EPA-540/1-86-058
of Emergency and
Remedial Response
Washington DC 20460
Off'ce of Research and Development
Office of Health and Environmental
Assessment
Environmental Criteria and
Assessment Office
Cincinnati OH 45268
Superfund
HEALTH EFFECTS ASSESSMENT
FOR SELENIUM (AND COMPOUNDS)
-------
EPA/540/1-86-058
September 1984
HEALTH EFFECTS ASSESSMENT
FOR SELENIUM (AND COMPOUNDS)
U.S. Environmental Protection Agency
Office of Research and Development
Office of Health and Environmental Assessment
Environmental Criteria and Assessment Office
Cincinnati, OH 45268
U.S. Environmental Protection Agency
Office of Emergency and Remedial Response
Office of Solid Waste and Emergency Response
Washington, DC 20460
-------
DISCLAIMER
This report has been funded wholly or In part by the United States
Environmental Protection Agency under Contract No. 68-03-3112 to Syracuse
Research Corporation. It has been subject to the Agency's peer and adminis-
trative review, and H has been approved for publication as an EPA document.
Mention of trade names or commercial products does not constitute endorse-
ment or recommendation for use.
11
-------
PREFACE
This report summarizes and evaluates Information relevant to a prelimi-
nary interim assessment of adverse health effects associated with selenium
(and compounds). All estimates of acceptable Intakes and carcinogenic
potency presented in this document should be considered as preliminary and
reflect limited resources allocated to this project. Pertinent toxlcologic
and environmental data were located through on-line literature searches of
the Chemical Abstracts, TOXLINE, CANCERLINE and the CHEMFATE/DATALOG data
bases. The basic literature searched supporting this document is current up
to September, 1984. Secondary sources of information have also been relied
upon in the preparation of this report and represent large-scale health
assessment efforts that entail extensive peer and Agency review. The
following Office of Health and Environmental Assessment (OHEA) sources have
been extensively utilized:
U.S. EPA. 1980a. Ambient Water Quality Criteria for Selenium.
Environmental Criteria and Assessment Office, Cincinnati, OH. EPA
440/5-80-070. PB 81-117814.
U.S. EPA. 1983a. Reportable Quantity for Selenium (and Com-
pounds). Prepared by the Environmental Criteria and Assessment
Office, Cincinnati, OH, OHEA for the Office of Solid Waste and
Emergency Response, Washington, DC.
The intent in these assessments 1s to suggest acceptable exposure levels
whenever sufficient data were available. Values were not derived or larger
uncertainty factors were employed when the variable data were limited in
scope tending to generate conservative (i.e., protective) estimates. Never-
theless, the interim values presented reflect the relative degree of hazard
associated with exposure or risk to the chemlcal(s) addressed.
Whenever possible, two categories of values have been estimated for sys-
temic toxicants (toxicants for which cancer is not the endpolnt of concern).
The first, the AIS or acceptable intake subchronlc, is an estimate of an
exposure level that would not be expected to cause adverse effects when
exposure occurs during a limited time interval (i.e., for an Interval that
does not constitute a significant portion of the Hfespan). This type of
exposure estimate has not been extensively used or rigorously defined, as
previous risk assessment efforts have been primarily directed towards
exposures from toxicants in ambient air or water where lifetime exposure is
assumed. Animal data used for AIS estimates generally Include exposures
with durations of 30-90 days. Subchronlc human data are rarely available.
Reported exposures are usually from chronic occupational exposure situations
or from reports of acute accidental exposure.
111
-------
The AIC, acceptable Intake chronic, 1s similar 1n concept to the ADI
(acceptable dally Intake). It 1s an estimate of an exposure level that
would not be expected to cause adverse effects when exposure occurs for a
significant portion of the lifespan [see U.S. EPA (1980b) for a discussion
of this concept]. The AIC 1s route specific and estimates acceptable
exposure for a given route with the Implicit assumption that exposure by
other routes 1s Insignificant.
Composite scores (CSs) for noncarclnogens have also been calculated
where data permitted. These values are used for ranking reportable quanti-
ties; the methodology for their development 1s explained 1n U.S. EPA (1983b).
For compounds for which there Is sufficient evidence of cardnogenlcity,
AIS and AIC values are not derived. For a discussion of risk assessment
methodology for carcinogens refer to U.S. EPA (1980b). Since cancer Is a
process that Is not characterized by a threshold, any exposure contributes
an Increment of risk. Consequently, derivation of AIS and AIC values would
be Inappropriate. For carcinogens, q-]*s have been computed based on oral
and Inhalation data 1f available.
1v
-------
ABSTRACT
In order to place the risk assessment evaluation 1n proper context,
refer to the preface of this document. The preface outlines limitations
applicable to all documents of this series as well as the appropriate Inter-
pretation and use of the quantitative estimates presented.
