,832,058
38,228,464
47,763,821
63,490,137
66,559,047
83,331,787
85,191,013
84,227,842
Off-Site
Releases
45,269,882
47,233,186
33,543,677
25,994,951
-25,840,954
22,780,860
17,297,544
27,250,630
35,789,554
33,004,908
20,670,921
Total On- &
Off-site
Releases
111,884,004
117,539,423
77,841,658
72,386,427
67,993,597
73,563,600
83,623,022
96,066,393
122,122,054
121,956,833
114,197,765
source: U.S. EPA (2000b).
External Review Draft — Manganese—April 20Q2
3-7
-------
Although the TRI can be useful in giving a general idea of release trends, the data are far
from exhaustive and have significant limitations. For example, only industries which meet TRI
criteria (at least 10 full-time employees and manufacture and processing of quantities exceeding
25,000 Ibs/yr, or use of more than 10,000 Ibs/yr) are required to report releases. These reporting
criteria do not account for releases from smaller industries. Threshold manufacture and
processing quantities also changed from 1988 to 1990 (dropping from 75,000 Ibs/yr in 1988 to
50,000 Ibs/yr in 1989 to its current 25,000 Ibs/yr in 1990), creating possibly misleading data
trends. Finally, the TRI data are meant to reflect releases and should not be used to estimate
general exposure to a chemical (U.S. EPA, 2000c, d).
In summary, manganese and many of its compounds are naturally occurring and fonnd at
low levels in soil, water, air, and food. Furthermore, manganese compounds are produced in the
United States from manganese ore and are in widespread use. Most ferromanganese is used in
steel production, while other manganese compounds are used in a variety of applications from
fertilizers and industrial products to water treatment. Recent statistics regarding import for
consumption indicate production and use are substantial (Table 3-1). Manganese and its
compounds are also TRI chemicals (Tables 3-4 and 3-5). Industrial releases have been reported
since 1988 in all 50 States. Off-site releases constitute a considerable amount of total releases,
with releases to land being the most significant on-site releases.
External Review Draft—Manganese—April 2002
3-8
-------
4.0 EXPOSURE FROM DRINKING WATER
4.1 Introduction
This chapter examines the occurrence of manganese in drinking water. No complete
national database exists regarding the occurrence of unregulated or regulated contaminants in
drinking water from public water systems (PWSs) collected under the Safe Drinking Water Act
(SDWA). In this chapter, existing federal and State data that have been screened for quality, ,
completeness, and representativeness are aggregated and analyzed. Populations served by PWSs
exposed to manganese are also estimated, and the occurrence data are examined for special
trends. To augment the incomplete national drinking water data and aid in the evaluation of
occurrence, information on the use and environmental release, as well as ambient occurrence of
manganese, is also reviewed.
4.2 Ambient Occurrence
To understand the presence of a chemical in the environment, an examination of ambient
occurrence is useful. In a drinking water context, ambient water is source water existing in
surface waters and aquifers before treatment. The most comprehensive and nationally'consistent
data describing ambient water quality in the United States are being produced through the United
States Geological Survey's (USGS) National Ambient Water Quality Assessment (NAWQA)
program. NAWQA, however, is a relatively young program and complete national data are not
yet available from the entire array of sites across the nation.
Data Sources and Methods
The USGS instituted the NAWQA program in 1991 to examine water quality status and
trends in the United States. NAWQA is designed and implemented in such a manner to allow
consistency and comparison among representative study basins located around the country, -
facilitating interpretation of natural and anthropogenic factors affecting water quality (Leahy and
Thompson, 1994).
The NAWQA program consists of 59 significant watersheds and aquifers referred to as
"study units." The study units represent approximately two-thirds of the overall water usage in
the United States and a similar proportion of the population served by public water systems.
Approximately one-half of the nation's land area is represented (Leahy and Thompson, 1994).
To facilitate management and make the program cost-effective, approximately one-third of
the study units at a time engage in intensive assessment for a period of 3 to 5 years. This is
followed by a period of less intensive research and monitoring that lasts between 5 and 7 years.
This way, all 59 study units rotate through intensive assessment over a ten-year period (Leahy and
Thompson, 1994). The first round of intensive monitoring (1991-96) targeted 20 watersheds and
the second round monitored 16 basins beginning in 1994.
External Review Draft — Manganese — April 2002
4-1
-------
Manganese is an analyte for both surface and ground water NAWQA studies, with a
Minimum Reporting Level (MRL) of 0.001 mg/L. Manganese occurrence in bed sediments and
aquatic biota tissue is also assessed, with MRLs of 4 mg/kg and 0.1 mg/kg, respectively.
Additional information on analytical methods used in the NAWQA study units, including
minimum reporting levels, are described by Gilliom and others (1998). i
Manganese data from the first two rounds of intensive NAWQA monitoring have
undergone USGS quality assurance checks and are available to the public through their NAWQA
Data Warehouse (USGS, 2001). EPA has analyzed these data after further data quality review
and occurrence results are presented below. The descriptive statistics generated from the
manganese NAWQA data broadly characterize the frequency of manganese detections by sample
and by site. Furthermore, detection frequencies above a Health Reference Level (HRL) of 0.3
mg/L are also presented for all samples, and by site. The HRL is a preliminary health effect level
used for this analysis (see Section 4.3 for further discussion of the HRL and its development).
The median and 99th percentile concentrations are included as well to characterize the, spread of
manganese concentration values in ambient waters sampled by the NAWQA program.
Results
Typical of many inorganic contaminants, manganese occurrence in ambient surface and
ground waters is high (Table 4-1). This is to be expected, considering that manganese constitutes
approximately 0.1% of the earth's crust (of the heavy metals, it is surpassed in abundance only by
iron), and the element and its compounds are used in many products. Significantly, potassium
permanganate is used in wastewater and drinking water treatment. !
Detection frequencies are consistently greater for surface water than for ground water,
possibly because surface waters are more likely to act as sinks for anthropogenic releases of
manganese. Median concentrations are also generally higher for surface water (median
concentration for all sites is 0.016 mg/L in surface water and 0.005 mg/L in ground water).
However, manganese detection frequencies > HRL are consistently higher in ground water, and
99th percentile ground water concentrations are as much as eight times larger than corresponding
99th percentile surface water concentrations. Locally high concentrations in ground water, higher
than any seen in surface water, are not surprising given the possibility of long contact tunes
between ground water and rocks enriched in manganese at a given location. Contact tunes
between surface waters and naturally occurring manganese are orders of magnitude shorter, hence
concentrations are lower. Furthermore, surface waters subject to large anthropogenic inputs of
manganese are more easily diluted by waters integrated from other parts of the watershed, where
manganese concentrations may be lower.
Table 4-1 illustrates that low-level manganese occurrence is ubiquitous. Surface water
detection frequencies by site are greater than 95% for all land use categories. Median
concentrations and HRL exceedances (by site) are greater in urban and agricultural basins
compared to basins characterized as mixed land use or foresl/rangeland. This distribution of
manganese occurrence is probably influenced by the wide use of manganese compounds in both
External Review Draft —Manganese—April 2002 ', 4-2
-------
industry and agriculture. Mixed land use basins are generally larger than either urban or
agricultural basins, and the lower occurrence in these basins may reflect some dilution of the
contaminant. The 99th percentile concentrations for surface water range from 0.4 mg/L to 0.8
mg/L. The frequency of detections exceeding the MRL and HRL by site for all sites are
approximately 96.9% and 10.2%, respectively. These figures indicate that, although manganese is
nearly ubiquitous in surface water, detections at levels of public health concern are relatively low.
For ground water, detections by site are higher in urban and forest/rangeland areas than in
mixed or agricultural lands. Over 80% of urban and forest/rangeland sites reported detections,
while approximately 63 to 64% of mixed and agricultural land use sites detected manganese. The
finding that ground water manganese occurrence is higher in forest/rangeland areas than in either
mixed or agricultural sites may result from natural variation in manganese occurrence in soil and
rock. Urban areas have the highest median and 99* percentile concentrations (0.015 mg/L and
5.6 mg/L, respectively), as well as the highest detection frequencies (by site: 85.3%) and HRL
exceedances (both by sample and by site: 17.2% and 21%, respectively) of manganese in
groundwater. These results suggest that urban releases of manganese and manganese compounds
can leach to ground water.
Detection frequencies and HRL exceedances by site for all ground water sites are
approximately 70.1% and 13.8%, respectively. Again, these figures suggest that, while
manganese occurrence in ground water is high, detections at levels of public health concern are
relatively low. . ;
Manganese was detected at 100% of NAWQA stream bed sediment sampling sites. The
median and 99th percentile concentrations in bed sediments are 1.1 mg/kg (dry weight) and 9.4
mg/kg (dry weight), respectively. The occurrence of manganese in stream sediments is pertinent
to drinking water concerns because, though many manganese compounds are either insoluble or
have low solubility and are transported in water as suspended sediment, some desorption of the
compound from sediments into water will occur through equilibrium reactions, although in very
low concentrations.
In aquatic biota tissue, detections are also 100% of all samples and sites (Table 4-2).
However, concentration percentiles for tissues are substantially lower than for bed sediments: the
median for biotic tissue is 0.01 mg/kg (dry weight) and the 99th percentile is 2.9 mg/kg (dry
weight). Significant manganese concentrations in aquatic biota tissues would imply a potential for
bioaccumulation. Although manganese was detected in aquatic biota tissues at 100% of samples
and sites, low concentration percentiles suggest that the element does not bioaccumulate
appreciably.
External Review Draft—Manganese — April 2002
4-3
-------
4.3 Drinking Water Occurrence l
National Inorganic and Radionuclide Survey (NIRS)
In the mid-1980s, EPA designed and conducted the National Inorganic and Radionuclide
Survey (NIRS) to collect national occurrence data on a select set of radionuclides and inorganic
chemicals being considered for National Primary Drinking Water Regulations. The NIRS
database includes 36 inorganic compounds (IOC) (including 10 regulated lOCs), 2 regulated
radionuclides, and 4 unregulated radionuclides. Manganese was one of the 36 lOCs monitored.
The NIRS provides contaminant occurrence data from 989 community PWSs served by
ground water. The NIRS does not include surface water systems. The selection of this group of
PWSs was designed so that the contaminant occurrence results are statistically representative of
national occurrence. Most of the NIRS data are from smaller systems (based on population-
served) and each of these statistically randomly selected PWSs was sampled a single time between
1984 and 1986. '
i
i
The NIRS data were collected from PWSs in 49 States. Data were not available for the
State of Hawaii. In addition to being statistically representative of national occurrence, NIRS
data are designed to be divisible into strata based on system size (population served by the PWS).
Uniform detection limits were employed, thus avoiding computational (statistical) problems that
sometimes result from multiple laboratory analytical detection limits. Therefore, the NIRS data
can be used directly for national contaminant occurrence analyses with very few, if any, data
quality, completeness, or representativeness issues. , i
Supplemental IOC Data
One limitation of the NIRS study is a lack of occurrence data for surface water systems.
To provide perspective on the occurrence of manganese in surface water PWSs relative to ground
water PWSs, SDWA compliance monitoring data that were available to EPA were reviewed from
States with occurrence data for both kinds of systems. ,
The State ground water and surface water PWS occurrence data for manganese used in
this analysis were submitted by States for an independent review of the occurrence of regulated
contaminants in PWSs at various times for different programs (U.S. EPA, 1999a). In the U.S.
EPA (1999a) review, occurrence data from a total of 14 States were noted. However;, because
several States contained data that were incomplete or unusable for various reasons, only 12 of the
14 States were used for a general overview analysis. From these 12 States, 8 were selected for
use in a national analysis because they provided the best data quality and completeness and a
balanced national cross-section of occurrence data. These eight were Alabama, California,
Illinois, Michigan, Montana, New Jersey, New Mexico, and Oregon. i
External Review Draft — Manganese—April 2002
4-4
-------
Table 4-1. Manganese Detections and Concentrations in Streams and Ground Water.
Surface Water
urban
mixed
agricultural
forest/rangeland
all sites
Ground Water
urban
mixed
agricultural
forest/rangeland
all sites
Detection frequency
>MRL*
% samples
% sites
Detection frequency
>^HRL*
%samples
%sites
Concentration percentiles
(all samples; mg/L)
medjan
99*
99.1 %
92:4%
96.3 %
90.9%
94.0 %
99.6 %
98.5%
97.2%
96.4%
96.9 %
4.6 %
1.3%
3.7 %
5.0%
3.0%
13.0 %
6.4%
12.3%
6.6%
10.2%
0.036
0.012
0.019
0.011
0.016
0.7
0.4
0.7
0.8
0.7
74.7%
56.9 %
61.4%
75.3 %
, 64.1%
85.3 %
62.9 %
64.0%
81.3 %
70.1 %
17.2 %
8.9%
11.9%
10.9%
12.8%
21.0%
9.0%
12.8 %
13.8 %
13.8 %
0.015
0.002
0.004
0.012
0.005
5.6
1.3
1.6
2.9
2.9
" The Minimum Reporting Level (MRL)for manganese in water is 0.001 mg/L and the Health Reference Level (HRL) is 0.3 mg/L. The HRL is a
preliminary health effect level used for this investigation.
Table 4-2. Manganese Detections and Concentrations in Bed Sediments and Aquatic
Biota Tissues (all sites).
• sediments'
aquatic biota tissues
Detection frequency
>MRL*
% samples
100 %
100 %
% sites
100 %
100%
Concentration percentiles
(aU samples; mg/kg dry weight)
median
1.1
0.01
99*
9.4
2.9
* The Minimum Reporting Levels (MRLs) for manganese in sediments and biota tissues are 4 ug/g and O.I fig/g, respectively.
External Review Draft — Manganese—April 2002
4-5
-------
Only the Alabama, California, Illinois, New Jersey, and Oregon State data sets contained
occurrence data.for manganese. The data represent more than 37,000 analytical results from
about 4,000 PWSs mostly during the period from approximately 1993 to 1997, though some
earlier data are also included. The number of sample results and PWSs vary by State.
Data Management
' .1
The data used in the State analyses were limited to only those data with confirmed water
source and sampling type information. Only standard SDWA compliance samples were used;
"special" samples, "investigation" samples (investigating a contaminant problem that would bias
results), or samples of unknown type were not used in the analyses. Various quality control and
review checks were made of the results, including follow-up questions to the States providing the
data. Many of the most intractable data quality problems encountered occurred with older data.
These problematic data were, in some cases, simply eliminated from the analysis. For example,
when the number of data with problems were insignificant relative to the total number of
observations, they were dropped from the analysis (for further details see U.S. EPA, 1999a).
Occurrence Analysis
The summary descriptive statistics presented in Table 4-3 for manganese are derived from
analysis of the NIRS data. Included are the total number of samples, the percent samples with
detections, the 99th percentile concentration of all samples, the 99th percentile concentration of
samples with detections, and the median concentration of samples with detections. The
percentages of PWSs and population served indicate the proportion of PWSs and PWS
population served whose analytical results showed a detections) of the contaminant (simple
detection, > MRL) .at any time during the monitoring period; or a detections) greater than half
the Health Reference Level (HRL); or a detections) greater than the HRL. The HRL used for
this analysis is 0.30 mg/L.
The HRL was derived for contaminants not considered to be "linear" carcinogens by the
oral route of exposure. EPA derived the HRL using an RED approach as follows: HRL = (RID x
70 kg)/2 L x RSC,
where:
RfD = Reference Dose; an estimated dose (mg/kg-day) to the human population
(including sensitive subgroups) that is likely to be without an appreciable risk of
deleterious effects during a lifetime. It can be derived from a NOAEL, LOAEL, or
benchmark dose, with uncertainty factors generally applied to reflect limitations of
the data used;
70 kg = The assumed body weight of an adult;
2 L = The assumed daily water consumption of an adult;
External Review Draft—Manganese—April 2002
4-6
-------
RSC = The relative source contribution, or the level of exposure believed to result
from drinking water when compared to other sources (e.g., air), and is assumed to
be 20% unless noted otherwise.
EPA used only the best available peer reviewed data and analyses in evaluating adverse
health effects. Health effects information is available for manganese in the Integrated Risk
Information System (IRIS). IRIS is an electronic EPA data base containing reviewed information
(both inside and outside of the Agency) on human health effects that may result from exposure to
various chemicals in the environment. These chemical files contain descriptive and quantitative
information on RfDs for chronic noncarcinogenic health effects and hazard identification, as well
as slope factors and unit risks for carcinogenic effects.
In Table 4-3, national occurrence is estimated by extrapolating the summary statistics for
manganese to national numbers for systems, and population served by systems, from the Water
Industry Baseline Handbook, Second Edition (U.S. EPA, 2000e). From the handbook, the total
number of ground water community water systems (CWSs) plus ground water non-transient, non-
community water systems (NTNCWSs) is 59,440, and the total population served by ground
water CWSs plus ground water NTNCWSs is 85,681,696 persons (see Table 4-3). To arrive at
the national occurrence estimate for the HRL, the national estimate for ground water PWSs (or
population served by ground water PWSs) is simply multiplied by the percentage for the given
summary statistic [i.e., the national estimate for the total number of ground water PWSs with
detections at the HRL of 0.30 mg/L (40,388) is Hie product of the percentage of ground water
PWSs with detections (68%) and the national estimate for the total number of ground water
PWSs (59,440)].
In Table 4-4, occurrence data on manganese directly submitted by the States of Alabama,
California, Illinois, New Jersey, and Oregon for .4 Review of Contaminant Occurrence in Public
Water Systems (U.S. EPA,, 1999a) were used to augment the NIRS study which lacked surface
water data. Included in the table are the same summary statistics as shown in Table 4-3, with
additional information describing the relative distribution of manganese occurrence between
ground water and surface water PWSs in the 5 States.
The State data analysis was focused on occurrence at the system level because a PWS
with a known contaminant problem usually has to sample more frequently than a PWS that has
never detected the contaminant. The .results of a simple computation of the percentage of samples
with detections (or other statistics) can be skewed by the more frequent sampling results reported
by the contaminated site. The system level of analysis is conservative. For example, a system
need only have a single sample with an analytical result greater than the MRL, i.e., a detection, to
be counted as a system with a result "greater than the MRL."
When computing basic occurrence statistics, such as the number or percent of samples or
systems with detections of a given contaminant, the value (or concentration) of the MRL can have
important consequences. For example, the lower the reporting limit, the greater the number of
detections (Ryker and Williamson, 1999). As a simplifying assumption, a value of half the
External Review Draft — Manganese—April 2002 4-7
-------
Table 4-3. Manganese Occurrence in Ground Water PWS of NIRS Survey.
Frequency Factors
Total number of samples/systems ;
99* percentile concentration (all samples)
Minimum Reporting Level (MRL)
99th percentile concentration of detections
Median concentration of detections
Total population
Occurrence by Samples/System
% Ground water PWSs with detections (> MRL)
Range of sampled States
% Ground water PWSs >Vi HRL
Range of sampled States
% Ground water PWSs > HRL
Range of sampled States
Occurrence by Population Served
% Ground water PWS population served with detections
Range of sampled States
% Ground water PWS population served > V4 HRL
Range of sampled States
% Ground water PWS population served > HRL (
Range of sampled States
Health
Reference
Level = 0.3
mg/L
989
0.63 mg/L
0.001 mg/L
0.72 mg/L
0.01 mg/L
1,482,153
67.9%
8.3-100%
6.1%
0-31.6%
3.2%
0-21.0%
55.4%
0.3-100%
4.6%
0-89.2%
2.6%
0-89.2%
National System
& Population
Numbers1
59,440
—
• —
~
; —
85,681,696
National
Extrapolation
HRL =
03 mg/L
140,388
. NA
3,606
: NA •
1,923
NA
! •
47,502,000
. NA
3,940,000
NA
2256,000
NA
' Total PWS and population numbers are from EPA
March 2000 Water Industry Baseline Handbook.
MRL = minimum reporting level
PWS= public water system
NA = not applicable
HRL = health reference level
— = no data
External Review Draft — Manganese—April 2002
4-8
-------
MRL is often used as an estimate of the concentration of a contaminant in samples/systems whose
results are less than the MRL. However, for these occurrence data mis is not straightforward.
This is in part related to State data management differences as well as real differences in analytical
methods, laboratories, and other factors.
The situation can cause confusion when examining descriptive statistics for occurrence.
Because a simple meaningful summary statistic is not available to describe the various reported
MRLs, and to avoid confusion, MRLs are not reported in the summary table (Table 4-4).
Additional Drinking Water Data From 1996 AWWA Survey
To augment the SDWA drinking water data analysis described above, results from a 1996
American Water Works Association (AWWA) survey are reviewed. The survey, called
WATER:/STATS, is a cooperative project of AWWA and AWWA Research Foundation. The
WATER:/STATS survey database stores results from the 1996 WATER-./STATS survey of water
utilities in the United States and Canada in terms of facilities, scale of operation, and major inputs
and outputs. A total of 794 AWWA member utilities responded to the survey with ground water
and/or surface water information. However, the actual number of respondents for each data
category varies because not all participants in the survey responded to every question.
4.4 Results
The NIRS data in Table 4-3 show that approximately 68% of ground water PWSs (an
estimate of approximately 40,000 systems nationally) had detections of manganese, affecting
about 55% of the ground water PWS population served (approximately 47.5 million people
nationally). At an HRL of 0.30 mg/L, approximately 6.1% of the MRS PWSs had detections > V*
HRL (about 3,600 ground water PWSs nationally), affecting approximately 4.6% of the
population served (estimated at 3.9 million people nationally). The percentage of NIRS PWSs
with detections > HRL of 0.30 mg/L was approximately 3.2% (about 1,900 ground water PWSs
nationally), affecting 2.6% of the population served (estimated at approximately 2.3 million
people nationally) (Table 4-3).
Drinking water data for manganese from the supplemental individual States vary among
States (Table 4-4). Manganese has not been required for monitoring under SDWA, though these
States had obviously conducted some monitoring. The number of systems with manganese data
for Illinois and Oregon is far less than the number of PWSs in these States. Hence, the extent to
which these data are representative is unclear. Alabama, California, and New Jersey have
substantial amounts of data and PWSs represented. Because the NIRS data only represent
manganese occurrence in ground water PWSs, the supplemental State data sets provide some
perspective on surface water PWS occurrence. For example, the median concentration of
detections for the States ranged from Oi02 mg/L to 0.15 mg/L, higher than the NIRS data (0.01
mg/L). For detections by PWSs, 3 of the 5 States (California, Illinois, and Oregon) had higher
ground water PWS detections.
External Review Draft — Manganese—April 2002
4-9
-------
Table 4-4. Occurrence Summary of Ground and Surface Water Systems by State for
Manganese.
Frequency Factors
Total number of samples
Number of ground water samples
Number of surface water samples
% Samples with detections
% Ground water samples with detections
V» Surface water samples with detections
99"* percentile concentration (all samples)
Minimum reporting level (MRL)
99* percentile concentration of detections
Median concentration of detections
Total number of PWSs
Number of ground water PWSs
Number of surface water PWSs
Total population served
Ground water population
Surface water population
Alabama
1,343
934
409
30.2%
28.1%
35.0%
0.13 mg/L
Variable1
0.56 mg/L
0.02 mg/L
434
365
69
3,662,222
1,820,214
1,837,743
California
31,998
29,923
2,075
16.5%
17.5%
1.9%
0.71 mg/L
Variable1
1.52 mg/L
0.15 mg/L
2,516
2,293
; 223
45,388,246
27,840,774
30,675,992
Illinois
344
275
69
44.2%
50.2%
20.3%
0.96 mg/L
Variable1
57 mg/L
0.04 mg/L
227
160
67
1,995,394
724,635
1,270,179
New Jersey
3,196 !
2,795 '.
401 i
39.7% i
40.6% !
33.7% :
0.42 mg/L
Variable1 :
0.89 mg/L ;
0.02 mg/L
1,179
1,147
32
7,472,565
2,386396 ;
3,687,076
Oregon
H72
90
82
39.5%
61.1%
15.9%
1.6 mg/L
Variable1
6.7 mg/L
O.OS mg/L
84
54
30
1,306,283
301,440
1,117,782
Occurrence by System :
% PWSs with detections (> MRL)
% Ground water PWSs with detections
% Surface water P WSs with detections
46.5%
41.6%
72.5%
Health Reference Level (HRL)= 0.3 mg/L
%PWSs>J*HRL
% Ground water PWSs > 'A HRL
% Surface water PWSs > 'A HRL
%PWSs> HRL
% Ground water PWSs > HRL
% Surface water PWSs > HRL
1.8%
1.4%
4.4%
0.9%
0.6%
2.9%
Occurrence by Population Served
% PWS population served with detections
% Ground water PWS population with
detections
% Surface water PWS population with
detections
Health Reference Level (HRL) = 0.3 mg/L
% PWS population > 'A HRL
% Ground water PWS population > 14 HRL
% Surface water PWS population > 'A HRL
% PWS population > HRL
% Ground water PWS population > HRL
% Surface water PWS population > HRL
71.9%
50.9%
73.4%
28.2%
i 29.8%
: n.7%
41.4%
50.6%
19.4%
53.5%
52.3% i
96.9%
46.4%
55.6%
30.0%
17.2%
18.5%
3.6%
: 10.1%
10.9%
; 1.8%
1
49.3%
66.2%
10.5%
5.9%
0.8%
0.7%
2.4%
0.1%
0.6%
34.8%
52.6%
; 4.4%
27.2%
42.8%
4.2%
9.3%
11.9%
3.0%
4.4%
5.0%
3.0%
5.8%
5.7%
9.4%
2.5%
2.5%
3.1%
36.5%
66.3%
19.5%
16.5%
29.1%
9.4%
14.7%
24.2%
9.4%
85.7%
70.4%
100.0%
13.1%
20.4%
0.0%
6.0%
9.3%
0.0%
58.0%
41.8%
56.8%
15.3%
10.4%
23.3%
9.1%
4.9%
14.5%
4.6%
19.9%
0.0%
3.2%
14.0%
0.0%
1 See text for details PWS =
MRL ™ minimum reporting level HRL
public water system
= health reference level
External Review Draft—Manganese — April 2002
4-10
-------
For simple detections, the supplemental State data show a range from 30% to 56% of
ground water PWSs (Table 4-4). These figures are lower than the NIRS ground water PWS
results: 68% > MRL (Table 4-3). The supplemental State data show considerably greater
percentages of simple detections for surface water PWSs, with higher variability as well:
12%-97% >MRL. Comparisons made between data for simple detections need to be viewed with
caution because of differences in MRLs between the State data sets and the NIRS study, and
among the States themselves (see Section 3).
The supplemental State data sets indicate ground water PWS detections > HRL of 0.30
mg/L between 0.6% and 11% (Table 4-4). Again, this range brackets the NIRS national average
of PWS > HRL of 0.30 mg/L (3.2%) (Table 4-4). Notably, surface water PWSs showed fewer
exceedances of the HRL than ground water PWSs at this higher concentration; ranging from 0%
to 3.1%.
Reviewing manganese occurrence by PWS population served shows that from 0. l%-43%
of the States' ground water PWS populations were served by systems with detections > HRL of
0.30 mg/L (Table 4-4). Comparatively, 2.6% of the NIRS ground water PWS population served
experienced detections > HRL of 0.30 mg/L (Table 4-3). Populations served by surface water
PWSs with detections > HRL of 0.30 mg/L ranged from 0%-14.5% among the five supplemental
States. Population figures for the supplemental States are incomplete and are only reported for
those systems in the database that have reported their population data. For manganese,
approximately 80% of the PWSs reporting occurrence data for these 5 States also reported
population data.
Occurrence in AWWA PWSs
The AWWA sponsored 1996 WaterStats Survey showed manganese occurrence above
levels at which health effects are expected to be realized to be relatively similar to that reported in
the NIRS data and the supplemental State data. Approximately 11% of the participating ground
water PWSs (serving about 5.1 million people) had maximum detections of manganese in raw
water greater than the HRL of 0.30 mg/L. The 99th percentile of concentration and the median
concentration were 9.0 mg/L and 0.09 mg/L, respectively. Surface water PWSs showed
comparable results with approximately 12.8% of survey respondents (serving about to 10.5
million people) having maximum detections of manganese in raw water greater than the HRL of
0.30 mg/L. The 99th percentile of concentration and the median concentration in raw surface
waters were 3.08 mg/L and 0.092 mg/L, respectively.
In finished ground water samples, approximately 3% of survey respondents (serving close
to 1.7 million people) had maximum detections of manganese greater than the HRL of 0.30 mg/L.
The 99th percentile concentration and the median concentration were 0.80 mg/L and 0.021 mg/L,
respectively. For finished surface water samples, approximately 1.5% of survey respondents
(about 1.7 million people) reported maximum detections greater than the HRL of 0.30 mg/L. The
99th percentile concentration and the median concentration in finished surface water samples were
0.64 mg/L and 0.013 mg/L, respectively.
External Review Draft — Manganese—April 2002 4-11
-------
4.5 Conclusion .
i
r
Manganese and its compounds are TRI chemicals. Industrial releases have been recorded
since 1988 in all 50 States. Off-site releases constitute a considerable amount of totalreleases,
with releases to land being the most significant on-site releases.
Low-level manganese occurrence in ambient waters and bed sediments monitored by the
USGS NAWQA program is ubiquitous, with detections approaching 100% of surface water sites
and greater than 62% of ground water sites. Stream bed sediments and aquatic biota tissues show
detections of 100% by sample and by site. Urban basins generally have more surface and ground
water manganese detections greater than the HRL than basins in other land use categories, and
higher median and 99th percentile concentrations. Although manganese detection frequencies are
high in ambient waters, stream bed sediments, and aquatic biota tissue, manganese occurrence at
levels of public health concern is low.
Manganese has been detected in ground water PWS samples collected through the NIRS
study. Occurrence estimates are relatively high with approximately 68% of all samples showing
detections affecting about 55% of the national population served. The 99th percentile
concentration of all samples is 0.63 mg/L. Exceedances of the HRL at 0.30 mg/L affect 2.6% of
the ground water PWS population served, or approximately 2.3 million people nationally.
Additional SDWA data from the States of Alabama, California, Illinois, New Jersey., and
Oregon, including both ground water and surface water PWSs, were examined through
independent analyses and also show substantial levels of manganese occurrence. These data
provide perspective on the NIRS estimates that only include data for ground water systems. The
supplemental State data show ground water systems reported higher manganese detections in 3 of
the 5 States (California, Illinois, and Oregon). If national data for surface water systems were
available, the occurrence and exposure estimates would be substantially greater than from NIRS
alone. ' ,
External Review Draft—Manganese—April 2002
4-12
-------
5.0 EXPOSURE FROM ENVIRONMENTAL MEDIA OTHER THAN WATER
5.1 Food
5.1.1 Concentrations of Manganese in Food
Table 5-1 summarizes mean manganese concentrations in 234 foods analyzed by the Food
and Drag Administration (FDA). Nuts and grains contain the highest manganese concentrations,
with values as high as 40 to 50 mg/kg reported. Fruits, vegetables, fish, poultry, meat, and eggs
tend to have intermediate concentrations. Manganese levels in milk tend to be low, with
concentrations of 10 and 30 micrograms per liter (jJ-g/L) reported for human and cow's milk,
respectively. In contrast, values of 50 to 300 ng/L have been reported for infant formula (Collipp
et al., 1983 as cited in U.S. EPA, 1996a).
Manganese has been detected in the muscle of fresh bluefin tuna (Thunnus ihynnus).
Hellou et al. (1992) as reported in ATSDR (2000), analyzed concentrations in 14 tuna samples
using inductively coupled plasma mass spectrometry. The level of manganese varied from 0.16 to
0.31 micrograms per gram (ng/g) dry weight, with a mean value of 0.22 (ig/g dry weight.
Black tea samples from the United Kingdom (UK) were found to have mean manganese
concentrations of 4.6 mg/L, 40% of which was bioavailable (Powell et al., 1998).
The issue of bioavailability is important to consider when assessing manganese levels in
foods, and is discussed further in the next section. For instance, the actual absorption of
manganese from ingested tea is limited by the presence of polyphenolic compounds (tannins) in
the tea which bind manganese (Freeland-Graves and Llanes, 1994). This explains the low
bioavailabilty of manganese in tea. Likewise, the relatively high levels of manganese in fruits,
nuts, grains, and vegetables, as well as in soy-based infant formula (discussed in Section 5.1.2),
are limited in their bioavailability by the presence-of phytic acids, oxalic acids, and fiber in these
foods (U.S. EPA, 1996a). In addition, high levels of calcium or magnesium ingestion may inhibit
manganese absorption, while persons with diets that are deficient in iron may experience increased
manganese absorption (U.S. EPA, 1996a).
5.1.2 Intake of Manganese From Food
General Population
Manganese is an essential nutrient. It is very unevenly distributed in foods. Although
manganese is rich in tea, whole grains, legumes, and nuts, it is found in negligible amounts in
meats, dairy products, sweets, refined grams, and most fruits. Thus, many individuals who do not
consume whole grains, nuts, certain fruits (pineapple), green leafy vegetables, and tea will
consume a "low manganese" diet - less than 2 mg per day (Davis et al., 1992). In addition,
women tend to consume less food than men; hence their intakes of individual nutrients, including
manganese, are often lower than those of men (Pennington et al., 1989),
External Review Draft—Manganese—April 2002 5-1
-------
The Food and Nutrition Board set an adequate intake level (AT) for manganese at 2.3
nag/day for men and 1.8 mg/day for women (Food and Nutrition Board, 2002; Trumbo et al.,
2001). The current recommendations for infants and children are 0.003 to 0.6 mg/day and 1.2 to
1.9 mg/day, respectively (Food and Nutrition Board, 2002). An adequate intake level is defined
as "a recommended intake value based on observed or experimentally determined approximations
or estimates of nutrient intake by a group (or groups) of healthy people that are assumed to be
adequate - used when an RDA cannot be determined." Some nutritionists feel that this level may
be too low. Freeland-Graves et al. (1987), as cited in U.S. EPA (1996a), have suggested a range
of 3.5 to 7 mg/day for adults based on a review of human studies.
Dietary habits have evolved in recent years to include a larger proportion of meats and
refined foods in conjunction with a lower intake of whole grains (Freeland-Graves, 1994; U.S.
EPA, 1996a). The net result of such dietary changes includes a lower intake of manganese. A
significant number of adult Americans, particularly women, may consume suboptimal amounts of
manganese (ATSDR, 2000; Pennington et al., 1986). On the other hand, it is not known whether
infants may ingest more than the Al for their age group as a result of the high manganese content
of prepared infant foods and formulas.
Table 5-1. Manganese Concentrations in Selected Foods"
TYPE OF FOOD
Nuts and nut products
Grains and grain products
Legumes
Fruits
Fruit juices and drinks ,
Vegetables and vegetable products
Desserts ':
Infant foods
Meat, poultry, fish and eggs
Mixed dishes •
Condiments, fats, and sweeteners
Beverages (including tea)
Soups i
Milk and milk products
RANGE OF MEAN
CONCENTRATIONS
(mg/kg)
18.21-46.83
0.42-40.70
2.24-6.73
0.20-10.38
0.05-11.47
0.42-6.64
0.04-7.98
0.17-4.83
0.10-3.99
0.69-2.98
0.04-1.45
0.00-2.09
0.19-0.65
0.02-0.49
a Adapted from ATSDR (2000) and Pennington et al. (1986).
External Review Draft — Manganese —April 2002
5-2
-------
Based on various surveys, the Food and Nutrition Board (2002) concluded that the
average manganese intake of adults eating western-type and vegetarian diets ranged from 0.7 to
10.9 mg/day (Food and Nutrition Board, 2002), and the median intakes for women and men
ranged from 1.6 to 2.3 mg/day (Food and Nutrition Board, 2002). The total dietary manganese
intake among individuals may vary greatly depending upon dietary habits. Individual intake
estimates for Canadian adult male blue-collar workers (n = 28) and garage mechanics (n = 37), as
determined by analysis of dietary records, ranged from 1.0 to 14 mg/day (Loranger and Zayed,
1995). The mean values in this study for manganese intake by blue-collar workers and mechanics
were 3.7 and 2.9 mg/day, respectively. It should be noted that FDA's Total Diet Study menus
used to measure the levels of several nutritional elements including manganese from 1982 to 1986
in Pennington et al. (1989) reflect "typical", American diets and contain less manganese than the
diets consumed by Canadian males.
The Food and Nutrition Board also set a tolerable upper intake level (UI) for manganese
at 11 mg per day for adults, based on the upper range of manganese intake for adults (see review
by Greger, 1999). An UI is defined as "the highest level of daily nutrient intake that is likely to
pose no risk of adverse health effects for almost all individuals in the general population. As
intake increases above the UL, the risk of adverse effects may increase." For shorter duration,
Davis and Greger (1992) reported that women given daily supplements of 15 mg manganese for
90 days experienced no adverse effects other than a significant increase in lymphocyte manganese-
dependent superoxide dismutase (Greger, 1998,1999; Food and Nutrition Board, 2002).
Based on a conservative range for manganese intake of 2 to 10 mg/day, U.S. EPA (1996a)
estimated a dietary manganese intake of 28.6 to 126 micrograms per kilogram per day (p-g/kg-
day). For children, assuming a manganese intake of 1.28 p-g/calorie (U.S. EPA, 1984; ATSDR,
2000) and a caloric intake of 1,000 calories/day for a 10 kg child, the estimated average daily
intake would be 128 (ig/kg-day. ,
Groups with Potential for High Manganese Intake from Food
Groups with potential for high intake of dietary manganese include vegetarians, heavy tea
drinkers, and infants. Vegetarians may consume a larger proportion of manganese-rich nuts,
grains, and legumes in their diet than the general population (U.S. EPA, 1996a). Manganese
intake by North American vegetarians has been estimated to be as high as 10 mg Mn/day (Gibson,
1994). However, many components of vegetarian diets, including phytates, tannins, oxalates, and
fiber, inhibit manganese uptake from the gastrointestinal tract. Consequently, the bioavailability
of manganese in vegetarian diets is uncertain. Johnson et al. (1991) studied the absorption of
radiolabelled manganese from various plant foods in adult men and women, and reported that
mean fractional absorption values from lettuce and spinach were 5.20 and 3.81%, respectively.
Mean fractional absorption from sunflower seeds was significantly less (1.71%), while that from
wheat was 2.16%. All percent absorption values from plant food were significantly less than
mean values from MnCl2 dissolved in water, which ranged from 7.74 to 10.24%.
External Review Draft — Manganese—April 2002
5-3
-------
Heavy tea drinkers may have a higher; manganese intake than the general population. An
average cup of tea may contain 0.4 to 1.3 mgimanganese (ATSDR, 2000); Consumption of three
cups of tea per day would therefore have the potential to double manganese intake for some
individuals. Again, however, it is likely that the high level of tannins in tea will result in reduced
manganese absorption (Freeland-Graves and Llanes, 1994).
Infants may ingest high levels of manganese from infant formulas or prepared baby foods,
although manganese absorption in infants is influenced by several variables, and the degree to
which absorption levels may be a health concern is unknown. Infant formulas contain 50 to 300
[ig/L manganese (Collipp et al., 1983 as cited in U.S. EPA, 1996a), compared to human milk
which contains 7 to 15 p.g/L manganese (U.S. EPA, 1996a). Assuming an intake of 742
milliliters (mL) of breast milk/day (U.S. EPAJ, 1996a), a breast-fed infant would have an estimated
daily manganese intake of 5.2 to 11.1 ng/day. An infant consuming the same volume of infant
formula would have an estimated daily manganese intake of 37.1 to 223 jig/day. Assuming an
average weight of 6 kg for an infant of age 6 months, the weight-adjusted average daily intake
would range from 0.87 to 1.85 |ig/kg-day for breast-fed infants. The corresponding weight-
adjusted intake for a formula-fed infant would be 6.2 to 37.2 |ig/kg-day. Generally, solid foods
are introduced at the age of 4 months. Once solid foods are introduced, the dietary intake of
manganese increases so substantially that the contribution of Mn intake from milk becomes less
significant.
In assessing infant exposure to manganese, however, one must also consider constituents
of infant formula and of breast milk which may affect manganese bioavailability. For instance,
formula made from soy protein contains high levels of phytic acids and vegetable proteins which
probably decrease the manganese bioavailability. If the formula is also iron-fortified, manganese
bioavailability may be further decreased, although studies on the inhibitory influences of iron have
produced conflicting results (Freeland-Graves^ 1994). Davidsson et al. (1989a) measured
absorption of radiolabelled manganese in adult humans given human milk, cow's milk, or soy
formula and found that fractional manganese absorption from human milk (8.2%) was
significantly higher than absorption from cow?s milk (2.4%) and soy formula (0.7%). Manganese
in infant formula is in the divalent state, the absorption of which cannot be regulated by the
lactoferrin receptors in the gut; breast milk manganese is in the trivalent form bound to lac toferrin,
and its absorption is thus regulated (U.S. EPA, 1996a). Davidsson et al. (1989a) suggested that
the lactoferrin in human milk as well as the higher calcium content in cow's milk contributed to
the difference in absorption. Dorner et al. (1989) observed similar differences in fractional
manganese retention in infants as those observed by Davidsson et al. (1989a) in adults. In the
infant study, a higher percentage of manganese was retained from ingested breast milk (41%) than
from cow's-milk formula (~19%). Therefore, imany factors probably control manganese
absorption from infant formula, and firm conclusions are difficult to make in the absence of more
direct data. Keen etal. (1986) demonstrated that fractional manganese uptake from human
breast milk and cow's milk were relatively high (~80% and ~89 %, respectively), whereas uptake
from soy formula was lowest (~60%) in rat pups.
External Review Draft — Manganese—April 2002
5-4
-------
It should be noted that Davidsson et al. (1989a) performed their studies in adults;
manganese body burden in infants may be additionally influenced by the fact that the biliary
excretion system, which is the primary route of manganese excretion, is not completely developed
in neonates (Lonnerdal, 1994). Studies in rats have further demonstrated that young animals
absorb significantly more manganese in the gut than do mature animals (Lonnerdal et al. 1987).
Also, animal studies have shown that manganese crosses the blood-brain barrier hi neonates at a
rate 4 times higher than that hi adults (Mena, 1974). However, the relevance of these studies to
humans is unknown, and few direct absorption data for manganese in human infants are available.
hi this context, it is noteworthy that Collipp et al. (1983) reported hair manganese levels that
increased significantly from birth (0.19 ug/g) to 6 weeks (0.865 ug/g) and 4 months (0.685 u-g/g)
of age in infants given formula, while infants given breast milk exhibited no significant increase
(0.330 |ig/g at 4 months). This study also reported that the average hair manganese level hi
children exhibiting learning disabilities was significantly increased (0.434 ug/g) compared to those
that exhibited normal learning ability (0.268
5.2 Air
5.2.1 Concentration of Manganese in Air
General Population
Table 5-2 summarizes nationally aggregated data collected between 1953 and 1982 for
manganese concentrations in ambient air of nonurban, urban, and source-dominated locations.
Average manganese concentrations for nonurban areas ranged from a high value of 60 nanograms
per cubic meter (ng/m3) determined in 1953-1957 to a low of 5 ng/m3 in 1982. Average
concentrations for urban areas ranged from 1 10 to 33 ng/m3 over the same period. Average levels
in source-domhiated locations varied widely, ranging from a high reading of 8,300 ng/m3 during
the 1965-1967 measurement period, to concentrations of 130 to 140 ng/m3 hi 1982. Although
differences hi sample collection and analytical methods complicate interpretation, these data
suggest that manganese concentrations hi ambient air decreased over the time period of record
(U.S. EPA, 1 984). This change has been attributed to installation of emissions controls hi the
metals industry (ATSDR, 2000). More recently, U.S. EPA (1990) has proposed an average
annual background concentration of 40 ng/m3 for urban areas, based on data for 24-hour average
concentrations hi 102 cities across the U.S.
Multiple local studies have estimated airborne manganese concentrations. A series of
Canadian studies evaluated total airborne manganese concentrations hi the home and workplace
(Sierra et al., 1995; Zayed et al., 1994, 1996). Table 5-3 summarizes the results of these studies.
Concentrations of manganese were determined by use of personal sampling devices. Mean levels
of manganese measured hi homes ranged from 7 to 12 ng/m3. Mean workplace concentrations
ranged from 12 to 44 ng/m3 for non-automotive workers (primarily office workers) and taxi
drivers. Automotive workers, such as auto mechanics, experienced mean workplace levels
ranging from 250 to 448 ng/m3. Sample sizes for these studies ranged from 9 to 35 individuals.
External Review Draft — Manganese—April 2002
5-5
-------
Table 5-2. Average Concentrations of Manganese in Ambient Air Sampled from
1953-1982a.
SAMPLING
LOCATION/YEAR
Nonurban
Urban
Source-dominated
1 CONCENTRATION (ng/m3)
1953-1957
60
110
No data
1965-1967
12
73
250-8,300
1982
5
33
130-140
•Source: ATSDR (2000) and U.S. EPA (1984). ;
Table 5-3. Manganese Levels in Air of Canadian Urban Locations as Determined by
Personal Exposure Monitoring.
OCCUPATION
Garage worker
Garage worker
Taxi driver
Taxi driver
Auto Mechanic
Auto Mechanic
Nonautomotive
Nonautomotive
Office worker
Taxi driver
LOCATION
Work
Home
Work
Home
Work
Home
Work
Home
Work
Work
DURATION
5 days
2 days
5 days
2 days J
4 weeks
4 weeks
4 weeks
4 weeks
7 days
7 days
N
10
10
10
10
35
35
30
30
23
9
MEAN
(ng/m3)
250
7
24
11
448
12
44
8
12
28
RANGE
(ng/m3)
9-2,067
4-27
6-69
4-22
10-6,673
6-63
11-1,862
5-87
2-44
8-73
REFERENCE
Zayed et al.
(1994)
Sierra et al.
(1995) '
Zayed et al.