The effects of oral selenium exposure have been studied relatively
thoroughly 1n experimental animals and man. Evidence suggests that selenium
1s an essential element. An oral AIS of 0.224 mg/day and an oral AIC of
0.21 mg/day have been estimated based on animal and human data respectively.
The AIC 1s In good agreement with NAS guidelines. A CS of 49 associated
with Increased neonatal mortality by oral exposure was calculated.
Data concerning Inhalation effects of selenium are limited. Human data
suggest an AIC of 0.07 mg/day for Inhalation exposure. Data were inadequate
to estimate an inhalation AIS.
-------
ACKNOWLEDGEMENTS
The Initial draft of this report was prepared by Syracuse Research
Corporation under Contract No. 68-03-3112 for EPA's Environmental Criteria
and Assessment Office, Cincinnati, OH. Or. Christopher DeRosa and Karen
Blackburn were the Technical Project Monitors and Helen Ball was the Project
Officer. The final documents 1n this series were prepared for the Office of
Emergency and Remedial Response, Washington, DC.
Scientists from the following U.S. EPA offices provided review comments
for this document series:
Environmental Criteria and Assessment Office, Cincinnati, OH
Carcinogen Assessment Group
Office of Air Quality Planning and Standards
Office of Solid Waste
Office of Toxic Substances
Office of Drinking Water
Editorial review for the document series was provided by:
Judith Olsen and Erma Durden
Environmental Criteria and Assessment Office
Cincinnati, OH
Technical support services for the document series was provided by:
Bette Zwayer, Pat Daunt, Karen Mann and Jacky Bohanon
Environmental Criteria and Assessment Office
Cincinnati, OH
The Initial draft of this report was prepared by Syracuse Research
Corporation under Contract No. 68-03-3112.
v1
-------
TABLE OF CONTENTS
1.
2.
3.
4.
5.
6.
ENVIRONMENTAL CHEMISTRY AND FATE
ABSORPTION FACTORS IN HUMANS AND EXPERIMENTAL ANIMALS . . .
2.1. ORAL
2.2. INHALATION
TOXICITY IN HUMANS AND EXPERIMENTAL ANIMALS
3.1. SUBCHRONIC
3.1.1. Oral
3.1.2. Inhalation
3.2. CHRONIC
3.2.1. Oral
3.2.2. Inhalation
3.3. TERATOGENICITY AND OTHER REPRODUCTIVE EFFECTS. . . .
3.3.1. Oral
3.3.2. Inhalation
3.4. TOXICANT INTERACTIONS
CARCINOGENICITY
4.1. HUMAN DATA
4.2. BIOASSAYS
4.3. OTHER RELEVANT DATA
4.4. WEIGHT OF EVIDENCE
REGULATORY STANDARDS AND CRITERIA
RISK ASSESSMENT
6.1. ACCEPTABLE INTAKE SUBCHRONIC (AIS)
6.1.1. Oral
6.1.2. Inhalation
.Page
1
4
... 4
7
. . . 8
, . . 8
... 8
10
... 10
... 10
... 15
, , , 16
... 16
... 17
... 18
... 22
... 22
... 22
... 25
26
... 28
... 29
... 29
... 29
... 30
-------
TABLE OF CONTENTS (cont.)
Page
6.2. ACCEPTABLE INTAKE CHRONIC (AIC) 30
6.2.1. Oral 30
6.2.2. Inhalation 32
6.3. CARCINOGENIC POTENCY (q^) 32
6.3.1. Oral 32
6.3.2. Inhalation 33
7. REFERENCES 34
APPENDIX: Summary Table for Selenium (and compounds) 52
-------
LIST OF ABBREVIATIONS
ADI Acceptable daily Intake
AIC Acceptaole Intake chronic
AIS Acceptable intake subchronic
bw Body weight
CAS Chemical Abstract Service
CS Composite score
EKG Electrocardiogram
FAA N-2-fluorenyl-acetamide
GI Gastrointestinal
GSH-Px Glutathione peroxidase
LOAEL Lowest-observed-adverse-effect level
LOEL Lowest-observed-effect level
MED Minimum effective dose
NOAEL No-observed-adverse-effect level
NOEL No-observed-effect level
PCB Polychlorinated biphenyl
ppb Parts per billion
ppm Parts per million
RVd Dose-rating value
RVe Effect-rating value
STEL Short-term exposure limit
TLV Threshold limit value
TWA Time-weighted average
ix
-------
1. ENVIRONMENTAL CHEMISTRY AND FATE
Selenium 1s an element belonging to Group VIA of the periodic table.
Elemental selenium has a CAS Registry number of 7782-49-2. It exists 1n
nature In several oxidation states: -2, 0, +4 and +6. Selected physical
properties of a few environmentally significant selenium compounds are given
1n Table 1-1.
Both natural and anthropogenic emissions are sources of selenium 1n the
environment; however, the anthropogenic sources far outweigh the natural
sources of selenium 1n the atmosphere. The main natural sources of selenium
In the atmosphere are continental dust flux and volcanic dust and gas flux
(Lantzy and MacKenzle, 1979). Coal combustion and copper production consti-
tute -90% of the overall atmospheric anthropogenic sources of selenium.
Glass manufacturing, selenium-recovery plants, burning of fuel oil, and
refuse burning are the other Important anthropogenic sources of selenium 1n
the atmosphere (NAS, 1976).
A small fraction of selenium may exist 1n the gaseous state In the atmo-
sphere; however, the predominant amount of atmospheric selenium 1s expected
to be present In the partlculate form (NAS, 1976). Although chemical reac-
tions 1n the troposphere may cause spedatlon of selenium, these processes
may not be directly responsible for the removal of atmospheric selenium.
Removal of selenium from the atmosphere may occur primarily through wet and
dry deposition (NAS, 1976). The atmospheric residence time of selenium Is
probably dependent on the partlculate diameter of selenium 1n the atmo-
sphere. It has been determined by Natusch et al. (1974) that selenium
emitted from coal-fired power plants (the primary source of atmospheric
selenium) remains most concentrated 1n the smallest resplrable particles.
-1-
-------
TABLE 1-1
Selected Physical Properties of a Few Selenium Compounds3
ro
i
Element/Compound Formula
Selenium hydride H?Se
Selenium Se
Cadmium selenlde CdSe
Sodium selenlte Na2Se03«5H20
Sodium selenate Na2Se04
Selenium dioxide Se02
Molecular
Weight
80.98
78.96
191.36
263.01
188.94
110.96
Specific
Gravity/Density
2.004 for liquid
at -41.5°C
4.81$° for hexa-
gonal bluish gray
metal
5.81J5
NA
3.213 g/cma
at 17.4°C
3.95]f>
Vapor
Water Solubility Pressure
(mm Hg)
3.77 g/100 ml 760 at
at 4°C -41.l°C
1nsolubleb 1 at 356°C
1nsolubleb NA
solubleb NA
84 g/100 ml NA
at 35°C
38.4 g/100 ml 1 at 157°C
at 14°C
aSource: Weast, 1980
bNo further Information regarding solubility of these compounds Is available In Weast, 1980
NA = Not available
-------
Therefore, the residence time of atmospheric selenium 1s expected to be a
few hours to several days.
The fate of selenium 1n aquatic media 1s dependent on the pH and oxida-
tion-reduction potential of water. Under anaerobic conditions or low pH or
both, Insoluble elemental selenium or metal selenlde Is formed. Under
2_
aerobic conditions or pH >6, water soluble selenlte (Se03 ) or selenate
2_
(SeO. ) Is formed. Sorptlon and copredpHatlon of the soluble sele-
nlte and selenate onto hydrous Iron and manganese oxide control the mobility
of the soluble selenium species. In a reducing aquatic environment, vola-
tile H_Se may be formed. Microorganisms present in sediments in bodies of
water may also convert selenium into volatile methylated products. The last
two processes may cause mobilization of selenium from the aquatic to the
atmospheric phase (Callahan et al., 1979). The bioconcentratlon factor for
selenium 1n freshwater and marine fish is -400 (Callahan et al., 1979).
The fate of selenium 1n soils 1s largely dependent on the pH and redox
potential of the soil. In acidic and poorly aerated soil, heavy metal
selenides are formed and exist 1n immobilized form 1n soil. In well aerated
2- 2-
and alkaline soils, selenium may be converted to Se03 and SeO. .
2_
The selenites (SeO,, ) may be immobilized by adsorption or complexation
or both with Iron and maganese hydroxides. The selenates, however, may
leach from soils into groundwater (MAS, 1976). Page (1981) detected
selenium at a median concentration of 2.0 ppb in all of the groundwater
samples collected from New Jersey.
-3-
-------
2. ABSORPTION FACTORS IN HUMANS AND EXPERIMENTAL ANIMALS
2.1. ORAL
In a study of the GI absorption of selenium, Thomson and Stewart (1973)
administered doses of ~5 yd [7SSe] selenomethlonine (<5 yg selenium)
by gavage to groups of 20 female Wistar rats. GI absorption ranged from
91-93%. Another group of 20 rats received intragastric (~5 yg) doses of
[75Se] selenite and GI absorption was estimated to range from 95-97%.
Subsequently, Thomson et al. (1975) administered 5 yd [7sSe] seleno-
cystine or ~2 yC.1 [75Se] selenomethionine to groups of 25 female Wistar
rats by gavage. Each dose contained not more than 5 yg selenium. Esti-
mated absorption factors from the GI tract were 81.1 and 86.4% for [75Se]
selenocystlne and [75Se] selenomethionine, respectively.
Thomson and Stewart (1974) measured GI absorption of selenium in three
women aged 33, 21 and 25 years. While fasting, each received an oral dose
of -10 yd [75Se] selenite containing <10 yg Se. Calculated GI
absorption rates for the women were 70, 64 and 44%, respectively, indicating
variation in absorption of selenite selenium.