(1996)
Automotive fuels in Canada and the U.S. contain the antiknock agent
methylcyclopentadienyl manganese tricarbonyjl (MMT). The allowable level of MMT in Canadian
gasoline is 0.062 grams per gallon (g/gal), which is double the allowable limit of 0.031 g/gal in the
U.S. (Davis, 1998). Combustion of MMT releases manganese to the atmosphere in the form of
manganese oxides, phosphates, and sulfates (see Section 3.2 above), and these compounds may
constitute a significant source of manganese contamination in urban environments. In Canada, a
car exhaust study determined that 4 to 41% of Mn in gasoline is emitted from the tailpipe,
External Review Draft -t- Manganese—April 2002
5-6
-------
depending on the vehicle and driving cycle (Ardeleanu et al., 1999). The fraction not emitted to
the atmosphere appears to remain in the engine (Ardeleanu et al., 1999).
Levels of unbumed MMT in air resulting from emission of residual MMT in vehicular
exhaust or evaporative emissions (e.g., at gas stations) are expected to be low. Although data are
limited, Zayed et al. (1999a) reported concentrations ranging from 0.4 ng/m3 to 12 ng/m3 when
measured in five different microenvironments in Montreal, Canada. The highest average
concentration of MMT in ambient air was measured at gas stations.
Use of MMT in gasoline has resulted in public health concerns related to the potential
health effects of increased manganese exposure. As a result, determination of the extent to which
MMT contributes to environmental levels of manganese (and ultimately to human exposure) has
been an area of active research. Several studies in Montreal, Canada have examined manganese
concentrations in ambient air in relation to motor vehicle traffic (Table 5-4). Loranger et al.
(1994a) found ambient manganese concentrations to be significantly correlated with traffic
density. Areas of intermediate and high traffic densities had ambient.manganese concentrations
above the natural background level in Montreal of 40 ng/m3 (Loranger and Zayed, 1994;
Loranger et al., 1994a).
Loranger et al. (1995) summarized modeling and empirical data relating atmospheric
manganese concentrations to combustion of gasoline containing varying concentrations of MMT
(Table 5-5). Estimated increases predicted by studies listed in the table but conducted prior to
1990 were characterized by Loranger et al. (1995) as being of limited use due to insufficient
information on methodology. Based on an estimated background level of 40 ng/m3 (calculated by
taking the average of data from 102 U.S. cities), U.S. EPA (1990) predicted that the potential
increase in ambient background manganese from the use of MMT would be 0.05P, where P is the
fraction of total manganese in fuel that is emitted in vehicular exhaust.
Canadian studies have addressed the fraction of total manganese concentration in air
associated with particulates of respirable size. Zayed et al. (1996) reported respirable manganese
(MnR) and total manganese (MnT) concentrations determined by personal exposure monitoring
of taxi drivers and office workers. Mean concentrations of MnR were 10 and 15 ng/m3 for office
workers and taxi drivers, respectively. Mean concentrations of MnT were 12 and 28 ng/m3 for
the same respective groups. Loranger and Zayed (1997a) measured concentrations of MnR and
MnT at two sites in Montreal with different vehicle traffic densities. MnR and MnT
concentrations adjacent to a heavily traveled (> 100,000 vehicles/day) road were 24 and 50 ng/m3,
respectively. Values for MnR and MnT at a site with lower traffic density (10,000 to 15,000
vehicles/day) were 15 and 27 ng/m3, respectively. Zayed et al. (1999a) measured mean
concentrations of respirable manganese ranging from 18 to 53 ng/rn3 in five microenvironments in
Montreal. The overall mean concentrations of respirable and total manganese were 36 ± 7 ng/m
and 103 ± 32 ng/m3, respectively. These data indicate that approximately 35 to 90% of total
manganese in urban air is respirable.
External Review Draft — Manganese—April 2002
5-7
-------
Table 5-4. Ambient Air Concentrations of Manganese in Relation to Traffic Density,
Montreal, Canada 1981-1994.
TRAFFIC DENSITY
(vehicles/day)
< 15,000
> 15,000
4,900
75,000
< 15,000
< 30,000
> 100,000
117,585
117,585
Mn
(ng/m3)
<40
(50% of samples)
;>40
(50% of samples)
26
36
20
i 50
i 60
|54
29-37
REFERENCE
Loranger et al. (1994a)
Loranger et al. (1994a)
Loranger et al. (1994b)
Loranger et al. (1994b)
Loranger and Zayed (1994)
Loranger and Zayed (1994)
Loranger and Zayed (1994)
Loranger et al. (1995a)
Loranger et al. (1995a)
Source: Zayed et al. (1999b)
External Review Draft — Manganese—April 2002
5-8
-------
Table 5-5. Estimated Atmospheric Mn Concentration in Relation to the Combustion of
MMT in Gasoline.
Mn concentration in gasoline
mg/lL
132.0
33.0
33.0
33.0
33.0
33.0
33.0
26.4
18.0
17.0
16.5
16.5
8.3
8.3
8.3
10.0
g/gal
0.5
0.125
0.125
0.125
0.125
0.125
0.125
0.100
0.068
0.064
0.063
0.063
0.031
0.031
0.031
0.038
Estimated
concentration from
MMT source rig/m3
~
335
—
-
~
-
—
20-200*
25*
20-200 .
70-140
20
17
150"
10-20
< 1-3"
2-29°
Ambient air
concentration from all
sources ng/m3
200-800
1,200-1,500°
2-250"
20-3,400°
70-720d
730-10,000°
120-3,630f
< l,000h
<500"
-
90-3,800*
, "
• -
55'
50-60"
34
-'
Reference
(Mena, 1974)
(Piver, 1974)
(Moran, 1975)
(U.S. EPA, 1975)
(U.S. EPA, 1975)
(U.S. EPA, 1975)
(U.S. EPA, 1975)
(TerHaaletal., 1975)
(Cooper, 1984)
(Abbott, 1987)
(HWC, 1978)
(Pierson et al., 1978)
(Ethyl Corp., 1990)
(U.S. EPA, 1990)
(U.S. EPA, 1991b)
(Lorangeretal., 1995)
Source: Table adapted from Loranger et al. (1995).
-• Annual average.
b 24-hour average.
c EPA model: 24-hour average; beside highway (1-500 m), 20% emission at the tailpipe.
11 Ethyl corp. model: 24-hour average, beside highway (1-500 m), 20% emission at the tailpipe.
' EPA model: hourly peak, beside highway (1-500 m), 20% emission at the tailpipe.
fEthyl corp. model: hourly peak, beside highway (1-500 m), 20% emission at the tailpipe.
8 Median value = 0.05, near roadway.
h Median value.
' Beside highways.
] Maximum monthly average.
k 30% emission at the tailpipe, mid-size car (20 mi/US gal).
1 Urban annual average background concentration = 0.04 jig m'3.
" SCREAM model, background concentration = 0.04 ug m"3.
" CALINE4 and ISCLT models: > 250 m beside expressway.
° CAUNE4 model: < 250 m beside expressway.
— = no data
External Review Draft—Manganese—April 2002
5-9
-------
Personal exposures (expressed as concentration in air) to airborne manganese were
measured before and after the introduction of MMT into 20% of the diesel fuel used in London
(Pfeifer et al., 1999). Concentrations of manganese encountered by office workers and taxi
drivers (10 subjects/occupation) were measured during 2-week periods in both 1995 (before
MMT introduction) and 1996 (after MMT introduction). Manganese concentrations reported for
office workers ranged from 2 to 239 ng/m3 and from'4 to 147 ng/m3 in 1995 and 1996,
respectively. Taxi drivers experienced exposure to concentrations of 4 to 44 ng/m3 and 9 to 36
ng/m3 in 1995 and 1996, respectively. Thus, neither occupational group experienced apparent
exposure to increased Mn after the introduction of MMT to gasoline. The greater exposure of
office workers to airborne manganese when compared to taxi drivers was an unexpected result.
The higher intake by office workers was attributed to manganese enrichment (approximately 10-
fold greater than in the general environment) of the particulate matter in subway tunnels. When
combined with elevated levels of particulates, manganese concentrations were estimated to be two
orders of magnitude higher in the underground microenvironment. While these results differed
from previous studies where, regardless of MMT use, taxi driver exposures to airborne
manganese were higher than office workers' exposures (Lynam et al., 1994; Zayed et al., 1994;
Riveros-Rosas et al., 1997), they are consistent with findings cited in Lynam et al. (1999) which
indicated that subway system commuters in Toronto, Canada had higher manganese exposures
than non-subway users.
The Particle Total Exposure Assessment Methodology (PTEAM) study provided
information on levels of airborne manganese in Riverside, CA [findings summarized in Davis
(1998)]. This study was conducted over a 7-w
-------
Table 5-6. Mean Manganese Exposures from 3-day Indoor, Outdoor and Personal Air
Samples.
Sample
Personal
Indoor Air
Outdoor Air
PMw-associated Mn (ng/m3)*
35.8
8.0
17.5
PMij-associated Mn (ng/m3)
13.1
5.5
9.7
Source: Pellizzari et al.(1999).
* Estimated from Figure 4 in Pellizarri et al. (1999).
Populations with Potential for High Exposure
Workers in certain occupations may be exposed to significantly higher manganese
concentrations than the general population. Historically, the production of manganese fumes or
manganese-containing dusts in the ferromanganese, iron and steel, dry cell battery manufacturing,
welding, and mining industries may result in workplace concentrations as much as 10,000-fold
higher than average ambient levels in air (ATSDR, 2000). ATSDR (2000) has noted that data for
current occupational levels of manganese exposure are not available. However, to be in
compliance with Occupational Safety and Health Administration (OSHA) regulations, manganese
levels in the workplace should not exceed the OSHA time-weighted average Permissible
Exposure Limit (PEL) of 1 mg/m3.
5.2.2 Intake of Manganese in Air
General Population
U.S. EPA (1990) has calculated an average annual atmospheric manganese background
concentration of 40 ng/m3 for urban areas, based on data for 24-hour average concentrations in
102 cities across the U.S. (U.S. EPA, 1990). Assuming an intake of 15.2 cubic meters per day
(mVdav) (U.S. EPA, 1996d), the average estimated daily intake for a 70 kg adult would be 8.7
ng/kg-day. The corresponding average daily intake for a 10 kg child would be 35 ng/kg-day if an
inhalation rate of 8.7 mVday (U.S. EPA, 1996d) is assumed. Alternatively, assuming a range of
ambient concentrations from 2 to 220 ng/m3 for rural and urban populations, and an inhalation
rate of 15.2 m3/day, the estimated daily intake range for a 70 kg adult would be 0.43 to 47.8
ng/kg-day. The daily intake for a 10 kg child would range from 1.74 to 122 ng/kg-day. These
calculated adult intakes are in general agreement with intakes calculated by others. Loranger and
Zayed (1997a) predicted a total manganese dose for adults of 1 to 50 ng/kg-day predicted for
two urban sites in Montreal, Canada, using Monte Carlo simulation. Zayed et al. (1999a)
calculated intakes of 5 to 15 ng/kg-day based on measurements of respirable manganese
concentrations at five sites in Montreal. .
External Review Draft — Manganese—April 2002
5-11
-------
Populations with Potential for High Exposure
Historically, workers in occupational settings such as manganese mining or
ferromanganese smelting have experienced the potential for high levels of manganese exposure.
Published estimates of current occupational exposure levels were not available in the materials
reviewed for this document. However, assuming a maximal legal concentration of 1 mg/m3 and
inhalation of 10 m3 of air over the course of a! work day, adults exposed to manganese in some
occupational settings may have a daily intake as high as 143,000 ng/kg-day (ATSDR, 2000).
5.3 Soil |
53.1 Concentration of Manganese in Soil
Manganese constitutes approximately 10.1% of the earth's crust, and is a naturally
occurring component of nearly all soils (ATSDR, 2000). Natural levels of manganese range from
less than 2 to 7,000 mg/kg, with a geometric mean concentration of 330 mg/kg (Shacklette and
Boerngen, 1984). The estimated arithmetic mean concentration is 550 mg/kg. Accumulation of
manganese occurs in the subsoil rather than oil the soil surface (ATSDR, 2000). An estimated
60-90% of soil manganese is associated with the sand fraction (WHO, 1981, as cited in ATSDR,
2000). :
5.3.2 Intake of Manganese in Soil
General Population
No published reports quantify exposure to manganese associated with soil ingestion.
Assuming a concentration range of < 2 to 7,000 mg/kg soil and average ingestion of 50 mg of
soil/day, the average manganese intake of a 70-kg adult would be 0.0014 to 5 (ig/kg-day. The
corresponding intake for a 10-kg child consuming 100 mg of soil/day would be 0.02 to 70 ng/kg-
day.
Populations with Potential for High Exposure
No highly exposed populations were identified with respect to soil intake.
5.4 Other Media
No published reports identify other sources of manganese exposure.
l
5.5 Summary of Exposure to Manganese in Media Other Than Water
Table 5-7 summarizes information on exposure to manganese in media other than water.
Inspection of data in this table reveals that ingestion of food contributes a major proportion of
manganese exposure. This observation is consistent with the findings of Loranger and Zayed
External Review Draft'— Manganese—April 2002
5-12
-------
(1995,1997b), who estimated that food contributed 95 to 99% of the multimedia dose of
manganese in Canadian studies. The contribution of soil as a source of manganese was not
evaluated in the 1995 study (Loranger and Zayed, 1995). However, as evident from Table 5-7,
soil ingestion has the potential to contribute substantially to intake in areas with naturally high or
anthropogenically enriched concentrations of soil manganese.
EPA has derived an oral reference dose (RfD) for manganese of 0.14 mg/kg-day and an
inhalation reference concentration (RfC) of 5 x 10's mg/m3 (see Section 8.1). These values can be
converted to daily doses (assuming a 70 kg adult inhaling 15.2 m3/day of air) of 10 mg and 7.6 x
10"4 mg manganese, respectively. Thus, the level of safe exposure determined for the inhalation
route is five orders of magnitude less than that determined for the oral route, reflecting the much
greater toxicity observed for inhaled versus ingested manganese. For exposure to manganese
from drinking water, EPA recommends applying an additional modifying factor of three to the
above RED, yielding 0.047 mg/kg-day (U.S. EPA, 1996a). This recommendation derives from
concern raised by the Kondakis study (1989) (see Sections 7.1.3 and 8.1) about the potential for
higher absorption of manganese from water, and also from consideration of potentially higher
absorption in fasting individuals and neonates, the latter of which may have higher absorption
rates and lower excretion rates of manganese than mature individuals (U.S. EPA, 1996a).
For drinking water, a National Secondary Drinking Water Regulation (or secondary
Maximum Contaminant Levels, s-MCL) for manganese also exists (0.05 mg/L) to prevent clothes
staining and taste problems. Secondary standards are non-enforceable guidelines regulating
contaminants that may cause aesthetic effects (such as color, taste or odor) or cosmetic effects
(such as skin or tooth discoloration) in drinking water. EPA recommends s-MCLs to water
systems but does not require systems to comply.
External Review Draft—Manganese — April 2002
5-13
-------
Table 5-7. Summary of Human Exposure to Manganese in Media Other than Water
(General Population).
PARAMETER
Concentration in
Medium
Estimated Average
Daily Intake (jig/kg-
day)
EXPOSURE MEDIUM
Food
Adult
Child
0.04-^7 mg/kg ;
1
28.6-126
0.87-37.2 J
(infant)
128 (child)
Air
Adult
Child
40 ng/m3
0.0087
0.034
Soil
Adult
Child
< 2-7,000 nig/kg
0.0014-5.0
0.02-70
External Review Draft \— Manganese—April 2002
5-14
-------
6.0 TOXICOKINETICS
The absorption, distribution, metabolism and excretion of manganese in the body are
reviewed, discussed, and summarized in Greger (1999), U.S. EPA (1984), Kies (1987), U.S. EPA
(1993), and ATSDR (2000). Age, chemical species, dose, route of exposure, and dietary
conditions all affect manganese absorption and retention (LSnnerdal et al., 1987). Uptake of
dietary manganese appears to be controlled by several dose-dependent processes: biliary
excretion, intestinal absorption, and intestinal elimination. Manganese absorbed in the divalent
form from the gut via the portal blood is complexed with plasma proteins that are efficiently
removed by the liver. Absorption of manganese via inhalation, intratracheal instillation, or
intravenus infusions bypasses the control processes by the gastrointestinal tract. After absorption
to the blood system by these alternate routes, manganese is apparently oxidized, and the trivalerit
manganese binds to transferrin. Transferrin-bound trivalent manganese is not as readily removed
by the liver, as are protein complexes with divalent manganese. Thus, manganese delivered by
these other routes would be available for uptake into tissues for a longer period of time than the
orally administered manganese, leading to quantitative differences in tissue uptake (Andersen et
al., 1999).
6.1 Absorption
Human Studies
The following sections discuss absorption of manganese following oral exposure only.
Recent studies show that significant differences exist in the amounts of manganese that are
absorbed across different exposure routes, with inhaled manganese being absorbed more rapidly
and to a greater extent than ingested manganese ( Roels et al., 1997; Tjalve et al. 1996).
Past manganese intake and iron, phosphorus, and calcium intake affect manganese
absorption in humans. Further, phytate, fiber, and polyphenols (tannins) in vegetable diet tend to
decrease manganese absorption (Greger, 1999; Greger and Snedeker, 1980). Manganese
speciation and the route of exposure also affect its absorption (Andersen et al., 1999; Tjalve et al.,
1996).
Mena et al. (1969) investigated gastrointestinal absorption of manganese in 11 healthy,
fasted human subjects. The subjects received 100 (iCi of ^MnClz with 0.200 mg stable 55MnCl2
(0.087 mg Mn) as a carrier. After 2 weeks of daily whole body counts, the absorption of 54Mn
was calculated to average approximately 3%. Comparable absorption values were found for
healthy manganese miners and ex-miners with chronic manganese poisoning. However,
enterohepatic circulation was not taken into account in this study. These values may therefore
underestimate absorption (U.S. EPA, 1993).
Thomson et al. (1971) reported a higher absorption rate of 54MnCl2 in segments of
jejunum and duodenum using a double-lumen tube. The mean absorption rate in eight subjects
was 27 ±3%. .
External Review Draft — Manganese—April 2002 6-1
-------
Schwartz et al. (1986) studied the absorption and retention of manganese over a 7-week
period in seven healthy male volunteers aged 22—32 years. Volunteers consumed 3,100-4,400
kcal/day which provided levels of manganese ranging from 12.0 to 17.7 mg Mn/day. The authors
noted that this level of manganese intake was higher than the 2 to 5 mg/day reported as being safe
and adequate by the Food and Nutrition Board of the National Research Council (NRC, 1989).
Assuming an adult body weight of 70 kg, this intake corresponds to 0.17 to 0.25 mg/kg-day.
During weeks 2 to 4, manganese absorption was -2.0 ± 4.9% of the intake. During weeks 5 to 7,
the reported absorption was 7.6 ± 6.3%. Despite the high level of intake, net retention of
manganese was not observed in these individuals. Fecal loss accounted for nearly all of the
ingested manganese, and in some cases exceeded the intake. A portion of this loss likely
represents biliary secretion of previously absorbed manganese.
Sandstrom et al. (1986) administered 450 mL of infant formula containing 0.050 mg Mn/L
to eight healthy subjects, aged 20 to 38 years. 1 The average absorption for seven of the subjects
was 8.4 ± 4.7%. The eighth subject was diagnosed with iron deficiency anemia, and absorbed
45.5%. Six additional subjects received 2.5 nig of manganese (as sulfate) in a multi-element
preparation. The mean absorption for the second group of subjects was 8.9 ± 3.2%.
• i
Davidsson et al. (1989b) studied whole-body retention of 54Mn in adult humans after
intake of radiolabeled infant formula. These authors observed reproducible retention figures at
day 10, after repeated administrations of the labeled formula to six subjects. Absorption ranged
from 0.8—16%. This range corresponds to a 20-fold difference between the highest and lowest
values. The mean value was 5.9±4.8%. Retention at day 10 ranged from 0.6-9.2%, with a mean
value of 2.9±1.8% when measured in 14 healthy individuals. These results suggest substantial
variation in absorption between individuals. ;
In addition, Davidsson et al. (1989a) studied manganese absorption from human milk,
cow's milk, and infant formulas inhuman adults using extrinsic labeling of the foods with MMn or
52Mn and measurements of whole-body retention. The fractional manganese absorption from
human milk (8.2%±2.9%) was significantly different when compared with cow's milk
(2.4%±1.7%) or soy formula (0.7%±0.2%). The total amount of absorbed manganese, however,
was significantly higher from the cow's milk formula as compared with human milk.
Several studies have reported a greater retention of manganese in the neonate than in
adults. In a study of the nutritional requirements for manganese in pre-term infants, Zlotkin and
Buchanan (1986) showed that 99% of the manganese given intravenously for 6 days was retained.
Mena (1969) observed that healthy adults absorb 3% of ingested manganese. Lonnerdal et al.
(1987) showed that manganese uptake from brush border membranes was higher in 14 day-old
rats than in 18 day-old rats. Although Rehnberg et al. (1985) found that younger animals had a
slower distal intestinal transit time than older animals ( potentially contributing to a higher
proportional uptake), Bell and Lonnerdal (1989} showed that the uptake rate was similar in pre-
and post-weanling animals suggesting that age-dependent differences in manganese retention were
not due to immature intestinal transport mechanisms.
External Review Draft \— Manganese—April 2002
6-2
-------
Domer et al. (1989) studied retention of manganese in breast-fed infants compared to pre-
term (-34-36 weeks gestational age) or full-term (2-17 weeks postgestational age) infants fed
cow's milk formulas. This study is unique in that it analyzed potential differences in infant
development on the intake and retention of manganese from different dietary sources. The
authors observed that full-term breast-fed infants retain approximately 41% of ingested
manganese from breast milk (containing 6.2 ug Mn/L). Manganese intake in the formula-fed
infants (14.2 ug/kg, full-term and 15.0 ug/kg, pre-term) was high relative to that of breast-fed
infants (1.06 ug/kg). Formula-fed infants also retained a higher absolute amount of manganese
from their diet compared to breast-fed infants (0.06,2.8, and 0.43 ug/kg retained in pre-term
formula-fed, full-term formula-fed, and breast-fed, respectively). These data indicate that the
percentage of manganese retained between the different food sources is not comparable; a higher
percentage of ingested manganese from breast milk is retained by the infant. Nevertheless,
formula-fed babies retain a larger total amount of manganese, due to the greater amount of
manganese present in the formula (77-99 ug/L). The data also indicate that pre-term infants had
an active excretory capacity for manganese obtained from formula, as compared to full-term
infants.
Because human breast milk contains low levels of manganese (4-10 ug/L; Arnaud and
Favier, 1995; Collipp et al. 1983; Dorner et al. 1989), it is suggested that the neonates' propensity
to retain greater amounts of manganese was an adaptive mechanism to insure that sufficient
amounts were available to the developing animal. Regardless of the mechanism (e.g., increased
uptake and/or decreased elimination), results from human and animal studies suggest increased
manganese retention hi the neohate. Neurological development in the rat is incomplete-at birth,
suggesting that there may be differential susceptibility to excess levels of manganese during this
critical developmental period. Although much of the nervous system is complete at birth in
humans, there is evidence that some discrete neurological functions undergo further development
after birth. The developmental stage in humans that is exactly comparable to the pre-weanling
age in rats is unclear. Although results from animal data suggest that elimination rates reach
adult levels by the age of weaning, the comparable period in human development at which
manganese uptake and elimination reaches that of an adult is unknown.
Factors that Affect Absorption in Humans
Bioavailability of ingested manganese is an important issue hi assessing the health hazard
of manganese. Multiple factors have been reported to affect the absorption of manganese by
humans, including chemical form, age, dose, route of exposure, and presence or deficiency of
other dietary components (Greger, 1999; Greger and Snedeker, 1980). Thomson et al. (1971)
and Gibbons et al. (1976) reported that the divalent form of manganese is absorbed most
efficiently. However, the efficiency of absorption also varies for different manganese salts. In this
regard, Bales et al. (1987) reported that manganese chloride was more efficiently absorbed than
the sulfate or acetate salts. ,
Presence of other dietary components may influence the absorption of manganese.
Calcium, for example, may inhibit the absorption of manganese. McDermott and Kies (1987)
External Review Draft — Manganese— April 2002 6-3
-------
suggested that this inhibition results from the influence of calcium on GI tract pH. Manganese is
more readily absorbed as the Mn(n) form. As the pH rises, conversion to the less absorbable
Mn(UT) and Mn(IV) forms is favored, and uptake is decreased. Alternatively, calcium and
manganese may compete for common absorption sites. The extent to which calcium effects on
absorption influence net manganese balance is uncertain. However, Spencer et al. (1979) did not
observe any significant effect of dietary calcium levels (from 200-800 nag/day) on manganese
balance in healthy males.
A strong association between dietary iron and manganese uptake has been noted in several
human studies. Thomson et al. (1971) observed that iron deficiency increased manganese
absorption. Davis and Greger (1992) reported that women consuming increased levels of non-
heme iron experienced decreased levels of serbm and urinary manganese. Finley et al. (1994)
observed that serum sodium ferritin concentration was negatively associated with manganese
absorption in young women consuming a manganese-adequate diet.
Finley (1999) demonstrated that iron status (assessed as serum concentrations of sodium
ferritin) may also affect manganese absorption and retention. Absorption (determined by
regression of whole body MMn counts) was assessed in women aged 20 to 45 years who were
categorized as having high (upper 10% of normal range, mean values 68 to 69 jig/L) or low
(lower 10% of normal range, mean values 8.7 to 8.9 ng/L) serum ferritin levels. Absorption was
determined under conditions of high (9.5 mg Mn/day) or low (0.7 mg Mn/day) dietary manganese
intake. Within a diet group, individuals with low ferritin absorbed 3- to 5-fold more manganese
(as a percentage of dose) than individuals with high ferritin. Manganese absorption was greatest
in women with low serum ferritin concentrations consuming the low manganese diet. The level
of dietary manganese had no significant effect on absorption in women with high ferritin
concentrations.
Phytate, a component of plant protein, may also interfere with manganese absorption.
Davies and Nightingale (1975) observed a decrease in manganese retention in the presence of
phytate. This result was attributed to the formation of a stable complex between manganese and
phytate in the intestinal tract. Bales et al. (1987) reported that cellulose, pectin, and phytate
reduced the plasma uptake of manganese in human subjects. These data suggest that the presence
of these components contributes to the decreased bioavailability of manganese from vegetarian
diets. However, Schwartz et al. (1986) found ho significant correlation between phytate intake
and manganese absorption in healthy males.
Ruoff (1995) conducted a literature review to determine the relative bioavailability of
manganese from water versus food. The calculated ratio following evaluation of a wide variety of
exposure scenarios in non-fasted subjects was 1.4. However, the difference in absorption
between the two media was not statistically significant. The ratio for fasted subjects was 2..0,
indicating that the absorption from drinking wkter is twice that from foods when the water is
consumed in the absence of partially digested foods in the gastrointestinal tract. A study that
directly measured the absorption of radiolabelled manganese from various manganese-rich plant
foods given to adult men and women after an overnight fast reported a significantly greater
External Review Draft — Manganese —April 2002
6-4
-------
percent absorption of MnCl2 from water compared to manganese absorption from lettuce,
spinach, sunflower seeds, or wheat (Johnson et al., 1991). In addition, different diets may have
different levels of constituents that affect manganese absorption. The greater levels of phytates,
tannins, oxalates, and fiber in vegetarian diets, for instance, are expected to have an inhibitory
effect on manganese uptake from the gastrointestinal tract. Johnson et al. (1991) reported mean
percent absorption values from lettuce and spinach of 5.20 and 3.81%, respectively, and from
sunflower seeds and wheat of 1.71 and 2.16%, respectively. Mean percent absorption values
from MnCl2 dissolved in water only (controls) ranged from 7.74 to 10.24%.
Animal Studies
There are studies using ^Mn-labeled manganese to estimate absorption by animals.
However, these studies measured the apparent absorption, not true absorption., because feeding
radioactive isotopes of manganese does not eliminate the problem that absorbed manganese is
very rapidly excreted through bile into the feces (Malecki et al., 1996). Thus, it is impossible to
separate non-absorbed manganese from secreted manganese without elaborate study designs.
When investigators used elaborate methodology in which ^Mri bound to albumin was injected
intraportally, true manganese absorption was calculated to be 8.2%, and 37% of the absorbed
manganese was excreted into the gut (Davis et al., 1993).
Greenberg et al. (1943) administered a single oral dose containing 0.1 mg of ^Mn-labeled
manganese (as chloride) to rats, and estimated that 3-4% was absorbed from the intestine.
Pollack et al. (1965) administered a single oral dose of MMn as chloride with 5 umol (0.27 mg
Mn) stable carrier to fasted rats and reported 2.5-3.5% absorption 6 hours after administration.
In separate studies, Rabar (1976) and Kostial et al. (1978) administered a single oral dose of ^Mn
as chloride, carrier free, to post-weaning non-fasted rats and reported 0.05% absorption 6 days
after administration. This low absorption value may reflect either loss of absorbed manganese
through fecal excretion, or the fact that the rats were not fasted (U.S. EPA, 1984).
Cikrt and Vostal (1969) showed that manganese is likely to be absorbed from both the
small and large intestine in rats. Factors reported to influence manganese absorption in animals
include dose, chemical form, and age. With respect to dose, Garcia-Aranda et al. (1983) studied
the intestinal uptake of manganese in adult rats and concluded that saturation of the absorptive
process occurred at higher levels of intake. Keen et al. (1986) observed that when suckling rats
were fed 0.5 mL of infant formula containing 5 or 25 mg Mn/mL, retention of manganese
decreased at the higher concentration.
Tissue levels of manganese may be influenced by the form of manganese administered in
the diet. Komura and Sakamoto (1991) administered manganese in soluble (manganese acetate or
manganese chloride) and relatively insoluble (manganese dioxide or manganese carbonate) forms
to male ddY mice. Weight gain was reduced in animals receiving the more soluble forms.
Manganese levels in the liver and kidney appeared to be higher in animals fed manganese acetate
or manganese carbonate. The statistical significance of these apparent differences was not
determined.
External Review Draft — Manganese — April 2002
6-5
-------
Keen et al. (1986) demonstrated a strong effect of age on intestinal manganese uptake
and retention. Sprague-Dawley rat pups were [fasted overnight and then intubated with 0.5 mL of
human milk containing 0.005 mg 54Mn/mL. ijdanganese retention was highest (^ 80%) in pups
less than 15 days old. In older pups (16-19 days old), the average retention was 40%. Keen et
al. (1986) also administered infant formula to rat pups. Soy formula typically contains a much
higher level of Mn than does human milk. The amount of manganese retained in 14-day old rat
pups was 25 times higher in animals given soy formula when compared with pups receiving
human milk.
Chan et al. (1987) demonstrated that developmental stage has a significant influence on
the absorption of manganese. Manganese absorption decreased in rat pups from age 9 days to 20
days. The observed decrease in manganese absorption was correlated with a switch in the site of
maximal absorption. The duodenum was more active in manganese uptake in younger rats, while
the jejunum became more important as the animals matured.
Little is known about the factors that determine the bioavailability of ingested manganese
in animals. Chan et al. (1982,1987) reported jdifferences in the concentration and chemical form
of manganese found in different miUc sources.; Human milk contained only 0.008 ± 0.003 mg
Mn/L, while bovine milk, infant formula and rat milk contained 0.030 ± 0.005, 0.073 ± 0.004, and
0.148 ± 0.018 mg Mn/L, respectively. However, absorption of manganese by suckling rats from
these four types of milk was comparable, suggesting that total concentration may not always be a
reliable indicator of bioavailable manganese. :Chan et al. (1982) determined that the chemical
form of manganese in infant formula is very different from that in human or cows' milk. Human
and cow's nrilk contain two and three manganese-binding proteins, respectively. All manganese
in milk from these sources is protein bound, while the manganese in infant formulas is in the form
,of soluble salts. The degree to which the association of manganese with protein influences
absorption is unknown, but is likely to be important.
i
Lonnerdal et al. (1987) reported that age, manganese intake and dietary factors affect
manganese absorption and retention in rats. Retention is very high during the neonatal period and
•decreases considerably with age. Decreased absorption with age apparently results from a
combination of decreased intestinal absorption and increased excretion in the bile. In young rat
pups, the bioavailability of manganese from various milk sources varied, with greater absorption
occurring from human and cow's milk formula than from soy formula. These differences were
less pronounced in older pups.
Several studies have explored the interrelationship among manganese, cobalt, and iron
uptake. Thomson et al. (1971) reported that iron and cobalt compete with manganese for the
same absorption sites. Competition was proposed to occur during uptake from the lumen into
mucosal cells and in the transfer from mucosA into other compartments. Rehnberg et al. (1982)
administered dietary Mn3O4 (450,1,150, or 4;000 mg/kg Mn) to young rats. These authors
amended the basal diets with varying levels of iron, and demonstrated that iron deficiency
promoted the intestinal absorption of manganese. Conversely, manganese absorption was
inhibited by large amounts of dietary iron. GJruden (1984) demonstrated that 3-week-old rat pups
External Review Draft — Manganese —April 2002
6-6
-------
given a high concentration of iron (0.103 mg Fe/L) in cow's milk absorbed 50% less manganese
than pups receiving the control milk (0,005 mg Fe/mL). This difference was not observed in rats
tested at 8,11,14, or 17 days of age, suggesting that the inhibition of manganese absorption by
iron has a rapid onset during the third week of life.
6.2 Distribution
Human Studies
Manganese is a normal component of human tissues and fluids. Information about the
distribution of manganese in humans is generally derived from post-mortem analyses of various
organs and tissues. The patterns observed in these analyses reflect the body and organ burden of
a lifetime intake of manganese. Cotzias (1958) and WHO (1981) reported a total of 12-20 mg
manganese in a normal 70 kg man. Sumino et al. (1975) reported an average of 8 mg among 15
male and 15 female cadavers with an average weight of 55 kg.
The highest concentrations of manganese in the body of persons without excessive
exposure are found in the liver, kidney, pancreas, and adrenal glands. Intermediate concentrations
occur in the brain, heart and lungs (Table 6-1) (ATSDR, 2000). The lowest concentrations of
manganese are observed in bone and fat. Some data suggest that tissues rich in mitochondria (for
example, liver, kidney, and pancreas) contain higher levels of manganese (Kato, 1963; Maynard
and Cotzias, 1955).
Manganese levels have been determined in human serum and blood. Serum concentrations
in healthy male and female subjects in Wisconsin were 1.06 |ig/L and 0.86 \ig/L, respectively
(Greger et al., 1990; Davis and Greger, 1992). Blood and serum levels of manganese in healthy
subjects living in the Lombardy region of Italy were 8.8 ± 0.2 u-g/L and 0.6 ±0.014 |ig/L,
respectively (Minoia et al., 1990).
A variety of factors have been reported to influence manganese levels in blood and blood
fractions. Hagenfeldt et al. (1973) found variations in plasma manganese concentrations in
women and suggested that the variation may be due to hormonal changes. Horiuchi et al. (1967)
and Zhernakova (1967) found no difference in the concentration of manganese, in the blood of
men and women. Slight seasonal (lower during summer and autumn) and diurnal (lower during
the night) variations in blood manganese concentrations have also been reported (U.S. EPA,
1984).
Three studies have addressed manganese distribution within human organs. Perry et al.
(1973) investigated manganese concentrations in different sections of the liver and found little
variation. Larsen et al. (1979) and Smeyers-Verbeke et al. (1976) studied the regional
distribution of manganese in the brain and reported the highest concentrations in the basal ganglia.
External Review Draft — Manganese—April 2002
6-7
-------
Table 6-1. Normal Manganese Levels in Human and Animal Tissues.
Tissue
Liver
Pancreas
Adrenals
Kidney
Brain
Lung
Heart
Testes
Ovary
Muscle
Spleen
Fat
Bone (rib)
Pituitary
Tissue concentrations (\ig Mn/g wet weight)
i
Humans
A
1.68
1.21
0.20
0.93
0.34
0.34
0.23
0.19
0.19
0.09
0.22
~
—
—
B
il.2
0.77
iO.69
iO.56
0.30*
0.22
0.21
0.20
;0.19
0.09
iO.08
0.07
0.06
~
Rats
C
2.6-2.9
—
2.9
0.9-1.0
0.4
—
~
0.4
~
—
0.3
~
--
0.5
Rabbits
D
2.1
1.6
0.67
1.2
0.36
0.01
0.28
0.36
0.60
0.13
0.22
—
—
2.4
Adapted from ATSDR (2000) ;
A Tipton and Cook (1963) ;
B Smnino et al. (1975)
C Rehnbergetal.(1982) " |
D Fore and Morton (1952) '
Average of cerebrum and cerebellum !
— No data :
Studies by Schroeder et al. (1966) and Widdowson et al. (1972) indicate that placental
transfer of manganese occurs in humans. Manganese levels in fetal and newborn tissues were
reported to be similar to adult levels, with the exception of higher concentrations observed in fetal
bone.
Animal Studies
Knowledge of manganese distribution patterns in animals was initially derived from
parenteral exposure studies which facilitated the use of radioactive manganese as a tracer. The
External Review Draft —Manganese—April 2002 • 6-8
-------
distributions of parentally (injected) and orally administered manganese are very different.
Cellular uptake of manganese is affected by the way in which manganese is transported in the
plasma. Injected manganese (and probably inhaled manganese as well), which is transported by
transferrin, is more apt to accumulate in the brain and cause toxicity than orally administered
manganese, which is transported from the gut to the liver by albumin (Andersen et al., 1999;
Davis et al., 1993). Davis et al. (1993) demonstrated that the distribution pattern of albumin-
bound, but not transferrin-bound, intraportally-injected manganese was similar to that of orally-
administered manganese.
Kato (1963) and Maynard and Cotzias (1955) suggested that mitochondria-rich tissues
such as liver, kidney, and pancreas contain higher levels of manganese. Distribution studies in
mice, rats, and monkeys have subsequently identified liver, kidney, and endocrine glands as
primary sites of manganese accumulation following parenteral exposure. Kato (1963), for
example, investigated distribution in.mice using radiolabeled manganese. High levels of
radioactive manganese were found in the liver, kidneys, and endocrine glands, with lesser amounts
detected in brain and bone. Dasturetal. (1969) administered an intraperitoneal dose of
radioactive manganese to rats, and subsequently found the highest concentrations of labeled
manganese in suprarenal, pituitary, liver, and kidney tissue. In general, these results are in
agreement with the patterns of manganese distribution observed in human tissues.
Dastur et al. (1971) observed a similar pattern of distribution in monkeys exposed to
manganese by intraperitoneal injection. The highest concentrations of manganese were found in
the liver, kidney and endocrine glands, as observed in rodents. Following treatment, manganese
levels in the central nervous system decreased more slowly than levels in other tissues. Suzuki et
al. (1975) injected monkeys subcutaneously with manganese, and subsequently found increased
tissue concentrations of manganese in endocrine and exocrine glands (thyroids, parotids, and gall
bladder) and in the nuclei of cerebral basal ganglia. Newland et al. (1989) noted substantial
accumulation in the pituitary gland of Macaco fascicularis and Cebus apella monkeys at low
cumulative doses.
Several studies have addressed regional distribution of manganese in the brain following
parenteral exposure. Newland and Weiss (1992) investigated distribution of manganese in the
brain of monkeys. Three Cebus monkeys received multiple intravenous doses of 5 or 10 mg/kg of
manganese chloride over the course of 450 days. Magnetic resonance imaging revealed darkening
of the globus pallidus and substantia nigra. This result is consistent with accumulation of
manganese in these regions.
Scheuhammer and Cherian (1981) reported the distribution of manganese in male rat brain
tissue with and without intraperitoneal exposure to 3 mg Mn/kg as manganese chloride. In
unexposed rats, the highest concentrations of manganese were found in the hyporaalamus,
colliculi, olfactory bulbs, and midbrain. In treated rats, all brain regions showed an increase in
manganese concentration, and the highest manganese concentrations were observed in the corpus
striatum and corpus callosum.
External Review Draft — Manganese—April 2002
6-9
-------
Autissier et al. (1982) reported that rats given a daily intraperitoneal dose of 10 mg/kg-day
manganese chloride for 4 months showed significant increases in the accumulation of manganese
in the brain. This dose was equivalent to 4.4 mg Mn/kg-day. The study showed a 359% increase
in the concentration of manganese in the brain stem, a 243% increase in the corpus striatum, and a
138% increase in the hypothalamus. i
The tissue distribution of manganese appears to be affected by co-exposure to other
metals. Shukla and Chandra (1987) exposed young male rats to lead (5 mg/L in drinking v/ater)
and/or manganese (1 or 4 mg/kg, by intraperitoneal injection) for 30 days. Exposure to individual
metals resulted in accumulation in all brain regions. Co-exposure to lead and manganese resulted
in increased levels of both metals, particularly in the corpus striatum. Administration of
manganese alone led to dose-dependent increased levels in liver, kidney and testis. Co-exposure
to lead further increased manganese accumulation in liver. The authors concluded that the
interaction of metals can alter tissue distribution of manganese, and that adverse health effects
may result from co-exposure to even low levels of metals.
.
The chemical form in which manganese is injected may influence the subsequent tissue
distribution of manganese. Gianutsos et al. (1985) demonstrated that blood and brain levels of
manganese in mice are increased following intraperitoneal injection of manganese chloride,
manganese oxide, or methylcyclopentadienyl manganese tricarbonyl (MMT). However, MnCl2
administration resulted in more rapid accumulation and ultimately higher levels of blood and brain
manganese. It was suggested that the differences seen among the three manganese compounds
result from the oxide and MMT forms being more hydrophobic. Hydrophobicity may cause
formation of a depot at the site of injection that retards absorption. Gianutsos et al. (1985) also
demonstrated that the exit of manganese from the brain is a slower process than its entry,
resulting in a long retention period and potential accumulation. A single injection of 0.4 mEq
Mn/kg resulted in a significant increase (> 2-fold) in brain levels within 1—4 hours. The increased
levels were maintained for at least 21 days. Brain manganese levels were especially sensitive to a
repeated dose regimen. Much greater accumulation occurred when the dose was divided into 10
injections given every other day as compared with a single injection. This observation may help
explain the slow onset of manganese neurotoxicity: acute exposure results in other organs serving
as the primary target, while chronic exposure results in gradually increasing brain levels with
subsequent neurotoxicity.
Distribution of manganese has also been investigated in oral exposure studies. Chan et al.
(1981) administered 278 mg/L manganese chloride in drinking water to rats for two years. At the
termination of the study, these investigators found a 31% increase in manganese concentration in
the brain and a 45% increase in the liver relative to control values. Assuming a body weight for
rats of 0.35 kg and water consumption of 0.049 L/day, the average daily dose of manganese in
this experiment was equivalent to 17 mg/kg-day.
Some oral exposure data suggest that developmental stage may influence the distribution
of manganese. The brain, for example, may be a site for preferential accumulation of manganese
in neonates. Kostial et al. (1978) observed that rat pups showed a greater accumulation of
External Review Draft — Manganese—April 2002
6-10
-------
manganese in the brain, but not in the liver, than did their mothers. The data of Rehnberg et al.
(1980,1981,1982) indicate that the neonatal brain reaches higher concentrations of manganese
than other tissues. The authors suggested that this pattern reflects a response to a nutritional need
for manganese in the developing brain.
Kontur and Fechter (1985) demonstrated placental transfer of manganese in Long-Evans
rats exposed via drinking water throughout gestation. Transfer was limited, with only 0.4% of
the administered manganese accumulating in a single fetus. Neonatal pups of exposed dams had
significantly increased levels of manganese in the forebrain. However, the increase was not
associated with any overt signs of toxicity.
Komura and Sakamoto (1993) investigated the subcellular distribution of Mn and the
binding characteristics of Mn to brain protein in male mice following administration of different
forms of manganese. Four different manganese compounds (MnCl2-4H2O, Mn(CH3COO)2»4H2O,
MnCO3, or MnOj) were administered in the diet at a concentration of 2,000 mg Mn/kg for 12
months. Each treatment group included 6 male mice. The control group received a diet
containing approximately 130 mg Mn/kg (form not specified). Assuming a food factor of 0.13,
the control and treatment dietary levels correspond to approximately average daily doses of 17
and 260 mg Mn/kg-day, respectively. Cerebral cortex concentrations of Mn were significantly
higher in mice receiving the relatively insoluble compounds MnCO3 and MnO2 than in controls.
The subcellular distribution of manganese in the striatum and the gel chromatographic profiles of
manganese were similar for all tested manganese compounds.
Roels et al. (1997) reported that repeated gavage dosing of rats (once weekly for 4 weeks)
with 24.3 mg Mn/kg (5% of the dose, or 1.22 mg/kg, was assumed to be absorbed by the study
authors) as MnCl2 resulted in significantly increased concentrations of the metal in blood (68%)
and brain cortex (22%) compared to saline controls but did not significantly increase striatal or
cortex Mn concentrations. Similar administration of MnO2 at the same dose level did not result in
significant increases of Mn in blood or any brain tissue. Further studies indicated that Mn from
MnCl2 was absorbed much more rapidly and reached a higher peak concentration in the
bloodstream of the dosed rats than did MnO2. The peak Mn blood level following gavage dosing
of MnCl2 was roughly twice that of the oxide and was reported 1 hour post-dosing, while that of
MnO2 was not reported until 144 hours post-dosing (Roels et al. 1997). These data indicate that
administered manganese can be distributed into the brain and the kinetics of uptake and
partitioning depend on the chemical form of the manganese compound.
6.3 Metabolism
As a metallic element, manganese does not undergo metabolic conversion to other
products. However, manganese has the potential to exist in several oxidation states in biological
systems: Circumstantial evidence from the study of manganese-containing enzymes and from
electron spin trapping experiments suggests that manganese undergoes conversion from Mn(H) to
Mn(III) within the body (ATSDR, 2000). The conversion from Mn(H) to Mn(HI) appears to be
External Review Draft — Manganese—April 2002
6-11
-------
catalyzed by the a-globulin protein ceruloplasmin (Andersen et al., 1999). This reaction may be
enhanced by the high affinity of the iron-transporting protein transferrin for Mn(m).