Currently, the concept of bloavallability is beginning to replace
empirical estimates of GI absorption associated with trace elements and
other nutrients. A biological endpoint, such as alleviation of an experi-
mentally reproducible deficiency syndrome, is chosen, and various forms of a
trace element are administered to test animals. The abilities of these test
compounds to alleviate or protect against a deficiency syndrome, compared to
the protective ability of a reference compound, are evaluated and the
results are usually expressed as a percentage of the biological activity of
the reference compound.
-4-
-------
Young et al. (1982) reviewed factors affecting bloavailabmty (hence,
presumably, uptake) of various forms of selenium 1n animals and humans.
Dietary factors Included the Intake level of the element and Its chemical
form; the presence or absence of promoters or Inhibitors such as ascorbic
add, phytate, fiber, sugars, fats and proteins; various mineral-mineral
Interactions and m1neral-macronutr1ent Interactions; type and degree of food
processing; and the concomitant 1ngest1on of certain drugs. They also sug-
gested that physiological factors such as nutritional state, physiological
states (growth, pregnancy, etc.) and pathological states can influence
selenium utilization. Biological factors such as Infectious agents or
social factors such as dietary habits also affect selenium uptake.
Gabrielsen and Opstvedt (1980) compared the availability of three
sources of selenium to restore GSH-Px activity In baby chicks, a particu-
larly sensitive experimental biological system. The ability of equal quan-
tities of selenium contained 1n fishmeal, soybean meal and selenomethionlne
to restore GSH-Px activity were 48, 18 and 78%, respectively, wjien compared
to GSH-Px activity restoration resulting from an equal amount of selenium
from sodium selenlte. These data seem to Indicate that selenium in seleno-
methionlne is much more readily absorbed than selenium in fishmeal or
soybean meal.
Apparent biological availability depends in part, on the biological
endpolnt chosen. Young et al. (1982) reviewed the data of Cantor et al.
(1975a,b) summarized in Table 2-1. From these data it 1s clear that
selenlte is more effective than selenomethionine In protecting chicks from
exudative diathesis but that selenomethionlne Is far more protective than
selenlte against pancreatic flbrosls, another manifestation of selenium
deficiency in chicks. It 1s impossible from these data to determine which
-5-
-------
TABLE 2-1
Comparison of Two Biological Endpolnts to Determine
BloavallablHty of Selenium*
Criterion: Protection Aqalnst
Selenium Compound
Selenlte
Selenocystine
Selenometh1on1ne
Exudative Diathesis
(bloavaHabllUy)
100
68-78
18-61
Pancreatic Flbrosls
100
-100
-400
*Source: Young et al., 1982
-6-
-------
of the selenium compounds tested was absorbed most completely. It can be
concluded that bloavailability tests reflect factors affecting metabolism
and action at the target as well as absorption.
Several studies of New Zealand women, summarized by Young et al. (1982),
Indicated that absorption of [75Se] selenomethionine was greater (96%)
compared to either [75Se] selenite (80%) or [7SSe] selenomethionine
(79%) 1n food. Young et al. (1982) administered [7SSe]-labeled selenite
to chickens by gavage periodically during a 42-day growing period. The
chickens were killed and the resulting [74Se]-labeled chicken meat with or
without additional selenium ([76Se]labeled selenite) was fed to four
volunteers. Those consuming the [74Se]-labeled chicken meat received 13.4
yg selenium and absorbed 80^5%. Subjects who received the labeled chicken
meat and added [76Se] selenite ingested 71.6 vq selenium and absorbed
30^11%. Thomson and Stewart (1973) and Thomson et al. (1975) indicated that
GI absorption of selenium from selenite and selenomethionine was >86% in
rats. In humans, Thomson and Stewart (1974) showed that absorption of
selenium as selenite was quite variable, but considerably less than the
absorption 1n rats. Apparently, 61 absorption of organic forms of selenium
in humans is more complete than absorption of selenium as selenite.
2.2. INHALATION
Pertinent data regarding pulmonary absorption of selenium following
inhalation exposure could not be located in the available literature. The
report by Glover (1967) associating urinary excretion of selenium with
levels of selenium In workroom air (Section 3.2.2.) is Indicative of the
absorption of selenium from the lung.
-7-
-------
3. TOXICITY IN HUMANS AND EXPERIMENTAL ANIMALS
3.1. SUBCHRONIC
3.1.1. Oral. Selenium 1s an essential trace element In livestock (Under-
wood, 1977), rat and chicken (U.S. EPA, 1980a) nutrition and 1s probably
also essential to human health (U.S. EPA, 1980a; Frost and Llsh, 1975; Harr,
1978). Frost and Llsh (1975) cited Its possible role In preventing sudden
Infant death syndrome, coronary heart disease, arthritis and cancer and Us
ability to reduce the toxlcity of other heavy metals, such as cadmium and
mercury. Flohe et al. (1973) found that selenium was associated with human
GSH-Px, supporting the essential role of selenium to human health. The NAS
(1980) determined an adequate and safe range for selenium Intake of 50-200
yg/day for an adult human.