A small fraction of absorbed manganese is present as the free ion. However, manganese •
readily forms complexes with a variety of organic and inorganic ligands. The complexes formed
include 1) low molecular weight complexes with bicarbonate, citrate or other ligands; 2) an
exchangeable complex with albumin; and 3) tightly bound complexes with proteins such as
transferrin and a2-macroglobulin. In addition, manganese can assume a structural role in
metalloproteins such as mitochondrial superoxide dismutase, pyruvate decarboxylase, and liver
arginase. Manganese also plays a catalytic or regulatory role in enzymatic reactions involving,
select hydrolases, dehydrogenases, kinases, decarboxylases and transferases.
6.4 Excretion
The primary route for elimination of manganese is to the feces through bile, as
demonstrated in several animal studies (Weigand et al., 1986; Davis et al., 1993; Malecki et al.,
1996). Fecal manganese concentration reflects both unabsorbed manganese and biliary secretion
of absorbed manganese. .
Human Studies ' \
i
The primary route for elimination of manganese is via the feces. Fecal manganese
concentration reflects both unabsorbed manganese and biliary secretion of absorbed manganese.
Price et al. (1970) determined the excretion pattern for preadolescent girls consuming 2.13
to 2.43 mg Mn/day. Approximately 1.66 to 2.23 mg Mn/day was excreted in the feces. In
contrast, only 0.01 to 0.02 mg/day was excreted in the urine. Results from other studies confirm
the importance of the fecal pathway for excretion. WHO (1981) and Newberne (1973) reported
that human excretion of manganese in urine, sweat, and milk is minimal. The normal level of
manganese found in urine of humans has been ;reported to be 1—8 ng/L, but values as high as 21
u-g/L have also been reported (U.S. EPA, 1984). Greger et al. (1990) reported urinary excretion
levels of 7.0 and 9.3 nmol Mn/g creatine/day (0.38 and 0.51 \ig Mn/g creatinine/day) for healthy
men and women, respectively. Urinary excretion of manganese was not responsive to oral intake
levels of manganese (Davis and Greger, 1992)!
A number of studies have addressed the kinetics of manganese excretion. Humans who
ingested tracer levels of radioactive manganese excreted the tracer with whole-body retention
half-times of 13 to 37 days (Mena et al., 1969;|Davidsson et al., 1989b; Sandstrom et al., 1986).
Sandstrom et al. (1986) gave volunteers a single oral dose of radioactive manganese and reported
a mean biologic half-life value of 13 days (range 6-30 days) for 14 subjects monitored on post-
exposure days 5-20, and a mean half-life of 34 days (range 26-54 days) for 6 subjects monitored
on post-exposure days 20—50. Two additional ^ubjects received manganese intravenously and
experienced a much slower turnover.
External Review Draft — Manganese — April 2002
6-12
-------
Mahoney and Small (1968) investigated the clearance of intravenously injected MnCl2 by
humans. These investigators observed a biphasic clearance pattern, with a rapid phase that lasted
4 days and a slow phase that lasted 39 days. Schroeder et al. (1966) reported a whole body
turnover rate in healthy adults of about 40 days, with a total body manganese content of about 20
mg.
Cotzias et al. (1968) injected manganese intravenously and reported values for biological
half-time of 37.5 days in healthy subjects, 15 days in healthy miners, and 28 days in subjects with
chronic manganese poisoning. These researchers also found that clearance by healthy subjects
averaged 25 days from the liver, 54 days from the head, and 57 days from the thigh, as measured
by external counting with a collimator. In healthy miners, liver clearance averaged 13 days; head
clearance averaged 37 days; and thigh clearance averaged 39 days. Subjects with chronic
manganese poisoning cleared manganese from the liver in 26 days, from the head in 62 days, and
from the thigh hi 48 days.
Finley (1999) demonstrated that iron status (assessed as serum concentrations of sodium
ferritin) may affect manganese excretion. Biological half-life (determined by regression of whole
body ^Mn counts) was assessed in women aged 20 to 45 years who were categorized as having
high (upper 10% of normal range, mean values 68 to 69 p-g/L) or low (lower 10% of normal
range, mean values 8.7 to 8.9 ng/L) serum ferritin levels. Biological half-life was determined
under conditions of high (9.5 mg Mn/day) or low (0.7 mg Mn/day) dietary manganese intake.
Subjects with low ferritin status consuming the low manganese diet had a mean biological half-life
that was more than twice the value determined for high ferritin status subjects consuming the
same diet (36.6 days versus 17.0 days). There was no effect of ferritin status on mean half-life for
subjects consuming the high manganese diet (13.0 and 11.8 days for low and high ferritin status
groups, respectively). .
Animal Studies
No studies of excretion following oral administration of manganese in animals were
identified.
Greenberg and Campbell (1940) reported that 90.7% of a 1 mg intraperitoneal dose of
radiolabeled manganese (54Mn) was found in rat feces within 3 days. In a subsequent study,
Greenberg et al. (1943) found that 27.1% of a 0.01 mg intraperitoneal dose of radiolabeled
manganese and 37.3% of a 0.1 mg dose were collected in rat bile within 48 hours. Tichy et al.
(1973) administered a 0.6 jig dose of manganese chloride to rats and reported that 27% was
excreted into the bile within 24 hours.
Klaassen (1974) demonstrated that bile is the main route of manganese excretion, and that
biliary excretion represents a major homeostatic mechanism for manganese levels in the body.
This investigator administered intravenous doses of 0.3,1.0, 3.0, or 10.0 mg Mn/kg to rats,
rabbits and dogs. Urinary excretion was low. As the dose increased, the excretion of manganese
into the bile also increased. The concentration of manganese in bile was 100 to 200 times higher
External Review Draft — Manganese—April 2002 6-13
-------
than in plasma at the three lower doses. However.
increase in excretion of manganese into the bile
was attained, suggesting that a saturable active transport mechanism may exist (U,
•, at the 10 mg dose there was no further
A maximum excretion rate of 8.5 fig Mn/min/kg
.S. EPA, 1984).
Britton and Cotzias (1966) and Suzuki (1974) found that an increase in dietary intake of
manganese decreased biologic half-times. Studies also indicate that the biologic half-time of
manganese in the brain of rats, mice and monkeys is longer than the half-life in the body (Suzuki,
1974;Dasturetal., 1969,1971). ' • |
In developmental studies of manganese excretion, neonatal mice, rats, and kittens were
found to rapidly accumulate manganese without excreting it during me first 18 days of life (U.S.
EPA, 1984). In contrast, when lactating rats and cats were given excessive doses of manganese
in drinking water (> 280 mg/L), then: offspring initiated excretion before the 16th day of life.
Although human and animal evidence indicates that most manganese is excreted to the
feces in bile, alternative routes for manganese lexcretion also exist. Experiments conducted by
Bertinchamps and Cotzias (1958), Kato (1963), and Papavasiliou et al. (1966) demonstrated
direct excretion of manganese through the intestinal wall. This route is most evident in the
presence of biliary obstruction or following high doses of manganese. Bertinchamps et al. (1966)
and Cikrt (1973) reported that in rats excretion of manganese occurred through the intestinal wall
into the duodenum, jejunum and terminal ileum. Burnett et al. (1952) demonstrated that
manganese excretion by dogs also occurs via the pancreatic juice. Other potential sources of fecal
manganese include intestinal secretions and the manganese present in sloughed off intestinal
microvillus cells. The fraction of total excretion attributable to these alternative pathways has not
been reported, but is expected to be relatively small when compared to biliary secretion.
External Review Draft — Manganese—April 2002
6-14
-------
7.0 HAZARD IDENTIFICATION
7.1 Human Effects
7.1.1 Case Reports
General Population
A number of investigators reported the toxicity of total parenteral (TPN) manganese in
humans, .especially on changes in brain MR! scans (Ejima et al., 1992; Fell et al., 1996; Mirowitz
and Westrich, 1992). These studies emphasize the difference in the effect of oral and parenteral
manganese. When administered parenterally, manganese bypasses the typical excretory
mechanisms in the gastrointestinal tract and liver and accumulates in the brain (Mirowitz and
Westrich, 1992).
In addition, there are a limited number of case reports describing the outcome of exposure
following accidental or intentional ingestion of manganese from potassium permanganate, a strong
oxidizing agent. Unspecified toxic effects were reported following ingestion of 2,4 mg/kg-day
potassium permanganate (0.83 mg Mn/kg-day) by a woman of unknown age and health status.
This information was reported in a 1933 French study cited in NIOSH (1984), and was not
available for review. Dagli et al. (1973) described a case in which oral ingestion of a 300 mg dose
of potassium permanganate (104 mg Mn) resulted in extensive damage to the distal stomach and
pyloric stenosis. Mahomedy et al. (1975) described two cases of methemoglobinemia following
ingestion of an unspecified amount of potassium permanganate which had been prescribed by
African tribal healers. Development of methemoglobinemia likely reflects the chemical oxidation
of heme iron.
Holzgraefe et al. (1986) reported neurological effects in an adult man who ingested
approximately 1.8 mg/kg-day of potassium permanganate (0.62 mg Mn) for 4 weeks. A
syndrome similar to Parkinson's disease developed after about 9 months. However, data in this
study are reported to be insufficient to establish causation (U.S. EPA, 1993). Bleich et al. (1999)
published a 14-year follow-up of this case report. Most of the symptoms originally noted
(including rigor, muscle pain, hypersomnia, increased libido, sweating, fatigue, and anxiety) had
improved, and the study authors noted that there appeared to be.no evidence for progression of
the parkinsonian syndrome as described by others (Huang et al., 1998).
Additional case reports suggest the potential for manganese toxicity following oral
exposure, but are difficult to assess quantitatively. One report involved a 59-year-old male who
was admitted to the hospital with classical symptoms of manganese poisoning, including dementia
and a generalized extrapyramidal syndrome (Banta and Markesbery, 1977). The patient's serum,
hair, urine, feces, and brain were found to have manganese "elevated beyond toxic levels." No
source of manganese exposure was identified for this individual. Exposure may have resulted
from the use of large quantities of vitamin and mineral supplements for 4 to 5 years. No
quantitative data were provided in this report.
External Review Draft —Manganese—April 2002 7-1
-------
Manganese intoxication was described Jin a 62 year-old male who received total parenteral
nutrition that provided 2.2 mg of manganese (form not stated) daily for 23 months (Ejima et al.,
1992), This level corresponds to a dose of approximately 0.023 mg Mn/kg-day for a 70 kg adult.
The patient's whole blood manganese concentration was elevated. The patient exhibited
dysarthria, mild rigidity^ hypokinesia with masked face, a halting gait, and severely impaired
postural reflexes, and the diagnosis of this condition was parkinsonism. Assuming an average
absorption of roughly 5% of an oral dose, the intravenous dose of 2.2 mg Mn/day would be
approximately equivalent to an oral intake of 40 mg Mn/day (U.S. EPA, 1993).
[•
Sensitive Populations
Individuals with impaired liver function or bile flow may represent potentially sensitive
subpopulations for manganese exposure. For example, Hauser et al. (1994) reported changes in
brain MRI scans in liver failure patients whichlwere identical to those observed in cases of
manganese intoxication. The patients (n=3) examined exhibited bilateral signal hyperintensity in
the globus palladi and substantia nigrae in Tl-weighted MRI and increased blood manganese
levels but had no history of increased exposure to manganese. Hauser et al. (1994) postulated
that impaired elimination of normal dietary manganese could result in manganese intoxication.
Devenyi et al. (1994) described a case study of an 8 year-old girl with Alagille's syndrome, an
autosomal dominant disorder characterized by neonatal cholestasis, intrahepatic bile duct paucity,
and end-stage liver disease. The patient exhibited a stable peripheral neuropathy, and for a period
of 2 months exhibited episodic, dystonic posturing, and cramping of her hands and arms. Whole
blood manganese level was elevated (27 \ig/L, in contrast to a normal range of 4 to 14 ng/L), and
cranial Tl-weighted magnetic resonance imaging (MRl) revealed symmetric, hyperintense globus
pallidi and subthalamic nuclei. These findings|were interpreted as indications of manganese
toxicity. Following liver transplantation, the patient's manganese levels returned to normal, her
neurological symptoms improved, and MRI results appeared normal. The interpretation of this
series of events was that: 1) progressive liver dysfunction resulted in inadequate excretion of
manganese into the bile, 2) subsequent accumulation of manganese resulted in neurotoxicity, and
3) liver transplantation restored biliary excretion and alleviated the symptoms.
* r
i.
Manganese has been identified as a possible etiologic agent in the occurrence of
neurological symptoms associated with hepatic encephalopathy (a brain disorder associated with
chronic liver damage). Medical evidence supporting an etiologic role has been summarized, by
Layrargues et al. (1998). Patients with chronic liver disorders such as cirrhosis experience a high
incidence of extrapyramidal symptoms resembling those observed in cases of occupational
manganism. Manganese concentration increases in the blood and brain of patients with chronic
liver disease and these changes are accompanied by pallidal hyperintensity on Tl-weighted MRI.
Autopsy data from ten patients who died in hepatic coma indicate that manganese levels are 2- to
7-fold higher in the globus pallidus of cirrhotic patients when compared to the general population.
Liver transplantation normalizes the pallidal MR signals and results in the disappearance of
extrapyramidal symptoms. Conversely, transjugular intrahepatic portosystemic shunting (a
procedure which increases the systemic availability of manganese) intensifies the pallidal MR
signal and results in deterioration of neurological function.
External Review Draft -j- Manganese—April 2002
7-2
-------
7.1.2 Short-term Studies
General Population
Kawamura et al. (1941) reported health effects resulting from the ingestion of manganese-
contaminated well water by 25 individuals. The source of contamination was identified as
leachate from approximately 400 dry cell batteries buried near the drinking water well. Chemical
analysis also revealed high levels of zinc in the well water. The length of exposure to manganese
was estimated to be 2 to 3 months. The concentration of manganese in the well was
approximately 14 mg Mn/L (as Mn3O4) when analyzed 7 weeks after the first case appeared. This
level corresponds to a dose of approximately 28 mg Mn/day (assuming a daily water intake of 2
L), or 0.5 mg Mn/kg-day (for a 60 kg adult). When reanalyzed 1 month later, the manganese
concentration had decreased by about 50%. Based on these measurements, retrospective
extrapolation suggests that the initial exposure level may have been 28 mg Mn/L or higher.
Assuming a daily water intake of 2 L, and an additional manganese intake from food of at least 2
mg/day, Ibis represents a dose of atleast 58 mg Mn/day. This intake of manganese is about 10 to
20 times the level considered to be safe and adequate by the Food and Nutrition Board of the
National Research Council (NRC, 1989). Assuming a body weight of 60 kg for an adult, this
intake level corresponds to a dose of 0.93 mg Mn/kg-day from drinking water. No information
on dietary intake was available.
Etealth effects reported by Kawamura et al. (1941) included lethargy, increased muscle
tonus, tremor and mental disturbances. Out of 25 people examined, 15 had symptoms. Five cases
. were considered severe, 2 cases were categorized as moderate and 8 cases were described as
mild. The most severe symptoms were observed in the elderly. Younger people were less
affected, and symptoms of intoxication were absent in young children (age 1 to 6 years). Three
deaths occurred, including one from suicide. Upon autopsy, the concentration of manganese in
the brain of one person was found to be 2 to 3 times higher than concentrations measured in two
control autopsies. Extreme macroscopic and microscopic changes were seen in the brain tissue,
especially hi the globus pallidus. The authors also reported elevated levels of zinc in the well
water, but concluded that the zinc appeared to have no relation to the observed symptoms or
tissue pathology. This conclusion was largely based on the observation of morphological changes
in the corpus striatum which are characteristic of manganese poisoning, but are not a feature of
zinc poisoning. ,
While toxicity in the Kawamura et al. (1941) study is attributed to manganese, several
aspects of the observed health effects are inconsistent with traits of manganism observed in
humans following chronic inhalation exposure. Inconsistencies include the rapid onset of
symptoms and rapid progression of the disease. Two adults who came to tend the members of
one family developed symptoms within 2 to 3 weeks. The course of the disease was very rapid,
progressing hi one case from initial symptoms to death in 3 days. Some survivors recovered prior
to significant decreases in the manganese concentration of the well water which resulted when the
batteries that caused the contamination were removed from the site. This pattern contrasts with
the longer latency period and irreversible damage caused by inhalation exposure to manganese.
External Review Draft—Manganese—April 2002 7-3
-------
These observations may represent differences in the pharmacokinetics of ingested versus inhaled
manganese, but there is little information to support this conclusion. Although these individuals
were clearly exposed to high levels of manganese, it is possible that additional factors contributed
to the observed effects (U.S. EPA, 1993; ATSDR 2000).
Sensitive Populations
Study data for sensitive populations were not identified in the materials reviewed for
preparation of this document
7.1.3 Long-Term and Epidemiological Studies
General Populations
Kondakis et al. (1989) conducted an eriidemiologic study of manganese in drinking water
in northwest Greece. Three areas with different levels of manganese in the drinking water supply
were chosen for this study. Area A had manganese concentrations of 3.6 to 14.6 ng/L, Area B
had concentrations of 81.6 to 252.6 jig/L, and Area C had concentrations of 1,800 to 2,300
|ig/L. The total population in the study areas ranged from 3,200 to 4,350 people. The study
included only individuals over the age of 50 drawn from a random sample of 10% of all
households. The sample sizes were 62,49, anil 77 for areas A, B, and C, respectively. The study
authors reported that "all areas were similar with respect to social and dietary characteristics," but
few details were provided. Kondakis et al. (1989) determined whole blood and hair manganese
concentrations in samples collected from study participants. The mean concentration of
manganese in hair was 3.51,4.49 and 10.99 |j,g/g dry weight for areas A, B and C, respectively.
Concentrations in hair differed significantly between areas C and A (p < 0.001). No significant
differences in whole blood manganese levels were observed among the three areas. However,
manganese concentration in blood is not considered to be a reliable indicator of manganese
exposure (U.S. EPA, 1993). ;
i
Kondakis et al. (1989) also administered a neurological examination which evaluated the
presence and severity of 33 symptoms (e.g., weakness/fatigue, gait disturbances, tremors,
dystonia) in all subjects. The results of the neurological examination were expressed as a
composite score. A higher neurological score indicated an increased frequency and/or severity of
the 33 evaluated symptoms. Results for the three geographic areas are summarized in Table 7-1.
Mean scores for both sexes combined were 2.7 (range 0-21) for Area A; 3.9 (range 0-43) for
Area B; and 5.2 (range 0-29) for Area C. The authors indicated that the difference in mean
scores for Area C versus Area A was statistically significant (Mann-Whitney Test, z = 3.16, p =
0.002, for both sexes combined), suggesting neurologic impairment in people living in Area C. In
a subsequent analysis, logistic regression indicated a significant difference between areas A and C
when both age and sex were taken into account (Kondakis, 1990).
External Review Draft -,— Manganese—April 2002
7-4
-------
Table 7-1. Mean Neurological Scores of Residents in Three Areas of Northwest Greece
with Different Levels of Manganese in Drinking Water (range is given in
parentheses).
Subject
Males
Females
Both
Area A
(3.6-14.6 jig Mn/L)
2.4(0-21)
3.0 (0-18)
2.7 (0-21)
Area B
(81.6-252.6 jig Mn/L)
1.6(0-6)
5.7(0-43)
3.9(0r43)
Area C
(1,800-2,300 ng Mn/L)
4.9 (0-29)
5,5(0-21)
5.2 (0-29)
Source: Kondakis et al. (1989)
Limitations to the Kondakis study have been noted by ATSDR (2000). These include: 1)
lack of clearly detailed descriptions of neurological signs and symptoms that reportedly increased
following manganese exposure, and 2) failure to describe procedures for avoiding bias when
evaluating subjective neurological scoring parameters. An additional shortcoming of this study is
the lack of quantitative exposure data (U.S. EPA, 1996a). The individuals examined by Kondakis
et al. (1989) also consumed manganese in their diet. The initial estimate of dietary intake was 10
to 15 mg/day based on high intake of vegetables (Kondakis, 1990). This figure was subsequently
revised to an estimate of 5 to 6 mg Mn/day (Kondakis, 1993), but data were not provided to
substantiate this estimate. Lack of dietary intake and water consumption data prevents
determination of a quantitative dose-response relationship for manganese toxicity in this study.
Nevertheless, this study raises concern for adverse neurological effects at estimated doses that are
not far from the range of essentiality (U.S. EPA, 1996a).
Although conclusive evidence is lacking, some investigators have linked increased intake
of manganese with violent behavior. Gottschalk et al. (1991) found significant increases hi the
level of manganese in the hair of convicted felons (1.62 ± 0.173 ppm in prisoners compared with
0.35 ± 0.020 ppm in controls). The study authors suggested that "a combination of cofactors,
such as the abuse of alcohol or other chemical substances, as well as psychosocial factors, acting
in concert with mild manganese toxicity may promote violent behavior." The number of potential
variables indicates that caution should be exercised in interpretation of these data.
Results from studies of an Aboriginal population in Groote Eylandt have been cited as
additional evidence for a relationship between elevated manganese exposure, violent behavior, and
adverse health effects. The soil on this Australian island is exceptionally high in manganese
(40,000 to 50,000 mg/kg), and the fruits and vegetables grown in the region are reported to
contain elevated concentrations of manganese. High alcohol intake, anemia, and a diet deficient in
zinc and several vitamins (Florence and Stauber, 1989) may contribute to increased uptake and
retention of manganese. The proportion of arrests in this native population is the highest in
Australia, and high incidences of stillbirths and congenital malformations, as well as a high
occurrence of Parkinson-like neurobehavioral syndrome, have been observed (Cawte and
Florence, 1989; Kilburn, 1987). Clinical symptoms consistent with manganese intoxication are
External Review Draft — Manganese — April 2002
7-5
-------
present in about 1% of the inhabitants. Quantitative data on oral intake have not been reported,
but elevated concentrations of manganese have been determined in the blood and hair of the
Aborigines (Stauber et al., 1987). However, Stauber et al. (1987) did not find a correlation
between hair levels of manganese and the severity of neurological symptoms in individuals.
A study of the neurologic status of the Aborigines in Groote Eylandt identified two
general syndromes. One syndrome is charactejrized by muscle atrophy and weakness, while the
other is characterized by ataxia and oculomotor disturbances (Kilburn, 1987). Although an
association of adverse health effects with elevated manganese exposure is suggested by these
observations, the small population of Groote Eylandt and the difficulty in defining an appropriate
control population have prevented the identification of statistically significant trends (U.S. EPA,
1993). I
i
Several of the studies above utilized hair analysis as a method for estimating exposure to
manganese. ATSDR (2000) has outlined several potential limitations to the use of hair analysis.
The normal cycle of hair growth and loss restricts its usefulness to a period of a few months
following exposure. External contamination of hair by dye, bleaching agents, or other materials
may result in values which are not representative of absorbed doses. The affinity of manganese
for pigmented tissue may result in variation of manganese concentration with hair color.
Goldsmith et al. (1990) investigated a Parkinson's disease cluster within southern Israel.
The prevalence of the disease was increased among persons 50 to 59 years old, suggesting an
early onset of the disease. Well water and soils in the region reportedly contained high levels of
manganese, although no quantitative data were provided. In addition, the manganese-containing
fungicide Maneb was commonly used in the area. However, several factors limit the use of this
study for evaluation of the human health effects of excess manganese exposure. Lack of
environmental concentration data prevented reliable estimation of exposure rates. Potentially
confounding factors included the high levels of aluminum, iron, and other metals in the soil and
water, and the use of the herbicide paraquat inline area (ATSDR, 2000). Paraquat is structurally
related to N-memyl-4-phenyl-l,2,3,6-tetrahydropyridme (MPTP), a piperidine derivative which
causes irreversible symptoms of parkinsonismiin humans.
Vierrege et al. (1995) investigated the neurological impact of chronic manganese exposure
via drinking water in a cross-sectional study of two proband cohorts in rural northern Germany.
The study population was drawn from the county Herzogtum Lauernburg in the northernmost
province of Germany. This region is characterized by agricultural and forestry activities but no
steel or mining industry. Many of the residents of this area draw their drinking water from wells,
and by law, the well water is routinely monitored for chemicals and bacteria. A survey was
conducted in 1991 and was combined with a c|oss-sectional investigation of a randomly selected
group of right-handed residents aged 40 years br older who had used their wells as the primarily
source of drinking water for a minimum of 10 years (range 10 to 40 years). Complete
documentation of manganese monitoring results for six years prior to the investigation was
required for study eligibility. Participants were assigned to two groups on the basis of manganese
concentration in their well water. Group A included individuals who continually ingested well
External Review Draft -f- Manganese—April 2002
7-6
-------
water containing between 0.300 and 2.160 mg Mn/L. Group B included individuals whose well
water manganese concentration had never exceeded 0.050 mg/L. Detailed information on medical
history, employment history, diet, alcohol consumption, drug use and smoking was collected by
intervievtr. Individuals in Group A were matched to individuals in Group B with respect to age,
sex, nutritional habits, and drug intake. Criteria for exclusion from the study included history of .
employment in the steel industry, adherence to dietary restrictions, history of CNS-relevant drug
use, diabetes meltitus, history of stroke, or treatment for psychiatric disorders. Conditions that
could affect performance on the neurological assessment tests (neurorthopedic impairment of
hand-finger function or poor vision) were also grounds for exclusion from the study.
A total of 164 eligible subjects was identified. Of these, 49 subjects were excluded for
failure to meet the health or water monitoring criteria. Group A included 41 subjects (21 male
and 20 female) with a mean age (± standard deviation) of 57.5 ± 10.3 years (range 41 to 84
years). Group B included 74 subjects with a mean age of 56.9 ± 11.8 years (range 41 to 86)
years. No dietary differences were evident between the two groups. Neurological status was
assessed by experienced personnel blinded to the group status of the subjects. Each participant
was evaluated for neurotoxicological symptoms by use of a modified German version of a
standardized symptoms list. Signs of parkinsonism were evaluated by the Columbia University
Rating Scale (CURS). Fine motor ability (each hand) was assessed using a conventional
apparatus ("Motorische Leistungsserie," MLS) and application of aiming, steadiness, line pursuit,
and tapping tests. Manganese status was evaluated by determination of manganese hi blood. The
concentration of manganese in hair or nails was not determined. :
; The results of neurological evaluations are summarized in Table 7-2. There were no
significant differences between groups for the mean item scores on the standardized symptoms list
or the CURS. MLS test results were obtained for 36 group A subjects (18 male and 18 female,
mean age 56.4 ±8.4 years, range 41 to 72 years) and 67 Group B subjects (35 men and 32
women, mean age 55.1 ± 9.9 years, range 41 to 72 years). Results of participants older than 72
years were not included in the statistical analysis of MLS data because normative information
from the general population have an upper age limit of 72 years. No significant differences were
observed between groups for any test when results were standardized to age-corrected values.
Mean blood manganese concentrations were 8.5 ± 2.3 ng/L and 7.7 ± 2.0 ng/L for groups A and
B, respectively. The blood manganese values did not differ significantly and both fell within the
normal range for the general (non-occupationally exposed) population. Separate analyses for
possible confounding factors did not reveal differences in clinical or instrumental test outcomes
related to high or low consumption of alcohol, mineral water, coffee, tea, tobacco, vegetables, or
fruit. Where cases of parkinsonism (n = 3) were encountered in this study, they occurred in the
low exposure group (Group B) and were considered to be typical Parkinson's disease and thus
unrelated to manganese exposure. The authors of this study concluded that there was no
evidence of an association between consumption of high concentrations of manganese in well
water and neurological impairment (including those suggestive of parkinsonism).
External Review Draft — Manganese—April 2002
7-7
-------
Table 7-2. Mean Neurological Scores of Residents in Germany Exposed to Different
Levels of Manganese in Well Water.
Assessment
Neurotoxicological
Symptom
Questionnaire
CURS
Parkinsonism
MLS Aiming
MLS Steadiness
MLS Line Pursuit
MLS Tapping
Measure ,
I
Item Number
,
Item Number
• i
Duration (sec) j
1
Errors (number) ;
Duration of errors (sec)
Errors (number)
Duration of errors (sec)
Total duration (sec) i
Rate (number) i
Exposure Group
Group A (High)
3.2 ± 3.0a
1.2 ±1.0
104.8 ± 9. lb
103.9 ±103.9
100.8 ± 10.6
106.4 ±7.6
102.3 ±8.1
104.3 ± 12.6
103.1 ±7.2
Group B (Low)
3.9 ±3.1
1.7 ±2.0
102.9 ± 10.0
103.1 ±7.9
100.2 ±10.5
106.6 ±8.0
103.1 ± 10.6
100.T± 15.5
103.9 ±10.5
1 Mean ± standard deviation
b MLS test results are for right hand •
Three potential limitations related to the ecologic design of this investigation were noted
by Vieregge etal. (1995). First, the investigators could not control for possible migration of
subjects with manganese-induced neurological disorders from the study area prior to the
investigation. However, Vieregge et al. (1995) stated on the basis of inquiries and general
experience in Hie region that a migration effect!was unlikely to be significant. Second, although
possible confounding by several dietary items or groups was evaluated and found to be non-
evident, confounding effects of nutrition (particularly in subjects working outside their home
residence) could not be completely excluded. Finally, blood manganese levels are thought to
primarily reflect current body burden of manganese rather than exposure.
Iwami et al. (1994) reported that the incidence of motor neuron disease (MND) in a small
town in Japan was positively correlated with a significantly increased manganese concentration in
local rice and a low magnesium concentration in the drinking water. This study, however, did not
provide good estimates of overall exposure to manganese in either the control population or the
population with MND.
External Review Draft — Manganese — April 2002
7-8
-------
Adverse neurological effects (decreased performance in school and in neurobehavioral
exams of the WHO core test battery) were reported in 11- to 13-year-old children who were
exposed to excess manganese through ingestion of well water and from wheat fertilized with
sewage water (He et al. 1994; Zhang et al. 1995). The exposed group consisted of 92 children
pair-matched to 92 controls from a nearby area. The groups were matched for age, sex, grade,
family income level, and parental education level; further, both groups lived on farms. The
average manganese concentration of the drinking water of the exposed group was 0.241 mg/L
compared to the control level of 0.04 mg/L. Although the study authors had drinking water data
from a 3-year period, it Was not clear how long the children were exposed prior to the study".
Further, the exposure data were not well-characterized; therefore ifwas not possible to establish a
cause-effect link between ingestion of excess manganese and preclinical neurological effects in
children. -
Sensitive Populations
Study data for sensitive populations were not identified in the materials reviewed for
preparation of this document.
7.1.4 Beneficial Effects
Manganese is a naturally-occurring element that is required for normal physiological
functioning in all animal species (U.S. EPA, 1996a). Manganese plays a role in bone
mineralization, metabolic regulation, protein and energy metabolism, protection of cells from
oxidative stress, and synthesis of mucopolysaccharides (ATSDR, 2000). Many of these roles are
achieved by participation of manganese as a catalytic or regulatory factor for enzymes, including
hydrolases, dehydrogenases, kinases, decarboxylases and transferases. In addition, manganese is a
structural component of the metalloenzymes mitochondrial superoxide dismutase, pyruvate
carboxylase, and liver arginase. Additional information on the biochemical and nutritional roles of
manganese in human health is available in Wedler (1994) and Keen et al. (1999). At present, the
• optimal levels of oral manganese exposure have not been well defined for humans (Greger, 1999).
Overt signs of manganese deficiency have been demonstrated in multiple animal species
(Keen et al., 1999). Biochemical effects observed in manganese-deficient animals include
alterations in carbohydrate, protein, and lipid metabolism. Physiological outcomes associated
with deficiency include impaired growth (Smith, 1944), skeletal abnormalities (Amdur et al.,
1944; Strause et al., 1986), impaired reproductive function in females, and testicular degeneration
hi males (Boyer et al., 1942). The molecular basis for these effects has not been established with
certainty, but may be related to the participation of manganese hi numerous enzymatic reactions.
In addition, the effect of manganese deficiency on mitochondrial superoxide dismutase activity
has functional consequences. Manganese-deficient rats experience more oxidation of
mitochondrial membranes of the heart and more formation of conjugated dienes than manganese-
adequate rats (Malecki and Greger, 1996). In another study, Gong and Amemiya (1996)
observed ultrastractural changes suggestive of oxidative damage hi the retinas,-of rats fed a
manganese-deficient diet for 12 to 30 months
External Review Draft — Manganese—April 2002
7-9
-------
Manganese is ubiquitous in human foods, and outright manganese deficiency has not been
observed in the general population. However^ observations reported by Doisy (1973) and
Friedman et al. (1987) indicate that manganese is an essential element for humans. Doisy (1973)
reported a decreased level of clotting proteins, decreased serum cholesterol, reddening of black
hair, retarded growth of hair and nails, and scaly dermatitis in a subject inadvertently deprived of
manganese. Friedman et al. (1987) administered a manganese-deficient diet to seven men for 39
days. Five of the seven subjects exhibited dermatitis at the end of the manganese-deficient period.
The development of dermatitis was attributed; to decreased activity of manganese-requiring
enzymes that are required for skin maintenance. The symptoms cleared rapidly when manganese
was restored to the diet. I
i ,
7.2 Animal Studies
, This section presents the results of manganese toxicity studies in animals. The first four
subsections provide study results by duration t>f exposure. In general, acute studies are those
which address exposure durations of 24 hoursj or less. Short-term studies have exposure
durations greater than 24 hours but less than approximately 90 days. The exposure duration of
subchronic studies is typically 90 days, and chronic studies are those in which exposure is longer
than 90 days. Some studies may fall into more than one exposure category since they measure
impacts over several exposure periods. The discussion of acute, short-term, subchronic and
chronic studies summarizes observed lexicological effects on all body systems. The remaining
subsections of Section 7.2 provide toxicological data related to specific organ systems and types
of endpoints, including neurotoxicity, developmental and reproductive toxicity, and
carcrnogemcity.
7.2.1 Acute Toxicity '
Oral Exposure
LD50 values determined for selected manganese compounds are summarized in Table 7-3. •
Oral LDgo values among Ihe water soluble manganese compounds ranged from 400 to 475 mg
Mn/kg for manganese chloride, and from 379 jto 810 mg Mn/kg for potassium permanganate. An
LD50 of 83 6 mg Mn/kg was reported for manganese acetate.
Age may be a factor in susceptibility to acute manganese toxicity. Kostial et al. (1978)
found that MnCl2 produced the greatest oral toxicity in4he youngest and oldest groups. Roth and
Adleman (1975) proposed that the increased susceptibility of older rats may result from a
decrease in adaptive responsiveness, which is characteristic of the aging process. Increased
susceptibility of younger rats may reflect high1 intestinal absorption and body retention of
manganese.
External Review Draft \— Manganese—April 2002
7-10
-------
Table 7-3. LD50 Values for Manganese Compounds.
Compound
Manganese
acetate
Manganese
chloride
Manganese
dioxide
Manganese
sulfate
Manganese
sulfate,
tetrahydrate
Manganese
nitrate
Methylcyclo-
pentadienyl
manganese
tricarbonyl
(MMT)
Potassium
permanganate
Species
rat
rat
rat
rat
mouse
guinea pig
rat
mouse
mouse
rat
mouse
mouse
mouse
rat
rat
rat
mouse
mouse
rat
rat
guinea pig
Route
oral
oral
oral
oral
oral
oral
i.p.
i.p.
i.v.
oral
i.p.
i.p.
i.p.
oral
oral
oral
oral
oral
oral
oral
oral
LD5»
(mg Mo/kg)
836
425
475
410
450
400
38
56
16
2,197
44
64
56 .
10
12
12
48
750
379
750
810
Reference
Smyth etal. (1969)
Shigan and Vitvickaja (1971)
Kostial etal. (1978)
Holbrook etal. (1975)
Shigan and Vitvickaja (1971)
Shigan and Vitvickaja (1971)
Franz (1962); Holbrook et al. (1975)
Franz (1962); Holbrook et al. (1975)
Larsen and Grant (1997)
Holbrook etal. (1975)
Bienvenu et al. (1963)
Yamamoto and Suzuki (1969)
Yamamoto and Suzuki (1969)
Hanzlik etal. (1980)
Hinderer (1979)
Hysell et al. (1974)
Hinderer (1979)
Shigan and Vitvickaja (1971)
Smyth etal. (1969)
Shigan and Vitvickaja (1971)
Shigan and Vitvickaja (1971)
i.p. = intraperitoneal
i.v. = intravenous
External Review Draft—Manganese — April 2002
7-11
-------
Parenteral Exposure
Manganese compounds administered by parenteral routes generally result in mortality at
lower doses. LD50 values for the intraperitoneal route ranged from 14 to 64 mg Mn/kg. Franz
(1962) andBienvenu et al. (1963) conducted comparative intraperitoneal toxicity studies, and
found that manganese is less toxic than many other metals. Jonderko (1965) found increased
serum calcium and decreased inorganic phosphorous in rabbits exposed mtramuscularly to 3.5 mg
Mn/kg. Details on the compound and the duration of exposure were not available.
Baxter et al. (1965) measured physiological parameters in manganese-treated rats
weighing 100 to 550 g. Measurements were made 1 to 72 hours after subcutaneous
adininistration of 5 to 150 mg of manganese as MnCl2 in saline. Levels of hemoglobin,
hematocrit, and mean corpuscular volume were significantly increased in rats receiving 150 mg
Mn/kg. A measurable response in these parameters occurred at 50 mg Mn/kg, while the peak
increase in these parameters occurred at 12 and 18 hours after dosing. The maximum response
occurred at 170 to 300 mg Mn/kg. Necrotic changes were noted in hepatic tissue 18 hours after a
single dose of 170 mg Mn/kg. •
Pancreatic endocrine function is affected by acute manganese exposure. Baly et al. (1985)
injected rats intraperitoneally with 40 mg Mn/kg. Manganese injection resulted in a decrease hi
plasma insulin levels, an increase hi plasma glucose levels, and a transitory increase in glucagon
concentration. j
Larsen and Grant (1997) administered a single intravenous dose of 150, 200, 300, or 400
Hmol/kg manganese chloride in saline to malejmice (5/group). These doses correspond to 8.2, 11,
16, and 22 mg Mn/kg, respectively. These stutiy authors reported an LD50 value of 300 jimol/kg
(16mgMn/kg).
7.2.2 Short-Term Studies
Oral Exposure \
Matrone et al. (1959) orally administerjed 2,000 ppm manganese as MnSO4»H2O to 6-
month-old anemic rabbits' for 6 weeks. The investigators also administered 125 ppm Mn as
MnSO4-H2O to anemic newborn pigs for 27 days. In each case, the investigators observed
decreased hemoglobin content hi the blood of treated animals. Hemoglobin depression hi baby
pigs fed up to 2,000 ppm manganese was overcome by a dietary supplement of 400 ppm iron.
Kimura et al. (1978) provided rats with diets supplemented with 564 nag/kg of manganese
as MnCl2 for 3 weeks. Assuming a food consumption factor of 0.05 above the dietary
background, this corresponds to a daily dose of 28 mg Mn/kg-day. The study authors reported
that brain serotonin levels were decreased hi manganese-treated rats. Monoamine oxidase activity
was unchanged, but L-amino-acid decarboxylase activity hi the brain was decreased by manganese
External Review Draft—Manganese—April 2002
7-12
-------
treatment. Histopathological analysis of the brain was not conducted. Blood serotonin levels
were increased in treated rats, and this change was accompanied by decreased blood pressure.
Shukla et al. (1978) administered a dose of 16 mg MnCl2-4H2CMcg (4.4 mg Mn/kg) in
drinking water (dose calculated by investigators) to rats for 30 days and evaluated the effect on
hepatic enzyme activity. Treated rats revealed significantly decreased succinic dehydrogenase,
alcohol dehydrogenase, and p-amylase activity when compared with controls. In contrast,
manganese exposure resulted in significantly increased activities of monoamine oxidase (MAO),
adenosine triphosphatase, arginase, glutamate-pyravate transaminase (= alanine aminotransferase,
or ALT), ribonuclease, glucose-6-phosphatase,.and a-amylase activity in the livers of treated rats.
Hietanen et al. (1981) also studied the effect of manganese on hepatic and extrahepatic
enzyme activities. Male Wistar rats were exposed to 0.5% Mn (as MnClj) in the drinking water
for 1,4, or 6 weeks. Assuming an average body weight of 0.35 kg and average water
consumption of 0.049 L/day (U.S. EPA, 1986d), this corresponds to an exposure of 0.7 mg
Mn/kg-day. Changes in the activity of several enzymes, including aryl hydrocarbon hydroxylase,
ethoxycoumarin 0-deethylase, and epoxide hydrase, were observed at 1 week but not at 6 weeks.
Enzyme activities were increased in the liver, and decreased in the intestines and kidney.
In a 14-day oral exposure study, NTP (1993) administered diets containing 0, 3,130,
6,250,12,500,25,000, or 50,000 ppm manganese sulfate monohydrate to F344 rats (5/sex/dose).
All rats survived the exposure period. Statistically significant differences in manganese-treated
rats included reduced body weight gain (57% decrease) and final body weight (13% decrease) in
the high-dose.males when compared to the control group. Decreased leukocyte and neutrophil
counts aind reduced liver weight were observed in high-dose males and females. The high-dose
groups also exhibited diarrhea during the second week of the study. Manganese concentrations in
the livers of animals receiving the 50,000 ppm diet were more than twice those of the controls.
The NO'AEL and LOAEL values based on decreased weight gain (males) and hematological
changes were approximately 650 and 1,300 mg Mn/kg-day, respectively.
NTP (1993) also administered diets containing 0, 3,130, 6,250,12,500,25,000, or 50,000
ppm manganese sulfate monohydrate to B6C3Fi mice (5/sex/dose) for 14 days. However, study
animals were poorly randomized at the beginning of the study, and no effects clearly attributable
to manganese exposure were identified.
Par enteral Exposure
Singh et al. (1974; 1975) administered 6 mg Mn/kg-day (as MnSO4«4H2O)
intraperitoneally to male IRTC rats for 25 days. Histopathological analysis of the livers revealed
mild congestion of central veins and sinusoids, and some focal necrosis in treated animals.
Scheuhammer and Cherian (1983) reported toxic effects in the pancreas resulting from
intraperitoneally injected manganese. The exposure duration was 30 days. Adverse effects
included a pancreatitis-like reaction. The authors suggested that this reaction was potentiated by
External Review Draft — Manganese—April 2002 7-13
-------
the presence of manganese in the peritoneal cavity, and would not occur as readily with
manganese administered by the oral route. j
Khandelwal et al. (1984) administered |6 mg Mn/kg-day (as MnCl2'4H2O) intraperitoneally
to male IRTC rats for 28 days. Activity of succinic dehydrogenase and cytochrome oxidase in
liver tissue were decreased after 28 days of manganese treatment.
i
Khan et al. (1997) administered 16 mg)kg-day MnCl2«4H2O in saline intravenously to male
beagle dogs (3/group). Treatment duration wa|s up to 4 hours/day for 4 days. Two of the three
dosed animals were in moribund condition, and were sacrificed for ethical reasons (one on day 3
and one on day 4). The third treated dog died on day 4. Symptoms prior to death included
vomiting, diarrhea, tremors, lethargy, reduced food intake, reduced blood pressure with reflex
tachycardia, and severe hepatotoxicity. j
7.23 Subchronic Studies
i
Oral Exposure '
Mitochondria-rich organs, such as the liver and pancreas, are hypothesized to be most
affected by excess manganese exposure. Wassermann and Wassermann (1977) reported
ultrastructural changes of the liver cells in rats| exposed to 200 mg/L of manganese chloride in
their drinking water for 10 weeks. Assuming ^ater consumption of 0.049 L/day and an average
body weight of 0.35 kg (U.S. EPA, 1986d), this level of exposure corresponds to an average daily
dose of approximately 12 mg Mn/kg-day. Increased metabolic activity was inferred from an
increased amount of rough endoplasmic reticulum, the occurrence of multiple rough endoplasmic
cisternae and prominent Golgi apparatus, and large Golgi vesicles filled with osmiophilic particles
in the biliary area of the liver cell. The authors attributed this apparent increase in metabolic
activity to biochemical processes related to the nutritional requirement for manganese, and
homeostatic processes triggered by increased exposure. They noted that other observed liver
effects, including the presence of glycogenosoines in the biliary area, groups of collagen fibers in
the Disse's spaces, and degenerative changes in some centrilobular liver cells, may either be direct
toxic phenomena or secondary responses to the effect exerted by manganese on other target
tissues. ATSDR (2000) evaluated these data and designated 12 mg Mn/kg-day as the NOAEL in
this study.
Carter et al. (1980) exposed young, iron-deficient rats to 400 to 3,550 ppm Mn as Mn3O4
for 32 weeks (route not specified). Manganese treatment resulted in decreased hemoglobin levels.
i
Leung et al. (1982) administered 1,000; 10,000, or 20,000 mg MnCl2-4H2O/L in drinking
water to female Wistar rats. Exposure was initiated at conception by administration of
manganese-containing drinking water to the dams, and continued through age 60 days. The
estimated doses were 38.9, 389, and 778 mg Mn/kg-day (U.S. EPA, 1993). Treated rats
exhibited liver necrosis and ultrastructural alterations that resembled human cholestasis. A
LOAEL of 38.9 mg Mn/kg-day was identified in this study based on hepatic necrosis.
External Review Draft — Manganese—April 2002
7-14
-------
In a 13-week study, NTP (1993) administered diets containing 0,1,600, 3,130, 6,250,
12,500, or 25,000 mg/kg manganese sulfate monohydrate above basal levels to F344 rats
(10/sex/dose).' The concentration of manganese in the control diets was approximately 92 mg/kg.
Mean daily intake of manganese sulfate monohydrate ranged from 98 mg/kg-day (32 mg Mn/kg-
day) for the low-dose to 1,669 mg/kg-day (542 mg Mn/kg-day) for the high-dose males. For
females, the range was 114 mg/kg-day (37 mg Mn/kg-day) for the low-dose group and 1,911
mg/kg-day (621 mg Mn/kg-day) for the high-dose group. No rats died during the study, and no
clinical or histopathology findings were attributed to manganese exposure. Females receiving
diets with ^6,250 mg/kg manganese sulfate experienced decreased body weight gain. Absolute
and relative liver weights were decreased in males receiving diets with z 1,600 mg/kg, and in
females in the highest dose group only. Hematological effects were also reported. All groups of
exposed males exhibited a significantly increased neutrophil count. Lymphocyte counts were
decreased in males receiving s 6,250 mg/kg in the diet and females in the three highest dose
groups. The low dose of 1,600 mg/kg (about 32 mg Mn/kg-day) was identified as the LOAEL
for this study, based on effects on liver weight and neutrophil counts hi male rats.