Elementary selenium 1n any of Its allotroplc forms 1s virtually Insol-
uble 1n water and reportedly harmless 1f Ingested (Shapiro, 1973); however,
other dietary constituents affect the toxldty of selenium compounds. Smith
(1939) found that 10 ppm selenium was very toxic to rats when fed a 10%
protein diet, but no signs of toxldty were noted when dietary protein was
Increased by an additional 20%.
Although the literature contains many references regarding the toxldty
of selenium (Schroeder et al., 1970), few quantitative studies of subchronic
oral toxldty of selenium compounds to humans or animals have been located
that relate effects to dosages administered. The purpose of this document
1s to evaluate concisely only those studies that provide data useful for
risk assessment.
Anspaugh and Roblson (1971) stated that rats and dogs exposed to dietary
levels of 5-10 yg selenlum/g may be expected to show evidence of chronic
selenium toxldty, such as Hver atrophy or necrosis, cirrhosis and hemor-
rhage, and a marked and progressive anemia with very low hemoglobin values.
-8-
-------
Vesce (1974) observed changes in endocrine glands, especially the ovaries,
pituitary and adrenals, in guinea pigs exposed orally to 5-12.5 mg sodium
selenide for two periods of 20 days each. No further descriptions of
protocol were available for these studies.
Tinsley et al. (1967) concluded that a dose of 0.5 rag/kg bw/day seemed
to be the threshold that would affect longevity in rats. The authors calcu-
lated that for a 200 g rat eating 10 g of feed/day, this intake constituted
a dietary concentration of 10 ppm. Harr et al. (1967) reported that
additions of 0.5-2 yg selenium/g of diet resulted in hepatomegaly in rats
(duration of exposure not specified). Effects at the lower dietary levels
were more pronounced when selenium was added to semi-purified rather than
commercial diets, suggesting that interaction of selenium with other trace
elements 1s Important In maintaining health. Harr and Muth (1972) stated
that the MED for liver lesions was 0.25 vg selenium/g and the MED for
effects on longevity and lesions of the heart, kidney and spleen was 0.75
vg/g. In these studies, chronic liver lesions (bile duct and parenchyma!
hyperplasia) were more prevalent in rats fed an unsupplemented commercial
diet than in those fed the purified (caseln-cerelose) diet supplemented with
selenium. Interpretation of these results is, therefore, difficult.
Halverson et al. (1966) fed weanling male Sprague-Dawley rats wheat
based diets containing 1.6, 3.2, 4.8, 6.4, 8.0, 9.6 or 11.2 yg selenium
(natural feedstuff and selenite)/g for 6 weeks. No effects were reported in
rats on diets containing up to 4.8 ppm selenium. At the dietary level of
4.8 ppm selenium, a depression 1n growth rate was noted in rats receiving
their selenium as selenite but not in rats receiving their selenium as
-9-
-------
natural selenium 1n selenlferous wheat. At dietary levels >6.4 ppm, reduced
feed Intake, Increased mortality, splenomegaly, Increased pancreas size,
reduced liver weight and anemia were reported. In this study the dietary
level of 3.2 ppm selenium was designated a NOEL.
Frost and L1sh (1975) administered a selenium-vitamin E combination by
capsule to seven volunteers for a period of 10 weeks. Another group of
seven volunteers received placebos. All 14 volunteers received placebos
during a 1-week pretreatment observation period. Those receiving selenium
received 0.5 mg Se/day for 3 weeks, 1.0 mg Se/day for the next 3 weeks and
2.0 mg Se/day for the remaining 4 weeks. The preparation presumeably
contained 0.5 mg selenite selenium and 100 ID of d-alpha tocopherol acid
succinate. Observations, which continued for another 6 weeks, consisted of
weekly physical examinations and "a battery of laboratory [tests] designed
to monitor vital organ function." No differences could be detected between
the drug and placebo.
3.1.2. Inhalation. Pertinent data regarding subchronic inhalation expo-
sure of humans or animals to selenium could not be located in the available
literature.
3.2. CHRONIC
3.2.1. Oral. Chronic oral exposure of humans to toxic levels of selenium
has occurred from well water or foodstuffs grown on selenlferous soils,
particularly in South Dakota, Wyoming, Nebraska and the People's Republic of
China. Additionally, it is expected that acute or chronic toxlcity may
occur as a result of people medicating themselves with over-the-counter
selenium preparations In an attempt to prevent cancer.
-10-
-------
8eath (1962) described symptoms of lassitude, total or partial alopecia,
discoloration of the skin and loss of fingernails 1n several people exposed
to well water containing 9 mg selenium/8.. Discontinued use of the contam-
inated water resulted 1n regrowth of the hair and nails and increased mental
alertness. Assuming an Intake of 2 8. of water/day, the well water
resulted 1n these effects at an intake of 18 mg/man/day.
Smith et al. (1936) conducted an ep1dem1ological study of chronic
selenium toxlcity In people living on selenlferous soil where animals had
been diagnosed as having alkali disease. Clinical signs in these people
Included bad teeth, jaundice, chloasma, vertigo, chronic GI disease,
dermatitis, changes 1n nails (unspecified), arthritis, edema (location
unspecified), lassitude and fatigue. Analysis of foodstuffs in the diets of
these affected people revealed an Intake of selenium of 0.1-0.2 mg/kg/day.
Nearly all urine samples of affected people contained measurable amounts of
selenium and 45% of these samples contained 0.2-1.33 pg/ms.. Urinary
content >0.2 yg/ma. correlated with severity of symptoms.
Yang et al. (1983) reported outbreaks of selenium intoxication 1n
several villages in the Hubei Province of the People's Republic of China.
In five villages 1n which severe outbreaks occurred (totalling 248 inhabi-
tants) the average Incidence was 49.2% with an incidence of 85.5% in the
most severely affected village. Evacuation and the concomitant change 1n
diet resulted 1n recovery. The cause of the selenosls appeared to be the
presence of high levels of selenium in surface coal in the air, which was
then leached from the coal into the soil and into surface and ground waters.
The outbreak was apparently caused by the Increased consumption of selenium
accumulating vegetables such as corn and turnip greens. Drinking water was
considered to contribute negligibly to the intake of selenium.
-11-
-------
Symptoms observed Included brHtleness of nails, loss of nails and hair,
prurKIs of the scalp, dermatitis characterized by hyperemla, edema and
eruptive blistering, and nervous symptoms such as peripheral anesthesia,
acroparesthesla and pain 1n the limbs. Eventually, exaggerated tendon
reflexes, numbness, convulsions, motor dysfunction progressing to paralysis
and hemlplegla developed. Mottled teeth and Increased tooth decay were also
reported, but Interpretation of these findings Is difficult because the area
was also found to be high 1n fluoride. These symptoms correlated with dally
Intakes of selenium ranging from 3.20-6.69 mg/day with an average of 4.99
mg/day, computed based on analysis of components of the diet.
An Increase In the Incidence of dental carles has been associated with
toxic levels of selenium In humans. Hadjlmarkos and Bonhorst (1961) and
Hadjlmarkos (1969) showed that urinary excretion of selenium was about twice
as high 1n children with a high Incidence of dental carles compared with
children with a low Incidence of carles. Other studies have shown an Incon-
sistent (Huhleman and Konlg, 1964) or only marginally significant (Ludwlg
and Biddy, 1969) relationship between prevalence of dental carles and
dietary levels of selenium. Unfortunately, no exposure data were available
1n the secondary source (U.S. EPA, 1980a) from which these studies were
summarized. An Increased Incidence of dental carles has been observed In
children In selenlferous communities where urine selenium levels averaged
0.33 mg/a (Tank and Storvick, 1960). Bowen (1972) exposed young monkeys,
Macaca 1rus. to drinking water containing 1 ppm selenium as sodium selenlte
for 5 years. Carious lesions developed twice as frequently and 3 times
faster 1n the selenium-treated monkeys as 1n the control group. Assuming
monkeys weigh 3.5 kg and drink 450 ma of water/day, 1 ppm selenium 1n the
drinking water corresponds to an Intake of 0.13 mg selen1um/kg/day. Buttner
-12-
-------
(1963) reported a positive correlation of dietary selenium with dental
carles In rats. It 1s easier to control dietary factors 1n laboratory rats
than 1n human populations, however, and other dietary factors may have been
related to the Incidence of dental carles 1n humans.
Several studies of chronic oral exposure of laboratory animals to
selenium have been conducted. Franke and Potter (1935) fed groups of nine
Wlstar rats diets containing 0, 22.3 or 33.5 ppm selenium as sodium sele-
nite. At 33.5 ppm selenium in the diet, decreased food consumption and
growth and Increased mortality (8/9 dead by day 359) were observed. At 22.3
ppm selenium in the diet, growth depression and increased mortality (5/9
dead by day 359) were noted. Necropsy of low-dose survivors revealed marked
evidence of liver necrosis and degeneration.
Nelson et al. (1943) fed diets containing 0 or 10 ppm added selenium as
ammonium potassium selenlde to groups of 18 rats for 24 months. Excessive
mortality (6/18 of controls, 12/18 of treatment rats) was reported.
Schroeder (1967) and Schroeder and MHchener (1971a) treated groups of
-100 Long-Evans BLU:LE rats with drinking water containing no added selenium
(controls), 2-3 mg/a of sodium selenite or 2-3 mg/s, of sodium selenate.