In a concurrent 13-week study, NTP (1993) administered diets containing 0, 3,130, 6,250,
12,500,25,000, or 50,000 mg/kg manganese sulfate monohydrate above basal levels to B6C3F,
mice (10/sex/dose). The concentration of manganese in the control diets was approximately 92
mg/kg. Mean daily intake of manganese sulfate monohydrate ranged from 328 mg/kg-day (107
mg Mn/kg-day) for the low-dose to 8,450 mg/kg-day (2,746 mg Mn/kg-day) for the high-dose
group. No deaths were attributed to manganese exposure. All groups of male mice and female
mice in the highest dose group exhibited significantly: decreased body weight gam. Relative and
absolute liver weights were decreased in males in the highest dose group. Both sexes receiving
the 50,000 mg Mn/kg diet exhibited decreased hematocrit and hemoglobin concentration. The
NTP report suggests that these findings may indicate microcytic anemia, which may have resulted
from a sequestration or deficiency of iron. Males receiving ^25,000 ppm also exhibited
significantly lower, leukocyte counts, although this finding was of questionable relevance to
manganese exposure. No clinical findings were attributed to manganese exposure. The LOAEL
for this study was 3,130 mg/kg-day (107 mg Mn/kg-day), based on significantly decreased body
weight gain in male mice. ,
Komura and Sakamoto (1991) supplemented mouse diets with different chemical forms of
manganese. Male mice (8/group) were exposed either to a control diet containing 130 mg Mn/kg,
or a diet supplemented with an additional 2,000 mg Mn/kg as MnQ2*4H2O,
Mn(CH3COO)2*4H2O, MnCO3, or MnO2. Assuming an average food consumption of 13% of
body weight, the average daily dose from the control diet was approximately 17 mg Mn/kg-day,
while the average daily dose from the manganese enriched diet was 276 mg Mn/kg-day. The
duration of treatment was 100 days. The mice were tested for spontaneous motor activity after
30 days. Blood and tissues were analyzed at the termination of the experiment. No significant
difference in food intake among groups was seen. Body weight gain and red and white blood cell
count was decreased in groups that received Mn(CH3COO)2«4H2O or MnCl2. Motor activity was
reduced in the MnCO3 group. Tissue manganese concentrations in groups receiving supplemental
External Review Draft — Manganese—April 2002
7-15
-------
manganese was 2 to 3 times that of controls. A LOAEL of 276 mg Mn/kg-day was identified in
this study'based on decreased weight gain and hematolbgical effects.
Par enteral Exposure
Suzuki et al. (1975) administered 250,500, or 1,000 mg of MnO2 in saline to monkeys
(Macaco mullatd) by subcutaneous injection. Injections were given once a week for 9 weeks.
The study authors reported a body weight of 4 kg for monkeys used in the study. Estimated time-
averaged doses correspond to 5.6,11, a^id 23 mg Mn/kg-day. At autopsy, manganese-treated
monkeys had irregular arrangement of hepatic i cords and lymphocytic infiltration.
7.2.4 Neurotoxicity '
Occupational studies of miners, industrial workers, and agricultural workers have
established injury to the central nervous system as the chief health effect associated with inhalation
exposure to manganese. High level exposure by this pathway typically results in a suite of
neurological effects collectively termed manganism. Chronic manganism associated with
inhalation exposure is characterized by an extrapyramidal syndrome with symptoms that are
somewhat similar to those observed in Parkinson's disease. One characteristic difference is the
"cock-walk"of the manganism patient, in which the patient'walks on his toes with his spine erect
and elbows flexed. Further, manganism patients do not often exhibit the "resting tremor" that
Parkinson's patients do, and they have a propensity for losing their balance and falling backwards.
The clinical course of manganism occurs in three phases: an initial phase of subjective and
nonspecific symptoms; an.intermediate phase pf evolving neurological symptoms related to
speech, dexterity, facial expression, and movement; and an established phase characterized by -
persistent, often irreversible neurological deficits (Chang, 1996). While MRI scans of the brains
of humans and non-human primates exposed to excess manganese indicate 'that the metal deposits
in the globus pallidus and to a lesser extent in the substantia nigra, degenerative lesions are limited
to the globus pallidus (Calne et al. 1994). An important question in the evaluation of health
effects associated with manganese in drinking water is whether similar neurotoxicological effects
occur following exposure by the oral route, j
i
Oral Exposure
\ '
Table 7-4 summarizes studies of the neurotoxic effects of manganese exposure. A single
study exists for evaluation of manganese exposure in primates by the oral route. Gupta et al.
(1980) administered 25 mg MnCl2«4H2O/kg orally to four male rhesus monkeys daily for 18
months. This level is equivalent to an average^ daily dose of 6.9 mg Mn/kg-day. Animals were
maintained on monkey pellets, two bananas/day, and tap water. The monkeys developed
muscular weakness and rigidity of the lower limbs. Histological analysis revealed degenerated
neurons in the substantia nigra and scanty neuromelanin granules in some of the pigmented cells.
i
Bonilla and Diez-Ewald (1974) noted mat chronic exposure of rats to manganese chloride
produces a marked decrease in brain biogenic amines, particularly dopamine.
External Review Draft •— Manganese—April 2002
7-16
-------
Singh et al. (1979) administered manganese (16 mg/kg in a 10% sucrose solution) alone or
in combination with ethanol to groups of 20 male albino rats for 30 days. Exposure to manganese
alone led to a 72% increase in manganese concentration in the brain (3.13 |ig/g dry weight versus
1.82 u.g/g for controls). This outcome was hot altered by ethanol exposure. There were no
morphologic changes in the brain tissue of any group. Significant alterations in activity were
reported for several brain enzymes. Manganese exposure resulted in significant increases in
monoamfme oxidase (p < 0.001), adenosine triphosphatase (p < 0.001), ribonuclease (p < 0.001),
and glutamate-oxaloacetate transaminase (= aspartate aminotransferase, or AST; p < 0.001).
Significant decreases were observed for succinic dehydrogenase (p < 0.02 and deoxyribonuclease
(p < 0.001). Concurrent exposure to eihanol resulted in a synergistic effect with some enzymes
and an antagonistic effect with others. No mechanism was proposed to explain the pattern
observed in the presence of ethanol.
Chandra et al. (1979) evaluated the neurological effects of manganese in mice exposed
from birth. Neonatal mice were initially exposed by nursing from dams given 5 mg/mL MnCl2 in
their drinking water. After weaning at 25 days, the mice received manganese in their drinking
water. Average exposures to manganese were determined to be 0.030 mg Mn/day for 60 days,
0.036 mg Mn/day through the 90th day, 0.075 mg Mn/day through the 120th day and 0.090 mg
Mn/day for the interval between 150 and 180 days. Assuming a body weight of 0.03 kg at
adulthood, the average daily dose at the termination of the experiment was approximately 3 mg
Mn/kg-day. Elevated levels of striatal dopamine, norepinephrine, and homovanillic acid were
observed, at 60 and 90 days of,age, with a concomitant increase in spontaneous locomotor
activity. Exposure past 90 days did not influence motor activity. Chandra et al. (1979) proposed
that the hyperactivity observed in these mice was an early behavioral effect of excess manganese
exposure that resulted from elevated dopamine and norepinephrine levels. The study authors
further suggested that the observed hyperactivity may be comparable to the psychomotor
excitement observed in the early stages of human manganism.
Gray and Laskey (1980) found that dietary exposure to 1,100 mg/kg manganese (as
Mn3O4) in rats for 2 months produced only reduced reactive locomotor activity. Assuming a body
weight of 0.35 kg, this level of exposure corresponds to an average daily dose of 55 mg Mn/kg-
day. Deskin et al. (1980) studied neurochemical alteration induced by manganese chloride in
neonatal CD rats. Rats were intubated with daily doses of 1,10, or 20 mg Mn/kg-day from birth
to 24 days old. Neurochemical components were subsequently analyzed in the hypothalamus and
corpus slriatum. Administration of 10 and 20 mg Mn/kg-day resulted in significantly elevated
manganese concentrations in both regions, but neurochemical alterations were observed only in
the hypothalamus. These alterations included a decrease in dopamine concentration and turnover.
The highest dose of manganese also resulted in a significant decrease in hypothalamic tyrosine
hydroxylase activity, and an increase in monoamine oxidase activity. Visible signs of toxicity
were not observed in any group.
Deskin et al. (1981) intubated rats with daily doses of 10,15 or 20 mg MnCl2«4H2O/kg
from birth to 24 days old. The authors reported a significant elevation of serotonin levels in the
hypothalamus, but not the striatum, following exposure to 20 mg Mn/kg.
External Review Draft — Manganese—April 2002
7-17
-------
CUD
C
at
O)
1
i
CJ
•a
o
I
I
i
02
H
Reference
42
8
w
CO
5
-
o
B
o>
JS
u
0
M
1
"3
J2
S
C/9
•a
1
A
Duration
I
•e
a
s
g.
B
o
•~
CO
CO
1
VO
_s< ,-< _S"3J-
^5 O *5 O
^ a J s
SH C^Tf 5C-
.a
d1
^Q
S
s
1
oo
^_l
Morganti et al. (.
1^1
CO
,
1
I
f
1— 1
1
<5
,*-3
S
S
o
1
[2 •
J.
Bonilla and Diez
(1974)-....-
t
I
i
_l_
CO
I
g !
7 months
i
o
f i
«0
bO |
111
|
(S '
0 J
C i
S ;
i
1
gs
oo
Deskin etal. (19
CO
CO
P5
^ 1
S-S
"7
""< 60
-T'S-
W)
1
tf
^3
s
1
^.
oo
Deskin etal. (19
CO
CO
&
m •$
J
o" »
t-H S
4)
60
1
O
a?
^
A
0
^H
S
1
at
i—i
•i
Chandra and Shi
(1981)
~
C/3
+
w
1
1
ts
ll
— "j-
bO
'§ i-
ll
a?
"s*"
A
o
B
S
1
^
t-H
OO
"3
ts
T
i
CO
28 months
L
1 1
a iif
^-< -^i-
bO
'§ t~<
ll
O
1
TJ"
A
D
B
2
1
^^
T-H
oo
T— t
t>
•a
~T"
§
»3
U CO
0 ^»
0
IPs!
— « •<*
00
.9
if
a?
**T
€>
(N
O
JB
s
1
OJ
oo
1— t
•a ••
•s
•s
'
•
CO
1
d*^.
60 Cj^
'-i ^
00
'§ >-l
II
d"q
^3 3!!!!
2 •*
1
oo
-------
-8
I
O
-—
A
cs
H
Reference
I
CNSEffi
1
I
o>
1
"2
Histologi
he
o
£
cu
pa
a •
Duratio
P
1
g4
a
cu
o
0
£
"3
CO
oo
•a
•s
•§
+
•S
.+
1
a»
o
•-I
0
1
0
H
§
1
^
00
o
Nachtman et al. (1
-
00
55
00
+ •
1
0
Jj
1 1
— < -^r
g,.
1.1
O
tcf
•,
^ 60
,
"S
j?
SJ
O 2
&
g
00
. •
u o,
hS ^fl
A •*
1
Singh etal. (1979;
+ ^
00
55
00
55 ..
1
o
bo
I
«o
*—H
txa
2 1
"8
Ifl
'o
Q>
0
55
1
y— N
Gupta etal. (1980
00
55
+
+
a
1
oo
r-f
es
O
2 °"
ft
CN Tf
4*->
i
' 0
w
^S"
•
0
JH
s
&•
w
o
s
-------
Chandra and Shukla (1981) exposed male albino rats to 1,000 mg/L MnCl2-4H2O (436 mg
Mn/L) in drinking water. Assuming water consumption of 0.049 L/day and an average adult body
weight of 0.35 kg, this level of exposure corresponds to an average daily dose of 61 mgMn/kg-
day. Levels of catecholamines, homovanillic acid, manganese, and the activity of monoamine
oxidase were determined in the corpus striatum at time intervals up to 360 days. The
investigators found initial increases in dopamine, norepinephrine, and homovanillic acid levels.
This initial increase was followed by a period of normal levels. After 300 days, a decrease in all
levels was observed. These changes were not correlated with the tissue concentration of
manganese. The authors suggested that the decreased locomotor activity observed during later
periods of manganese exposure may be related to lowered dopamine and norepinephrine levels in
the brain, and that this stage of chronic toxicity may correspond to-the later neurologic phase of
motor dyskinesia in humans. Ali et al. (1981) conducted concurrent behavioral studies, and
found an initial increase in spontaneous locomotor activity followed by a decrease during later
periods of manganese exposure. I
Lai et al. (1981) exposed female Wistar rats to 1,000 mg/L MnCl2»4H20 (280 mg Mn/L)
in drinking water. Exposure was initiated at mating. Pups were exposed in utero by
administration of manganese in drinking water to dams via maternal milk during nursing, and by
inclusion in drinking water after weaning. Groups of rats were exposed to manganese for over 2
years and were either 2 months or 24 to 28 months of age at examination. Assuming a body
weight of 0.35 kg and water consumption of 0.049 L/day, the average daily dose forrats at
adulthood was approximately 39 mg/kg-day. The brains were dissected and analyzed for activity
of glutamic acid decarboxylase (GAD), choline acetyltransferase (ChAT), and acetylcholinesterase
(AChE). GAD, ChAT, and AChE are neurochemical markers for the GABA and cholinergic .
systems, and had previously been implicated in manganese toxicity (Sitaramayya et al., 1974;
Bonilla, 1978a, b). Adverse effects of chronic manganese exposure on the activity of GAD,
ChAT, or AChE were not apparent in 2-month-old rats. The study authors reported that
lifetime exposure to manganese produced effects that counteracted age-related decreases in GAD,
ChAT, and AChE. \
\
Leung et al. (1981) analyzed the same groups of rats used by Lai et al. (1981) for
monoamine oxidase (MAO) activity. MAO is a key enzyme in oxidative degradation of
neurotransmitter amines. The only effect observed following exposure of 2-month-old rats to
manganese was a small decrease in the neurotransmitter serotonin in the cerebellum. No
significant differences were observed in manganese-treated 24- to 28-nionth-old rats.
I
Lai etal. (1982a) examined the effectsiof manganese exposure on male Wistar rats. The.
rats Were initially exposed to manganese in utero. Following weaning, the rats were exposed to
1,000 mg MnCl2-4H2O/L (280 mg Mn/L) in drinking water for either 70 to 90 days or 100 to 120
days after birth. Assuming an adult weight of 0.35 kg, and water consumption of 0.049 L/day,
this level corresponds to a dose of approximately 39 mg/kg-day. Levels of dopamine,
noradrenaline, serotonin, and choline were determined. A significant decrease was seen in the
uptake of dopamine by synaptosomes isolated from the hypothalamus, striatum and midbrain in
70- to 90-day-old rats. No effect was observecl in the 100- to 120-day-old rats. Choline levels
External Review Draft'— Manganese—April 2002
7-20
-------
were higher in 70- to 90-day-bld-exposed rats and lower in 100- to 120-day-old-exposed rats
when compared with controls. The authors suggested that this finding may reflect involvement of
both the dopaminergic and choliriergic systems in manganese toxicity. They concluded that,
although the rat may not serve as an ideal model for understanding the neurotoxic effects of
manganese, neurochemical effects are discernible when analyses are made at the appropriate
period.
Lai et al. (1982b) investigated the effect of manganese exposure on the developmental
profile of acetylcholinesterase activity in different regions of the brain. Female Wistar rats were
exposed to manganese chloride tetrahydrate provided in drinking water at a concentration of
1,000 mg/L. Exposure was initiated at conception. Male offspring were weaned onto drinking
water containing 1,000 mg/L manganese chloride tetrahydrate and exposed for up to 60 days.
Enzyme activity in the cerebral cortex, striatum, midbrain, pons and medulla, hypothalamus, and
cerebellum was determined at 5,12,20, 30, and 60 days after birth. The developmental profile of
the enzyme differed in the various regions. Activity was detected earlier in the more caudal
regions, except in the cerebellum where there was no increase. Exposure to manganese from
conception did hot influence the developmental profile of acetylcholinesterase activity.
Gianutsos and Murray (1982) studied changes in the concentrations of dopamine and
GABA in mice exposed to MnCl2 in the diet. A 1% concentration of MnCl2 was administered in
the diet to an unspecified number of male CD-I mice for 1 month. This level of exposure
corresponds to 568 mg Mn/kg-day. The concentration was increased to 4% for an additional 5
months. During this period, the average daily dose was 2,272 mg Mn/kg-day. Dopamine content
in the striatum and in the olfactory tubule at 6 months was reduced compared with controls (p <
0.05), GAB A content of the striatum was increased (p< 0.05). Apparent increases in the
substantia nigra area and a decrease in the cerebellum were not statistically significant. No
changes in neurotransmitter levels were observed when assays were conducted after 1—2 months
of exposure.
Morganti et al. (1985) conducted a behavioral study using male ICR strain Swiss mice.
The mice were fed powdered Charles River's RMH 300 diet that contained 1,000 mg MnO2/kg.
This dietary concentration corresponds to approximately 632 mg Mn/kg. The mice consumed 5 g
of food daily. Assuming a body weight of 0.03 kg (U.S. EPA, 1986d), this level of exposure
corresponds to an average daily dose of 105 mg/kg-day. Neurobehavioral evaluation began after
16 weeks of feeding and continued at 2-week intervals for 30 weeks. The endpoints evaluated
were open field and exploratory behavior, passive avoidance learning, and rotarod performance (a
measurement of balance and coordination). Multivariate analysis of variance (2 treatments and 8
samples by week of exposure) was used to test for intergroup differences. No significant
behavioral differences were apparent in any treatment group. In contrast, Morganti et al. (1985)
observed significant effects in mice exposed to manganese by inhalation for 7 hours/day, 5
days/week, at levels greater than 50 mg Mn/m3. The duration of exposure was 16 to 32 weeks.
This level of inhalation exposure was considered by the authors to be comparable to the oral
exposure.
External Review Draft—Manganese—April 2002
7-21
-------
Ali et al. (1985) studied the effect of dietary protein on manganese neurotoxicity. Rats
received either a normal diet (21% casein) or a low protein diet (10% casein). Half of each
dietary group served as a control while the other half received MnCl2«4H2O (3,000 mg Mn/L) in
the drinking water for 90 days. Assuming an adult body weight of 0.35 kg and water
consumption of 0.049 L/day, this corresponds to an average daily dose of 420 mg/kg-day. The
low-protein diet was associated with decreased levels of brain dopamine (DA), norepinephrine
(ME), and serotonin. Manganese exposure resulted in a marked increase in DA and NE levels,
which were more pronounced in the low-protein group. A significant decrease in serotonin levels
following manganese exposure occurred only in the low-protein group. Weaned Fj pups of
treated rats exhibited the same pattern of effects. The study authors concluded that protein
deficiency can increase vulnerability of rats to jthe neurotoxic effects of manganese.
Nachtman et al. (1986) studied the behavioral effects of chronic manganese exposure.
Male Sprague-Dawley rats were administered 0 or 1 mg MnCl2«4H2O/mL in drinking water for
65 weeks. Assuming a body weight of 0.35 kg and water consumption of 0.049 L/day, this
corresponds to an average daily dose of 39 mg Mn/kg-day. The treatment did not result in any
change in body weight. The manganese-exposed rats exhibited a significant increase in locomotor
activity during weeks 5 to 7. However, the effects were transient, and by 8 weeks the activities
had returned to control levels. Treated rats examined at 14 and 29 weeks were found to be more
responsive to the effects of J-amphetamine (a locomotor stimulant that works primarily by
releasing dopamine) than were controls. There was no difference between manganese-treated rats
and controls at 41 or 65 weeks. The investigators concluded that manganese exposure may result
in a transient increase in dopaminergic function, as evidenced by increased spontaneous and d-
amphetamine-stimulated locomotor activity.
Kristensson et al. (1986) studied the effect of manganese on the developing nervous
system of young rats. Starting at 3 days of age, Sprague-Dawley rats received a daily dose of 150
mg Mn/kg-day (as MnCy by gavage. The treatment continued until the rats reached 44 days of
age. At days 15 to 22 there was a large but transient increase (7- to 40-fold) of manganese in the
brain, and the rats displayed a rigid and unsteady gait. By 44 days, the rats appeared normal and
brain manganese levels had declined to approximately 3 times the control level. Histological
analysis revealed no abnormalities in the brains of manganese-exposed rats. Axonal growth and
the axon-myelin relation were normal. A second group of rats was treated for 15 days. At fids
time point, half the rats were sacrificed and half were maintained untreated until sacrifice at 60
days of age. The rats were subsequently analyzed for brain content of dopamine and its
metabolites, including 2,4-dihydroxyphenylacetic acid and homovanillic acid (HVA), and
serotonin and its major metabolite 5-hydroxyindolacetic acid. Only HVA levels in the
hypothalamus and striatum were affected by manganese treatment. Significantly decreased HVA
levels were seen at the 15-day sacrifice. Similar decreases in rats treated for 15 days and allowed
to recover until 60 days of age were not observed. The investigators concluded that divalent
manganese has a very low degree of toxicity for the developing nervous system in rats, but that
longer-term exposure to more active manganese compounds may cause severe damage to certain
neurologic pathways. In addition, the investigators emphasized that rodents may not be
External Review Draft — Manganese—April 2002
7-22
-------
appropriate for comparison with primates. Unpublished studies, where monkeys exposed to
manganese oxide developed severe motor disturbances, were cited as the basis for this conclusion.
Eriksson et al. (1987) studied the effect of long-term manganese exposure on biogenic
amine levels in rat brain. Starting at 20 days of age, groups of male Sprague-Dawley rats were
provided with drinking water containing 10 g/L manganese chloride (MnCl2«4H20) for 60, iOO,
165, or 265 days. This concentration corresponds to 2,777 mg Mn/L. Assuming an adult body
weight of 0.35 kg and water consumption of 0.049 L/day, this level of exposure results in an
average daily dose of approximately 390'mg Mn/kg-day. There were no clinical signs of toxicity.
Following 60 days of exposure, manganese concentration in the striatum was estimated to be 1.3
to 2.0 mg/kg, in contrast to control levels of 0.4 to 0.5 mg/kg. Levels of dopamine, 3,4-
dihydroxyphenylacetic acid, homovanillic acid, serotonin and 5-hydroxyindoleacetic acid were
determined in discrete regions of the caudate-putamen. Rats exposed for 60 and 165 days
showed significantly increased levels of dopamine (DA) and 3,4-dihydroxyphenylacetic acid, but
these alterations were not seen in rats exposed for 100 or 265 days. This suggests an increased
synthesis and turnover of dopamine that is reversible, even with continuous manganese exposure.
This study identifies a LOAEL of 390 mg Mn/kg-day based on increased levels of dopamine at 60
days.
Kontur and Fechter (1988) intubated neonatal Long-Evans rats daily with 0,25, or 50
mg/kg-day manganese chloride (MnCl2 • 4H20) for 14 or 21 days. These doses correspond to 6.9
and 13.9 mg Mn/kg-day. The level of manganese in the brain was increased at both 14 and 21
days, but was greater at 14 days. Monoamine and metabolite levels were not altered in any brain
region by manganese treatment. The study authors^suggested that the different results reported
by different laboratories may be attributable to species or strain differences, the dosing regimen or
vehicle, the route of administration, or the time points chosen for testing. These data suggest a
NOAEL of 6.9 mg Mn/kg-day for this study, based on the absence of effect on monoamine levels.
These collective studies suggest that preclinical neurological effects are possible in the
human following oral exposure; however, there are dissimilarities in the spectrum of responses
between rodent and primate models of toxicity that preclude a determination of the oral dose
range that might be expected to induce these preclinical effects. Further, conflicting data
concerning responses in humans and confounding factors in the limited human epidemiological
studies prevent determination of any dose-response effect in humans exposed to manganese
excesses via ingestion.
Par enter al Exposure
Although deficiencies exist in experimental design (U.S. EPA, 1984), primate studies by
parenteral routes of exposure have reported extrapyramidal signs and histologic lesions similar to
those described in humans. Mella (1924) treated four rhesus monkeys with MnCl2 for 18 months.
Two monkeys served as controls. The treated monkeys received gradually increasing doses of
MnCl2 by intraperitoneal injection on alternate days. The doses started at 5 mg and reached a
maximum of 25 mg per injection. The monkeys exhibited uncontrolled, involuntary movements
External Review Draft — Manganese—April 2002 7-23
-------
(chorea) followed by rigidity, disturbances of inotility, fine hand tremors, and finally, confaacture
of the hands. Histological changes were reported in the putamen, the caudate, and the globus
pallidus. , ;
Suzuki et al. (1975) exposed monkeys subcutaneously to 39.5, 79.0 or 158.0 mg Mn/kg as
MnO2 once a week for 9 weeks and found the latency of neurologic signs (tremors, excitability,
choreifonn movement, loss of equilibrium, anci contracture of hands) inversely related to
cumulative dose. Signs appeared earlier when higher doses were administered, but the severity of
symptoms was not completely dose-related. Ijhe estimated daily doses in this experiment were
5.6, ll,and23mgMn/kg-day. |
i •
Olanow et al. (1996) reported damage to the globus pallidus and substantia nigra in
monkeys that were dosed intravenously with doses as low as 4.4 mg Mn/kg/week (for 7 weeks).
The brain damage was accompanied by neuromuscular toxiciry including bradykinesia, rigidity,
facial grimacing, and abnormal posturing of the limbs. Newland and Weiss (1992) administered
repeated daily intravenous doses of manganese to Cebus monkeys so that the monkeys received
cumulative doses of 5 or 10 mg/kg for 450 days. The dosings were separated by at least one
week. The authors observed that single intravenous doses of 5 or 10 mg Mn/kg-day resulted in a .
significant increase in the number of incomplete responses of dosed monkeys to a spring-loaded
test device that measured physical exertion through a rowing motion. The increase in incomplete
responses occurred within a few days after dosing began. Further, action tremor was observed in
the monkeys who had received cumulative doses of 40 mg/kg or higher; however, dystonia was
never observed. ;
Eriksson et al. (1992) subcutaneously injected two monkeys with 0.4 g doses
(0.253 g Mn) in water. Eleven doses were administered over 4 months, followed by a final dose
at 12 months. Both animals developed an unsteady gait and exhibited hypoactive behavior. PET
scans indicated that degeneration of dopamineirgic nerve endings occurred following Mn
intoxication.
Additional studies have examined the neurotoxic effects of manganese administered by
parenteral routes in non-primate species. Mustafa and Chandra (1971) and Chandra (1972)
reported paralysis of the hind limbs in rabbits administered 169 mg Mn/kg (as MnO^
intratracheally. The paralysis developed after a period of 18 to 24 months. Examination of the of
affected animals brains showed widespread neuronal loss and neuronal degeneration in the
cerebral cortex, caudate nucleus, putamen, substantia nigra and cerebellar cortex. These findings
are reminiscent of the characteristic histopathologic and neurologic consequences of manganism
found in exposed workers (U.S. EPA, 1993). A marked decrease in brain catecholamine levels
and related enzyme activity was also noted.
Histopathologic evaluations of exposed rats by Chandra and Srivastava (1970), Chandra
et al. (1979) and Shukla and Chandra (1976) found scattered neuronal degeneration in the
cerebral and cerebellar cortex. Daily intraperitoneal administration of 2 to 4 mg Mn/kg for <; 120
days appeared to be the threshold for the appearance of microscopic lesions. These studies also
External Review Draft — Manganese —• April 2002
7-24
-------
demonstrated an association between the maximum number of degenerated neurons and
maximum manganese concentration in the brain.
Scheuhammer (1983) treated male Sprague-Dawley rats intraperitoneally for 30 days with
either 3.0 mg Mn/kg or an equal volume of 0.9% NaCl. Assuming an average adult body weight
of 0.35 kg, this treatment corresponds to an average daily dose of 8.6 mg Mn/kg-day. Following
sacrifice, the pancreas was removed, fixed in 10% buffered formalin, and subsequently processed
for light microscopy. Significant pathological changes were observed in pancreatic tissue from
manganese-exposed rats. The changes were characterized by a pancreatitis-like reaction
consisting of expanded interacinar spaces, a thickened connective tissue capsule with
imaginations of fibrotic connective tissue septa extending into the body of the gland, the presence
of an inflammatory infiltrate of neutrophils, lymphocytes, macrophages, and the separation of
groups of acini from the body of the pancreas with occasional destruction of acinar cells. Other
peritoneal organs did not exhibit pathological changes. This study suggests that intraperitoneally
injected Mn(H) exerts a selective toxicity on pancreatic tissue. Therefore, the study author
cautioned against use of intraperitoneal injection as the route of administration for chronic Mn
neurotoxicity studies.
Brouillet et al. (1993) administered 0, 0.5,1, or 2 umol of MnCl2 in deionized water to
male rats by a single intrastriatal injection. Assuming a body weight of 0.35 kg for an adult rat,
these doses correspond to 0, 0.077,0.171, and 0.314 mg Mn/kg. Each treatment group
contained 9 to 10 rats. The lowest dose produced a significant reduction in dopamine, but had no
effect on the other neurochemical markers examined. Doses of 0.171 and 0.314 mg Mn/kg
produced a reduction in dopamine levels, changes in neurochemical markers, and indications of
impaired oxidative metabolism.
7.2.5 Developmental/Reproductive Toxicity
Developmental Studies
Studies are limited regarding developmental toxicity in humans following oral exposures
to manganese. Kilburn (1987) reported an increased incidence in birth defects and stillbirths in a
small population of indigenous peoples in Groote Eylandt, Australia. Although the area was rich
in manganese deposits and ingestion of excess amounts of the metal was suspected, the study
suffered from a lack of exposure data, small sample sizes, and no suitable control group. Further,
inhalation exposure to manganese could not be ruled out. Studies by He et al. (1994) and Zhang
et al. (1995) suggest that oral exposures to excess manganese can possibly result in increased
neurological deficits measured as poorer performance in school and on standardized
neurobehavioral exams. These studies also suffer from a lack of adequate exposure data and the
potential presence of confounding factors, such as exposure to other potential neurotoxicants and
possible inhalation exposure to manganese.
External Review Draft — Manganese—April 2002
7-25
-------
Developmental studies conducted in animals are summarized in Table 7-5. These studies
suggest that manganese is a potential developmental toxicant, but additional studies that are better
controlled are necessary in order to determine how potent it is.
! • •
Several studies have reported developmental effects in animal models following oral
administration of manganese. Jarvinen and AnlstrSm (1975) exposed female rats to 4,24,54,
154, 504, or 1,004 mg Mn/kg (as manganese sulfate heptahydrate) in the diet for 8 weeks after
weaning and during pregnancy. No signs of embryotoxicity or fetotoxicity were observed.
Increases in the whole body content of manganese in fetuses and in liver manganese content of
the dams were reported at dietary levels above 154 mg Mn/kg. No increase in liver manganese
content was observed in non-pregnant females: Chandra and Shukla (1978) administered bolus
doses of 1 mg Mn/kg-day to neonatal rats for 60 days. Neuronal degeneration and increased
monoamine oxidase were reported on days 15 and 30 of the study, but no clinical or behavioral
signs of manganese neurotoxicity were reported.
Several studies have measured changes in brain chemistry in neonatal rats following oral
exposure to manganese. Deskin et al. (1980,1981) dosed rat pups via gavage with MnCl2 in 5%
sucrose for 24 days starting on the first postnatal day. The administered doses in the earlier study
were 0,1,10, and 20 mg Mn/kg-day. Decreased dopamine levels in the hypothalamus were
reported at the two highest doses, and decreased tyrosine hydroxylase levels and increased
monoamine oxidase activity (perhaps due to increased levels of the enzyme) were reported in the
hypothalamus at the highest dose. No other changes in brain chemistry were reported in the
hypothalamus, and no other brain section was affected. In the latter study, doses of 0,10,15, and
20 mg Mn/kg-day were administered. Hypothalamic serotonin was observed to be increased at
the highest dose; the level of this transmitter was unaffected in the striatum. Lai et al. (1984)
reported small decreases in choline acetyltransjferase activity hi the cerebellum and midbrain of 2-
month-old rats that had been exposed to 40 mg Mn/kg-day from conception, throughout
gestation, and throughout life. Other neuronal|enzymes (e.g., glutamic acid decarboxylase,
acetylcholinesterase) were unaffected.
Kristensson et al. (1986) dosed 3-day old male rat pups with 150 mg Mn/kg-day (in
water) for 41 days. The authors reported a transient ataxia (days 15-22), which was resolved by
the end of the dosing period, in the pups. Manganese levels in the blood and brain (brain levels
were increased 7-40 fold) were elevated significantly over controls in 15- and 20-day old pups;
brain levels had decreased to approximately 3-fold over control levels in 43-day old pups.
Homovanillic acid (metabolite of dopamine) cpncentrations were decreased in the striatum and
hypothalamus, but not in other brain regions; no other monoamines or their metabolites were
affected.
External Review Draft — Manganese—April 2002
7-26
-------
Table 7-5. Developmental Effects of Exposure to Manganese.
Compound
MnSO4
•7H20
MnCl2
MnCl2
Mn3O4
MnCl2
Mn3O4
MnCl2
Mn3O4
MnCl2
Species
Rat
(female)
Rat
Rat
Mouse
(Male)
Rat
Rat
Rat
Rat
Rat
Route
Oral
(diet)
Bolus
(in water)
Oral
(gavage)
Oral
(diet)
Oral
(gavage)
Oral
(diet)
Oral
(drinking
water)
Oral
(drinking
water)
Oral
(drinking
water)
Dose
(n»g
Mn/kg-
day)
0
4
24
54
154
504
1,004
1
0
1
10
20
1,050
0
10
15
20
0
350
1,050
3,500
240
40
0
68
136
232
Effect
Increased manganese
concentration in fetus and
maternal liver; no indications of
embryo- or fetotoxicity
Neuronal degeneration; increased
monoamine oxidase; no
indications (clinical or behavioral)
of neurotoxicity
Decreased dopamine; decreased
tyrosine hydroxylase; increased
monoamine oxidase activity
(all changes in hypothalarnus
only)
Decreased preputial gland,
seminal vesicle, and testes growth
Increased hypothalamic serotonin
Decreased serum testosterone;
decreased sperm count; decreased
testes weight; prevented normal
decrease in serum FSH
Delayed air righting reflex;
delayed age of eye opening;
delayed development of auditory
startle
Decreased chloline
acetyltransferase activity in
cerebellum and midbrain
Decreased water consumption and
decrease in weight gain in two
highest dose groups; no changes
in catecholamine or startle
response in the exposed pups
Reference
Jarvinen and
Ahlstrdm
(1975)
Chandra and
Shukla(1978)
Deskin et al.
(1980)
Gray and
Laskey (1980)
Deskin et al.
(1981)
Laskey et al.
(1982)
Ali et al.
(1983)
Lai et al.
(1984)
Konturand
Fechter(1985)
External Review Draft — Manganese—April 2002
7-27
-------
Table 7-5. Developmental Effects of Exposure to Manganese (continued).
Compound
Mn3O4
MnCl2
MnCl2»
4H20
MnCl2
MnCl2
MnCl2
MnCl2
MnCl2
MnCl2
Species
Rat
Rat
(male)
Mouse
Rat
(female)
Rabbit
(female)
Mouse
Rat
Mouse
Rat
Rat
Route
Oral
(gavage)
Oral
(drinking
water)
Subcutaneous
injection
Oral
(gavage)
Subcutaneous
injection
Intravenous
Intravenous
Oral
(drinking
water)
Oral'
(drinking
water)
Dose 1
(mg
Mn/kg-
day)
i
0 i
71
214
150
I
0
0.56 ;
1.1
2.2
4.4
0
11
22
33
50
|
1
I
0
0.27 x 1C'3
1.1 xiO'3
2.2 x 10-3
0.3 i
1.6
I
0
350
1,420 :
11
22
Effect
Decreased serum testosterone
following 7 days of hCG induction
Transient ataxia; decreased
striatal and hypothalamic
homovanillic acid concentrations
Decreased weight gain/food
consumption; increased late
resorptions; reduced fetal body
weight; increased incidence of
morphological defects
Delayed skeletal and internal
organ development and increased
external malformations in rat pups
delivered by Caesarean section.
No effects in rabbit
Late resorptions; postimplantation
loss; skeletal anomalies; reduced
fetal body weight
Increased incidence of skeletal
malformations including
angulated or irregularly shaped
clavicle, femur, fibula, humerus,
ilium, radius, scapula, tibia,
and/or ulna
Increased fetal weight at low dose;
decreased fetal weight at high
dose; fetal skeletal abnormalities
at high dose
Thinning of cerebral cortex;
absence of convincing brain
histopathological or behavioral
evidence from perinatal
manganese exposure on the brain
Decreased body weight gain;
increased response to auditory
stimulus
Reference
Laskey et al.
(1985)
Rristensson et
al. (1986)
Sanchez et al.
(1993)
Szakmary et
al. (1995)
Colomina et
al. (1995)
Treinen and
Blazak(1995)
Grant et al.
(1997)
Pappas et al.
(1997)
Dorman et al.
(2000)
External Review Draft -r- Manganese—April 2002
7-28
-------
A few studies have measured reproductive endpoints in the developing rodent.
Manganese administered to pre-weanling male mice at a dose of 1,050 mg Mn/kg-day beginning
on postnatal day 15 resulted in the decreased growth of reproductive organs (preputial gland,
seminal vesicle, and testes) measured on days 58, 73, and 90 but did not affect body growth or
liver or kidney weights (Gray and Laskey, 1980). Laskey et'ai. (1982) administered dietary
manganese at doses of 0, 350,1,050, and 3,500 mg Mn/kg-day to male and female rats fed a diet
either adequate or deficient hi iron. Males and females were mated during days 90-100 of the
study; testes weights of male offspring fed the iron-deficient diet were decreased as compared to
controls at day 40 at the highest two doses and at day 100 at the intermediate dose. While 40-
day-old weanling rats did not exhibit any treatment-related hormonal changes, exposed rats
showed a dose-related decrease in serum testosterone at 60-100 days of age (when age-related
increases were expected), and no increase in serum luteinizing hormone was observed. The
normal decrease in serum follicle stimulating hormone (FSH) from 60 to 100 days was prevented
by manganese exposure. Epididymal sperm count was decreased by the treatment only when
given with the iron-deficient diet.
In an additional study measuring the effects of manganese exposure on the developing
reproductive system, Laskey et al. (1985) administered 0, 71, and 214 mg Mn (as Mn3O4)/kg-day
via gavage to pre-weanling rats on postnatal days 1-21. The study authors measured serum levels
of FSHa LH, and testosterone in the pups at 21 or 28 days of age. Manganese exposure did not
affect endogenous or stimulated serum levels of FSH or LH, nor did it affect endogenous or acute
human ehorionic gonadotropin (hCG)-induced serum testosterone at 2 hours. Serum testosterone
was decreased following 7 days of hCG induction, however. The delayed decrease in
testosterone was hypothesized by the study authors to be a result of an unknown manganese-
induced effect .on the Leydig cell.
AH et al. (1983) evaluated potential changes in developmental endpoints in rat pups after
administering excess manganese hi drinking water to pregnant dams fed a normal or low-protein
diet. Manganese exposure was started 90 days prior to mating and continued throughout
gestation and nursing. The offspring of dams who had ingested 240 mg Mn/kg-day exhibited
delayed, air righting reflexes. Significant delays hi the age of eye opening and the development of
auditory startle were reported in pups from dams ingesting protein-deficient diets. No decreases
hi body weight or brain weight were reported hi the offspring of rats fed normal-protein diets.
Kontur and Fechter (1985) exposed pregnant Long-Evans rats to 0, 5,000,10,000, or
20,000 mg/L of manganese chloride hi drinking water throughout the gestation period. Rats hi
the 10,000 and 20,000 mg/L groups showed reduced water intake and a significant decrease hi
weight gain. A significant decrease hi birth weight was observed hi the 20,000 mg/L group. At
one day of age, pups from the 10,000 and 20,000 mg/L groups had increased manganese levels hi
the forebrain, although there was no difference hi the extent of accumulation between the two
groups. The increased manganese levels were not associated with any changes hi catecholamine
function or startle response hi the exposed pups. The authors concluded that manganese is not
particularly toxic to developing rats, perhaps as a result of limited placental transfer.
External Review Draft — Manganese — April 2002
7-29
-------
The developmental effects of manganese have also been evaluated following parenteral
administration. Sanchez et al. (1993) investigated the embryotoxic and teratogenic potential of
manganese during organogenesis. Pregnant Swiss mice received daily subcutaneous injections of
0,2,4, 8 or 16 mg/kg-day of MnCl2-4H2O on days 6 to 15 ,of gestation. These doses correspond
to 0,0.56,1.1,2.2, or 4.4 mg Mn/kg-day, respectively. Dams were sacrificed on gestational day
18. Significant reductions in weight gain and food consumption were reported in dams receiving
8 mg/kg-day and above, and treatment-related deaths were reported at 16 mg/kg-day. A
significant increase in the number of late resorptions was observed at doses of 4 mg/kg-day and
higher, and reduced fetal body weight and an increased incidence of morphological defects; were
reported at doses of 2.2 mg Mn/kg-day and higher. No difference was seen in the incidence of
individual or total malformations in treated groups when compared with controls. A NOAEL of
1.1 mg/kg-day was identified by the study authors for maternal toxicity. A NOAEL of 0.56
mg/kg-day was identified for embryo/fetal toxicity
Pappas et al. (1997) assessed behavioral, neurohistological, and neurochemical endpoints
in rats exposed to manganese from conception to weaning. The investigators administered 0,
2,000, or 10,000 mg Mn/L as manganese chloride in drinking water to female rats (10/group) and
their litters from conception until postnatal day (PND) 30. The average daily consumption of
manganese during gestation was 350 and 1,420 mg/kg-day, respectively, for the two manganese
treatment groups. No effects were observed on pregnancy or birth parameters and no physical
abnormalities were evident in the offspring of treated dams. The findings reflect a lack of effects
on reproductive capability. Fifty male pups from each treatment group were subsampled for
behavioral tests (10 to 22 per group), histopathology (6 to 8 per group) and neurochemical
analyses (6 to 8 per group). The rats exposed to 10,000 mgMn/L showed a 2.5-fold increase in
brain cortical Mn levels. They also experienced reduced weight gain during PND 9 to 32, and
were hyperactive at PND 17. Behavioral tests were conducted on pups from, all groups at PND
17, 90 or 95. No significant differences in performance were noted for the radial arm maze,
elevated plus apparatus, or Morris water maze behavioral tests. Both doses resulted in thinning of
the cerebral cortex. The observed thinning may have been a consequence of either perinatal
malnutrition or a direct effect on cortical development. Brain monoamine levels and choline
acetyltransferase activity were unaffected by manganese exposure. Tyrosine hydroxylase
immunohistochemistry indicated that dopamine neurons of the substantia nigra were intact. Glial
fibrillary acidic protein immunoreactivity, an indicator of neuronal damage, was not increased in
cortex, caudate nucleus or hippocampus. The authors emphasized that the most noteworthy
result of this study was the absence of convincing histopathological and behavioral evidence for
persistent effects of perinatal manganese exposure on the brain.
Grant et al. (1997) failed to observe any effects of manganese exposure on weight gain,
gross malformations, or skeletal malformations in the offspring of pregnant rats dosed via gavage
with 22 mg Mn/kg-day on gestational days 6-17. Another study indicates a lack of persistent
developmental effects from oral manganese exposure during gestation. Szakmary et al. (1995)
reported the developmental effects of manganese administered via gavage to pregnant rats
throughout gestation and to pregnant rabbits through organogenesis (gestation day 6-20) at doses
of 0,11,22, and 33 mg/kg-day. No developmental effects in the rabbit were observed. The
External Review Draft•— Manganese — April 2002
7-30
-------
highest dose resulted in retardation of development of the skeleton and internal organs of the rat,
as well as a significant increase in external malformations (e.g., clubfoot) in pups delivered by
caesarean section. These effects, however, were not observed in 100-day-old offspring of dams
that had been similarly dosed, indicating that the developmental effects were self-correcting.
Manganese treatment did not affect the following endpoints in either the pup group that was
surgically delivered or the group born live: ears, teeth, eyes, forward motion, clinging ability, body
posture, correction reflex, or negative geotaxis reflex.
In a more recent study, Dorman et al. (2000) dosed neonatal CD rats with 11 or 22 mg
Mn (in water)/kg-day for 21 days from birth to weaning. The high dose resulted in decreased
body weight gain in the pups and affected brain neuroehemistry. Manganese treatment induced a
significant increase in the amplitude of response to an auditory stimulus but did not affect motor
activity, performance in a passive avoidance task, or brain histopatholpgy.
Colomina et al. (1995) conducted a study to determine which gestation day is most critical
for developmental toxicity of manganese in mice. The investigators administered a 50 mg/kg dose
of manganese chloride by subcutaneous injection once during the period between gestation days 9
and 12. Late resorptions, post-implantation loss, and skeletal anomalies increased in all treatment
groups. Significant reductions in fetal body weight occurred following exposure on gestation day
9 or 10, indicating these days were most critical.
Treinen and Blazak (1995, abstract only) dosed female Sprague-Dawley rats (15/group)
intravenously with 0, 5,20, or 40 nmol/kg MnCl2 on days 6 to 17 of gestation. These doses
correspond to approximately 0, 0.27,1.1, or 2.2 jig Mn/kg-day. Treatment resulted in an
increased incidence of skeletal malformations (doses which elicited effects were not reported).