The experiment was Intended to continue throughout the Hfespan of the
treated animals as a carcinogeniclty bioassay. Male rats were especially
sensitive to selenium as selinlte and 50% of the males In the selenite group
had died within 2 months. The remaining males in this group were given
sodium selenate for the remainder of the experiment. Females in the sele-
nite group were not similarly affected and remained on the sodium selenite
treatment. No other parameters of toxiclty were reported.
Fltzhugh et al. (1944) fed diets containing 5 ppm selenium as selenif-
erous corn in a relatively protein-deficient diet to rats for 24 months.
-13-
-------
After 2 months and 1 year, respectively, mortality was 0 and 11% 1n the
control rats and 11 and 17% In the treated rats. Although mortality In the
treated rats after 1 year did not appear to differ significantly from
mortality among control rats, after 2 months the Incidence of mortality was
much higher among treated rats compared with controls. Since a high protein
diet provides partial protection against selenium toxicity (Smith, 1939),
these data suggest that the ability of a low protein diet to exacerbate
toxldty of selenium may be manifested more during the period of maximal
growth when dietary requirements for protein are high, rather than after
maturity.
Concurrently, FHzhugh et al. (1944) fed rats diets containing 10, 20 or
40 ppm selenium as ammonium potassium selenlde. The Incidence of mortality,
liver cirrhosis and liver tumors was Increased In all treated rats compared
with controls. The authors concluded, however, that selenlde selenium was
only half as toxic as an equivalent dose of selenium from selenlferous corn.
As mentioned previously, higher Intakes of selenium have been associated
with an Increased Incidence of dental carles. In an unpublished experiment
reported by Frost and L1sh (1975), selenium was given to monkeys by capsule
at 0, 20, 64 or 200 yg/Kg/day. The preparation contained 0.5 mg selenium
as the selenlte and 100 III vitamin E as d-alpha tocopherol add sucdnate.
After treatment for 55 weeks, three males and two females from each group
were sacrificed for macro- and microscopic pathological examinations. The
remaining monkeys were treated until week 58 and were sacrificed at week 70.
Physical examinations, behavioral observations, survival and body weight
gains, EKG and blood pressure determinations, ophthalmologlc examinations,
gross and microscopic pathological examinations, absolute and relative organ
weights and a host of clinical chemistry and hematology parameters failed to
reveal any signs of toxldty.
-14-
-------
3.2.2. Inhalation. Hamilton (1974) reported chronic toxUHy 1n workers
due to occupational exposure to selenium. Symptoms Including a strong
garlic odor to the breath, sweat and urine, upper respiratory and GI Irrita-
tion, cold-Uke symptoms, lacMmatlon and a metallic taste 1n the mouth were
reported by workers exposed to selenium In a copper refinery. Hamilton
(1974) also reported other cases of selenium toxldty related to Industrial
exposure; however, the level of exposure was not quantHated 1n any of these
reports.
Bellies (1981) suggested that selenium dioxide may be responsible for
toxldty In Industrial exposures. It 1s probably formed from heating
selenium, and then forms selenlous acid 1n water or sweat. The acid Is
Irritating and was probably the cause for the evacuation of all workers and
the hospltallzatlon of two who experienced severe headache and respiratory
discomfort as a result of exposure to fumes from an aluminum smelting opera-
tion that had been charged with selenium-contaminated, used rectifier plates
(Clinton, 1947).
Glover (1967) measured urinary selenium concentration In workers at a
selenium rectifier plant over a 5-year period. Workers were exposed to
selenium 1n air at concentrations that varied with the particular process 1n
which they were employed. Although the air concentrations were analyzed
only sporadically, there seemed to be a correlation between air concentra-
tions and urinary concentrations of selenium. Workers exposed to undeter-
mined levels of airborne selenium complained of skin rashes (contact derma-
titis), garlic odor of the breath, Indigestion and Indefinite sodopsycho-
loglcal effects. Although data were few, mortality among the selenium
exposed workers did not appeared to be different from mortality among the
-15-
-------
general population. Glover (1967, 1970) recommended a maximum allowable
urinary concentration of selenium of 0.1 mg/i, which he believed to corre-
spond approximately to an air concentration of 0.1 mg/m3.
3.3. TERATOGENICITY AND OTHER REPRODUCTIVE EFFECTS
3.3.1. Oral. Although no reports conclusively link teratogenicity of
selenium or Its compounds to humans, Robertson (1970) suggested that sele-
nium may be a teratogen 1n man. Rosenfeld and Beath (1964) suggested that
malformed Infants of Indian women 1n Colombia may have resulted from Ingest-
ing grains containing "toxic" levels of selenium. Robertson (1970) evalu-
ated Information on the possible association between abnormal pregnancies
and exposure of women to selenlte. Of one possible and four certain preg-
nancies studied, one went to term and resulted 1n a bilateral clubfoot
Infant. No Information on the other pregnancies was available.