The observed malformations included angulated or irregularly shaped clavicle, femur, fibula,
humerus, ilium, radius, scapula, tibia, and/or ulna. - v
Grant et al. (1997) administered 6 or 30 \aaol MnCl2/kg-day to female mice (24/grpup) by
intravenous injection on gestation days 6 to 17. These doses correspond to approximately 0.3
and 1.6 mg Mn/kg-day. The experiment was terminated on gestation day 20. No significant
differences were noted in manganese-treated mice for number of corpora lutea, implantation sites,
pre-or post-implantation losses, or number of viable fetuses per litter. Fetal weight was
significantly increased (p< 0.05) at the lower dose, and significantly decreased (p< 0.05) at the 1.6
mg/kg-day dose. Skeletal abnormalities were noted in the fetuses of dams receiving the higher
intravenous dose. In contrast, no increase in skeletal abnormalities was observed in the fetuses of
mice administered 400 nmol of MnCl2 (approximately 22 mg Mn/kg-day) by oral gavage daily
from days 6 to 17 of gestation.
One in vitro developmental study was located. Hanna et al. (1996, abstract only) cultured
two-stage mouse embryos in media containing varying concentrations of essential and non-
essential minerals, including manganese. Embryos were incubated in culture media containing
0.05-200 uM manganese for 72 hours. Both essential and nonessential minerals were
embryotoxic at relatively low doses.
External Review Draft — Manganese—April 2002 7-31
-------
Reproductive Studies
Some inhalation data from occupational exposure studies suggest that male reproductive
dysfunction is a primary endpoint of manganese toxicity. Toxicity is manifested in symptoms
including loss of libido and impotence (U.S. EPA, 1996a). Some evidence indicates that the
hypothalamus and pituitary are sites of manganese accumulation (see Section 6.2), suggesting
disturbance of the hypothalamic-pituitary-gonadal axis hormones as a potential mechanism for
reproductive effects. No human reproductive data for oral manganese exposure are available in
the current literature. Reproductive studies in animals orally exposed to manganese are described
below. Results of these studies are summarized in Table 7-6.
Chandra and colleagues consistently reported degenerative changes in the seminiferous
tubules in the testes after parenteral exposure of rats and rabbits to manganese (Chandra., 1971;
Shukla and Chandra, 1977; Imam and Chandra, 1975; Chandra et aL, 1973,1975). However,
similar changes were not observed in subchronlc or chronic studies in mice or rats (NTP, 1993).
Gray and Laskey (1980) exposed male mice to 1,100 mg Mn/kg as Mn3O4 in a casein diet
from gestation day 15 to 90 days of age. Assuming a food consumption factor of 0.13 (U.S.
EPA, 1986d), the estimated daily dose at the termination of the study would be approximately
143 mg/kg-day. Sexual development was retarded, as indicated by decreased weight of testes,
seminal vesicles andpreputial glands. Reproductive performance was not evaluated.
Laskey et al. (1982) exposed Long-Evans rats to 0,400,1,100 or 3,550 mg Mn/kg (as
Mn3O4) in the diet from day 2 of mother's gestation to 224 days of age. Assuming a food
consumption factor of 0.05 (U.S. EPA, 1986d), the average daily dose at the termination of the
study was 0,20, 55, or 177 mg Mn/kg-day. The investigators observed a dose-related decrease
in serum testosterone concentration (without a concomitant increase in serum luteinizing hormone
concentration), and reduced fertility at the highest dose. Testes weight, number of ovulations,
resorption and preimplantation deaths, litter size, and fetal weights were unaffected by manganese
exposure.
i
Laskey et al. (1985) conducted studies to assess the effect of manganese on hypothalamic,
pituitary and testicular function. Long-Evans fat pups (8/litter) were dosed by gavage from day 1
to day 21 with a 50% sucrose solution containing particulate Mn3O4. The average daily dose of
manganese was calculated to be 0, 71 or 214 nig Mn/kg-day. Assessments of the hypothalamic,
pituitary, or testicular functions were determined by measuring the endogenous or stimulated
serum concentrations of follicle-stimulating hormone, luteinizing hormone, and/or testosterone at
21 or 28 days of age. Body, testes, and seminal vesicles weight and tissue concentrations of Mn
were also evaluated. Effects attributed to manganese included slight decreases in body and testes
weights, and a reduction in serum testosterone. There was no indication of hypothalamic or
pituitary dysfunction. The authors suggested that the decrease in testosterone level resulted from
manganese-induced damage of Leydig cells.
External Review Draft—Manganese—April 2002
7-32
-------
Table 7-6. Reproductive Effects of Exposure to Manganese.
Compou
nd
Mn3O4
Mn3O4
Mn3O4
MnCl2
MnCl2»
4H20
MnCl2»
4H20
MnSO4
MnO2
Spec
ies
Mous
e
Rat
Rat
Rat
Rat
*Rabbi
t
Rat
Rabbi
t
Rout
e
Oral
(diet)
Oral
(diet)
Oral
(gavag
e)
i.p.
i.p.
i.v.
i.p.
i.t.
Dose
143 mg
Mn/kg-day
20 mg Mn/kg-
day
55
177
71 mg Mn/kg-
day
214
8mg/kg-day
15 mg/kg-day
3.5 mg/kg
6mgMn/kg
250 mg/kg
single dose
Effect
Decreased weight of testes, seminal
vesicles and preputial glands after 90
days.
Dose-related decrease in serum
testosterone concentration. Reduced
fertility at 3550 ppm after 224 days.
Decreased body and testes weights.
Reduction in serum testosterone.
Degenerative changes in approx.
50% of seminiferous tubules after
150 and 180 days.
Increased Mn in testes; decreased
nonprotein sulfhydryls and decreased
activity of glucose-6-phosphate
dehydrogenase and glutathione
reductase after 15-45 days.
Inhibition of succinic dehydrogenase
in seminiferous tubules after 5 days.
Morphologic changes were not
apparent.
Increased Mn in testes after 25-30
days. Degenerative changes in 10%
of seminiferous tubules.
Destruction and calcification of the
seminiferous tubules at 8 months.
Infertile females.
Reference
Gray and
Laskey
(1980)
Laskey et al.
(1982)
Laskey et al.
(1985)
Chandra
(1971)
Shukla and
Chandra
(1977)
Imam and
Chandra
(1975)
Chandra et
al. (1975)
Chandra et
al. (1973)
i.p, = intraperitoneal; i.v. = intravenous; i.t. = intratracheal
External Review Draft—Manganese—April 2002
7-33
-------
Studies exist, however, that report no adverse reproductive effects in female rats following
oral manganese exposure. Pappas et al. (1997) dosed pregnant rats with up to 620 mg Mn/kg-
day (as MnCLj) throughout gestation. No treatment-related effects were reported in dam health,
litter size, or sex ratios of the pups. The study 'did not include more extensive analysis of female
reproductive organs. Grant et al. (1997) administered 22 mg Mn/kg-day (as MnCy via gavage to
pregnant dams on gestation days 6-17. No treatment-related effects were reported in dams as
measured by mortality, clinical signs, food and water intake, or body weights.
7.2.6 Chronic Toxicity
NTP (1993) investigated the chronic toxicity of manganese in a 2-year oral exposure
study. Concentrations of 0,1,500,5,000 or 15,000 mg/kg manganese sulfate monohydrate were
administered in the diet to male and female F344 rats (70/sex). These dietary concentrations
resulted in doses ranging from 30 to 331 mg Mn/kg-day for males, and 26 to 270 mg Mn/kg-day
for females. Ten rats/group were sacrificed at 9 and 15 months. Survival of males in the high-
dose group was significantly decreased starting at week 93 of the study, and death was attributed
to advanced renal disease associated with manganese exposure. Food consumption was similar
for all groups. However, by the end of the study, high-dose males exhibited a mean body weight
that was 10% lower than controls. No clinical'findings or effects on hematologic or clinical
chemistry parameters were attributed to manganese exposure in any group. Tissue concentrations
of manganese were elevated in the livers of mid- and high-dose males, concurrent with a decrease
in hepatic iron concentrations. Renal disease in high-dose males was the only pathological effect
noted. No increases in tumor incidence were attributed to manganese exposure.
The chronic oral toxicity of manganese was evaluated in mice in a concurrent study
conducted by NTP (1993). Concentrations of 0,1,500, 5,000, or 15,000 mg/kg manganese
sulfate monohydrate were administered in the diet to B6C3F! mice (70/sex) in a 2-year oral
exposure study. These dietary concentrations were reported to be equivalent to doses ranging
from 63 mg Mn/kg-day to 722 mg Mn/kg-day for male mice, and from 77 mg Mn/kg-day to 905
mg Mn/kg-day for female mice. Interim sacrifices of 11 mice/group were made at 9 and 15
months. No clinical findings or effects on survival were observed in any group of mice. Mean
body weights of males were not affected. Female mice had a dose-related decrease in mean body
weight after week 37. The final mean body weights for the low-, mid- and high-dose females
were 6%, 9% and 13% lower than controls, respectively. No differences were seen in food
consumption for any group. No effects were reported on hematologic parameters. Tissue
concentrations of manganese were significantly elevated in the livers of all exposed females, and in
high-dose males. Elevated manganese concentration was associated with decreased hepatic iron.
7.2.7 Carcinogenicity
The carcinogenicity of ingested manganese was evaluated in concurrent 2-year oral
exposure studies conducted in mice and rats by NTP (1993). An overview of these studies is
provided below. No other studies of manganese carcinogenicity via the oral route were identified.
External Review Draft—Manganese — April 2002
7-34
-------
Groups of rats were exposed to dietary levels of manganese sulfate monohydrate that
resulted in intakes ranging from 30 to 331 mg Mn/kg-day for males and 26 to 270 mg Mn/kg-day
for females. At the termination of the study, no manganese-related increase in any tumor type
was observed (NTP, 1993).
In a parallel study, NTP (1993) administered 0,1,500,5,000, and 15,000 mg/kg
manganese sulfate monohydrate in the diet to B6C3Fj mice (70/sex) for 2 years. These dietary
concentrations resulted in intakes ranging from 63 to 722 mg Mn/kg-day for males and from 77 to
905 mg Mn/kg-day for females. The estimated manganese intake in the high-dose mice was
approximately 107 times greater than the recommended dietary allowance. Incidence of thyroid
folliculax cell hyperplasia was significantly greater in high-dose male and female mice than hi
controls. The incidence of follicular cell adenomas is summarized in Table 7-7. In males, tumors
were observed only at the highest dose (6% incidence). The highest incidence of tumors in
females was also observed at the highest dose. No significant differences in tumor incidence
relative to the controls were observed for either sex. The follicular cell tumors were seen only at
the termination of the study (729 days), and their number was only slightly increased relative to
the historical control range in female B3C6Fj mice (0 to 9% historical range versus 10% tumor
incidence in high-dose females). Hence, the relevance of these findings to human carcinogenesis
is questionable. The issues of concern are: ,1) the large intake of manganese required to elicit a
response seen only at the end of the study, and 2) tumor frequencies that are not significantly
different from historical controls. While NTP (1993) has concluded that the marginal increase hi
thyroid adenomas of the mice was equivocal evidence of carcinogenicity, others have questioned
the relevance of these data to human carcinogenicity (U.S. EPA, 1993).
Table 7-7. Follicular Cell Tumor Incidence in B6C3F, Mice.
Sex
Males
Females
Concentration of MnSO4»H2O in Diet
Control
0/50
2/50
Low
0/49
1/50
Medium
0/51
0/49
High
3/50
5/51'
Source: NTP (1992)
Other studies reporting positive results for carcinogenicity are summarized in Table 7-8.
Stoner et al. (1976) tested manganese sulfate in a mouse lung adenoma screening bioassay. These
investigators exposed 6- to 8-week-old Strain A/Strong mice of both sexes (10/sex) to 6,15 or 30
mg MnSO4/kg via intraperitoneal injection. Doses were administered three times a week for a
total of 21 injections. The cumulative doses were 132, 330 and 660 mg MnSO4/kg. These doses
corresponded to 42.9,107.2 and 214.4 mg Mn/kg, Observation continued for 22 weeks after the
dosing period, and the mice were sacrificed at 30 weeks. Table 7-9 summarizes the results of this
study. The percentage of mice with tumors was elevated at the highest dose level, but the
difference was not significant (Fisher Exact test) when compared with the vehicle controls. An
apparent increase in the average number of pulmonary adenomas per mouse was noted both at the
External Review Draft — Manganese — April 2002
7-35
-------
Table 7-8. Summary of Carcinogenicity Studies Reporting Positive Findings for
Selected Manganese Compounds'.
Compound
Manganese
chloride-
Manganese
sulfate
Manganese
acety-lacetonate
(MAA)
Species
Mouse
Mouse
Mouse
Rat
Route
i.p.
s.c.
i.p.
i.m.
Dose
0.1 mL
ofl%
0.1 mL
ofl%
0% (control)
660 mg/kg
Omg/kg
1,200
mg/kgb
Omg/kg
Duration
(weeks
intermit-
tent)
26
26
8
26
Results
41% - Lymphosarcomas
67% - Lymphosarcomas
24% - Lymphosarcomas
67% - Lung adenomas
31—37% - Lung ademomas
40% (males)
Fibrosarcomas
24% (females)
Fibrosarcomas
4 % (control males and
females)
Reference
DiPaolo
(1964)
Stoner et al.
(1976)
Furst(1978)
i.p. — intraperitoneal; s.c. = subcutaneous; i.m. = intramuscular
• Source: U.S. EPA (1984)
External Review Draft I— Manganese—April 2002
7-36
-------
Table 7-9. Pulmonary Tumors in Strain A Mice Treated with Manganese Sulfate ».
Total Dose
Group
Untreated
control
Solvent
control
(0.85%
Nad)
Treated
Treated
Treated
20 mg
urethane"
mg MnSO4/kg
0
0
132
330
660
0
mg
Mn/kg
0
0
42.9
107.2
214.4
0
Mortality
1/20
1/20
1/20
0/20
2/20
2/20
Mice with
Lung Tumors
(%)
6/19 (31)
7/19(37)
7/19(37)
7/20(35)
12/18(67)
18/18(100)
Average Number
Tumors/Mouse11
0.28 ±0.07
0.42 ±0.10
0.47 ±0.11
0.65 ±0.15
1.20 ±0.49
21.6±2.81
Yahiec
NA
NA
NS
NS
0.05d
MR
" Source: Stoner et al. (1976)
bX±S.E.
c Student t-test
d Fisher Exact Test p = 0.068
c Single intraperitoneal injection
NA = Not applicable; NS = Not significant; NR = Not reported
middle and high doses, but the increase was significant only at the high dose (660 mg MnSO4/kg)
(Student's t-test, p < 0.5). Although these study results are suggestive of carcinogenic activity,
they do not conclusively meet the positive response criteria (increase in the mean number of
tumors per mouse and an observable dose-response relationship) for the interpretation of lung
tumor data in this mouse strain (Shimken and Stoner, 1975).
Furst (1978) injected F344 rats intramuscularly with manganese acetylacetonate and
observed an increased incidence of fibrosarcomas at the injection site, but did not observe
increased tumor incidence at other sites.
7.3 Other Key Data
7.3.1 Mutagenicity/Genotoxicity
In Vivo Studies
No studies or reports were identified which describe mutagenic or genotoxic effects in
humans following oral exposure to manganese. Table 7-10 summarizes the results of the most
External Review Draft—Manganese—April 2002
7-37
-------
Table 7-10. Genotoxicity of Manganese In Vivo.
Species (test
system)
Compound
End Point
Route
Results
Reference
Nonmammalian
systems:
Drosophila
melanogaster
Drosophila
melanogaster
Drosophila
melanogaster
MnSO4
MnSO4
MnCl2
Sex-linked
recessive lethal
Sex-linked
recessive lethal
i
Somatic mutation
Feeding
Injection
Feeding
Injection
Soaking
larvae
-
-
-
Valencia et
al. (1985)
NTP (1993)
Rasmuson
(1985) .
Mammalian systems: j
Albino rat
(bone marrow cells)
(spertnatogonial cells)
Albino mouse
MnCl2
MhSO4
KMnO4
Chromosomal
aberrations
1
Chromosomal
aberrations
Chromosomal
aberrations
Oral
Oral
Oral
-
+
+
Dikshith and
Chandra
(1978)
Joardar and
Sharma
(1990)
recent in vivo mutagenicity and genotoxicity studies in animals. Results from additional studies
are noted in the text below. i
Studies of genotoxicity in animals have shown mixed results. The bone marrow cells of
rats receiving a 50 mg/kg oral dose of manganese (as manganese chloride) showed an increased
incidence of chromosomal aberrations (30.9%) compared with those of control animals (8.5%)
(Mandzgaladze, 1966; Mandzgaladze and Vasakidze, 1966). However, Dikshith and Chandra
(1978) administered repeated oral doses of manganese chloride (0.014 mg/kg-day) to male rats
for 180 days and did not observe significant chromosomal damage in bone marrow or
spermatogonial cells. >
Joardar and Sharma (1990) administered oral doses of manganese sulfate (approximately
102,202, and 610 mg/kg) and potassium permanganate (65,130, and 380 mg/kg) to male Swiss
albino mice for three weeks. Both compounds were clastogenic, with manganese sulfate being
more potent. The frequencies of chromosomal; aberrations in bone marrow cells and microriuclei
were significantly increased by both salts. There was also a statistically significant, dose-
dependent enhancement of sperm-head abnormalities. A LOAEL of 23 mg Mn/kg-day was
identified for this effect by ATSDR (2000). \
External Review Draft — Manganese — April 2002
7-38
-------
The divalent manganese ion (Mn II) interacts with DNA and chromosomes (Kennedy and
Bryant, 1986; Yamaguchi et al., 1986). In cultured mammalian cells, both MnCl2 and KMnO4
produced chromosome aberrations, including breaks, exchanges and fragments (Umeda and
Nishimura, 1979). DNA-strand breaks have also been induced by manganese in Chinese hamster
ovary calls and human diploid fibroblasts (Hamilton-Koch et al., 1986; Snyder, 1988). Tests for
induction of chromosomal aberrations and sister chromatid exchanges in cultured Chinese hamster
ovary cells were positive for manganese sulfate monohydrate in the absence of S9 metabolic
activation. In the presence of S9, only the sister chromatid exchange test was positive (NTP,
1993).
Tests for mutagenicity in Drosophila melanogaster have given negative results.
Manganese sulfate monohydrate did not induce sex-linked recessive lethal mutations in germ cells
of male Drosophila treated by feeding or injection (Valencia et al., 1985, as reported in NTP,
1993). Treatment of D. melanogaster with manganese chloride by soaking did not induce
somatic mutation (Rasmuson, 1985).
In Vitro Assays
Table 7-11 summarizes the results of the most recent in vitro mutagenicity and
genotoxiciry studies. Additional results from early studies are included in the text below.
Manganese chloride was mutagenic in Escherichia coli (Demerec et al., 1951; Durham
and Wyss, 1957; Zakour and Glickman, 1984), Photobacteriumfischeri (Ulitzer and Barak,
1988) and Serretia marcescens (Kaplan, 1962). Both positive (Nishioka, 1975) and negative
. (Kanematsu et al., 1980) results have been reported for the Bacillus subtilis recombination assay.
Positive (Pagano and Zeiger, 1992; Wong, 1988) and negative results (Wong, 1988) have also
been reported for manganese chloride in the Salmonella typhimurium reversion assay. Assays in
mammalian cell lines were positive for gene mutation in mouse lymphoma cells (Oberley et al.,
1982) and enhancement of transformation in Syrian hamster embryo cells (Casto et al., 1979). An
assay for DNA damage in human lymphocytes gave negative results with metabolic activation,
and positive results without activation (De Meo et al., 1991).
Manganese sulfate gave positive results in the T4 bacteriophage mutation test (Orgel and
Orgel, 1965), and the B. subtilis recombination assay with S9 activation (Nishioka, 1975).
Pagano and Zeiger (1992) obtained positive results for mutagenicity in S. typhimurium strain
TA97. In contrast, manganese sulfate monohydrate was not mutagenic in S. typhimurium strains
TA98, TA100, TA1535, or TA1537, either with or without exogenous metabolic (S9) activation
when assayed by Mortelmans et al. (1986). Results for strain TA97 were negative when assayed
with S9 activation, and equivocal when assayed without metabolic activation. Assays in
eukaryotic test systems were positive for mutagenicity in S. cerevisiae (Singh, 1984) and
chromosomal aberrations in Chinese hamster ovary (CHO) cells (NTP, 1993). Manganese sulfate
gave negative results when assayed for induction of sister chromatid exchange in CHO cells
(NTP, 1993).
External Review Draft — Manganese—April 2002
7-39
-------
Table 7-11. Genotoxicity of Manganese In Vitro.
Species (test system)
Compound
End Point
Strain
Results
With S9
Activation
Without S9
Activation
Prokaryotic
organisms: ;
Salmonella
typhimuriian
Salmonella
typhimuriian
Salmonella
typhimuriian
Salmonella
typhimurium
Photobacterium
fischeri
(bioluminescence test)
Escherichia coll
Bacteriophage (E. coli
lysis)
Bacillus subtilis
(recombination assay)
MnCl2
MnSO4-H2
0
MnSO4
MnCl2
MnCl2
MnCl2
MnSO4
MnCl2
Mn(NO3)2
MnSO4
Mn(CH3CO
0)2
KMnO4
Gene
mutation
Gene
mutation
Gene
mutation
Gene
mutation
Gene
mutation
(restored
luminescence)
Gene
mutation
Gene
mutation
Inhibition of
growth in
recombination
deficient
mutant (Rec")
compared to
wild type
(Rec+)
TA98
TA102
TA1535
TA1537
TA97
TA98
TA100
TA1535
TA1537
TA97
TA100
TA102
Pf-13
(dark
mutant)
KMBL
3835
T4
M45
(Rec-)
-
-
ND
ND
ND
ND
ND
ND
ND
+
-f
-/+
+
+
+
-f
+
+
+
+
Reference
Wong (1988)
Mortelmans et
al. (1986)
Pagano and
Zeiger (1992)
DeMeo et al.
(1991)
Ulitzur and
Barak (1988)
Zakour and
Glickman
(1984)
Orgel and
Orgel (1965)
Nishioka
(1975)
External Review Draft j— Manganese—April 2002
7-40
-------
Table 7-11 (continued)
Species (test system)
B. subtilss
(recombination assay)
Compound
MnCl2
Mn(NO3)2
Mn(CH3CO
0)2
End Point
Inhibition of
growth in
recombination
deficient
mutant (Rec")
compared to
wild type
(Rec*)
Strain
M45
(Rec~)
Results
WithS9
Activation
ND
Without S9
Activation
Reference
Kanematsu et
al. (1980)
Eukaryotic organisms:
Fungi:
Saccharcnnyces
cerevisiae
Mammalian cells:
Mouse lymphoma
cells
Mammalian cells:
Syrian hamster
embryo cells
Mammalian cells:
Human lymphocytes
(Single-cell gel assay)
Mammalian cells:
Chinese hamster ovary
cells
MnSO4
MnCl2
MnCl2
MnCl2
MnSO4
Gene
conversion,
reverse
mutation
Gene
mutation
Enhancement
ofSA7
transformatio
n
DNA damage
Chromosomal
aberrations
Sister
chromatid
exchange
D7
L5178Y
TK*.
ND
ND
ND
-
+
+
+
+
+
+
+
Singh (1984)
Oberley et al.
(1982)
Casto et al.
(1979)
DeMeo et al.
(1991)
NTP (1993)
Notes:
- = negative results
+ = positive results
-/+ = equivocal results
ND = no data available
DNA = deoxyribonucleic acid
MnSO4*H2O = manganese (If) sulfate monohydrate
Mn(CH3COO)2 = manganous acetate
MnCl2 = manganous chloride
Mn(NO3)2 = manganous nitrate
MnSO4 = manganous sulfate
Rec = recombination
Source: Modified from ATSDR (2000)
External Review Draft — Manganese —April 2002
7-41
-------
Comparatively little data are available tiiat describes the genotoxic potential of other
manganese compounds. Manganese oxide (Mri3O4) was not mutagenic in S. typhimurium or S.
cerevisiae (Simmon and Ligon, 1977). Data obtained for manganese nitrate (Mn(NO3)2) in the B.
sttbtilis recombination assay were inconsistent between studies (Nishioka, 1975; Kauematsu et al.,
1980). Manganese acetate (Mn(CH3OO)2) was mutagenic in the B. subtilis recombination assay
without exogenous metabolic activation, and gave negative results with activation (Nishioka,
1975; Kanematsu et al., 1980).
7.3.2 Immunotoxicity
Immunotoxicity and lymphoreticular effects do not appear to be significant outcomes of
oral exposure to manganese. A single report describes effects in this category following oral
exposure. NTP (1993) administered diets containing 0,1,600, 3,130, 6,250,12,500, or 25,000
mg/kg manganese sulfate monohydrate to F344 rats (10/sex/dose) in a 13-week study. Based on
measured feed consumption, the study authors determined that the mean intake of manganese
sulfate monohydrate ranged from 110 to 1,700 mg/kg-day (equal to about 36 to 553 mg Mn/kg-
day) for males, and from 115 to 2,000 mg/kg-day (equal to about 37 to 621 mg Mn/kg-day) for
females. Increased fieutrophil counts were noted at 32 mg Mn/kg-day in male rats. Decreased
leukocyte counts were noted at 155 mg Mn/kg-day in female rats.
i
Studies in animals exposed to manganese chloride by intraperitoneal or intramuscular
injection suggest that manganese can affect several immunological cell types (ATSDR, 2000).
Observed effects include stimulation of macrophage and natural killer cell activity in mice (Rogers
et al., 1983; Smialowicz et al., 1985,1987). Other effects include alteration of the responsiveness
of lymphoid cells to mitogens and inhibited antibody production in response to a T-cell antigen
(Hart, 1978; Lawrence, 1981; Srisuchart et al., 1987). The significance of these findings for
human immune function is presently unknown.
i
7.3.3 Hormonal Disruption
No reports describing hormonal disruption associated with manganese exposure were
located. i
I
7.3.4 Physiological or Mechanistic Studies
.
Biochemical and Physiological Role
Manganese is a naturally-occurring element that is required for normal physiological
functioning in all animal species (U.S. EPA, 1996a). It plays a role in bone mineralization,
metabolic regulation, protein and energy metabolism, protection of cells from oxidative stress, and
synthesis of mucopolysaccharides (ATSDR, 2000). Many of these roles are achieved by
participation of manganese as a catalytic or regulatory factor for enzymes, including hydrolases,
dehydrogenases, kinases, decarboxylases and transferases. In addition, manganese is a structural
component of the metalloenzym.es mitochondria! superoxide dismutase, pyruvate carboxylase, and
External Review Draft—Manganese — April 2002
7-42
-------
liver arginase. Studies conducted to determine the biochemical and nutritional roles of manganese
in human health are reviewed in greater detail by Wedler (1994) and Keen et al. (1999).
The frequency of occurrence and consequences of manganese deficiency are issues of
some debate (Keen et al., 1999). However, observations reported by Doisy (1973) and Friedman
et al. (1987) suggest that manganese is an essential element for humans. Doisy (1973) reported
decreased levels of clotting proteins, decreased serum cholesterol, reddening of black hair,
retarded growth of hair and nails, and scaly dermatitis in a subject inadvertently deprived of
manganese. Friedman et al. (1987) administered a manganese-deficient diet to seven men for 39
days. Five of the seven subjects exhibited dermatitis at the end of the manganese-deficient period.
The development of dermatitis was attributed to decreased activity of manganese-requiring
enzymes that are required for skin maintenance. The symptoms cleared rapidly when manganese
was restored to the diet.
Manganese deficiency has been experimentally induced in multiple animal species.
Outcomes associated with manganese deficiency in animals include impaired growth (Smith,
1944), skeletal abnormalities (Amdur et al., 1944; Strause et al., 1986), impaired reproductive
function in females and testicular degeneration in males (Boyer et al., 1942), ataxia (Hurley et al.,
1961), altered metabolism of carbohydrates (Baly et al., 1988; Hurley et al., 1984) and lipids
(Abrams et al., 1976), and decreased cholesterol synthesis and excretion (Davis et al., 1990;
Kawano et al., 1987). The biochemical basis for these effects has not been established with
certainty, but it may be related to the participation of manganese in numerous enzymatic
reactions. • •
Low serum manganese levels are associated with several disease states, including epilepsy,
exocrine pancreatic insufficiency, multiple sclerosis, cataracts, and osteoporosis (Freeland-Graves
and Llanes, 1994). In addition, the metabolic disorders phenylketonuria and maple syrup urine
disease, genetic disorders of ammo acid metabolism, are associated with poor manganese status
(U.S. EPA, 1996a).
Mechanisms ofNeurotoxicity
The central nervous system (CNS) has been identified as the major target of manganese
toxiciry (U.S. EPA, 1993; ATSDR, 2000). The blood-brain barrier (BBB) is a major regulator of
the (CNS) milieu, and the rate and extent of manganese transfer across the BBB may be a
determinant of manganese neurotoxicity (Aschner and Aschner, 1991). The mechanism by which
manganese crosses the BBB to gain access to neuronal tissue has not been fully elucidated, but
may be a function of binding to transferrin (Aschner and Aschner, 1991). In the portal
circulation, manganese as Mn(H) initially binds to alpha-2-macroglobulin, and this complex
cannot cross the BBB. The Mn(II)-alpha-2-macroglobulin complex is transported by the
bloodstream to the liver (Tanaka, 1982), where a small fraction of the circulating Mn(Q) may be
oxidized to Mn(HI). The iron-transporting protein transferrin has been shown to also bind
Mn(IH), and may be responsible for its transport into the brain. The observation that some of the
regions of the brain that accumulate manganese (e.g., globus pallidus, striatum, and substantia
External Review Draft — Manganese—April 2002
7-43
-------
nigra) receive neuronal input from the transferrin
putamen supports this argument. Both of these
i-rich nucleus accumbens and the caudate-
regions are rich in transferrin receptors.
Additional evidence for the transferrin transport hypothesis was provided by an
experiment in which rats were given a 6-hour intravenous administration of ferric-hydroxide
dextran complex (Aschner and Aschner, 1990). The uptake of radiolabeled manganese into the
brain was significantly (p < 0.05) inhibited following the administration of the iron complex as
compared with rats administered iron-free dextran. It was concluded that iron homeostasis may
play an important role in the regulation of manganese transport across the BBB, since both metals
are transported by transferrin and may be competing for binding sites.
Once manganese has crossed the BBB j several neurotransmitter systems in the brain
appear to be potential targets for manganese tpxicity. The primary targets appear to be the
monoamines, including dopamine, noradrenaline and serotonin (Neff et al., 1969; Mustafa and
Chandra, 1971). The ammo acid neurotransmitter v-amino butyric acid (GABA) may also be
affected (Gianutsos and Murray, 1982). Effects on neurotransmitters may be both specific and
highly localized. Manganese neurotoxicity, for example, is reportedly associated with a selective
depletion of dopamine in the striatum, a site of manganese accumulation (Neff et al., 1969;
Bernheimer et al., 1973).
A resemblance exists between the symptoms of manganism and Parkinsonism, a condition
characterized by loss of dopaminergic neurons in the substantia nigra and globus pallidus. In
addition, several clinical features of manganism respond favorably to therapy with L-dopa in a
manner similar to patients with Parkinson's disease (Mena et al., 1970) although long-term
response of manganism patients to L-dopa has not been observed (ATSDR, 2000; Came et al.
1994). However, despite some similarities in pymptoms, a comparative study of a 52-year-old
worker exposed to manganese in an ore crushing plant and a patient with Parkinson's disease did
not reveal any similarity in neuropathology (\jamada et al., 1986). Barbeau (1984), Calne et al.
(1994), and Pal et al. (1999) have summarized the similarities and differences between manganism
and Parkinsonism. These researchers have noted that manganism characteristically occurs in
phases of increasing severity and that sufferers exhibit dystonia (disordered tonicity of muscles),
symptoms of extrapyramidal dysfunction such as bradykinesia (extreme slowness of movements
and reflexes), monotonic speech, and an expressionless or even grimacing face. Although the
altered gait and fine tremor are common to both Parkinsonism and manganism, the syndromes are
different in that manganism patients sometimes have psychiatric disturbances early in the onset of
the syndrome, have a tendency to fall backwards, do not have the Lewy bodies in the substantia
nigra that are commonly found in Parkinson's1 patients. Further, fluorodopa positron emission
tomography (PET) scans are normal in manganism patients but not in individuals with Parkinson's
disease (ATSDR, 2000). !
Mapping studies by Yamada et al. (1986) indicate that most of the neuronal degeneration
attributed to manganese exposure lies close to monoamine cell bodies and pathways.
Histopathology in manganese-exposed primates shows more widespread deposition of the metal,
with intense signaling observed in both the globus pallidus and substantia nigra using MRI
External Review Draft — Manganese—April 2002
7-44
-------
(Newland and Weiss, 1992). Studies in humans indicate that excess manganese in the brain
deposits primarily in the globus pallidus (Fell et-al. 1996; Kafritsa et al. 1998; Ono et al. 1995)
and damage to the human brain from manganese deposition may be limited to that region. In a
study mat supports these findings, the globus pallidus exhibited atrophy in an autopsy performed
on a worker with inhalation-related manganese poisoning (Yamada et al., 1986).
Although there is consensus that the monoaminergic systems, particularly the
dopamiaergic system, are affected by excess exposure to manganese, the precise mechanism of
action remains obscure. One hypothesis proposes that oxidation of dopamine plays a key role in
manganese neurotoxicity. Manganese (IE) has been shown to oxidize dopamine to its cyclized O-
quinone (cDAoQ) (Archibald and Tyree, 1987). This irreversible process ultimately results in
decreasisd dopamine levels. The formation of cDAoQ may subsequently initiate the generation of
reactive oxygen species, leading to oxidative stress and cell death (Segura-Aguilar and Lind,
1989).
An alternative hypothesis for manganese toxicity proposes an effect on brain cytochrome
P-450 activity. Liccione and Maines (1989) demonstrated a high sensitivity of rat striatal
mitochondria to manganese-induced increases in cytochrome P-450 activity. These authors
speculated that the increase in mixed function oxidase activity may trigger an increase in the
formation of active oxygen species (e.g., superoxide anions) that exert a harmful effect on
dopamiaergic pathways.
Other mechanistic studies have identified tyrosine hydroxylase (TOH), the rate limiting
enzyme in dopamine synthesis, as a potential target in manganese-induced neurochemical effects.
Bonilla (1980) and Chandra and Shukla (1981) found that changes in TOH activity in the
presence of manganese closely paralleled dopamine levels. Qato and Maines (1985) determined
that alterations in the activity of TOH and other monooxygenases may be related to manganese-
induced alterations in brain heme metabolism.
Manganese toxicity may be selectively associated with adverse effects on mitochondria.
Maynard and Cotzias (1955) originally proposed the mitochondrion as the target organelle for
manganese cytotoxicity, with adverse effects expressed primarily as disruption of Ca(II)
homeostasis. Mn(H) preferentially accumulates in the mitochondria in regions of the brain
associated with neurological symptoms and manganism. Once inside the mitochondria, Mn(II)
disrupts oxidative phosphorylation. The fundamental role of mitochondrial energy metabolism in
manganese neurotoxicity has been highlighted by the studies of Aschner and Aschner (1990) and
Gavin et al. (1992), as cited in U.S. EPA (1996a) and ATSDR (2000).
The results of Brouillet et al. (1993) confirm that manganese impairs mitochondrial
oxidative metabolism. In addition, their findings indicate that manganese neurotoxicity involves
an JV-methyl-D-aspartate receptor-mediated process similar to that observed for some other
mitochondrial toxicants. Manganese may thus produce neuronal degeneration by an excitotoxic
process secondary to its ability to disrupt oxidative energy metabolism.
External Review Draft — Manganese—April 2002
7-45
-------
7.3.5 Structure-Activity Relationship
Information on structure-activity relatioriships is not available for manganese.
7.4 Hazard Characterization
7.4.1 Synthesis and Evaluation of Major Noncancer Effects
Manganese is an ubiquitous element that is essential for normal physiological functioning
in all animal species. The biochemical basis for this requirement is most likely the participation of
manganese as a structural component or catalytic cofactor for many enzymes. The Adequate
Intake levels for manganese range from 0.003to 0.6,mg/day for infants from birth to 6 months,
0.6 mg/day for infants from 7 months to 1 year; 1.2 mg/day for children aged 1-3 years, 1.5 to 1.9
mg/day for children aged 4-13 years, and from 1.6 to 2-3 mg/day for adolescents and adults (Food
and Nutrition Board, 2002). Although outright manganese deficiency has not been observed in
the general population, sub-optimal intake may be of concern for some individuals.
In contrast to the beneficial effects of manganese as a nutrient, excess exposure to
manganese may be associated with toxic effects. At present, the optimal level of oral exposure to
manganese is not well defined (Greger, 1999). i
Ingested manganese appears to be primarily absorbed in the Mn(II) form, and may
compete with iron arid-cobalt for common absorption sites. Absorption varies among individuals
and is also influenced by dietary factors. Absorption of 3 to 10% of ingested dietary manganese is
considered to be representative of the general population (U.S. EPA, 1996a). Iron deficiency
enhances the absorption of manganese in animals (U.S. EPA, 1984). Uptake of dietary
manganese may be reduced in the presence of other dietary components such as calcium and
phytate.
Once absorbed, manganese has the potential to accumulate in mitochondria-rich tissues,
including liver, pancreas, and kidney. Lesser amounts accumulate in brain and bone. Manganese
is efficiently removed from the blood by the liver and released into bile. Biliary secretion
represents the major pathway for manganese transport to the intestine, and studies in humans
indicate that manganese is primarily excreted in the feces. The rate of excretion responds
efficiently to increased manganese intake. The rate of biliary secretion acts in concert with
absorptive processes to establish homeostatic control of manganese levels in the body. As long as
physiological systems are not overwhelmed, humans appear to exert efficient homeostatic control
over manganese levels, so that levels in the body are kept relatively constant despite moderate
variations in intake. Manganese is also reabsorbed in the intestine through enterohepatic
circulation (Schroeder et al. 1966). \
While it is apparent that exposure to excess manganese can result in increased tissue
levels, the interrelationships between oral exposure levels, tissue accumulation, and health effects
in humans are not completely understood. Epidemiological studies of workers exposed by
External Review Draft f- Manganese—April 2002
7-46
-------
inhalation to manganese dusts and fumes have identified the central nervous system (CNS) as the
primary target for chronic manganese toxicity by the inhalation route (U.S. EPA, 1993). Both
Mn(III) and Mn(H) have been associated with the neurotoxic effects of manganese. While some
researchers note the similarities in CNS effects occurring following manganese exposure and in
Parkinson's disease (dystonia, rigidity, bradykinesia), there are significant differences in the two
diseases. For example, manganism patients exhibit a less-frequent resting tremor than do
Parkinson's patients, extrapyramidal symptoms including fixed expression or a facial grimace,
active tremor (particularly in tile upper body), a "cock-walk" in which the patient walks on the
toes with the back stiff and the elbows flexed, a propensity to fall backwards (especially when
pushed), and a failure to respond to dopaminomimetics (Barbeau, 1984; Came et al., 1994; Pal et
al., 1999).
iSeveral investigators have proposed a link between elevated oral manganese intake by
humans and neurological symptoms resembling manganism (Kawamura et al., 1941; Kilbum,
1987; Kondakis et al., 1989; Goldsmith et al., 1990). Results from these studies are described in
detail in Section 7.1. In each case, the data from these studies were insufficient to establish that
manganese was the causative factor (ATSDR, 2000). The evidence for a similar pattern of
neurotoxicity in humans following oral exposure is therefore considered equivocal.
Numerous studies have investigated manganese neurotoxicity in rodent models. However,
the utility of rodent studies for evaluating the potential neurotoxic effects of manganese in humans
has been questioned. Although biochemical and behavioral evidence of neurological effects has
been observed, signs of impaired motor function resembling those seen in humans are usually not
detected. In particular, studies of rodents exposed to manganese by drinking water or food have
been unable to produce the characteristic signs of extrapyramidal neurologic disease seen in
humans. In contrast, chronic administration of manganese to monkeys by oral (one study) or
parenteral routes (two studies) has resulted in neurological signs consistent with chronic
manganism. The failure to reproduce these signs in rodent studies may result from differences in
manganese accumulation and distribution between rodents and primates. The dietary requirement
for manganese in rodents, for example, is estimated to be 100 times higher than in humans. In
addition, neurotoxic effects in humans are associated with manganese accumulation in
neuromelanin-rich regions of the brain, and the homologous regions in rats and mice lack this
pigment Although primates are likely to be better models of the neurological manifestations of
manganese intoxication than rodent species, sufficient data from well-designed oral studies are not
currently available.
An additional drawback to animal studies of manganese neurotoxicity is the inability to
identify certain psychological or neurobehavioral signs. Overt neurological impairment in humans
is often preceded by psychological symptoms such as irritability and emotional lability. Since
accurate dose-response relationships based on neurobehavioral endpoints are generally not
available from animal studies, neurochemical responses have been examined as alternative
indicators of neurotoxicity. Such studies have been conducted on the assumption that since the
toxic manifestations of chronic manganese exposure resemble Parkinsonism, altered biogenic
amine metabolism in the CNS may be one of the underlying mechanisms. However the patterns
External Review Draft — Manganese—April 2002 7-47
-------
of neurochemical response reported following manganese exposure are not consistent among
studies. Although manganese exposure is generally thought to result in decreased dopamine
concentrations, some studies report increased ojr fluctuating levels. The effect of manganese on
dopamine levels, for example, appears to be age-dependent. Neonatal rats and mice exposed to
manganese from birth to 15 or 30 days of age have an increased levels of dopamine and
norepinephrine in the brain (Chandra et al., 1979; Cotzias et al., 1976; Shukla et al., 1980).
Further, temporal changes in dopamine neurochemistry have been observed with prolonged or
continuous manganese treatment and it is not established how these time-related changes affect
manganese-induced neurotoxicity.
Route of administration is also an issue of concern in evaluating the results of animal
studies. Scheuhammer (1983), for example, determined that intraperitoneal injection is not the
route of choice for studies of manganese exposure that are longer than 30 days in duration,
especially for investigations of neurotoxicity. Intraperitoneally administered manganese appears
to have a selectively toxic effect on the pancreas. This effect may make it difficult to distinguish
between subtle neurochemical changes resulting directly from manganese exposure, and changes
that are secondary to cellular damage in the pancreas. In addition, U.S. EPA (1984) noted that
results from parenteral studies are of limited value in predicting the reproductive hazards of
ingested manganese. At least one study exists,1 however, that shows the differential uptake and
distribution of manganese administered via injection compared to oral dosing. Roels et al. (1997)
investigated the uptake and distribution of manganese (as either MnO2 or MnCl2) in rats following
intra peritoneal injection or gavage dosing. Manganese concentrations were not increased in the
blood or brain following administration of 4 weekly doses of 1.22 mg Mn/kg of the dioxide via
gavage; following i.p. dosing, manganese concentrations were significantly increased in the blood,
striatum, cerebellum and cortex. Steady-state blood manganese concentrations were increased to
similar levels by both gavage and i.p. dosing of MnCl2. Gavage dosing of the dichloride
significantly increased the cortex manganese concentrations, but not that of the other two regions.
Intra peritoneal dosing of the compound increased the manganese levels in the striatum and
cortex, but not the cerebellum. These data indicate that depending on the compound, injection
administration of manganese results in higher blood and brain concentrations of the metal than
does gavage administration. ;
t
Toxic effects of oral manganese exposiire have also been reported in the hematopoietic,
cardiovascular, reproductive, and digestive systems in animals. Hematological and biochemical .
outcomes vary depending on age and iron status, with young or iron-deficient animals more likely
to exhibit adverse effects. Other effects observed following manganese exposure include reduced
body weight and reduced liver weight. Animal studies suggest that manganese is not a potent
developmental toxicant. ,
Infants have been identified as a potentially sensitive subpopulation for excess manganese
exposure. This determination reflects evidence for higher levels of manganese retention in the
brains of neonates than in adults, although the relationship between manganese accumulation in
the neonatal brain and toxiciry remains unclear (U.S. EPA, 1993). Additional concerns include
evidence for greater extent of manganese transport across the blood-brain barrier, the high
External Review Draft -r— Manganese — April 2002
7-48
-------
concentration of manganese in some infant formulas, and evidence suggestive of a possible link
between manganese exposure and learning disabilities. Although a causal relationship has not
been established for elevated manganese intake and learning disabilities, a need for further
research in this area has been noted (U.S. EPA, 1993).
Other potentially sensitive subpopulations for manganese exposure have been identified.
In general, these are groups who may have greater potential for increased body burdens due to
increased absorption or altered clearance mechanisms. The list includes pregnant women, elderly
persons, iron- or calcium-deficient individuals, and individuals with impaired liver function.
7.4.2 Synthesis and Evaluation of Carcinogenic Effects
The carcinogenic potential of ingested manganese has not been systematically evaluated in
epidemiological studies.
Data from animal studies are also limited. Currently, one of the few adequately designed
investigations is the 2-year oral exposure study conducted by the National Toxicology Program
(NTP, 1993). Groups of F344 rats (70/sex) were provided with diets containing 0,1,500, 5,000,
or 15,000 ppm manganese sulfate monohydrate. These dietary concentrations were reported to
be equivalent to an intake ranging from 30 to 331 mg Mn/kg-day for males, and 26 to 270 mg
Mn/kg-day for females. No increase in any tumor type could be attributed to manganese
exposure.
In a concurrent study, B6C3F, mice were administered 0,1,500, 5,000, or 15,000 mg/kg
manganese sulfate monohydrate (NTP, 1993). These:dietary concentrations were reported to be
equivalent to an intake ranging from 63 to 722 mg Mn/kg-day for males and 77 to 905 mg
Mn/kg-day for females. Compared to controls, the incidences of thyroid follicular cell hyperplasia
were significantly greater in high-dose males and in females at all dose levels. The incidence of
follicular cell adenomas in high-dose males (6%) was slightly greater than the range of historical
incidence in NTP studies of follicular cell adenomas in male B6C3Fj mice (0-4%). In high-dose
females, the incidence of follicular cell adenomas (10%) was also slightly above the historical
control range (0-9%). Follicular cell tumors were seen only at the termination of the study (729
days). NTP (1993) reported that the manganese intakes in the high-dose mice were 107 times
greater than the recommended dietary level. While NTP (1993) concluded that these data
provided "equivocal evidence of carcinogenic activity in mice," U.S. EPA (1993) questioned the
relevance of these findings to human carcinogenesis. The basis for concern was 1) the large dose
of manganese required to elicit a response observed only at the end of the study, and 2) tumor
frequencies that were not statistically different from historical controls.
Three additional studies address the carcinogenicity of manganese. DiPaolo (1964) found
that a larger percentage of DBA/1 mice exposed subcutaneously and intraperitoneally to
manganese chloride developed lymphosarcomas when compared to controls. A comprehensive
evaluation of these data was not possible, however, because they were published in an abstract
form which lacked sufficient experimental detail. Stoner et al. (1976) found a higher frequency of
External Review Draft — Manganese—April 2002 7-49
-------
lung tumors in strain A/Strong mice administered manganese sulfate intraperitoneally as compared
to controls. Although these results are suggestive of carcinogenic activity, they fail to meet the
positive response criteria for the interpretation of lung tumor data in this strain of 1) an increase in
the mean number of tumors per mouse, and 2) an observable dose-response relationship (Sliimkin
and Stoner, 1975). In the third study, Furst (1978) injected F344 rats intramuscularly with
manganese acetylacetonate. An increased incidence of fibrosarcomas was observed at the
injection site. Increased tumor incidence was not observed at other sites. When evaluated as a
group, these studies do not provide convincing1 evidence for carcinogenicity of manganese.
Both negative and positive results have been obtained in assays for the genotoxic effects
of manganese. Mutagenicity assays in multiple tester strains of Salmonella typhimurium gave
predominately negative results for manganese jsulfate monohydrate and manganese chloride when
tested with or without exogenous metabolic activation by S9 fraction (Wong, 1988; DeMeo et al.,
1991; Pagano and Zeiger, 1992; NTP, 1993). Neither compound induced mutations in
Drosophila melanogaster as evaluated by sex-jlinked recessive lethal or somatic mutation assays
(Rasmuson, 1985; Valencia et al., 1985; NTP, t1993). Dikshith and Chandra (1978) did not
observe increased incidence of chromosomal aberrations in rat bone marrow or spermatogonial
cells following oral administration of manganese chloride.
In addition to the negative results described above, positive results for manganese
compounds have been obtained in some assays for genotoxicity. Manganese sulfate induced
sister chromatid exchange and chromosomal aberrations in vitro in Chinese hamster ovary cells,
and induced chromosomal aberrations in vivo in albino mice following oral administration
(Joardar and Sharma, 1990). Manganese compounds also induced or enhanced mutation,
transformation, chromosomal aberrations, and DNA damage in some assays conducted in
mammalian cell lines (Casto et al., 1979; Oberly et al., 1982; DeMeo et al., 1991; NTP, 1993),
bacteria (Orgel and Orgel, 1965; Nishioka, 1975; Zakour and Glickman, 1984), and yeast (Singh,
1984). Although these results suggest that manganese may have genotoxic potential, there are
presently no epidemiological or unequivocal animal data to suggest that manganese is
carcinogenic. . ;
[
7.4.3 Mode of Action and Implications in Cancer Assessment
i
The molecular mechanisms responsible for the toxicity of manganese have not been
identified with certainty. Most effort has focused on identification of mechanisms mediating the
toxic effects observed in the central nervous system. Multiple researchers have proposed that
elevated levels of Mn(n) and Mn(HI) trigger the production of free radicals, reactive oxygen
species, and other cytotoxic metabolites in brain tissue. Generation of these reactive species is
hypothesized to occur via the oxidation or turnover of intracellular catecholamines, impacts on
mitochondrial metabolism, or stimulation of cytochrome P-450 activity. Manganese may also
influence transport systems, enzyme activity and receptor function in the brain and other organs.
At the present time, there is no evidence to link these proposed mechanisms of action to
carcinogenic potential.
External Review Draft •— Manganese — April 2002
7-50
-------
7.4.4 Weight of Evidence Evaluation for Carcihogenicity
The weight of evidence for manganese carcinogenicity is currently rated as Group D (Not
Classifiable) using the criteria of the U.S. EPA Guidelines for Carcinogen Risk Assessment (U.S.
EPA, 1986a). The classification of Group D was verified on May 25,1988 by the CRAVE Work
Group of the U.S. EPA. The basis for this determination is the inadequacy of existing studies for
assessment of manganese carcinogenicity (U.S. EPA, 1996a).
Manganese has not yet been evaluated using the criteria of the U.S. EPA Proposed
Guidelines for Carcinogen Risk Assessment (U.S. EPA, 1996b). However, evaluation of
currently available data suggests that the appropriate descriptor for manganese would be
"inadequate data for an assessment of human carcinogenic potential." This descriptor is
appropriate when the available information is judged to be inadequate to perform an assessment of
human carcinogenic potential. It is applied when there are conflicting data on carcinogenicity or
in situations where there is a paucity of data.
7.4.5 Sensitive Populations
Sensitive populations are defined as those which will exhibit an enhanced or altered
response to a chemical when compared with most persons exposed to the same concentration of
chemical in the environment. Factors that can contribute to this altered response include genetic
composition, age, developmental stage, health status, substance use history, and nutritional status.
These factors may alter the function of detoxification and excretory processes, or compromise the
function of target organs. In general, the elderly with declining organ function and infants and
children with developing organs are expected to be more sensitive to toxic substances than
healthy adults.
7.4.6 Potential Childhood Sensitivity
Neonates have been identified as a potentially sensitive subpopulation for manganese
exposure. This determination reflects observations in human (Zlotkin and Buchanan, 1986) and
animal ([Keen et al., 1986; Kostial et al., 1978; Rehnberg, et al. 1980) studies that suggest that
neonates retain higher levels of administered manganese than adults.
In adults, manganese concentrations are retained within a narrow range by the ability of
excretion systems to match the intake of this, element (Fechter, 1999). The process responsible
for manganese excretion is generally believed to require a significant time period to mature into
the adult pattern, with adult patterns of excretion developing at about the time of weaning
(Fechter, 1999). During this period of development, the young organism might be susceptible to
manganese toxicity if exposed to high levels in the diet or via environmental contamination.
Data with respect to fetal accumulation are not numerous, but appear to consistently
demonstrate that manganese is transported across the placenta to a limited extent (Fechter, 1999).
When all available data are examined, it appears that the fetus is relatively protected from
External Review Draft — Manganese—April 2002 7-51
-------
manganese accumulation when maternal exposure occurs at relatively low doses. Under
conditions of high maternal exposure, manganese accumulation also appears to be limited
(Fechter, 1999). The mechanism underlying this lack of accumulation is unknown, but may
reflect increased maternal excretion, limited uptake across the placenta, or fetal elimination.
i
The greatest concern for developmental susceptibility has been generated by data which
suggest the existence of a period prior to weaning when the neonate is unable to eliminate
manganese. Fechter (1999) reassessed data in the published literature and concluded that the
available literature does not support a toxicokinetic basis, for accumulation in the fetal or neonatal
organism [relative to the adult organism], under conditions of excess exposure to manganese.
"While the available data indicate that manganese does reach brain tissue, currently available
evidence does not support a clear regional distribution.
j
Kaur et al. (1980) found that younger neonates and 19-day fetuses were more susceptible
to manganese toxicity than older rats. Studies with ^Mn indicated that manganese was localized
to the liver and brain in younger animals, and there was more manganese per unit weight in
younger animals when compared with older animals.
Collipp et al. (1983) found that hair manganese levels in newborn infants increased
significantly from birth (0.19 |Ag/g) to 6 weeks of age (0.885 |ig/g) and 4 months of age (0.685
pg/g) when the infants were given formula. In contrast, there was no significant increase in
babies who were breast-fed (0.330 fig/g at 4 months). These results were attributed to the
difference in manganese content between infant formula and breast milk. Human breast milk is
relatively low in manganese (7 to 15 ng/L), while levels in infant formulas are 3 to 100 times
higher. Collipp et al. (1983) further reported that the level of manganese in the hair of learning
disabled children (0.434 Hg/g) was significantly increased in comparison to samples from normal
children (0.268 u-g/g). ;
There is at least one study reporting different responses in manganese-treated neonatal
animals compared to treated adults (Dorman et al., 2000). Pups were administered MnCl2 in
water at 11 or 22 mg Mn/kg for 21 days by mouth and were dosed starting after birth, postnatal
day 1 (PND 1), until weaning, PND 21. At PND 21, the effect of manganese treatment on motor
activity, learning and memory (passive avoidance task), evoked sensory response (acoustic startle
reflex), brain neurochemistry, and brain pathology was evaluated. Manganese treatment at the
highest dose was associated with decreased body weight gain in pups, although the authors
indicated absolute brain weight was not significantly altered. There were no statistically
significant effects on motor activity or performance in the passive avoidance task. However,
manganese treatment induced a significant increase in amplitude of the acoustic startle reflex.
Significant increases in striatal DA and DOPAC concentrations were also observed in the high-
dose treated neonates. No pathological lesions were observed in the treated pups. No effects on
body weight or behavior were observed in treated adult animals in this study. The authors
indicated that these results suggest that neonatal rats are at greater risk than adults for
manganese-induced neurotoxicity when compared under similar exposure conditions. This study,
along with evidence for increased absorption and reduced elimination in the neonate, suggests that
External Review Draft -\- Manganese — April 2002
7-52
-------
the very young may be more susceptible to the harmful effects of manganese exposure due to
differences in toxicokinetics.
Other investigators have reported an association between elevated hair levels of
manganese and learning disabilities in children (Barlow and Kapel, 1979; Pihl and Parkes, 1977).
Although no causal relationship has been established for learning disabilities and manganese
intake, further research in this area is warranted (U.S. EPA, 1993). The studies by He et al.
(1994) and Zhang et al. (1995) reported increased manganese levels in hair of school-age children
exposed to excess levels of manganese in drinking water and food stuffs. These studies conflict
with the Kawamura et al. (1941) study which showed that children were not adversely affected by
ingesting excess levels of manganese. The more recent studies differ in design, however, because
they measured early preclinical neurological effects of manganese overexposure. The older
studies did not have the sensitivity to measure such effects; this may explain why children were
not previously identified as a sensitive population. None of the studies in children provide
adequate exposure levels or properly control for confounding factors; therefore, they are not
strong enough to indicate that children are more sensitive than adults. They do confirm the need
for additional studies to investigate the possibility that children may be more susceptible than
adults to the effects of manganese overexposure.
High levels of manganese in infant formulas may also be of concern since Lonnerdal et al. (1987)
reported increased absorption and retention of manganese in neonatal animals. Manganese has
also been shown to cross the blood-brain barrier, with the rate of penetration in animal
experiments being 4 times higher in neonates than in adults (Mena, 1974). Dieter et al. (1992)
stated mat "if there were a toxicological limit to manganese according to the principles of
preventive health care, then it would have to be set at 0.2 mg/L of manganese in water for infants
as a group at risk."
7.4.7 Other Potentially Sensitive Populations
U.S. EPA (1996a) has identified additional sensitive subpopulations for manganese
exposure. In general, these are groups who may have greater potential for increased body
burdens due to increased absorption or altered clearance mechanisms. The list includes pregnant
women, elderly persons, iron- or calcium-deficient individuals, and individuals with impaired liver
function.
External Review Draft —^ Manganese — April 2002
7-53
-------
-------
8.0 DOSE-RESPONSE ASSESSMENT
8.1 Dose-Response for Noncancer Effects
8.1.1 RfD Determination
Choice of Principal Study and Critical Effect
Manganese is an essential trace element that is required for normal physiologic function in
humans and animals. Excess exposure to manganese, particularly via the inhalation route, is
associated with neurotoxicological symptoms that resemble parkinsonism. Thus, derivation of the
RfD must consider issues of both essentiality and toxicity.
The RfD is not based on rodent studies, because rodents do not exhibit the same
neurologic deficits that humans do following exposure to manganese. For example, manganese at
high doses induces parkinson-like symptoms in humans and primates, but not in rodents. Because
of the species difference in the response to manganese exposure, rodents are not good models for
manganese toxicity studies. More details on this can be seen in IRIS (USEPA, 1996a).
The reference dose (RfD) is based on the extensive information available for the dietary
intake of manganese by human populations (U.S. EPA, 1996a). Freeland-Graves et al. (1987)
reviewed human studies and proposed an estimated safe and adequate daily dietary intake of 3.5
to 7 mg for adults. WHO (1973) reviewed data on adult diets and concluded on the basis of
manganese balance studies that 2 to 3 mg/day, is an adequate daily intake and 8 to 9 mg/day is
"perfectly safe."
Dose-Response Assessment and Method of Analysis
'The current RfD for manganese was derived from information gathered in dietary surveys
of manganese exposure. In various surveys, manganese intakes of adults eating western-type and
vegetarian diets ranged from 0.7 to 10.9 mg per day (Freeland-Graves, 1994; Gibson, 1994 as
cited by Food and Nutrition Board, 2002). Depending on individual diets, a normal intake may be
well over 10 mg per day, especially from a vegetarian diet Based on this information, the U.S.
EPA (1996a) considers a dietary intake of 10 mg/day to be safe for a 70 kg adult. Thus, the
resulting dose of 0.14 mg/kg-day represents a NOAEL for chronic human consumption of
manganese in the diet (U.S. EPA, 1996a).
Application of Uncertainty and Modifying Factors
U.S. EPA (1996a) has recommended use of an uncertainty factor of 1 for derivation of the
manganese RfD. This recommendation is based on the following considerations. Manganese is
an essential trace element for human health. The information used to derive the RfD was
collected from many large human populations consuming normal diets over an extended period of
time. The available data suggest that as long as physiological systems are not overwhelmed,
External Review Draft — Manganese—April 2002 8-1
-------
humans exert effective homeostatic control over manganese so that body burden is kept relatively
constant when concentration of manganese in the diet varies.
U.S. EPA (1996a) has recommended the use of a modifying factor of 3 when assessing -
exposure to manganese from drinking water. U.S. EPA (1996a) has outlined four reasons for this
recommendation: ;
• While toxicokinetic data suggest that there is no significant difference in absorption of
manganese from food versus water, uptake of manganese from water appears to be
greater hi fasted individuals. |
l
• The study by Kondakis et al. (1989) raises concern for possible adverse health effects
associated with a lifetime consumption of drinking water containing 2 mg/L of
manganese.
i
• Evidence exists that neonates absorb more manganese from the gastrointestinal tract,
and excrete less of the absorbed manganese. Additional evidence suggests that
absorbed manganese more easily crosses the blood-brain barrier in neonates.
However, this evidence comes from animal studies; similar absorption studies in
human neonates have not been performed, although Collipp et al. (1983) observed
increased hair manganese levels in infants fed prepared formula compared with infants
fed breast milk.
• Infant formula typically contains a much higher concentration of manganese than
human or cows* milk. Powdered formula reconstituted with drinking water represents
an additional source of manganese intake for a potentially sensitive population.
These potential impacts on children, when considered hi conjunction with the likelihood
that the most adverse effects of manganese (e.g., those seen in manganese miners or others with
chronic overexposure to inhaled manganese) are likely to be irreversible and not manifested for
many years after exposure, warrant caution until more definitive data are available (U.S. EPA,
1996a). Recent data indicate, however, that in contrast to the symptoms of manganism,
preclinical neurological effects of inhalation exposure of occupational workers to excess
manganese are reversible (Roels et al. 1999). Similarly, symptoms of oral exposure to excess
manganese in compromised individuals (e.g., individuals with liver disease who could not excrete
manganese in the bile) were resolved when the exposure to excess manganese was decreased
(Devenyi et al. 1994; Fell et al. 1996). These data indicate that the human body can recover from
certain adverse effects of overexposure to manganese if the exposure is stopped and the body can
clear the excess. Significant uncertainty still exists, however, concerning at what level of
manganese intake these preclinical neurological symptoms might occur.
External Review Draft •— Manganese—April 2002
8-2
-------
The RfD for chronic exposure to manganese in drinking water is therefore calculated as follows:
RfD = 0.14 mg/kg-day = 0.047 mg/kg-day
1x3
where:
0.14 mg/kg-day = Chronic NOAEL for dietary manganese.
1 = Uncertainty factor.
3 = Recommended uncertainty factor for exposure in drinking water
8.1.2 RfC Determination
The inorganic manganese compounds predominating in drinking water are non-volatile.
Inhalation of manganese during use of drinking water for residential activities is therefore not
expected to be a significant pathway of exposure or toxicity.
U.S. EPA (1996a) has derived an inhalation Reference Concentration (RfC) for
manganese of 5 x 10"s mg/m3.
Choice of Principal Study and Critical Effect
The RfC for manganese (U.S. EPA, 1996a) was derived using data from two
epidemiological studies of workers exposed to manganese dioxide dust in occupational studies
(Roels et al., 1987; Roels et al., 1992). The critical effect was impairment of neurobehavioral
function, as assessed by medical questionnaire, audio-verbal short-term memory, visual simple
reaction time, hand steadiness, and eye-hand coordination. .
Dose-Response Characterization and Method of Analysis
The toxicity data for manganese were evaluated using the conventional NOAEL/LOAEL
approach. Neither of the principal studies identified a NO AEL. The LOAEL from the Roels et
al. (1992) is derived from an occupational-lifetime integrated respirable dust (IRD) concentration
of manganese dioxide (based on 8-hour time-weighted average [TWA] occupational exposures
for various job classifications, multiplied by individual work histories in years). This LOAEL is
expressed as mg Mn/m3-year. The IRD concentrations ranged from 0.040 to 4.433 mg Mn/m3-
year, with a geometric mean of 0.793 mg Mn/m3-year and a geometric standard deviation of
2.907. The geometric mean concentration (0.793 Mn/m3-year) was divided by the average
duration of manganese dioxide exposure (5.3 years) to obtain a LOAEL TWA of 0.15 mg Mn/m3-
year. The LOAEL (Human Equivalent Concentration, HEC) is 0.05 mg/m3.
External Review Draft — Manganese—April 2002
8-3
-------
The LOAEL indentified in the Roels et al. (1987) study is based on an 8-hour TWA
occupational exposure. The TWA of total airborne manganese dust ranged from 0.07 to 8.61
mg/m3, and the median was 0.97 mg/m3. The L1OAEL(HEC) is 0.34 mg/m3.
Application of Uncertainty and Modifying Factors
No modifying factor was used in derivation of the RfC. A composite uncertainty factor of
1,000 was used and reflects a factor of 10 for protection of sensitive individuals, a factor of 10 for
use of a LOAEL, and a factor of 10 for database limitations. The factor of 10 for database
limitations reflects an exposure period of less than chronic duration, lack of developmental data,
and potential but unquantified differences in the toxicity of different forms of manganese.
8.2 Dose-Response for Cancer Effects
Manganese is currently classified as a Group D chemical—NOT CLASSIFIABLE as to
HUMAN CARCINOGENICITY. This category is assigned to chemicals for which there is
inadequate human and animal evidence of carcinogenicity, or for which no data are available.
There are presently no human data to suggest an association of oral manganese exposure with
increased cancer incidence. Data collected from a 2-year oral exposure study in rats did not
reveal evidence for carcinogenic activity (NTPj 1993). Data collected from a 2-year oral
exposure study in mice revealed an apparent increase in tumor incidence at the highest dose
administered, but only near the end of the study (NTT, 1993). The observed increase was not
significantly different from the historical control incidence. These results are considered to be
equivocal. Based on the absence of any significant cancer response, a quantitative cancer dose-
response assessment for manganese will not be conducted.
Manganese has not yet been evaluated using the criteria of the U.S. EPA Proposed
Guidelines for Carcinogen Risk Assessment (U.S. EPA, 1999b). However, based on available
data it is likely that the appropriate descriptor fpr manganese would be "Data are inadequate for
assessment of human carcinogenic potential." This descriptor is appropriate when there is a
paucity of data on carcinogenic effects, or when the data are conflicting.
External Review Draft 4- Manganese —April 2002
8-4
-------
9.0 RISK DETERMINATION AND CHARACTERIZATION OF RISK FROM
DRINKING WATER
9.1 Regulatory Determination for Chemicals on the CCL
The Safe Drinking Water Act (SDWA), as amended in 1996, required the Environmental
Protection Agency (EPA) to establish a list of contaminants to aid the Agency in regulatory
priority setting for the drinking water program. EPA published a draft of the first Contaminant
Candidate List (CCL) on October 6,1997 (62 FR 52193, U.S. EPA, 1997). After review of and
response to comments, the final CCL was published on March 2,1998 (63 FR 10273, U.S. EPA,
1998). The CCL grouped contaminants into three major categories as follows:
Regulatory Determination Priorities - Chemicals or microbes with adequate data to
support a regulatory determination,
Research Priorities - Chemicals or microbes requiring research for health effects, analytical
methods, and/or treatment technologies,
Occurrence Priorities - Chemicals or microbes requiring additional data on occurrence in
(drinking water.
The March 2,1998 CCL included one microbe and 19 chemicals in the regulatory
determiaation priority category. More detailed assessments of the completeness of the health,
treatment, occurrence and analytical method data led to a subsequent reduction of the regulatory
determination priority chemicals to a list of 12 (one microbe and 11 chemicals) which was
distributed to stakeholders in November 1999.
SDWA requires EPA to make regulatory determinations for no fewer than five
contaminants in the regulatory determination priority category by August, 2001. In cases where
the Agency determines that a regulation is necessary, the regulation should be proposed by
August 2003 and promulgated by February 2005. The Agency is given the freedom to also
determine that there is no need for a regulation if a chemical on the CCL fails to meet one of three
criteria established by SDWA and described in Section 9.1.1.
9.1.1 Criteria for Regulatory Determination
These are the three criteria used to determine whether or not to regulate a chemical on the
CCL:
The contaminant may have an adverse effect on the health of persons,
External Review Draft — Manganese — April 2002
9-1
-------
The contaminant is known to occur or there is a substantial likelihood that the
contaminant will occur in public water systems with a frequency and at levels of public
health concern, I
I
In the sole judgment of the administratc-r, regulation of such contaminant presents a
meaningful opportunity for health risk reduction for persons served by public water
systems.
The findings for all criteria are used in making a determination to regulate a contaminant.
As required by SDWA, a decision to regulate cpmmits the EPA to publication of a Maximum
Contaminant Level Goal (MCLG) and promulgation of a National Primary Drinking Water
Regulation (NPDWR) for that contaminant. The Agency may determine that there is no need for
a regulation when a contaminant fails to meet one of the criteria. A decision not to regulate is
considered a final Agency action and is subject to judicial review. The Agency can choose to
publish a Health Advisory (a nonregulatory action) or other guidance for any contaminant on the
CCL independent of the regulatory determination.
9.1.2 National Drinking Water Advisory Council Recommendations
In March 2000, the EPA convened a Working Group under the National Drinking Water
Advisory Council (NDWAC) to help develop an approach for making regulatory determinations.
The Working Group developed a protocol for analyzing and presenting the available scientific
data and recommended methods to identify and.document the rationale supporting a regulatory
determination decision. The NDWAC Working Group report was presented to and accepted by
the entire NDWAC in July 2000.
•
Because of the intrinsic difference between microbial and chemical contaminants, the
Working Group developed separate but similar protocols for microorganisms and chemicals. The
approach for chemicals was based on an assessjment of the impact of acute, chronic, and lifetime
exposures, as well as a risk assessment that includes evaluation of occurrence, fate, and dose
response. The NDWAC Protocol for chemicals is a semi-quantitative tool for addressing each of
the three CCL criteria. The NDWAC requested that the Agency use good judgement in balancing
the many factors that need to be considered in making a regulatory determination.
i
The EPA modified the semi-quantitative NDWAC suggestions for evaluating chemicals
against the regulatory determination criteria and applied them in decision making. The
quantitative and qualitative factors for manganese that were considered for each of the three
criteria are presented in the sections that follow.
9.2 Health Effects
The first criterion asks if the contaminant may have an adverse effect on the health of
persons. Because all chemicals have adverse effects at some level of exposure, the challenge is to
define the dose at which adverse health effects are likely to occur, and estimate a dose at which
External Review Draft — Manganese —April 2002 9-2
-------
adverse health effects are either not likely to occur (threshold toxicant), or have a low probability
for occurrence (non-threshold toxicant). The key elements that must be considered in evaluating
the first criterion are the mode of action, the critical effect(s), the dose-response for critical
effect(s), the RfD for threshold effects, and the slope factor for non-threshold effects.
A full description of the health effects associated with exposure to manganese is presented
in Chapter 7 of this document and summarized below in Section 9.2.2. Chapter 8 and Section
9.2.3 present dose-response information.
9.2.1 Health Criterion Conclusion
The available toxicological data indicate that manganese has the potential to cause adverse
health effects in humans and animals at high doses. The primary route of exposure to toxic levels
of manganese is through the inhalation of manganese dust. An increased potential exists for
inhalation and ingestion exposure to manganese as a result of the use of MMT in fuels. Zayed et
al. (1999) measured airborne manganese concentrations (as MMT, respirable, and total
manganese) in five different microenvironments around Montreal, Canada. The authors
determined that the average daily exposure to respirable manganese was 0.010 u.g/kg-day and had
a low contribution to air, food, and water. Oral exposure to levels of toxicological concern is .
rare. In humans, neurological effects are the most likely manifestation of manganese toxicity.
There is no information available regarding the carcinogenicity of manganese in humans, and
animal studies have reported mixed results. Manganese is classified as Group D, or Not '
classifiable as to human carcinogenicity. The Reference Concentration (RfC) for manganese is 5 x
10"5 mg/m3 (U.S. EPA, 1998a) which is derived using data from two epidemiological studies of
workers exposed to manganese dioxide dust in an occupational setting (Reels et al., 1987; Roels
et al., 1992). The critical effect was impairment of neurpbehavioral function. The current RfD
for manganese in food is 0.14 mg/kg-day; and for drinking water, 0.047 mg/kg-day. Despite the
fact that it is possible for manganese to elicit some toxic effects at very high doses, the database is
too uncertain, especially related to children and other sensitive populations. Based on the
occurrence of adverse effects in humans and animals, the evaluation for Criterion #1 is positive.
9.2.2 Hazard Characterization and Mode of Action Implications
The primary health effect of manganese exposure is neurotoxicity, which is characterized
at high doses by ataxia, increased anxiety, dementia, a "mask-like" face, general extrapyrimidal
syndrome, or manganism, a syndrome similar to Parkinson's disease. The precise mechanisms of
manganese neurotoxicity are not known, although the observed effects of manganese on the
globus pallidus region of the brain suggest that a likely mechanism involves impairment of
dopaminergic function. Preclinical adverse neurological effects have been reported at much lower
doses than those resulting in manganism, however. Therefore, the possibility exists that any
potential neurological effects resulting from environmental exposures to manganese would likely
be more comparable to these subtle, though potentially significant, changes in neurological
function.
External Review Draft — Manganese—April 2002
9-3
-------
Studies in humans and animals are mixbd, but most animal studies indicate that children
are a potentially sensitive subpopulation based pn decreased excretion in the neonate (Lonnerdal,
1994). Additional potentially sensitive sub-populations include the elderly, pregnant women,
iron-deficient individuals, and individuals with impaired liver function.
Because the primary route of elimination for manganese is biliary excretion, persons with
impaired liver function may be especially susceptible to manganese toxicity (Layrargues et al.,
1998). Persons in a state of iron deficiency may also experience greater susceptibility to
manganese absorption and toxicity (Finley, 1999; Finley et al., 1994). In addition, infants and
neonates, in which the capacity for excretion through the bile is not fully developed, may also be
potentially susceptible to manganese toxicity (Lonnerdal, 1994). Although animal studies have
indicated an increased potential in neonates for! gastrointestinal absorption of manganese, as well
as decreased excretion potential, the degree to which these findings apply to human infants is
unknown. Dorman et al. (2000) have shown, however, that there is increased sensitivity for
neurotoxic effects following manganese exposure in neonatal rats compared to adult rats.
Because manganese is an essential nutrient in developing infants, however, the potential adverse
effects from manganese deficiency may be of greater concern than potential toxicity from over-
exposure.
r
An added complication is the fact that many inhibitors of manganese absorption, such as
phytates and plant fiber, are common in the diet and may thus lower the actual absorption of
ingested manganese. Also, manganese absorption from foods that are potentially high sources
may be inhibited by other factors such as the presence of co-occurring plant proteins that bind
manganese and decrease its bioavailability. Thus, although the manganese content in the soy-
based formula is higher than manganese content in human milk, the actual absorption of
manganese in the formula may not be substantially greater since it is prepared with soy milk,
which is high in phytate and vegetable protein. Data exist, however, that argue against this
possibility. For example, Keen et al. (1986) demonstrated in rat pups that manganese uptake from
human breast milk and cow^s milk was higher (-80% and -89 %, respectively) than that from soy
formula (~60%), but the absolute amount of manganese retained from soy was 25 times the
amount retained from human milk. Dorner et al. (1989) also reported increased retention of
manganese in full-term human infants fed cowls-milk formulas compared to breast-fed infants.
Human milk and cow's milk contain different proteins that bind manganese. In some cases, the
presence of these proteins may enhance manganese transport across the gut wall and hence
increase absorption. If infant formula is prepared with contaminated water, then it is possible that
the manganese will remain in a soluble form which may be more easily absorbed. More data are
needed on the various factors affecting manganese absorption hi infants before a confident
determination can be made. Other instances in which high dietary levels of manganese may not
necessarily correspond to high dose levels include vegetarian diets (many vegetables contain high
manganese levels but also high fiber and phytate levels) and possibly tea drinkers (tea also
contains nigh manganese levels accompanied by high levels of tannin, another inhibitor of
manganese absorption). ' i •
External Review Draft — Manganese—April 2002
9-4
-------
Several studies have explored the level of manganese intake which may be considered safe
in humans. The Food and Nutrition Board (2002) set an adequate intake level for manganese of
2.3 mg/day for men and 1.8 mg/day for women (Food and Nutrition Board, 2002; Trumbo et al.,
2001). The Food and Nutrition Board also set a tolerable upper intake level of 11 mg Mn/day for
adults based on the Greger (1999) review, which suggested that people eating western-type and
vegetarian diets may have intakes as high as 10.9 mg/day (Food and Nutrition Board, 2002).
Further,, for short-term duration, Davis and Greger (1992) found that daily intake of 15 mg/day
for 90 days resulted hi no adverse effects in women; the only effect seen was an increase in
superoxide dismutase activity.
No significant exposure-related neurological effects were seen in a cohort in Germany
exposed for up to 40 years to manganese in their well water at levels as high as 2.160 mg/L (0.3
to 2.160 mg/L; Vierrege et al., 1995). On the other hand, a study in Greece which examined
older populations chronically exposed to well water containing up to around 2 mg/L found effects
on neurological function in the high-exposure group (Kondakis et al., 1989); however, this study
did not adequately account for potential bias in subjective neurological test scores. Neither study
reported the dietary or other sources of manganese intake.
9.2.3 Dose-Response Characterization and Implications in Risk Assessment
The dose-response relationship for neurological effects of manganese by ingestion is not
well-characterized in animals or humans, but epidemiological data for humans indicate that intakes
as high as 11 mg/day (0.16 mg/kg-day) may not cause adverse effects in adult humans.
Additional evidence suggests a safe level as high asd5 mg/day (0.21 mg/kg-day for adult), based
on a study in which women received daily supplements of 15 mg manganese for 90 days and
exhibited only an increase in lymphocyte manganese-dependent superoxide dismutase, but no
measured adverse effects (Davis and Greger, 1992). Characterizing dose-response in humans is
complicated by the fact that manganese is an essential nutrient, and therefore some minimal level
of intake is necessary for good health. There are many reports of toxicity to humans exposed to
manganese by inhalation; much less is known, however, about oral intakes resulting in toxicity.
Rodents do not provide a good experimental model for manganese toxicity and only one limited
Study in primates by the oral route of exposure is available (Gupta et al., 1980).
A review of acute animal toxicity studies of manganese indicates that the manganese has
low to moderate oral toxicity. For example, the oral LD50 values for manganese compounds in
rats are in the range of 400 to 2,000 mg Mn/kg. Some animal studies have also reported
developmental and reproductive effects at high doses for some manganese compounds, but most
data from oral exposure suggest that manganese has a low developmental toxicity.
EPA has calculated an RfD for manganese. The RfD for manganese in food is 0.14
mg/kg-day, based on dietary surveys that have reported that, for an average 70 kg adult, having a
daily manganese intake of 10 mg presents no adverse effect. For drinking water, EPA
recommends to 'apply a modifying factor (MF) of 3 to yield a value of 0.047 mg/kg-day. This
modifying factor is meant to address the concern raised by the epidemiology study (Kondakis et
External Review Draft — Manganese—April 2002 9-5
-------
al., 1989), and a potential higher absorption of manganese in water, especially when drinking
fluids early in the morning, when the gut is empty. EPA has medium confidence in the RfD for
manganese.
9.3 Occurrence in Public Water Systems
The second criterion asks if the contaminant is known to occur or if there is a substantial
likelihood that the contaminant will occur in public water systems with a frequency and at levels
of public health concern. In order to address this question, the following information was
considered:
i
• Monitoring data from public water systems
• Ambient water concentrations and releases to the environment
Environmental fate
Data on the occurrence of manganese in public drinking water systems were the most
important determinants in evaluating the second criterion. EPA looked at the total number of
systems that reported detections of manganese,! as well those that reported concentrations of
manganese above an estimated drinking water health reference level (HRL). For noncarcinogens
the estimated HRL risk level was calculated from the RfD assuming that 20% of. the total
exposure would come from drinking water. For carcinogens, the HRL was the 10"6,risk-level.
The HRLs are benchmark values that were used in evaluating the occurrence data while the risk
assessments for the contaminants were being developed.
The available monitoring data, including indications of whether or not the contamination is
a national or a regional problem, are included in Chapters 4 of this document and are summarized
below. Additional information on production, use, and environmental fate are found in Chapters
2and3. |
9.3.1 Occurrence Criterion Conclusion
\
The available data for manganese production and use indicate a fairly stable trend for both.
While release of manganese to surface water is variable within a wide range of values, release of
manganese compounds to surface water is increasing. Releases of manganese and manganese
compounds to land are generally decreasing, while releases of manganese to air are decreasing
and air emissions of manganese compounds are stable (Tables 3-4 and 3-5). MMT in gasolines
provides a relatively new environmental source of manganese exposure. Recent testing suggests
that when very low levels of MMT are combusted (i.e., concentrations comparable to the
currently allowed levels), manganese is emitted primarily as manganese phosphate and sulfate.
Data on the occurrence of manganese in air resulting from combustion of MMT and other sources
are presented in Section 4.2. Pfeifer et al. (1999) determined that two occupational groups, office
workers and taxi drivers were exposed to comparable concentrations of manganese both before
External Review Draft —|- Manganese — April 2002
9-6
-------
and after MMT was present in fuels. These data, however, are counter to other modeling data
that indicate that taxi drivers are exposed to increased concentrations of manganese as a result of
MMT use (Lynam et al., 1994; Zayed et al., 1994; Riveros-Rosas et al., 1997). Modeling data
from five microenvironments in Canada indicate that with the currently acceptable levels of MMT
allowed in fuel, little impact to air and surface water concentrations of manganese is expected
from the use of MMT in fuels (Zayed et al., 1999). Monitoring data indicate that manganese is
infrequently detected in public water supplies. When manganese is detected, it rarely exceeds the
HRL or a value of one-half the HRL. Further, because manganese is an essential nutrient, the
risks of over-exposure must be weighed against the risks of manganese deficiency. Based on
these data, it is unlikely that manganese will occur in public water systems at frequencies or
concentration levels that are of public health concern. Therefore, the evaluation for Criterion #2
is negative.
9.3.2 Monitoring Data
Drinking Water
Occurrence data for manganese in drinking water are presented and analyzed in Chapter 4
of this document. Estimates of exposed populations are derived in Section 4.3. The National
Inorganic and Radionuclide Survey (NIRS) data represent 49 States. Data were not available for
the State of Hawaii. Since NIRS data lack occurrence information for surface water systems,
occurrence data on manganese exposure from the States of Alabama, California, Illinois, New
Jersey, and Oregon were used to obtain information on surface water.
At a health reference level (HRL) of 0.3 mg/L, approximately 6.2% of the NIRS PWSs
had detections greater than one-half the HRL (about 3,700 ground water PWSs nationally),
affecting approximately 4.6% of the population served (estimated at 4.0 million people
nationally).
The percentage of NIRS PWSs with detections greater than the HRL of 0.3 mg/L was
approximately 3.6% (about 2,200 ground water PWSs nationally), affecting 2.7% of the
population served (estimated at approximately 2.3 million people nationally).
irhe supplemental State data sets indicate that ground water PWS detections greater than
the HRL, of 0.3 mg/L are between 0.6% and 12%. Again, the NIRS national average is within
this range, with 3.6% of PWSs greater than the HRL. Notably, surface water PWSs showed
fewer exceedances of the HRL than ground water PWSs at this higher concentration, ranging
from 0% to 3%. Extrapolating national population exposures from these limited data sets is not
possible because exposure to manganese through surface water is not quantified beyond the five
States shown. However, exposure estimates incorporating surface water sources would certainly
be larger than the estimates provided here for groundwater sources.
External Review Draft — Manganese—April 2002
9-7
-------
Ambient Water \
The National Ambient Water Quality Assessment (NAWQA) program was begun in 1991
by the United States Geological Survey (USGS) to monitor water quality in representative study
basins located around the country. This program, which consists of 59 significant watersheds and
aquifers, was described in Chapter 4 of lids document hi regard to its use for monitoring ambient
levels of manganese in surface and ground wafers. The Minimum Reporting Level (MRL) in
water is 0.001 mg/L, while the MRLs in sediments and aquatic biota tissue are 4 mg/kg and 0.1
mg/kg, respectively. J
The data indicate that manganese is ubiquitous in surface and ground waters, presumably
as a result-of its natural occurrence in the earth's crust. The frequency of detection above the
HRL is generally higher in ground water than in surface water, but the median concentration in
sites reporting a detection is higher in surface water (0.016 mg/L in surface water versus 0.005
mg/L in ground water). Overall, the data indicate that, while manganese is nearly ubiquitous in
surface and ground water, detections at levels lof concern to public health are relatively few.
Manganese has been universally detected in stream sediments and aquatic biota tissues at
low levels. Manganese is not thought to bioaccumulate in tissues to any significant degree, and
desorption from sediments into the water cohurm is also limited by the insolubility of most
manganese compounds.
\ ... , ~ •
9.3.3 Use and Fate Data
Manganese is a naturally occurring element and is commonly found in soil, water, air, and
food, generally as a component of over 100 mineral compounds. Most manganese ore is
imported to the United States, with the amount increasing from 308 thousand metric tons in 1984
to 535 thousand metric tons in 1999. Most of this ore is smelted to produce ferromanganese,
which is used in steel production. Manganese compounds have a variety of other uses in industry
and agriculture, as described in Table 3-3 of this document.
Examination of data from the Toxic Release Inventory (TRI), shown in Tables 3-4 and 3-5
of this document, indicates that releases of manganese to water varied between 89 thousand and
2.4 million pounds for the period 1988 to 1998. Data for manganese compounds reveal an
increasing trend in surface water discharges, from 681 thousand to 4.5 million pounds for the
same period. j
Once released to the environment, manganese is readily deposited in the soil and taken up
by plants, whereupon it may enter the food chain. Significant bioaccumulation is not expected to
occur. Manganese is an essential nutrient in the diet, so some rninimal intake is necessary for
good health. Manganese particles may also become airborne, and some manganese compounds
are soluble in water. Manganese compounds may also adsorb to sediment surfaces and precipitate
out of solution.
External Review Draft — Manganese—April 2002
9-8
-------
Manganese, in the form of potassium permanganate, may be used in drinking water
treatment for oxidation and disinfection purposes (ANSI/NSF, 2000), in addition to its use in
industrial wastewater purification and odor abatement (ATSDR, 2000; U.S. EPA, 1984). The .
adsorption properties of some manganese compounds may cause them to be more prevalent in
certain types of soils or sediments.
9.4 Risk Reduction
The third criterion asks if, in the sole judgment of the Administrator, regulation presents a
meaningful opportunity for health risk reduction for persons served by public water systems. In
order to evaluate this criterion, EPA looked at the total exposed population, as well as the
population exposed above the estimated HRL. Estimates of the populations exposed and the
levels to which they are exposed were derived from the monitoring results. These estimates are
included in Chapter 4 of this document and summarized in Section 9.4.2 below.
lii order to evaluate risk from exposure through drinking water, EPA considered the net
environmental .exposure in comparison to the exposure through drinking water. For example, if
exposure to a contaminant occurs primarily through ambient air, regulation of emissions to air
provides a more meaningful opportunity for EPA to reduce risk than regulation of the
contaminant in drinking water. In making the regulatory determination, the available information
on exposure through drinking water (Chapter 4) and information on exposure through other
media (Chapter 5) were used to estimate the fraction that drinking water contributes to the total
exposure. The EPA findings are discussed in Section 9.4.3 below.
In making its regulatory determination, EPA also evaluates effects on potential sensitive
populations, including the fetus, infants and children. The sensitive population considerations are
included in Section 9.4.4.
9.4.1 Risk Reduction Criterion Conclusion
Approximately 47.5 million people are served by ground water public water systems with
detections greater than the MRL. More than 2.3 million of these individuals are served by
systems with detections greater than the HRL. Manganese is an essential nutrient that is common
and necessary in the diet. The estimated daily exposure to manganese from public water systems
is far below the expected daily intake from the diet, and also far below the level determined to be
safe and adequate. When average daily intakes from drinking water are compared with intakes
from food, air and soil, drinking water accounts for a relatively small proportion of manganese
intake. On the basis of these observations, the impact of regulating manganese concentrations in
drinking water on health risk reduction is likely to be small. Therefore, the evaluation for
Criterion #3 is negative.
External Review Draft — Manganese—April 2002
9-9
-------
9.4.2 Exposed Population Estimates
i
Estimates of exposed populations were derived in Chapter 4. National population
estimates for manganese exposure were derived using summary statistics from the National
Inorganic and Radionuclide Survey (NIRS), which lacked surface water data, with supplemental
surface water occurrence data that had been separately submitted to EPA from five States, An
estimated 47.5 million people in the U.S. are served by public water systems supplied from
ground water with detections of manganese above the minimum reporting level (MRL). An
estimated 4.0 million people (4.6% of the population) are served by ground water with levels
above one-half the health reference level (HRL) of 0.3 mg/L, and an estimated 2.3 million people
(2.7% of the population) are served by ground water with levels above the HRL. It should be
noted that these estimates are based on very limited and outdated data. The possibility exists that
the number of people served by ground water with Mn levels that are above the HRL could be
higher than these estimates; however, the data are lacking at this time to develop a more timely
assessment.
9.43 Relative Source Contribution
Relative source contribution analysis compared the magnitude of exposure expected via
drinking water to the magnitude of exposure fibm intake of manganese from other media such as
food, air, and soil. To perform this analysis, intake of manganese from drinking water must be
estimated. Occurrence data for manganese are presented in Chapter 4 of this document.
According to the NIRS data (Table 4-1), the median and 99th percentile concentrations for
manganese in ground water public water supplies were above the MRL of 0.001 mg/L. This is
not surprising considering the ubiquity with which manganese is present in the earth's crust.
[
•
Taking the median concentration of detections from the NIRS data (0.01 mg/L), and
assuming a daily intake of 2 L of drinking wat^r by a 70 kg adult, the average daily dose would be
0.02 mg/person-day or 2.8 x W4 mg/kg-day. The corresponding dose for a 10 kg child
consuming 1 L/day of drinking water would be 0.01 mg/child-day or 1.0 x 10° mg/kg-day. These
values are far below those expected from a norinal diet (2.9-12.6 x 10'2 mg/kg-day for adulls, 1.3
x 10'1 mg/kg-day for children, see Table 9-1 below), and are also less than the levels determined
by the National Academy of Sciences to be safe and adequate. The NAS determined that a daily
intake of 2.3 mg Mn is adequate for men and 1.8 mg is adequate for women, while the daily adult
intake expected from drinking water containing 0.01 mg/L Mn is 0.02 mg Mn. The NAS also
determined that a daily intake of 1.9 mg Mn is j adequate for boys and 1.6 mg is adequate for girls,
while the daily intake expected from drinking water containing 0.01 mg/L Mn is 0.01 mg for
children. The NAS has proposed that the Adequate Intake (AI) for manganese is 1.8 to 2.3
mg/day for adults (Food and Nutrition Board, 2002).
External Review Draft — Manganese—April 2002
9-10
-------
Table 9-1. Comparison of Average Daily Intake from Drinking Water and Other Media3
Mediium
Drinking Water*
Food
Air
Soil
Adult (jig/kg-day)
0.29
28.6-126
0.0087
0.0014-5.0
Child (ng/kg-day)
1.0
128.0
(0.87-37.2 for infants)
0.034
0.02 - 70
See Chapter 5 for derivation of intakes from media other than water
' based on median values
9.4.4 Sensitive Populations
'The sensitive populations identified for manganese include persons with impaired
detoxification and excretory function, such as infants and the elderly. Individuals with damaged
or impaired liver function may be particularly sensitive.
9.5 Regulatory Determination Decision
As stated in Section 9.1.1, a positive finding for all three criteria is required in order to
make a determination to regulate a contaminant. While there is evidence that manganese may
have adverse health effects in humans at high doses through inhalation, the evidence for adverse
effects through oral exposure at low or moderate levels is less compelling. Because manganese is
an essential nutrient, concern over potential toxic effects from high oral exposure must be
balanced against concern for adverse effects from manganese deficiency should intake be too low.
Manganese has been found to occur in an estimated 2,200 ground water public water systems
representing more than 2.3 million people exposed (2.7% of the population) to levels at or above
0.3 mg/L. The Agency believes that a meaningful opportunity for health risk reduction does not
exist for persons served by public water systems because the average dietary intake of manganese
exceeds the contribution normally found in public drinking water systems. Thus, based on the
evaluation of available data using the criteria described above, the regulatory determination is
"Do not regulate".
External Review Draft—Manganese—April 2002
9-11
-------
-------
10.0 REFERENCES
Abbott, P. 1987. Methylcyclopentadienyl manganese tricarbonyl (MMT) in petrol: The
toxicological issues. Sci. Total Environ. 67:247-255.
Abrams, E^J.W. Lassiter, W.J. Miller, et al. 1976. Effect of dietary manganese as a factor
affecting 54Mn absorbtion in rats. Nutr. Rep. Int. 14:561-565 (as cited in ATSDR, 2000).
Ali, M.M., G.S. Shukla, D.K. Saxena, et al. 1981. Behavioral dysfunctions and central
neurotransmitters in manganese exposed rats. J. Environ. Biol. 2:29-39 (as cited in ATSDR,
2000).
Ali, M.M.; R.C. Murthy, D.K. Saxena, et al. 1983. Effect of low protein diet on manganese
neurotoxicity: I. Developmental and biochemical changes. Neurobehav Toxicol Teratol 5:377-
383.
Ali, M.M., R.C. Murthy, S.K. Mandal, et al. 1985. Effect of low protein diet on manganese
neurotoxicity: III. Brain neurotransmitter levels. Neurobehav. Toxicol. Teratol. 7(5):427-431 (as
cited in U.S. EPA, 1993).
Amdur, M.O., L.C. Norris and G.F. Heuser. 1944. The need for manganese in bone
development by the rat. Proc. Soc. Exp. Biol. Med. 59:254-255 (as cited in ATSDR, 2000).
Andersen, M.E., J.M. Gearhart, H.J. Clewell, HI, et al. 1999. Pharmacokinetic data needs to
support risk assessments for inhaled and ingested manganese. Neurotoxicol. 20:161-172.
ANSI/MSF. 2000. American National Standards Institute, NSF International. Drinking water
treatment chemicals - Health effects. ANSI/NSF 60-2000. Ann Arbor, MI.
Archibald, F.S. and C. Tyree. 1987. Manganese poisoning and the attack of trivalent manganese
upon catecholamines. Arch. Biochem. Biophys. 2:638-650.
Ardeleanu, A., S. Loranger, G. Kennedy, et al. 1999. Emission rates and physico-chemical
characteristics of Mn particles emitted by vehicles using methylcyclopentadienyl manganese
tricarbonyl (MMT) as an octane improver. Water Air -Soil Pollut. 115:411-427.
Arnaud, J. And A. Favier. 1995. Copper, iron, manganese and zinc contents in human colostrum
. and transitory milk of French women. Sci Total Environ 159:9-15 (as cited in ATSDR, 2000).
Aschner, M. and J.L. Aschner. 1990. Manganese transport across the blood-brain barrier:
relationship to iron homeostasis. Brain. Res. Bull. 24:857-860 (as cited in U.S. EPA, 1993).
Aschner, M. and J.L. Aschner. 1991. Manganese neurotoxicity: cellular effects and blood-brain
barrier transport. Neurosci. Biobehav. Rev. 15(3):333-340 (as cited in ATSDR, 2000).
External Review Draft — Manganese —April 2002
10-1
-------
ASNS. 1999. Nutrient Information: Manganese. American Society for Nutritional Sciences. .
Available at http://www.nutrition.org/nutinfo
t
ATSDR. 1997. Toxicological Profile for Manganese (Update). Draft for Public Comment. U.S.
Department of Health and Human Services, Public Health Service, Agency for Toxic Substances
and Disease Registry, Atlanta, GA.
ATSDR. 2000. Toxicological Profile for Manganese (Final). U.S. Department of Health and
Human Services, Public Health Service, Agency for Toxic Substances and Disease Registry,
Atlanta, GA. [
Autissier, N., L. Rochette, P. Dumas, et al. 1982. Dopamine and norepinephrine turnover in
various regions of the rat brain after chronic manganese chloride administration. Toxicology
24(2):175-182 (as cited in U.S. EPA, 1993). I
i •
!
Bales, C.W., J.H. Freeland-Graves, P.H. Lin, et al. 1987. Plasma uptake of manganese:
Influence of dietary factors. In: Kies, C., ed., Nutritional Unavailability of Manganese.
Washington D.C.: American Chemical Society, pp. 112-122 (as cited in U.S. EPA, 1993).
t
Baly, D.L., B. Lonnerdal and C.L. Keen. 1985. Effects of high doses of manganese on
carbohydrate homeostasis. Toxicol. Lett. 25:95-102 (as cited in U.S. EPA, 1993).
Baly, D.L., B. Lonnerdal and C.L. Keen. 1988. Mechanism of decreased insulinogenesis in
manganese-deficient rats. Decreased insulin mRNA. FEES Lett. 239:55-58 (as cited in ATSDR,
2000). I
Banta, R.G. and W.R. Markesbery. 1977. Elevated manganese levels associated with dementia
and extrapyramidal signs. Neurology 27(3):213-216 (as cited in U.S. EPA, 1993).
Barbeau, A. 1984. Manganese and extrapyramidal disorders (a critical review and tribute to Dr.
George C. Cotzias). Neurotoxicology 5(1): 131-35 (as cited in ATSDR, 2000).
Barlow, PJ. and M. Kapel. 1979. Hair metal analysis and its significance to certain disease
conditions. 2nd Annual Trace Metals Health Seminar, Boston, MA (as cited in U.S. EPA, 1993).
Baxter, D.J., W.O. Smith and G.C. Klein. 1965. Some effects of acute manganese excess in rats.
Proc. Soc. Exp. Biol. Med. 119:966-970 (as cited in U.S. EPA, 1993).
l1
Bernheimer, H., W. Birkmayer, O. Hornykiewicz, et al. 1973. Brain dopamine and the
syndromes of Parkinson and Huntington: Clinical, morphological and neurochemical correlations.
J. Neurol. Sci. 20:415-455 (as cited in U.S. EPA, 1993).
External Review Draft 4- Manganese—April 2002
10-2
-------
Bertinchamps, A.J. and G.C. Cotzias. 1958. Biliary excretion of manganese. Fed. Proc. 17:428
(as cited in U.S. EPA, 1984; 1993).
Bertinchamps, A.J., S.T. Miller and G.C. Cotzias. 1966. Interdependence of routes excreting
manganese. Am. J. Physiol. 21 l(l):217-224 (as cited in U.S. EPA, 1984; 1993).
Bienvenu, P., C. Noire and A. Cier. 1963. Comparative general toxicity of metallic ions. A
relation with the periodic classification. Rech. Serv. Sante Armees, Lyons, France 256:1043-
1044 (as cited in U.S. EPA, 1984; 1993).
Bonilla, E. and M. Diez-Ewald. 1974. Effect of L-DOPA on brain concentration of dopamine
and homovanillic acid in rats after chronic manganese chloride administration. J. Neurochem.
22(2):297-299 (as cited in U.S. EPA, 1993).
Bonilla, E. 1978a. Flameless atomic absorption spectrophotometric determination of manganese
in rat brain and other tissues. Clin. Chem. 24:471-472.
Bonilla, E. 1978b. Increased GAB A content in caudate nucleus of rats after chronic manganese
chloride administration. J. Neurochem. 31(2):551-552 as cited in U.S. EPA, 1993).
Bonilla, E. 1980. L-tyrosine hydroxylase activity in the rat brain after chronic oral administration
of manganese chloride. Neurobehav. Toxicol. 2:37-41 (as cited in U.S. EPA, 1984; 1993).
Boyer, P.D., J.H. Shaw and P.H. Phillips. 1942. Studies on manganese deficiency in.the rat. J.
Biol. Chem. 143:417-425 (as cited in ATSDR, 2000).
Britton, A.A. and G.C. Cotzias. 1966. Dependence of manganese turnover on intake. Am. J.
Physiol. 211(1):203-206.
Brouillet, E.P., L. Shinobu, U. McGarvey, et al. 1993. Manganese injection into the rat striatum
produces excitotoxic lesions by impairing energy metabolism. Exp. Neurol. 120(l):89-94 (as
cited in ATSDR, 2000).
Burnett, W.T., Jr., R.R. Bigelow, A.W. Kimball, et al. 1952. Radio-manganese studies on the
mouse, rat and pancreatic fistula dog. Am. J. Physiol. 168:620-625 (as cited in U.S. EPA, 1984;
1993).
Calne, D.B., N.S. Chu, C.C. Huang, et al. 1994. Manganism and idiopathic parkinsonism:
similarities and differences. Neurology 44:1583-1586.
Carter, S.D., J.F. Hein, G.L. Rehnberg, et al. 1980. Chronic manganese oxide ingestion in rats:
hematological effects. J. Toxicol. Environ. Health 6(1):207-216 (as cited in U.S. EPA, 1993).
External Review Draft — Manganese—April 2002
10-3
-------
Casto, B.C., J. Meyers and J.A. DiPaolo. 1979. Enhancement of viral transformation for the
evaluation of the carcinogenic or mutagenic potential of inorganic metal salts. Cancer Res.
39:193-198 (as cited in ATSDR, 2000). ;
[
Cawte, J. and M.T. Florence. 1989. A manganic milieu in North Australia: Ecological
manganism: Ecology; diagnosis; individual susceptibility, synergism, therapy, prevention, advice
for the community. Int. J. Biosocial Med. Res. 11:43-56 (as cited in U.S. EPA, 1993).
Chan, A.W.K., J.C.K. Lai, M.J. Minski, et-al. 1981. Manganese concentration in rat organs:
Effect after life-long manganese treatment. Bipchem. Soc. Trans. 9:229 (as cited in U.S. EPA,
1993). I .
Chan, W.Y., J.M. Bates, Jr. and O.M. Rennert.: 1982. Comparative studies of manganese binding
in human breast milk, bovine milk and infant formula. J. Nutr. 112(4):642-651 (as cited in U.S.
EPA, 1993). \
Chan, W.Y., M.H. Raghib and O.M. Rennert. 1987. Absorption studies of manganese from milk
diets in suckling rats. In: Kies, C., ed., Nutritional Bioavailability of Manganese. Washington,
D.C.: American Chemical Society, pp. 80-89 (as cited in U.S. EPA, 1993).
Chandra, S.V. and S.P. Srivastava. 1970. Experimental production of early brain lesions in rats
by parenteral administration of manganese chlpride. Acta Pharmacol. Toxicol. 28(3): 177-183 (as
cited in U.S. EPA, 1984; 1993).
Chandra, S.V. 1971. Cellular changes induced by manganese in the rat testis - preliminary
results. Acta Pharmacol. Toxicol. 29(1):75-80 (as cited in U.S. EPA, 1984; 1993).
Chandra, S.V. 1972. Histological and histochjemical changes in experimental manganese
encephalopathy in rabbits. Arch. Toxikol. 29(J):29-38 (as cited in U.S. EPA, 1984; 1993).
Chandra, S.V., R. Ara, N. Nagar, et al. 1973. Sterility in experimental manganese toxicity. Acta
Biol. Med. Ger. 30(6):857-862 (as cited in U.S. EPA, 1993).
Chandra, S.V., D.K. Saxena and M.Z. Hasan. 11975. Effect of zinc on manganese induced
testicular injury in rats. Ind. Health 13:51-56 | (as cited in U.S. EPA, 1993).
Chandra, S.V., and G.S. Shukla. 1978. Manganese encephalopathy in growing rats. Environ Res
15:28-37 (as cited in ATSDR, 2000).
Chandra, S.V., G.S. Shukla and D.K. Saxena. J1979. Manganese-induced behavioral dysfunction
and its neurochemical mechanism in growing mice. J. Neurochem. 33(6): 1217-1221 (as cited in
U.S. EPA, 1993).
External Review Draft — Manganese—April 2002
10-4
-------
Chandra, S.V. and G.S. Shukla. 1981. Concentrations of striatal catecholamines in rats given
manganese chloride through drinking water. J. Neurochem. 36(2):683-687 (as cited in U.S.
EPA, 1993).
Chang, L,W. 1996. Toxico-neurology and neuropathology induced by metals. In: Chang, L.W.,
ed,. Toxicology of Metals. Boca Raton, FL: CRC Press, Inc., pp. 511-536.
ChemLDplus. 2000. Division of Specialized Information Services, National Library of Medicine
(NLM)nttp://chem.sis.nmi.nih.gov/cheniidplus/
Cikrt, MLand J. Vostal. 1969. Study of manganese resorption in vitro through intestinal wall.
Int. Z. Klin. Pharmakol. Ther. Toxikol. 2(3):280-285 (as cited in U.S. EPA, 1984; 1993).
Cikrt, M. 1973. Enterohepatic circulation of ^Cu, ^Mn and 203Hg in rats. Arch. Toxikol.
31(1):51-59 (as cited in U.S. EPA, 1984; 1993).
Clayton, C., E. Pellizzari, R. Rodes, et al. 1999. Estimating distribution of long-term particulate
matter and manganese exposures for residents of Toronto, Canada. Atmos. Environ. 33:2515-
2526.
Collipp, P.J., Chen, S.Y., and Maitinsky, S. 1983. Manganese in infant formulas and learning
disability. Ann. Nutr. Metab. 27 (6): 488-494. (As cited in U.S. EPA, 1993).
Colomina, M.T., J.L. Domingo, J.M. Llobet, et al. 1995. Embryotoxicity and fetotoxicity of
manganese in mice: variability with the day of exposure. Toxicologist 15(1):160 [abstract].
Cooper, W. 1984. The health implications of increased manganese in the environment resulting
from the combustion of fuel additives: a review of the literature. J. Toxicol. Envir. Hlth. 14:23-
46.
Cotzias, G.C. 1958. Manganese in health and disease. Physiol. Rev. 38:503-533 (as cited in
U.S. EPA, 1984; 1993).
Cotzias, G.C., K. Horiuchi, S. Fuenzalida, et al. 1968. Chronic manganese poisoning. Clearance
of tissue manganese concentrations with persistence of the neurological picture. Neurology
18(4):376-382 (as cited in U.S. EPA, 1984; 1993).
Cotzias, G.C., S.T. Miller, P.S. Papavasiliou, et al. 1976. Interactions between manganese and
brain dopamine. Med. Clin. North Am. 60(4):729-738 (as cited in U.S. EPA, 1993).
Dagli, A.J., D. Golden, M. Finkel, et al. 1973. Pyloric stenosis following uigestion of potassium
permanganate. Digest Dis. 18:1091-1094 (as cited in U.S. EPA, 1984,1993).
External Review Draft — Manganese—April 2002
10-5
-------
the rat, with special reference to the C.N.S. Q.
EPA, 1993).
Dastur, D.K., D.K. Manghani, K.V. Raghavendran, et al. 1969. Distribution and fate of Mn54 in
J. Exp. Physiol. 54(3):322-331 (as cited in U.S.
Dastur, D.K., D.K. Manghani and K.V. Raghavendran. 1971. Distribution and fate of 54Mn in
the monkey: Studies of different parts of the central nervous system and other organs. J. Clin.
Invest. 50:9-20 (as cited in U.S. EPA, 1993).
t
I
Davidsson, L., A. Cederblad, B. Ldnnerdal, et al. 1989a. Manganese absorption from human
milk, cow's milk, and infant formulas in humans. Am. J. Dis. Child 43(7):823-827.
I
Davidsson, L., A. Cederblad, B. Lonnerdal, et al. 1989b. Manganese retention in man: a method
for estimating manganese absorption in man. ^m. J. Clin. Nutr. 49(1): 170-179.
I
Davies, N.T. and R. Nightingale. 1975. The effects of phytate on intestinal absorption and
secretion of zinc, and whole-body retention of Zn, copper, iron and manganese in rats. Br. J.
Nutr. 34(2):243-258 (as cited in U.S. EPA, 1993).
I
Davis, C.D., D.M. Ney and J.L. Greger. 1990.1 Manganese, iron and lipid interactions in rats. J.
Nutr. 120(5):507-513 (as cited in ATSDR, 2000).
Davis, C.D. and J.L. Greger. 1992. Longitudinal changes of manganese-dependent superoxide
dismutase and other indexes of manganese and iron status in women. Am. J. Clin. Nutr.
55(3):747-752 (as cited in ATSDR, 2000). j .
Davis, C.D., E..A .Malecki, and J.L. Greger. 1^92. Interactions among dietary manganese., heme
iron, and non-heme iron in women. Am. J. Clin. Nutr. 56:926-932.
Davis, C.D., L. Zech, J.L. Greger. 1993. Manganese metabolism in the rats: An improved
methodology for assessing gut endogenous losses. Proc. Soc. Exp. Biol. Med. 202:103-108.
i
Davis, J.M., A.M. Jarabek and D.T. Mage. 1998. The EPA health risk assessment of
methylcyclopentadienyl manganese tricarbonyl (MMT). Risk Anal. 18(1):57-70.
i
De Meo, M., M. Laget, M. Castegnaro, et al. 1991. Genotoxic activity of potassium
permanganate in acidic solutions. Mutat. Res. 260(3):295-306 (as cited in ATSDR, 2000).
Demerec, M., G. Bertani and J. Flint. 1951. A survey of chemicals for mutagenic action oaE.
co//. Am. Nat. 85:119-136 (as cited in WHO, |l981; U.S. EPA, 1993).
Deskin, R., S.J. Bursian and F.W. Edens. 1980. Neurochemical alterations induced by
manganese chloride in neonatal rats. Neurotoxicology 2:65-73 (as cited in U.S. EPA, 1993).
External Review Draft — Manganese—April 2002
10-6
-------
Deskin, R., S. J. Bursian and F. W. Edens. 1981. The effect of chronic manganese administration
on some neurochemical and physiological variables in neonatal rats. Gen. Pharmacol. 12(4):279-
280 (as cited in U.S. EPA, 1993).
Devenyi, A.G., T.F. Barron and A.C. Mamourian. 1994. Dystonia, hyperintense basal ganglia,
and high whole blood manganese levels in Alagille's syndrome. Gastroenterology 106(4):1068-
1071 (as cited in U.S. EPA, 1999a). '
Dieter, H.H., W. Retard, J. Simon,.et al. 1992. Manganese in natural mineral waters from
Germany. Die Nahrung 36:477-484 (as cited in U.S. EPA, 1993).
Dikshith, T.S. and S.V. Chandra. 1978. Cytological studies hi albino rats after oral
administration of manganese chloride. Bull. Environ. Contam. Toxicol. 19(6):741-746 (as cited
in ATSDR, 2000).
DiPaolo, J.A. 1964. The potentiation of lymphosarcomas in mice by manganese chloride
. [Abstract]. Fed. Proc. 23:393 (as cited in U.S. EPA, 1984,1993).
Doisy, E.A. 1973. Effects of deficiency in manganese upon plasma levels of clotting proteins and
cholesterol in man. Trace Element Metabolism. In: Hoekstra, W.G., J.W. Suttie, A.E. Ganther, et
al., eds. Animals-2,2nd eds. Baltimore: University Park Press, pp. 668-670 (as cited in ATSDR,
2000),
Dorman, D.C., M.F. Struve, D. Vitarella, et al. 2000. Neurotoxicity of manganese chloride in
neonatal and adult CD rats following subchronic (21-day) high-dose oral exposure. J Appl Tox
20:179-187.
Dorner, K., S. Dziadzka, A. Hohn, et al. 1989. Longitudinal manganese and copper balances in
young infants and preterm infants fed on breast-milk and adapted cow's milk formulas. Br J
Nutrition 61:559-572.
Durham, N.N. and O. Wyss. 1957. Modified method of determining mutation rates in bacteria.
J. Bacteriol. 74:548-552 (as cited in WHO, 1981; U.S. EPA, 1993).
Ejima, A., T. Imamura, S. Nakamura, etal. 1992. Manganese intoxication during total
parenteral nutrition [letter]. Lancet 339: 426 (as cited in U.S. EPA, 1993).
Eriksson, H., S. Lenngren and E. Heilbronn. 1987. Effect of long-term administration of
manganese on biogenic amine levels in discrete striatal regions of rat brain. Arch. Toxicol.
59(6):426-431 (as cited in U.S. EPA, 1993).
Eriksson, EL, J. Tedroff, K.A. Thuomas, et al. 1992. Manganese induced brain lesions in
Macacafascicularis as revealed by positron emission tomography and magnetic resonance
imaging. Arch. Toxicol. 66(6):403-710 (as cited in ATSDR, 2000).
External Review Draft — Manganese—April 2002
10-7
-------
Ethyl Corporation. 1990. The case for an environmentally beneficial fuel additive. Baton Rouge,
USA. !
Fechter, L.D. 1999. Distribution of manganese in development. Neurotoxicology 20:197-201.
Fell, JJM.E., A.P. Reynolds, N. Meadows, et al. 1996. Manganese toxicity in children receiving
long-term parenteral nutrition. Lancet 347:1218-1221.
Finley, J.W. 1999. Manganese absorption and retention by young women is associated with
serum ferritin concentration. Am. J. Clin. Nutrl 70:37-43.
i
Finley, J. And C.D. Davis. 1999. Manganese deficiency and toxicity: Are high or low dietary
amounts of manganese cause for concern? Biofactors 10(1): 15-24.
!
Finley, J.W., P.E. Johnson and L.K. Johnson. 1994. Sex affects manganese absorption and
retention by humans from a diet adequate in manganese. Am. J. Clin. Nutr. 60(6):949-955.
Florence, T.M. and J.L. Stauber. 1989. Manganese catalysis of dopamine oxidation. Sci. Total.
Environ. 78:233-240 (as cited in U.S. EPA, 1993).
Food and Nutrition Board. 2002. Dietary Reference Intakes: Vitamin A, Vitamin K, Arsenic,
Boron, Chromium, Copper, Iodine, Iron, Manganese, Molybdenum, Nickel, Silicon, Vanadium,
and Zinc. Washington DC: National Academy;Press. pp. 10-1-10-22.
Fore, H. and R.A. Morton. 1952. Manganese in rabbit tissues. Biochem. J. 51:600-603 (as
cited in ATSDR, 2000). i
Franz, R.D. 1962. Toxicities of some trace metals. Naunyn-Schmiedebergs Arch. Exp. Path.
Pharmakol. 244:17-20 (as cited in U.S. EPA, 1984,1993).
|
Freeland-Graves, J. 1994. Derivation of manganese estimated safe and adequate daily dietary
intakes. In: Mertz, W., C.O. Abernathy and S.S. Olin, eds., Risk Assessment of Essential
Elements. Washington, D.C.: ILSI Press, pp. 237-252.
Freeland-Graves, J.H., C.W. Bales and F. Behmardi. 1987. Manganese requirements of humans.
In: Kies, C., ed., Nutritional Bioavailability of Manganese. Washington, D.C.: American
Chemical Society, pp. 90-104 (as cited in U.S. EPA, 1999a).
Freeland-Graves, J.H. and C. Llanes. 1994. Models to study manganese deficiency. In: Klimis-
Tavantzis, D.J., ed., Manganese in Health and Disease. Boca Raton, FL: CRC Press, Inc., pp.
59-86. !
External Review Draft -r- Manganese—April 2002
10-8
-------
Friedman, B.J., J.H. Freeland-Graves, C.W. Bales, et al. 1987. Manganese balance and clinical
observations in young men fed a manganese-deficient diet. J. Nutr. 117:133-143 (as cited in
U.S. EPA, 1993; ATSDR, 2000).
Furst, A. 1978. Tumorigenic effect of an organomanganese compound on F344 rats and Swiss
albino mice. J. Natl. Cancer Inst. 60(5): 1171-1173. (As cited in U.S. EPA, 1984,1993).
Garcia-Aranda, J.A., R.A. Wapnir and F. Lifshitz. 1983. In vivo intestinal absorption of
manganese in the rat. J. Nutr. 113(12):2601-2607 (as cited in U.S. EPA, 1993).
Gavin, C.E., K.K. Gunter and T.E. Gunter. 1992. Mn2+ sequestration by mitochondria and
inhibition of oxidative phosphorylation. Toxicol. Appl. Pharmacol. 115(1): 1-5 (as cited in
ATSDR, 2000).
Gianutsos, G. and M.T. Murray. 1982. Alterations in brain dopamine and GABA following
inorganic or organic manganese administration. Neurotoxicology 3(3):75-81 (as cited in U.S.
EPA, 1993).
Gianutsos, G., M.D. Seltzer, R. Saymeh, et al. 1985. Brain manganese accumulation following
systemic administration of different forms. Arch. Toxicol. 57(4):272-275 (as cited in U.S. EPA,
1993).
Gibbons, R.A., S.N. Dixon, K. Hallis, et al. 1976. Manganese metabolism in cows and goats.
Biochim.. Biophys. Acta 444:1-10 (as cited in U.S. EPA, 1993).
Gibson, R.S. 1994. Content and bioavailability of trace elements in vegetarian diets. AmJClin
Nutr 59(5 Suppl):1223S-1232S.
Gilliom, R.J., D.K. Mueller, and L.H. Nowell. 1998. Methods for comparing water-quality
conditions among National Water-Quality Assessment Study Units, 1992-95. U.S. Geological
Survey Open-File Report 97-589. Available on the Internet at: URL:
http://ca.water.usgs.gov/pnsp/rep/ofr97589/. Last modified October 09,1998.
Goldsmith, J., Y. Herishanu, J. Abarbanel, et al. 1990. Clustering of Parkinson's disease points
to environmental etiology. Arch. Environ. Health 45:88-94 (as cited in ATSDR, 2000).
Gong, H. and T. Amemiya. 1996. Ultrastructure of retina of manganese-deficient rats. Invest.
Ophthalmol. Vis. Sci. 37:1967-1974.
Gottschalk, L.A., T. Rebello, M.S. Buchsbaum, et al. 1991. Abnormalities in hair trace elements
as indicators of aberrant behavior. Compr. Psychiatry 32(3):229-237 (as cited in U.S. EPA,
1993).
External Review Draft — Manganese—April 2002
10-9
-------
Grant, D., W.F. Blazak and G.L. Brown. 1997
The reproductive toxicology of intravenously
administered MnDPDP in the rat and rabbit. ActaRadiol. 38(4 Pt 2):759-769.
Gray, L.E., Jr. and J.W. Laskey. 1980. Mulitvariate analysis of the effects of manganese on the
reproductive physiology and behavior of the male house mouse. J. Toxicol. Environ. Health
6:861-867 (as cited in U.S. EPA, 1993). • ! '
Greenberg, D.M. and W.W. Campbell. 1940. Studies in mineral metabolism with the aid of
induced radioactive isotopes. IV. Manganese. Proc. Natl. Acad. Sci. 26:448-452 (as cited in
U.S. EPA, 1984,1993). |
Greenberg, D.M., D.H. Copp and E.M. Cuthbertson. 1943. Studies in mineral metabolism with
the aid of artificial radioactive isotopes. VII. The distribution and excretion, particularly by way of
the bile, of iron, cobalt, and manganese. J.Biol. Chem. 147:749-756 (as cited in U.S. EPA,
1984,1993). I • .
l
Greger, J.L. 1998. Dietary standards for manganese: Overlap between nutritional and
toxicological studies. J.Nutr. 128:3688-3718.
Greger, J.L. 1999. Nutrition versus toxicology of manganese in humans: Evaluation of potential
biomarkers. Neurotoxicology 20:205-212.
. Greger, J.L. and SJM.Snedeker. 1980. Effect of dietary protein and phosphorus levels onthe
utilization of zinc, copper, and manganese by adult males. J.Nutr. 110:2243-2253.
Greger, J.L., C.D. Davis, J.W. Suttie, et al. 1990. Intake, serum concentrations, and urinary
excretion of manganese by adult males. Am. J. Clin. Nutr. 51(3):457-461.
Gruden, N. 1984. The influence of iron on manganese metabolism in the first three weeks of rat's
life. Nutr. Rep. Int. 30:553-557 (as cited in U.S. EPA, 1993).
Gupta, S.K., R.C. Murthy and S.V. Chandra. 1980. Neuromelanin in manganese-exposed
primates. Toxicol. Lett. 6:17-20 (as cited in U.S. EPA, 1993).
Hagenfeldt, K., L.O. Plantin and E. Diczfalusy. 1973. Trace elements in the human
endometrium. 2. Zinc, copper and manganese levels in the endometrium, cervical mucus and
plasma. Acta Endocrinol. (Copenh.) 72(1):115-126 (as cited in U.S. EPA, 1984,1993).
Hamilton-Koch, W., R.D. Snyder and J.M. Lavelle. 1986. Metal-induced DNA damage and
repair in human diploid fibroblasts and Chinese hamster ovary cells. Chem. Biol. Interact.
59(l):17-28 (as cited in U.S. EPA, 1993). !
External Review Draft — Manganese—April 2002
10-10
-------
Hanna, L., J.M. Peters, L.M. Wiley, et al. 1996. Comparative effects of essential and
nonessential metals on preimplantation mouse embryo development [abstract]. Faseb. J.
10(3):A783.
Hanzlik, R.P., R. Stitt and G.J. Traiger. 1980. Toxic effects of methylcyclopentadienyl
manganese tricarbonyl (MMT) in rats: Role of metabolism. Toxicol. Appl. Pharmacol. 56:353-
360 (as cited in U.S. EPA, 1994).
Hart, D.A. 1978. Evidence that manganese inhibits an early event during stimulation of
lymphocytes by mitogens. Exp. Cell. Res. 113:139-150 (as cited in ATSDR, 2000).
He, P., D. Liu, G. Zhang, et al. 1994. [Effects of high-level manganese sewage irrigation on
children's neurobehavior. Chung Hua Yu Fang I Hsueh Tsa Chih] 28:216-218. (Chinese) .
Health aind Welfare Canada. 1978. Methylcyclopentadienyl manganese tricarbonyl (MMT): As
an assessment of the human health implications of its use as a gasoline additive. Environmental
Health Directorate. Health Protection Branch, Ottawa, Canada. 78-EHD-21.
Hetjmancik, M., A.C. Peters, J.D. Toft, et al. 1987a. The chronic study of manganese sulfate
monohydrate (CAS No. 10034-96-5) in F344 rats. Report to National Toxicology Program,
Research Triangle Park, NC, by Battelle's Columbus Laboratories, Columbus, OH.
Hetjmancik, M., A.C. Peters, J.D. Toft, et al. 1987b. The chronic study of manganese sulfate
monohydrate (CAS No. 10034-96-5) in B6C3F1 mice. Report to National Toxicology Program,
Research Triangle Park, NC, by Battelle's Columbus Laboratories, Columbus, OH.
Hellou, J., L. Fancy and J. Payne. 1992. Concentrations of twenty-four elements in bluefin tuna,
Thunnus thynnus, from the Northwest Atlantic. Chemosphere 24(2):211-218.
Hietanen, E., J. Kilpio and H. Savolainen. 1981. Neurochemical and biotransformational enzyme
responses to manganese exposure in rats. Arch. Environ. Contam. Toxicol. 10(3):339-345 (as
cited in U.S. EPA, 1993).
Hinderer, R.K. 1979. Toxicity studies of methylcyclopentadienyl manganese tricarbonyl (MMT).
Am. Ind. Hyg. Assoc. J. 40:164-167 (as cited in U.S. EPA, 1994).
Holbrook, D.J., Jr., M.E. Washington, H.B. Leake, et al. 1975. Studies on the evaluation of the
toxicity of various salts of lead, manganese, platinum, and palladium. Environ. Health Perspect.
10:95-101 (as cited in U.S. EPA, 1984,1993).
Holzgraefe, M., W. Poser, H. Kijewski, et al. 1986. Chronic enteral poisoning caused by
potassium permanganate: A case report. J. Toxicol. Clm. Toxicol. 24:235-244 (as cited in
ATSDR, 2000).
External Review Draft — Manganese—April 2002
10-11
-------
Horiuchi, K., S. Horiguchi, N. Tanaka, et al. 1967. Manganese contests in the whole blood,
urine and feces of a healthy Japanese population. Osaka City Med. J. 13(2):151-163 (original in
Japanese).
i _
Hurley, L.S., D.E. Wolley and P.S. Timiras. 1961. Threshold and pattern of electro shock
seizures in ataxic manganese-deficient rats. Proc. Soc. Exp. Biol. Med. 106:343-346 (as cited in
ATSDR, 2000). i
Hurley, L.S., C.L. Keen and D.L. Baly. 1984. Manganese deficiency and toxicity: effects on
carbohydrate metabolism in the rat. Neurotoxicology 5(1):97-104 (as cited in ATSDR, 2000).
Hysell, D.K., W. Moore, Sr., J.F. Stara, et al. 1974. Oral toxicity of methylcyclopentadienyl
manganese tricarbonyl (MMT) in rats. Environ. Res. 7:158-168 (as cited in U.S. EPA, 1994).
[
Imam, Z. and S.V. Chandra. 1975. Histochernical alterations in rabbit testis produced by
manganese chloride. Toxicol. Appl. Pharmacol. 32(3):534-544 (as cited in U.S. EPA, 1984,
1993).
i
Iwami, O., T. Watanabe, T., C.S., Moon, et al. [ 1994. Motor neuron disease on the Kii Peninsula
of Japan: excess manganese intake from food coupled with low magnesium in drinking water as a
risk factor. Sci. Total Environ. 149:121-135 (as cited in ATSDR, 2000).
Jarvinen, R. and A. AhlstrSm. 1975. Effect of the dietary manganese level on tissue manganese,
iron, copper, and zinc concentrations in female rats and their fetuses. Med. Biol. 53:93-99 (as
cited in U.S. EPA, 1984,1993). :
Joardar, M. and A. Sharma. 1990. Comparisori of clastogenicity of inorganic Mn administered in
cationic and anionic forms in vivo. Mutat. Res. 240(3): 159-163 (as cited in U.S. EPA, 1993).
Johnson, P.E., G.I. Lykken and E.D. Korynta. 1991. Absorption and biological half-life in
humans of intrinsic MMn tracers from foods of plant origin. J. Nutr. 121:711-717.
Jonderko, G. 1965. Calcium, manganese, inorganic phosphorus, sodium, potassium and iron
level in the blood serum in the acute experimental manganism. Med. Pr. 16(4):288-292 (original
in Polish) (as cited in U.S. EPA, 1993). i
Kafritsa, Y., J. Fell, S. Long, et al. 1998. Long term outcome of brain manganese deposition in
patients on home parenteral nutrition. Arch. Djis. Child 79:263-265.
Kanernatsu, N., M. Hara and T. Kada. 1980. Rec assay andmutagenicity studies on metal
compounds. Mutat. Res. 77:109-116 (as cited in ATSDR, 2000).
External Review Draft — Manganese —April 2002
10-12
-------
Kaplan, R.W. 1962. Problems testing pharmaceutical products, additives and other chemicals for
their mutagenic action. Naturwissen-schaften 49:457-462 (original in German) (as cited in
WHO, 1981; U.S. EPA, 1993).
Kato, M. 1963. Distribution and excretion of radiomanganese administered to the mouse. Q. J.
Exp.Physiol. 48:355-369 (as cited in U.S. EPA, 1984, 1993).
Kaur, G., S.K. Hasan and R.C. Srivastava. 1980. The distribution of manganese-54 in fetal,
young and adult rats. Toxicol. Lett. 5:423-426 (as cited in U.S. EPA, 1993).
Kawamura, C.L., H. Ikuta, S. Fukuzimi, et al. 1941. Intoxication by manganese in well water.
Kitasato Arch. Exp. Med. 18:145-169 (as cited in U.S. EPA, 1993,1999a).
Kawano, J., D.M. Ney, C.L. Keen, et al. 1987. Altered high density lipoprotein composition in
manganese-deficient Sprague-Dawley and Wistar rats. J.Nutr. 117:902-906 (as cited in
ATSDR,2000).
Keen, C.L., J.G. Bell and B. Lonnerdal. 1986. The effect of age on manganese uptake and
retention from milk and infant formulas hi rats. J.Nutr. 116(3):395-402 (as cited in U.S. EPA,
1993).
Keen, C.L., J.L. Ensunsa, M.H. Watson, et al. 1999. Nutritional aspects of manganese from
experimental studies. Neurotoxicology 20:213-224.
Kennedy, S.D. and R.G. Bryant. 1986. Manganese deoxyribonucleic acid binding modes:
Nuclear magnetic relaxation dispersion results. Biophys. J. 50:669-676 (as cited in U.S. EPA,
1993).
Khan, K.N., 3M. Andress and P.F. Smith. 1997. Toxicity of subacute intravenous manganese
chloride administration in beagle dogs. Toxicol. Pathol. 25(4):344-350.
Khandelwal, S., M. Ashquin and S.K. Tandon. 1984. Influence of essential elements on
manganese intoxication. Bull. Environ. Contam. Toxicol. 32(1): 10-19 (as cited in U.S. EPA,
1993).
Kies, C. 1987. Nutritional Bioavailablity of Manganese. American Chemical Society,
Washington, D.C. (as cited in U.S. EPA, 1993).
Kilburn, C.J. 1987. Manganese, malformations and motor disorders: findings in a manganese-
exposed population. Neurotoxicology 8(3):421-429 (as cited hi U.S. EPA, 1993).
Kimura, M., N. Yagi and Y. Itokawa. 1978. Effect of subacute manganese feeding on serotonin
metabolism in the rat. J. Toxicol. Environ. Health 4(5-6):701-707 (as cited in U.S. EPA, 1993).
External Review Draft — Manganese—April 2002
10-13
-------
Klaassen, C.D. 1974. Biliary excretion of manganese in rats, rabbits, and dogs. Toxicol. Appl.
Pharmacol. 29(3):458-468 (as cited in U.S. EPA, 1993).
I
Komura, J. and M. Sakamoto. 1991. Short-term oral administration of several manganese
compounds in mice: physiological and behavioral alterations caused by different forms of
manganese. Bull. Environ. Contam. Toxicol. 46(6):921-928.
Komura, J. and M. Sakamoto. 1993. Subcellular and gel chromatographic distribution of
manganese in the mouse brain: Relation to the chemical form of chronically-ingested manganese.
Toxicol. Lett. 66(3):287-294 (as cited in ATSDR, 2000).
I
I
Kondakis, X.G., N. Makris, M. Leotsinidis, et al. 1989. Possible health effects of high
manganese concentration in drinking water. Arch. Environ. Health 44(3): 175-178 (as cited in
U.S. EPA, 1993).
Kondakis, X.G. 1990. Professor, University of Patras, Greece. Letter to S. Velazquez, U.S.
EPA, Cincinnati, OH. August 23 (as cited in U.S. EPA, 1993).
f. . ,
Kondakis, X.G. 1993. Professor, University of Patras, Greece. Letter to S. Velazquez, U.S.
EPA, Cincinnati, OH. June 7 (as cited in U.S. EPA, 1993).
Kontur, P.J. and L.D. Fechter. 1985. Brain manganese, catecholamine turnover, and the
development of startle in rats prenatally exposed to manganese. Teratology 32(1): 1-11 (as cited
inU.S.EPA,1993). :.
Kontur, P.J. and L.D. Fechter. 1988. Brain regional manganese levels and monoamine
metabolism in manganese-treated neonatal rats. Neurotoxicol. Teratol. 10(4):295-303 (as cited
inU.S.EPA,1993). • !
i
Kostial, K., D. Kello, S. Jugo, et al. 1978. Influence of age on metal metabolism and toxicity.
Environ. Health Perspect. 25:81-86 (as cited in U.S. EPA, 1993).
Kristensson, K., H. Eriksson, B. Lundh, et al. 1986. Effects of manganese chloride on the rat
developing nervous system. Acta Pharmacol. Toxicol. (Copenh.) 59(5):345-348 (as cited in
U.S. EPA, 1993). . I
Lai, J.C., T.K. Leung and L. Lim. 1981. Brain regional distribution of glutamic acid
decarboxylase, choline acetyltransferase, and acetylcholinesterase hi the rat: effects of chronic
manganese chloride administration after two yjears. J. Neurochem. 36(4): 1443-1448 (as cited in
U.S. EPA, 1993). ;
Lai, J.C., TJC. Leung, J.F. Guest, et al. 1982a. The effects of chronic manganese chloride
treatment expressed as age-dependent, transient changes in rat brain synaptosomal.uptake of
amines. J. Neurochem. 38(3):844-847 (as cited in U.S. EPA, 1993).
External Review Draft \— Manganese—April 2002
10-14
-------
Lai, J.C., T.K. Leung and L. Lim. 1982b. The ontogeny of acetylcholinesterase activities in rat
brain rejgions and the effect of chronic treatment with manganese chloride. J. Neurochem.
39(6):1767-1769 [abstract] (as cited in U.S. EPA, 1993).
Lai, J.C., T.K. Leung, and L. Lim. 1984. Differences in the neurotoxic effects of manganese
during development and aging: Some observations on brain regional neurotransmitter and non-
neurotransmitter metabolism in a developmental rat model of chronic manganese encephalopathy.
Neurotoxicology 5:37-47 (as cited in ATSDR, 2000).
Larsen, L.E. and D. Grant. 1997. General toxicology of MnDPDP. Acta Radiol. 38 (4 Pt
2):770-779. . •• • '
Larsen, N.A., H. Pakkenberg, E. Damsgaard, etal. 1979. Topographical distribution of arsenic,
manganese, and selenium in the normal human brain. J. Neurol. Sci. 42 (3):407-416 (as cited in
U.S. EPA, 1993).
Laskey, J.W., G.L. Rehnberg, J.F. Hein, et al. 1982. Effects of chronic manganese ( Mn3O4)
exposure on selected reproductive parameters in rats. J. Toxicol. Environ. Health 9(4):677-687.
Laskey, J.W., G.L. Rehnberg, J.F. Hein, et al. 1985. Assessment of the male reproductive
system in the preweanling rat following Mn3O4 exposure. J. Toxicol. Environ. Health 15(2):339-
350 (as cited in U.S. EPA, 1993).
Lawrence, D. A. 1981. Heavy metal modulation of lymphocyte activities. I. In vitro effects of
heavy metals on primary humoral immune responses. Toxicol. Appl. Pharmacol. 57:439-451 (as
cited in ATSDR, 2000).
Layrargues, G.P., C. Rose, L. Spahr, et al. 1998. Role of manganese hi the pathogenesis of
portal-systemic encephalopathy. Metab. Brain Dis. 13(4):311-317.
Leahy, P.P., and T.H. Thompson. 1994. The National Water-Quality Assessment Program.
U.S. Geological Survey Open-File Report 94-70. 4 pp. Available on the Internet at:
http://water.usgs.gov/nawqa/NAWQA.OFR94-70.html. Last updated August 23,2000.
Leung, T.K., J.C. Lai and L. Lim. 1981. The regional distribution of monoamine oxidase
activities towards different substrates: effects in rat brain of chronic administration of manganese
chlorideiand of ageing. J. Neurochem. 36(6):2037-2043 (as cited in U.S. EPA, 1993).
Leung, T.K., J.C. Lai and L. Lim. 1982. The effects of chronic manganese feeding on the
activity of monoamine oxidase in various organs of the developing rat. Comp. Biochem. Physiol.
71C (2):b23-228 (as cited in U.S. EPA, 1993).
External Review Draft—Manganese—April 2002
10-15
-------
Liccione, JJ. and M.D. Maines. 1989. Manganese-medicated increase in the rat brain
mitochondrial cytochrome P-450 and drug metabolism activity: Susceptibility of the striatum. J.
Pharmacol. Exp. Ther. 248:222-228 (as cited in U.S. EPA, 1993).
LSnnerdal, B., C.L. Keen, J.G. Bell, et al. 1987. Manganese uptake and retention: Experimental
animal and human studies. M: Kies, C., ed., Nutritional Bioavailability of Manganese. American
Chemical Society, Washington, D.C. pp. 9-20 (as cited in U.S. EPA, 1993).
L6nnerdal,B. 1994. Manganese nutrition of infants. In: KJimis-Tavantzis, D.J., ed., Manganese
in Health and Disease. CRC Press, Boca Raton, FL, pp. 175-191.
Loranger S. and J. Zayed. 1994. Manganese and lead concentrations in ambient air and emission
rates from unleaded and leaded gasoline between 1981 and 1992 in Canada: A comparative study.
Atmos. Environ. 28:1645-1651. \
Loranger, S. and J. Zayed. 1995. Environmental and occupational exposure to manganese: A
multimedia assessment. Int. Arch. Occup. Environ. Health. 67(2):101-110.
i
Loranger, S. and J. Zayed. 1997a. Environmental contamination and human exposure to
airborne total respirable manganese in Montreal. J. Air Waste Mngt. Assoc. 47(9):983-989
[Abstract]. !
Loranger, S., and J. Zayed. 1997b. Environmental contamination and human exposure
assessment to manganese in the St. Lawrence River ecozone (Quebec, Canada) using an
environmental fate/exposure model: GEOTOX. Environ. Res. 6(1-2):105-119.
Loranger, S., J. Zayed and E. Forget. 1994a. Manganese contamination in Montreal in relation
with traffic density. Water Air Soil Pollut. 74;385-396.
Loranger, S., G. Demers, G. Kennedy, et al. 1994b. The pigeon (Columba livid) as a monitor of
atmospheric manganese contamination from mobile sources. Arch. Environ. Contam. Toxicol.
27:311-317. ;
Loranger, S., J. Zayed and G. Kennedy. 1995. Contribution of methylcyclopentadienyl
manganese tricarbonyl (MMT) to atmospheric|Mn concentration near expressway: dispersion
modeling estimations. Atmos. Environ. 29:591-599.
Lynam D., G. Pfeifer, B. Fort, et al. 1994. Atmospheric exposure to manganese from use of
methylcyclopentadienyl manganese tricarbonyl (MMT) performance additive. Sci. Total Environ.
146/147:103-109.
Lynam, D.R., J.W. Roos, G.D. Pfeifer, et al. 1999. Environmental effects and exposures to
manganese from use of methylcyclopentadienyl manganese tricarbonyl (MMT) in gasoline.
Neurotoxicology 20(2-3):145-150.
External Review Draft — Manganese—April 2002 10-16
-------
Mahpmedy, M.C., Y.H. Mahomedy, P.A.S. Canhan, et al. 1975. Methaemoglobinaemia
following treatment dispensed by witch doctors. Anaesthesia 30:190-193 (as cited in U.S. EPA,
1984,1993).
Mahoney, J.P. and W.J. Small. 1968. Studies on manganese. 3. The biological half-life of
radiomanganese in man and factors which affect this half-life. J. Clin. Invest. 47(3):643-653 (as
cited in U.S. EPA, 1984,1993).
Malecki E.A. and J.L. Greger. 1996. Manganese protects against heart mitochondrial lipid
peroxidation in rats fed high levels of polyunsaturated fatty acids. J. Nutr. 126:27-33.
Malecki, E.A., G.M.Radazanowski, TJ.Radazanowski, D.D.Gallahler, and J.L.Greger. 1996.
Biliary manganese excretion in conscious rats is affected by acute and chronic manganese intake
but not by dietary fat. J.Nutr. 126:489-498.
Mandzgaladze, R.N. and M.I. Vasakidze. 1966. The effect of small doses of manganese
compounds, nitrogenous organomercury pesticides and some anticoagulants in white rat bone
marrow cells. Vopr. Gig. Tr. Profpatol. 10:209-212 (original in Russian) (as cited in WHO,
1981; U.S. EPA, 1993).
Mandzgaladze, R.N. 1966. On the mutagenic properties of manganese compounds. Vopr. Gig.
Tr. Profpatol. 10:225-226 (original in Russian) (as cited in WHO, 1981; U.S. EPA, 1993).
Matrone, G., R.H. Hartman and A.J. Clawson. 1959. Studies of a manganese iron antagonism in
the nutrition of rabbits and baby pigs. J.Nutr. 67:309-317 (as cited in U.S. EPA, 1993).
Maynard, L.S. and G.C. Cotzias. 1955. The partition of manganese among organs and
intracellular organelles of the rat. J. Biol. Chem. 214:489-495 (as cited in U.S. EPA, 1984,
1993).
McDennott, S.D. and C. Kies. 1987. Manganese usage in humans as affected by use of calcium
supplements. In: Kies, C., ed., Nutritional Bioavailability of Manganese. American Chemical
Society, Washington, D.C. pp. 146-151 (as cited in U.S. EPA, 1993).
Mella, H. 1924. The experimental production of basal ganglion symptomatology in macacus
rhesus. Arch. Neurol. Psych. 11:405-417 (as cited in U.S. EPA, 1993).
Mena, I., K. Horiuchi and K. Burke. 1969. Chronic manganese poisoning: Individual
susceptibility and absorption of iron. Neurology 19:1000-1006 (as cited in ATSDR, 2000).
Mena, I., J. Court, S. Fuenzalida, et al. 1970. Modification of chronic manganese poisoning.
Treatment with L-dopa or 5- OH tryptophane. N. Engl. J. Med. 282(1):5-10 (as cited in U.S.
EPA, 1993).
External Review Draft — Manganese—April 2002
10-17
-------
Mena,L 1974. The role of manganese in human disease. Ann. Clin. Lab. Sci. 4(6):487-491 (as
cited in U.S. EPA, 1993).
Merck. 1983. The Merck Index, 10th Ed. Merck & Co., Inc. Rahway,N.J.
Minoia, C., E. Sabbioni, P. Apostoli, et al. 1990. Trace element reference values in tissues from
inhabitants of the European community. I. A study of 46 elements in urine, blood and serum of
Italian subjects. Sci. Total Environ. 95:89-105 (as cited in ATSDR, 2000).
Mirowitz, S.A. and TJ. Westrich. 1992. Basal ganglia! signal intensity alterations: Reversal
after discontinuation of parenteral manganese administration. Radiol. 185:535-536.
Moran, J. 1975. The environmental implications of manganese as an alternative antiknock.
#750926, SAE Publications Division, Michigan.
I
Morganti, J.B., B.A. Lown, C.H. Stineman, et al. 1985. Uptake, distribution and behavioral
effects of inhalation exposure to manganese (MnO2) in the adult mouse. Neurotoxicology
6(1):1-15 (as cited in U.S. EPA, 1993).
l
Mortehnans, K., S. Haworth, T. Lawlor, et al. 1986. Salmonella mutagenicity test. II. Results
from the testing of 270 chemicals. Environ. Mutagen. 8 (Suppl. 7): 1-119 (as.cited in NTP,
1993), ;
l
\
Mustafa, SJ. and S.V. Chandra. 1971. Level^of 5-hydroxytryptamine, dopamine and
norepinephrine in whole brain of rabbits in chronic manganese toxicity. J. Neurochem.
18(6):931-933 (as cited in ATSDR, 2000).
Nachtman, J.P., R.E. Tubben and R.L. Commissaris. 1986. Behavioral effects of chronic
manganese administration in rats: locomotor activity studies. Neurobehav. Toxicol. Teratol.
8(6):711-715 (as cited in U.S. EPA, 1993).
Neff, N.H., R.E. Barrett and E. Costa. 1969. Selective depletion of caudate nucleus dopamine
and serotonin during chronic manganese dioxide administration to squirrel monkeys.
Experimentia 25:1140-1141 (as cited in U.S. EPA, 1993).
l
Newbeme, P.M. 1973. Input and disposition 6f manganese in man. In: Medical and Biologic
Effects of Environmental Pollutants: Manganese. National Academy of Sciences, Washington,
D.C. pp. 77-82 (as cited in U.S. EPA, 1984,1993).
t
Newland, M.C., T.L. Ceckler, J.H. Kordower, et al. 1989. Visualizing manganese in the primate
basal ganglia with magnetic resonance imaging. Exp. Neurol. 106(3):251-258 (as cited in
ATSDR, 2000). • ! '
External Review Draft — Manganese—April 2002
10-18
-------
Newland, M.C. and B. Weiss. 1992. Persistent effects of manganese on effortful responding and
their relationship to manganese accumulation in the primate globus pallidus. Toxicol. Appl.
Pharmacol. 113(l):87-97 (as cited in ATSDR, 2000).
NIOSH. 1984. National Institute for Occupational Safety and Health. Registry of Toxic Effects
of Chemical Substances. Prepared by Tracor Jitco, Inc., under Contract Number 210-81-8101.
Rockville, MD.
Nishioka, H. 1975. Mutagenic activities of metal compounds in bacteria. Mutat. Res. 31:185-
189 (as cited in ATSDR, 2000).
NRC. 1989. Recommended Daily Allowances, 10th ed. Food and Nutrition Board, National
Research Council, Washington, D.C. National Academy Press.
NTP. 1992. Toxicology and Carcinogenesis Studies of Manganese (EL) Sulfate Monohydrate
(CAS no. 10034-96-5) in F344/N Rats and B6C3F, Mice (Feed Studies). Draft Technical
Report. NTP Tech. Rep. Ser. 428. National Toxicology Program, Research Triangle Park, NC
(as cited in U.S. EPA, 1993). -
NTP. 1993. Toxicology and Carcinogenesis Studies of Manganese (n) Sulfate Monohydrate
(CAS no. 10034-96-5) in F344/N Rats and B6C3Fi Mice (Feed Studies). NTP Tech. Rep. Ser.
428. National Toxicology Program, Research Triangle Park, NC.
Oberleys TJ., C.E. Piper and D.S. McDonald. 1982. Mutagenicity of metal salts in the L5178Y
mouse rymphoma assay. J. Toxicol. Environ. Health 9:367-376 (as cited in ATSDR, 2000).
Olanow, C.W., P.F. Good, H. Shinotoh, et al. 1996. Manganese intoxication in the rhesus
monkey: A clinical, imaging, pathologic, and biochemical study. Neurology 46:492-498.
Ono, J., K. Harada, and R. Kodaka. 1995. Manganese deposition in the brain during long-term
parenteral nutrition. J Parent Enter Nutr 19:310-312 (as cited in ATSDR, 2000).
Orgel, A. and L.E. OrgeL 1965. Induction of mutations in bacteriophage T4 with divalent
manganese. J.Mol. Biol. 14:453-457 (as cited in ATSDR, 2000).
Pagano, D.A. and E. Zeiger. 1992. Conditions for detecting the mutagenicity. of divalent metals
in Salmonella typhimurium. Environ. Mol. Mutagen. 19(2): 139-146 (as cited in ATSDR,
2000).
Papavaslliou, P.S., S.T. Miller and G.C. Cotzias. 1966. Role of liver in regulating distribution
and excretion of manganese. Am. J. Physiol. 211(1):211-216 (as cited in U.S. EPA, 1984,
1993).
External Review Draft — Manganese—April 2002
10-19
-------
Pappas, B.A., D. Zhang, C.M. Davidson, et al. 1997. Perinatal manganese exposure: behavioral,
neurochemical, and histopathological effects in the rat. Neurotoxicol. Teratol. 19(l):17-25.
Pellizzari, E., C. Clayton, C. Rodes, et al. 1999. Particulate matter and manganese exposures in
Toronto, Canada. Atmos. Environ. 33:721-734.
Pennhigton, J.A., B.E. Young, D.B. Wilson, et[al. 1986. Mineral content of foods and total
diets: The selected minerals in food survey. J. Am. Diet Assoc. 86:876-891.
Pennington, JA., B.E. Young, and D.B. Wilson. 1989. Nutritional elements in the U.S. diet:
Results for the Total Diet Study, 1982-1986. J, Am. Diet Assoc. 89:659-664.
Perry, H.M., Jr., E.F. Perry, J.E. Purifoy, et al. 1973. A comparison of ultra- and interhepatic
variability of trace metal concentrations in normal men. In: Trace Substances in Environmental
Health, Proc. Univ. Missouri 7th Annual Conf. University of Missouri, Columbia, MO. pp. 281-
288 (as cited in U.S. EPA, 1993).
Pfeifer G., R. Harrison and D. Lynam. 1999. Personal exposures to airborne metals in London
taxi drivers and office workers in 1995 and 1996. Sci. Total Environ. 235:253-260.
i
Pierson, W., D. McKee, W. Brachaczek, et al. 1978. Methylcyclopentadienyl manganese
tricarbonyl: effect on manganese emissions from vehicles on the road. J. Air Pollut. Control
Assoc. 28:692-693. ;
l
Pihl, R.O. and M. Parkes. 1977. Hair element content in learning disabled children. Science
198:204-206 (as cited in U.S. EPA, 1993). '
Piver, W. 1974. Potential dilemma: The methods of meeting automotive exhaust emission
standards of the Clean Air Act of 1970. Environ. Health Perspect. 8:165-190.
Pollack, S., J.N. George, R.C. Reba, et al. 1965. The absorption of nonferrous metals in iron
deficiency. J. Clin. Invest. 44:1470-1473 (asicited in U.S. EPA, 1993).
Powell, J., T. Borden and R. Thompson. 1998. In vitro mineral availability from digested tea: A
rich dietary source of manganese. Analyst 12j3:1721 -1724.
Price, N.O., G.E. Bunce and R.W. Engel. 197Q. Copper, manganese and zjnc balance in
preadolescent girls. Am. J. Clin. Nutr. 23:258^260 (as cited in U.S. EPA, 1993).
I
Qato, M.K. and M.D. Maines. 1985. Regulation of heme and drug metabolism activities in the
brain by manganese. Biochem. Biophys. Res. jCommun. l28(l):18-24 (as cited in U.S. EPA,
1993). !
External Review Draft -— Manganese—April 2002
10-20
-------
Rabar, I. 1976, Some factors influencing manganese metabolism in rats. M.Sc. Thesis, Univ.
Zagreb., Zagreb, Yugoslavia (as cited in U.S. EPA, 1984,1993).
Rasmuson, A. 1985. Mutagenic effects of some water-soluble metal compounds in a somatic
eye-color test system in Drosophila melanogaster. Mutat. Res. 157:157-162 (as cited in
ATSDR.2000).
Rehnberg, G.L., J.F. Hein, S.D. Carter, et al. 1980. Chronic manganese oxide administration to
preweanling rats: manganese accumulation and distribution. J. Toxicol. Environ. Health
6(l):217-226.
Rehnberg, G.L., J.F. Hein, S.D. Carter, et al. 1981. Chronic ingestion of Mn3O4 by young rats:
tissue accumulation, distribution, and depletion. J. Toxicol. Environ. Health 7(2):263-272 (as
cited in U.S. EPA, 1993).
Rehnberg, G.L,, J.F. Hein, S.D. Carter, et al. 1982. Chronic ingestion of Mn3O4 by rats: tissue
accumulation and distribution of manganese in two generations. J. Toxicol. Environ. Health
9(2):175-188 (as cited in U.S. EPA, 1993).
Rehnberg, G.L., J.F. Hein, S.D. Carter, et al. 1985. Age-dependent changes in gastrointestinal
transport and retention of particulate manganese oxide in the rat. J Toxicol Environ Health
16:887-899.
Riveros-Rosas H., G. Pfeifer, D. Lynam, et al. 1997. Personal exposure to elements in Mexico
City air. Sci. Total Environ. 198:79-96.
Roels, H., R. Lauwerys, P. Genet, et al. 1987. Relationships between external and internal
parameters of exposure to manganese in workers from a manganese oxide and salt producing
plant. Am. J. Ind. Med. 11:297-305 (as cited in U.S. EPA, 1999a).
Roels, H.A., P. Ghyselen, J.P. Buchet, et al. 1992. Assessment of the permissible exposure level
to manganese in workers exposed to manganese dioxide dust. Br. J. Ind. Med. 49(l):25-34 (as
cited in U.S. EPA, 1999a).
Roels, H.A., G. Meiers, M. Delos, et al. Influence of the route of administration and the chemical
form (MnCl2, MnO2) on the absorption and cerebral distribution of manganese in rats. Arch.
Toxicol. 71:223-230.
Roels, EL A., M.I. Ortega-Eslava, E. Ceulemans, et al. 1999. Prospective study on the
reversibility of neurobehavioral effects in workers exposed to manganese dioxide.
Neurotoxicology 20:255-271.
Rogers, R.R., R.J. Garner, M.M. Riddle, et al. 1983. Augmentation of murine natural killer cell
activity by manganese chloride. Toxicol. Appl. Pharmacol. 70:7-17 (as cited in ATSDR, 2000).
External Review Draft — Manganese—April 2002
10-21
-------
I
Roth, G.S. and R.C. Adleman. 1975. Age-related changes in hormone binding by target cells and
tissues: Possible role of altered adaptive responsiveness. Exp. Gerontol. 10:1-11 (as cited in
U.S. EPA, 1993). \
Ruoff,W.L. 1995. Relative bioavailability of manganese ingested in food or water. In:
Proceedings: Workshop on the Bioavailability and Oral Toxicity of Manganese. Sponsored by the
U.S. Environmental Protection Agency, Cincinnati, OH. August 30-31,1994 (as cited in U.S.
EPA,1999a). '•
\
Ryker, S.J. and A.K. Williamson. 1999. Pesticides in Public Supply Wells of Washington State.
U.S. Geological Survey Fact Sheet 122-96.
\
Sanchez, D.J., J.L. Domingo, J.M. Llobet, et al. 1993. Maternal and developmental toxicity of
manganese in the mouse. Toxicol. Lett. 69(l):45-52.
SandstrSm, B., L. Davidsson, A. Cederblad, et al. 1986. Manganese absorption and metabolism
in man. Acta Pharmacol. Toxicol. (Copenh.) 59(Suppl 7):60-62.
Scheuhammer, A.M. and M.G. Cherian. 1981. The influence of manganese on the distribution of
essential trace elements. I. Regional distribution of Mn,Na,K,Mg,Zn,Fe, and Cu in rat brain
after chronic Mn exposure. Toxicol. Appl. Pharmacol. 61(2):227-233 (as cited in U.S. EPA,
1993). [
Scheuhammer, AM. 1983. Chronic manganese exposure in rats: histological changes in the
pancreas. J. Toxicol. Environ. Health 12(2-3):353-360 (as cited in U.S. EPA, 1993).
Scheuhammer, A.M. and M.G. Cherian. 1983r The influence of manganese on the distribution of
essential trace elements. II. The tissue distribution of manganese, magnesium, zinc, iron, and
copper in rats after chronic manganese exposure. J. Toxicol. Environ. Health 12(2-3):361-370
(as cited in U.S. EPA, 1993). !
Schroeder, H.A., J.J. Balassa and I.H. Tipton. 1966. Essential trace metals in man: Manganese.
Astudyinhomeostasis. J. Chronic Dis. 19(5):545-571 (as cited in U.S. EPA, 1993; ATSDR,
2000). I
Schwartz, R., BJ. Apgar and E.M. Wein. 1986. Apparent absorption and retention of Ca, Cu,
Mg, Mn, and Zn from a diet containing bran. Am. J. Clin. Nutr. 43:444-455 (as cited in U.S.
EPA, 1993).
Segura-Aguilar, J. and C. Lind. 1989. On the mechanism of the Mn3+-induced neurotoxicity of
dopamine: Prevention of quinone-derived oxygen toxicity by DT diaphorase and superoxide
dismutase. Chem. Biol. Interact. 72(3):309-324 (as cited in U.S. EPA, 1993).
External Review Draft — Manganese —April 2002
10-22
-------
Shacklette, H.T., and J.G. Boerngen. 1984. Element oncentrations in soils and other surficial
materials of the coterminous United States. USGS Paper 1270. Washington, DC: US Govern.
Printing Office, p. 6.
Shigan, S.A. and B.R. Vitvickaja. 1971: Experimental substantiation of permissible residual
concentrations of potassium permanganate in drinking water. Gig. Sanit. 36:15-18 (as cited in
U.S. EPA, 1993).
Shimkin, M.B. and G.D. Stoner. 1975. Lung tumors in mice: Application to cartinogenesis
bioassay. Adv. Cancer Res. 21:1-58 (as cited in U.S. EPA, 1993).
Shroeder, H.A., J.J. Balassa and I.H. Tipton. 1966. Essential trace metals in man: Manganese. A
• study in homeostasis. J. Chron. Dis. 19:545-571.
Shukla, G.S. and S.V. Chandra. 1976. Manganese induced morphological and biochemical
changes in the brain of iron deficient rats. Ind. Health 14:87-92 (as cited in U.S. EPA, 1993).
Shukla, G.S. and S.V. Chandra. 1977. Levels of sulfhydryls and sulfhydryl-containing enzymes
in brain, liver and testis of manganese treated rats. Arch. Toxicol. 37(4):319-325 (as cited in
U.S. EPA, 1993).
Shukla, G.S., S. Singh and S.V. Chandra. 1978. The interaction between manganese and ethanol
in rats. Acta Pharmacol. Toxicol. (Copenh.) 43(5):354-362 (as cited in U.S. EPA, 1993).
Shukla, G.S., M.P. Dubey and S.V. Chandra. 1980. Managanese-induced biochemical changes in
growing versus adult rats. Arch. Environ. Contam. Toxicol. 9(4):383-391 (as cited in U.S.
EPA, 1993).
Shukla, G.S. and S.V. Chandra. 1987. Concurrent exposure to lead, manganese, and cadmium
and their distribution to various brain regions, liver, kidney, and testis of growing rats. Arch.
Environ. Contam. Toxicol. 16(3):303-310 (as cited in U.S. EPA, 1993).
Sierra, P., S. Loranger, G. Kennedy, et al. 1995. Occupational and environmental exposure of
automobile mechanics and non-automotive workers to airborne manganese arising from the
combustion of methylcyclopentadienyl manganese tricarbonyl (MMT). Am. Ind. Hyg. Assoc. J.
56(7):713-716.
Simmon, V.F. and S. Ligon. 1977. In vitro microbiological mutagenicity studies of Ethyl
Corporation compounds. Interim report. Stanford Research Institute, California (as cited in
WHO, 1981; U.S. EPA, 1993).
Singh, I., 1984. Induction of gene conversion and reverse mutation by manganese sulphate and
nickel sulphate in Saccharomyces cerevisiae. Mutat Res. 137:47-49 (as cited in ATSDR,
2000).
External Review Draft — Manganese—April 2002
10-23
-------
Singh, J., R. Husain, S.K. Tandon, et al. 1974. Biochemical and histopathological alterations in
early manganese toxicity in rats. Environ. Physiol. Biochem. 4(l):16-23 (as cited in U.S. EPA,
1984,1993).
Singh, J., S.V. Chandra and S.K. Tandon. 1975. Chelation in metal intoxication II: In vitro and
in vivo effect of some compounds on brain, liver, and testis of rats treated with manganese
sulphate. Bull. Environ. Contain. Toxicol. 14(4):497-503 (as cited in U.S. EPA, 1984,1993).
. I
Singh, S., G.S. Shukla, R.S. Srivastava, et al. 1979. The interaction between ethanol and
manganese in rat brain. Arch. Toxicol. 41(4):307-316 (as cited in U.S. EPA, 1993).
Sitaramayya, A., N. Nagar and S.V. Chandra. :1974. Effect of manganese on enzymes in rat
brain. Acta Pharmacol. Toxicol. (Copenh.) 35(3): 185-190 (as cited in U.S. EPA, 1993).
i
Smeyers-Verbeke, J., P. Bell, A. Lowenthal, et al. 1976. Distribution of Mn in human brain
tissue. Clin. Chim. Acta 68(3):343-347.
Smialowicz, R.J., R.W. Luebke, R.R. Rogers, et al. 1985. Manganese chloride enhances natural
cell-mediated immune effector cell function: Effects on macrophages. Immunopharmacology
9:1-11 (as cited in ATSDR, 2000). j
Smialowicz, R.J., R.R. Rogers, M.M. Riddle, et al. 1987. Effects of manganese, calcium,
magnesium, and zinc on nickel-induced suppression of murine natural killer cell activity. J.
Toxicol. Environ. Health 20:67-80 (as cited in ATSDR, 2000).
f
Smith, S.E., M. Medlicott and G.H. Ellis. 1944- Manganese deficiency in the rabbit. Arch.
Biochem. Biophys. 4:281-289 (as cited in ATSDR, 2000).
I
Smyth, H.F., C.P. Carpenter, C.S. Weil, et al. 1969. Range-finding toxicity data: List VII. J.
Am. Ind. Hyg. Assoc. 30:470-476 (as cited ini U.S. EPA, 1993).
Snyder, R.D. 1988. Role of active oxygen species in metal-induced DNA strand breakage in
human diploid fibroblasts. Mutat. Res. 193(3):237-246 (as cited in U.S. EPA, 1993).
Spencer, H., C.R. Asmussen, R.B. Holtzman, et al. 1979. Metabolic balances of cadmium,
copper, manganese, and zinc in man. Am. J. Ctlin. Nutr. 32(9): 1867-1875 (as cited in U.S. EPA,
1993). i
i
Srisuchart, B., M.J. Taylor and R.P. Sharma. 1987. Alteration of humoral and cellular immunity
in manganese chloride-treated mice. J. Toxicol. Environ. Health 22:91-99 (as cited in ATSDR,
2000).
External Review Draft — Manganese —April 2002
10-24
-------
Stauber, J.L., T.M. Florence and W.S. Webster. 1987. The use of scalp hair to monitor
manganese in aborigines from Groote Eylandt. Neurotoxicology 8(3):431-435 (as cited in U.S.
EPA, 1993).
Stoner, G.D., M.B. Shimkin, M.C. Troxell, et al. 1976. Test for carcinogenicity of metallic
compounds by the pulmonary tumor response in strain A mice. Cancer Res. 36(5): 1744-1747
(as cited in U.S. EPA, 1984,1993).
Strause, L.G., J. Hegenauer, P. Saltman, et al. 1986. Effects of long-term dietary manganese and
copper deficiency on rat skeleton. J.Nutr. 116:135-141 (as cited in ATSDR, 2000).
Sumino, K., K. Hayakawa, T. Shibata, et al. 1975. Heavy metals in normal Japanese tissues.
Amounts of 15 heavy metals in 30 subjects. Arch. Environ. Health 30(10):487-494 (as cited in
U.S. EPA, 1984,1993).
Suzuki, Y. 1974. Studies on excessive oral intake of manganese, n. Minimum dose for
manganese accumulation in mouse organ. Shikoku Acta Med. 30:32-45 (as cited in U.S. EPA,
1993).
Suzuki, Y., T. Mouri, K. Nishiyama, et al. 1975. Study of subacute toxicity of manganese
dioxide in monkeys. Tokushima J. Exp. Med . 22:5-10 (as cited hi U.S. EPA, 1984,1993).
Szakmary, E., G. Ungvary, A. Hudak, et al. 1995. Developmental effect of manganese hi rat and
rabbit. Cent EurJOccup Environ Med 1:149-159.
Tanaka, Y. 1982. Manganese: Its possible significance hi childhood nutrition hi relation to
convulsive disorders. J. Am. Coll. Nutr. 1:113 (as cited hi U.S. EPA, 1993).
Ter Haar, G., M. Griffing, M. Brandt, et al. 1975. Methylcyclopentadienyl manganese
tricarbonyl as an antiknock: Composition and fate of manganese exhaust products. J. Air Pollut.
Control Assoc. 25:858-860.
Thomson, A.B., D. Olatunbosun and L.S. Valverg. 1971. Interrelation of intestinal transport
system for manganese and iron. J. Lab. Clin. Med. 78(4):642-655 (as cited hi U.S. EPA, 1993).
Tichy, M., M. Cikrt and J. Havrdova. 1973. Manganese binding in rat bile. Arch. Toxikol.
30(3):227-236 (as cited in U.S. EPA, 1984,1993).
Tipton, I.H. and M.J. Cook. 1963. Trace elements hi human tissue. Part II. Adult subjects from
the United States. Health Phys. 9:103-145 (as cited in ATSDR, 2000).
Tjalve, H., J. Henriksson, J. Tallvist, et. al. 1996. Uptake of manganese and cadmium from the
nasal mucosa into the central nervous system via olfactory pathway -in rats. Pharmacol. Toxicol.
79:347-356.
External Review Draft—Manganese—April 2002
10-25
-------
Treinen, K. A. and W.F. Blazak. 1995. Developmental toxicity of WIN 59010-2 in Sprague-
Dawley rats. Toxicology 15:160-161 [Abstract].
I
Ulitzur, C.E. and M. Barak. 1988. Detection of genotoxicity of metallic compounds by the
bacterial bioluminescence test. J.BioLChemj 2:95-99 (as cited in ATSDR, 2000).
Umeda, M. and M. Nishimura. 1979. Inducibility of chromosomal aberrations by metal
compounds 'in cultured mammalian cells. Mutat. Res. 67:221-223 (as cited in U.S. EPA, 1993).
U.S. EPA. 1975. Scientific and technical assessment report on manganese. North Carolina: U.S.
Environmental Protection Agency, Office of Research and Development. EPA 600/6-75-002.
U.S. EPA. 1984. Health Assessment Document for Manganese. U.S. Environmental Protection
Agency, Office of Health and Environmental Assessment, Environmental Criteria and Assessment
Office. EPA 600/8-83-013F. Cincinnati, OHL
I
U.S. EPA. 1986a. Guidelines for Carcinogenic Risk Assessment. U.S. Environmental Protection
Agency. Fed. Reg. 51(185):33992-34003. |
t '
I
U.S. EPA. 1986b. Guidelines for the Health Risk Assessment of Chemical Mixtures. U.S.
Environmental Protection Agency. Federal Register 51(185):34014-34025.
i
U.S. EPA. 1986c. Guidelines for Mutagenicity Risk Assessment. U.S. Environmental Protection
Agency. Federal Register 51(185):34006-340i2.
U.S. EPA. 1986d. Reference Values for Risk Assessment. U.S. Environmental Protection
Agency. Prepared by the Office of Health an4 Environmental Assessment, Environmental Criteria
and Assessment Office. Cincinnati, OH.
I
U.S. EPA. 1988. Recommendations for and Documentation of Biological Values for Use in Risk
Assessment. U.S. Environmental Protection Agency, Environmental Criteria and Assessment
Office, Office of Health and Environmental Assessment, Cincinnati, OH. EPA/600/6-87-008.
NTIS PB88-179874/AS, February 1988. j
U.S. EPA. 1990. Comments on the use of metiiylcyclopentadienyl manganese tricarbonyl in
unleaded gasoline. North Carolina: U.S. Environmental Protection Agency, Office of Research
and Development. i
\
U.S. EPA. 1991a. Guidelines for Developmental Toxicity Risk Assessment. U.S. Environmental
' Protection Agency. Federal Register 56:63798-63826.
U.S. EPA. 1991b. Summary of workshop discussions at the manganese and
methylcyclopentadienyl manganese tricarbonyl (MMT) conference, 12-15 March 1991. North
Carolina: U.S. Environmental Protection Agency, Office of Research and Development.
External Review Draft — Manganese—April 2002
10-26
-------
U.S. EPA. 1993. Drinking Water Criteria Document for Manganese. Final Draft. U.S.
Environmental Protection Agency, Environmental Criteria and Assessment Office, Office of
Health and Environmental Assessment. ECAO-CIN-D008. Cincinnati, OH.
U.S. EPA. 1994a. Drinking Water Criteria Document for Manganese. U.S. Environmental
Protection Agency, Office of Water, Research and Development. September, 1993. Updated:
March, 1994.
U.S. EPA. 1994b. Peer Review and Peer Involvement at the U.S. Environmental Protection
Agency. Signed by the U.S. EPA Administrator, Carol A. Browner, June 7.
U.S. EPA 1995. Use of the Benchmark Dose Approach in Health Risk Assessment. U.S.
Environmental Protection Agency. EPA/630/R-94/007.
U.S. EPA. 1996a. Manganese. Integrated Risk Information System. (IRIS). U.S.
Environmental Protection Agency. Available at http://www.epa.gov/iris. Last revised December
1,1996.
U.S. EPA 1996b. Proposed Guidelines for Carcinogen Risk Assessment. U.S. Environmental
Protection Agency, Office of Research and Development, Washington, D.C. EPA/600/P-
92/003C.
U.S. EPA. 1996c. Guidelines for Reproductive Toxicity Risk Assessment. U.S. Environmental
Protection Agency, Office of Research and Development, Washington, D.C. EPA/630/R-96/009.
U.S. EPA. 1996d. Exposure Factors Handbook. Vol. I-General Factors. EPA/600/8-89/043.
Washington, DC.
U.S. EPA. 1996e. Emergency Planning and Community Right-to-Know Section 313, List of
Toxic Chemicals. Available .on the internet at: http://www.epa.gov/tri/chemls2.pdf. Last modified
March 23,2000. T.ink to site at: http://www.epa.gov/tri/chemical.htm
U.S. EPA. 1997. U.S. Environmental Protection Agency. Announcement of the Draft Drinking
Water Contaminant Candidate List; Notice. Fed. Reg. 62(193):52193. October 6.
U.S. EPA. 1998. U.S. Environmental Protection Agency. Announcement of the Drinking Water
Contaminant Candidate List; Final Rule. Fed. Reg. 63(274): 10273. March 2.
U.S. EPA. 1998a. Guidelines for Neurotoxicity Risk Assessment. U. S. Environmental
Protection Agency. Federal Register 63(93):26926-26954.
U.S. EPA. 1998b. Science Policy Council Handbook: Peer Review. U.S. Environmental
Protection Agency, Office of Science Policy, Office of Research and Development, Washington,
D.C. EPA/100/B-98/001.
External Review Draft — Manganese —April 2002
10-27
-------
U.S. EPA. 1999a. A Review of Contaminant Occurrence in Public Water Systems. EPA
Report/816-R-99/006. U.S. Environmental Protection Agency, Office of Water, 78 pp.
U.S. EPA. 1999b. U.S. Environmental Protection Agency. Guidelines for Carcinogen Risk
Assessment. National Center for Environmental Assessment, Risk Assessment Forum,
Washington, D.C. NCEA-F-0644. My 1999. Available on-line at:
http://www.epa.gov/ncea/raf/car2sab.htm
U.S. EPA. 2000a. What is the Toxic Release Inventory. Available on the Internet at:
http://www.epa.gov/tri/general.htm Last modified February 28,2000.
U.S. EPA. 2000b. TR1Explorer: Trends. Available on the Internet at:
http://www.epa.gov/rriexplorer/trends.htm Last modified May 5, 2000.
I
U.S. EPA. 2000c. TRI Explorer: Are Year-to-Year Changes Comparable? Available on the
Internet at: www.epa.gov/triexplorer/yearsum.htm Last modified May 5,2000.
U.S. EPA. 2000d. The Toxic Release Inventory (TRI) and Factors to Consider when Using TRI
Data. Available on the Internet at: http://www.epa.gov/tri/tri98/98over.pdf. Last modified
August 11,2000. Link to site at: http://www.epa.gov/rri/tri98
i
U.S. EPA. 2000e. Water Industry Baseline Handbook, Second Edition (Draft). March 17.
USGS. 2000. Mineral Commodity Summaries, February, 2000 - Manganese. Available on the
Internet at: http://mmerals.usgs.gov/mmerals/pubs/commodity/manganese/420300.pdf
I
USGS. 2001. USGS National Water Quality Assessment Data Warehouse. Available on the
Internet at: http://infotrek.er.usgs.gov/pls/nawqa/nawqa.home. Last updated April 19, 2001
j
Valencia, H., J.M. Mason, R.C. Woodruff, et a}. 1985. Chemical mutagenesis testing in
Drosophila. HI. Results of 48 coded compounds tested for the National Toxicology Program.
Environ. Mutagen. 7:325-348 (as cited in ATSDR, 2000).
[
Vieregge, P., B. Heinzow, G. Korf, et al. 1995, Long term exposure to manganese in rural well
water has no neurological effects. Can. J. Neurol. Sci. 22:286-289.
Wassermann, D. and M. Wassermann. 1977. The ultrastructure of the liver cell in subacute
manganese administration. Environ Res. 14(3):379-390 (as cited in U.S. EPA, 1993).
Wedler, F.C. 1994. Biochemical and nutritional role of manganese: An overview. In:Klimis-
Tavantzis, D.J., ed., Manganese in Health and Disease. Boca Raton, FL: CRC Press, Inc., pp. 1-
36 (as cited in ATSDR, 2000). i
External Review Draft — Manganese —April 2002
10-28
-------
Weigand E., M. Kirchgessner, and U. Helbig. 1986. Trad absorption and endogenous fecal
excretion of manganese in relation to its dietary supply in growing rats. Biol. Trace Elem. Res.
10:265-279.
WHO. 1973. Trace Elements in Human Nutrition: Manganese. Technical Report Service, 532.
World Health Organization, Geneva, Switzerland.
WHO. 1981. Environmental Health Criteria 17. Manganese. World Health Organization,
Geneva, Switzerland.
Widdowson, E.M., H. Chan, G.E. Harrison, et al. 1972. Accumulation of Cu,Zn,Mn, Grand
Co in the human liver before birth. Biol. Neonate 20(5):360-367.
Wong,P.K. 1988. Mutagenicity of heavy metals. Bull. Environ. Contam. ToxicoL 40:597-603
(as cited in ATSDR, 2000).
Yamada, M., S. Ohno, I. Okayasu, et al. 1986. Chronic manganese poisoning: A
neuropathological study with determination of manganese distribution in the brain. Acta
Neuropathol. (Berl.) 70:273-278 (as cited in U.S. EPA, 1993).
Yamaguchi, M., K. Inomoto and Y. Soketa. 1986. Effect of essential trace metals on bone
metabolism in weanling rats: Comparison with zinc and other metals' actions. Res. Exp. Med.
186:337-342 (as cited in U.S. EPA, 1993).
Yamamoto, H. and T. Suzuki. 1969. Chemical structure of manganese compounds and their
biological effects. Jap. Assoc. Ind. Health, Fukuoka City, Japan, March 28-31 (as cited in U.S.
EPA, 1984,1993).
Zakour, R.A. and B.W. Glickman. 1984. Metal-induced mutagenesis in the lad gene of
Escherichia coli. Mutat. Res. 126:9-18 (as cited hi ATSDR, 2000).
Zayed, J., M. Gerin, S. Loranger, et al. 1994. Occupational and environmental exposure of
garage workers and taxi drivers to airborne manganese arising from the use of
methylcyclopentadienyl manganese tricarbonyl (MMT) hi unleaded gasoline. Am. Ind. Hyg.
Assoc. J. 55(l):53-58.
Zayed, J., M. Mikhail, S. Loranger, et al. 1996. Exposure of taxi drivers and office workers to
total and respirable manganese in an urban environment. Am. Ind. Hyg. Assoc. J. 57(4):376-380.
Zayed, J, C. Thibault, L. Gareau, et al. 1999a. Airborne manganese particulates and
methylcyclopentadienyl manganese tricarbonyl (MMT) at selected outdoor sites in Montreal.
Neurotoxicology 20(2-3):151-158.
External Review Draft — Manganese—April 2002
10-29
-------
Zayedj J., A. Vyskocil and G. Kennedy. 1999b. Environmental contamination and human
exposure to manganese-contribution of methylcyclopentadienyl manganese tricarbonyl in unleaded
gasoline. Int. Arch. Occup. Environ. Health. 72:7-13.
Zhang, G., D. Liu, and P. He. 1995. [Effects of manganese on learning abilities in school
children]. Chung HuaYu Fang I HsuehTsaChih 29:156-158.
i
Zhernakova, T.V. 1967. Correlation betweenjiron, manganese and copper content in the blood
serum of healthy individuals. Bull. Exp. Biol. iMed. 63:47-48 (as cited in U.S. EPA, 1984.,
1993).
Zlotkin, S.H. and B.E. Buchanan. 1986. Manganese intakes in intravenously fed infants:
Dosages and toxicity studies. Biol. Trace Elem. Res. 9:271-279 (as cited in ATSDR, 2000).
External Review Draft — Manganese—April 2002
10-30
-------
APPENDIX A: Abbreviations and Acronyms
ACGIH
ATSDR
CAS
CCL
CERCLA
CMR
CWS
DWEL
EPA
EPGRA
GW
HA
HAL
HRL
IOC
IRIS
MRL
NAWQA
NCOD
NIOSH
NIRS
NPDES
NPDWR
NHS
NTNCWS
ppm
PWS
RCRA
SARA Title HI
SDWA
SDWIS
SDWIS FED
STORET
SW
TRI
UCM
UCMR
UMRA
URCIS
U.S. EPA
USGS
- American Conference of Governmental Industrial Hygienists
- Agency for Toxic Substances and Disease Registry
- Chemical Abstract Service
- Contaminant Candidate List
- Comprehensive Environmental Response, Compensation &
Liability Act
- Chemical Monitoring Reform t
- Community Water System
- Drinking Water Equivalent Level
- Environmental Protection Agency
- Emergency Planning and Community Right-to-Know Act
-ground water
- Health Advisory
- Health Advisory Level
- Health Reference Level
- inorganic compound
- Integrated Risk Information System
- Minimum Reporting Level
- National Water Quality Assessment Program
- National Drinking Water Contaminant Occurrence Database
- National Institute for Occupational Safety and Health
- National Inorganic and Radionuclide Survey
- National Pollution Discharge Elimination System
- National Primary Drinking Water Regulation
- National Technical Information Service
- Non-Transient Non-Community Water System
- part per million
- Public Water System
- Resource Conservation and Recovery Act
- Superfund Amendments and Reauthorization Act
- Safe Drinking Water Act
- Safe Drinking Water Information System
- the Federal Safe Drinking Water Information System
- Storage and Retrieval System
- surface water
- Toxic Release Inventory
- Unregulated Contaminant Monitoring
- Unregulated Contaminant Monitoring Regulation/Rule
- Unfunded Mandates Reform Act of 1995
- Unregulated Contaminant Monitoring Information System
- United States Environmental Protection Agency •
- United States Geological Survey
External Review Draft — Manganese—April 2002
A-l
-------
ligfL
mg/L
>MCL
>MRL
• micrograms per liter
• milligrams per liter
• percentage of systems with exceedances
• percentage of systems with detections
External Review Draft — Manganese—April 2002
A-2
-------
APPENDIX B: Complete NBRS Data for Manganese
NIRS Data - Manganese Occurrence in Public Water Systems (HRL = 0.3 mg/L)
Statti
AK
AL
AR
AZ
CA
CO
CT
DE
FL
GA
IA
10
IL
IN
KS
ICY
LA
MA
MD
ME
Ml
MN
MO
MS
MT
NC
NO
NE
NH
NJ
MM
NV
NY
OH
OK
OR
PA
PR
Rl
SO
SD
TN
TX
UT
VA
vr
WA
Wl
WV
WY
Total
# Samples
8
8
9
14
60
10
23
10
56
23
28
12
46
19
6
8
26
7
6
7
25
19
21
26
11
44
19
19
10
6
7
2
57
25
12
S
36
1
1
18
8
9
74
10
30
12
52
30
8
•.
989
# Samples
>MRL
7
4
6
5
26
7
18
10
29
9
22
1
34
18
3
6
24
6
5
6
22
17
16
21
5
33
19
10
8
2
£
1
32
19
6
B
28
1
1
11
~t
6
51
/
25
8
31
24
»
j
672
%
Samples
>MRL
87.50%
50.00%
66.67%
35.71%
43.33%
70.00%
78.26%
100.00%
51.79%
39.13%
78.57%
8.33%
73.91%
94.74%
50.00%
75.00%
92.31%
85.71%
83.33%
85.71%
88.00%
89.47%
76.19%
80.77%
45.45%
75.00%
100.00%
52.63%
80.00%
33.33%
71.43%
50.00%
58.14%
76.00%
50.00%
62.50%
77.78%
100.00%
100.00%
81.11%
87.50%
88.89%
68.92%
40.00%
83.33%
66.67%
59.62%
80.00%
3750%
100.00%
67.95%
# Detects
>1BHRL
2
1
2
5
1
2
1
2
3
1
2
6
3
1
3
j
1
4
1
• 7
2
2
2
1
1
60
% Detects
> 1/2 HRL
25.00%
0.00%
0.00%
7.14%
3.33%
0.00%
0.00%
0.00%
0.00%
0.00%
17.88%
0.00%
2.17%
10.53%
16.67%
25.00%
11.54%
14.29%
0.00%
0.00%
B.00%
31.58%
14.29%
0.00%
9.09%
0.00%
15.79%
15.79%
0.00%
0.00%
14.29%
0.00%
7.02%
0.00%
0.00%
12.50%
19.44%
0.00%
0.00%
0.00%
25.00%
0.00%
0.00%
0.00%
0.00%
16.67%
5.77%
3.33%
12.50%
0.00%
6.07%
# Detects
>HRL
1
1
1
4
1
1
1
1
4
1
1
2
2
1
2
4
1
2
1
32
% Detects
>HRL
12.50%
0.00%
0.00%
7.14%
1.67%
0.00%
0.00%
0.00%
0.00%
0.00%
1429%
0.00%
2.17%
526%
16.87%
12.50%
0.00%
0.00%
0.00%
0.00%
0.00%
21.05%
4.78%
0.00%
9.09%
0.00%
1053%
1053%
0.00%
0.00%
1429%
0.00%
3.51%
0.00%
0.00%
0.00%
11.11%
0.00%
0.00%
0.00%
12.50%
0.00%
0.00%
0.00%
0.00%
16.67%
0.00%
0.00%
12.50%
0.00%
324%
Min Value
(mafl.)
< 0.00
< 0.00
< 0.00
< 0.00
< O.OO
< 0.00
< 0.60
0.00
< 0.00
< 0.00
-< 0.00
< 0.00
< 0.00
< 0.00
< 0.00
< 0.00
< osa
< 0.00
< 0.00
< 0.00
•: 0.00
< 0.00
< 0.00
< 0.00
< 0.00
< 0.00
0.00
< 0.00
< 0.00
< 0.00
< 0.00
< 0.00
< 0.00
< 0.00
< 0.00
< . 0.00
< 0.00
0.01
0.03
< 0.00
< 0.00
< 0.00
< 0.00
< 0.00
< 0.00
< 0.00
< 0.00
< 0.00
< 0.00
0.02
< 0.00
99% Value
(mg/L)
050
0.05
0.06
0.58
0.65
0.13
0.09
0.08
0.03
O.05
1.34
0.13
0.36
0.33
0.83
050
0.25
0.19
0.05
0.04
020
033
1.22
0.09
0.33
0.09
0.63
124
0.11
0.09
0.38
0.00
0.40
0.13
0.08
0.17
0.8S
0.01
0.03
0.07
0.72
0.08
0.13
0.02
0.13
0.33
0.18
0.18
0.76
0.09
0.63
Max Value
(mgfL)
0.50
0.05
0.06
0.58
0.65
0.13
0.09
0.08
0.03
0.05
1.34
0.13
0.36
0.33
0.83
0.50
025
0.19
0.05
0.04
020
0.63
122
0.09
0.33
0.09
0.63
124
0.11
0.09
0.38
0.00
0.40
0.13
0.08
0.17
0.86
0.01
0.03
0.07
0.72
0.08
0.13
0.02
0.13
0.33
0.18
0.18
0.78
0.09
1.34
M!n
Detects
(mg/L)
0.02
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.13
0.00
0.01
0.01
0.00
0.00
0.00
0.00
0.00
0.00
0.01
0.00
0.00
0.00
0.00
0.00
0.00
0.01
0.01
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.01
0.03
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.02
0.00
Median
Detects
(mg/L)
0.05
0.01
0.01
0.00
0.01
0.00
0.01
0.01
0.00
O.O2
0.01
0.13
O.01
0.03
0.07
0.02
0.01
0.00
0.02
0.01
0.02
0.09
0.00
0.01
0.07
0.01
0.01
0.05
0.05
0.05
0.02
0.00
O.03
0.02
0.00
0.01
0.02
0.01
0.03
0.01
0.06
0.00
0.02
0.00
0.01
0.00
0.01
0.02
0.10
0.02
0.01
PWS= PuMc Water Systems; GW= Ground Water (PWS Source Water Type); SW= Surface Water (PWS Source Water Type); MRL= Minimum Reporting Limit
(for laboratory analyses)
The Health Reference Level (HRL) is the estimated health effect level as provided by EPA for preliminary assessment for this work assignment
"% > HRL" Indicates the proportion of systems with any analytical results exceeding the concentration value of the HRL.
The Health Reference Level (HRL) used for Manganese is 028 mg/L. This is a draft value for working review only.
Manganese data were analyzed using two different HRLs and are, therefore, Bsted separately.
External Review Draft — Manganese — April 2002
B-l
-------
------- |