The chicken embryo Is extremely sensitive to selenium. HatchabUHy of
eggs Is reduced by dietary concentrations of selenium too low to cause
toxldty 1n other farm animals. The eggs are fertile but often produce
grossly deformed embryos lacking eyes and beaks and having deformed wings
and feet (Carlson et a!., 1951; Franke et al., 1936; Franke and Tully, 1935;
Gruenwald, 1958; Palmer et al., 1973). Deformed embryos were also produced
by Injection of selenlte Into the air cell of normal, fertile eggs of both
chickens (Franke et al., 1936) and turkeys (Carlson et al., 1951).
The consumption of selenlferous diets by rats (Franke and Potter, 1935;
Rosenfeld and Beath, 1964), pigs (Wahlstrom and Olson, 1958), sheep (Rosen-
feld and Beath, 1964) and cattle (Dlnkel et al., 1963) has been shown to
Interfere with normal fetal development and to cause fetal malformations.
-16-
-------
These malformations are a part of the total syndrome designated "alkali
disease" and, In general, seem to occur at dietary levels that cause other
toxic manifestations (Underwood, 1977). These reports are not useful 1n
risk assessment because no exposure data were given.
In an attempt to shorten the time Involved 1n testing the chronic toxlc-
1ty of chemicals, Schroeder and Mltchener (1971b) performed a multigenera-
tlon reproduction experiment 1n rats and mice using trace elements and other
metals with known toxlclty. Groups of five pairs of Charles River CO mice
were fed a purified diet containing minimal (0.056 ppm) selenium. Treat-
ment, Initiated at weaning, was by the addition of 3 ppm selenium (as sele-
nate) to doubly delonlzed drinking water. Control mice received the same
diet and doubly delonlzed drinking water. Pairs of mice were allowed to
breed to 6 months of age. The next generation was bred from F, , F,.
and F, offspring. The F- generation was obtained from F- and F_.
Utters. The experiment was terminated when it was obvious that the strain
was dying out or when the 3rd generation had been weaned.
Parameters of toxlclty evaluated were Intervals between Utters, age at
which first Utter was produced, sex ratio at birth, number of runts, number
of stillborn, failure to breed sucessfully, congenital abnormalities and
maternal death. By these criteria, selenium was toxic to mice. The strain
began to die out at the 3rd generation; seven pairs failed to breed; three
that delivered produced a total of 23 offspring of which 16 were runts. In
all, selenium In mice resulted 1n 93 runts of 389 offspring delivered live,
one stillborn Utter, death of 23 offspring shortly after birth, a male to
female ratio 1n individual offspring ranging from 1.27-1.65 and one maternal
death. Congenital malformations were not reported for selenium.
3.3.2. Inhalation. Studies associating teratogenlcity with Inhalation
exposure to selenium could not be located In the available literature.
-17-
-------
3.4. TOXICANT INTERACTIONS
Interactions between selenium and several toxic heavy metals have been
demonstrated. Moxon (1938) alleviated the toxicity associated with 15 ppm
dietary selenium (seleniferous grain, selenite or selenocystine) by provid-
ing -5 mg arsenic as sodium arsenate/j. of drinking water. The presence of
arsenic appeared to enhance excretion of selenium into the GI tract (Ganther
and Baumann, 1962), at least in part by increasing biliary excretion
(Levander and Baumann, 1966). Presumably, administration of selenium could
protect against toxicity induced by arsenic.
Selenium has been shown to protect against the toxic effects of cadmium
upon various reproductive organs of rats. Kar et al. (1960) and Mason and
Young (1967) demonstrated that testicular damage due to cadmium could be
prevented by the simultaneous administration of selenium. Kar et al. (1959)
and Parizek et al. (1968) found that selenium would prevent cadmium-induced
damage to the non-ovulating ovary in the rat. Parizek et al. (1968) and
Parizek (1964) found that administration of selenium prevented placental
necrosis caused by exposure to small amounts of cadmium near parturition.
The teratogenicity associated with cadmium and arsenic has been reduced by
selenium (Holmberg and Perm, 1969; Perm, 1972). Parizek et al. (1968) and
Gunn et al. (1968) greatly reduced the mortality rates of rats exposed to
lethal doses of cadmium by the simultaneous administration of selenium.
Selenium salts have been shown to protect the kidney and intestines of
rats exposed to lethal doses (0.02 mmole/kg) of mercury (Levander and
Argrett, 1969). [The protective effects of selenium appeared to be asso-
ciated with decreased urinary excretion of mercury (Parizek et al., 1971).]
-18-
-------
Levander and Argrett (1969) reported that mercury Increased the retention of
selenium In the blood, kidneys and spleen. Parlzek et al. (1971) found that
transplacental movement of mercury 1n pregnant rats was decreased by sele-
nium and In lactatlng rats less mercury was secreted Into the milk. Also,
the bloavanabllity of selenium In rats was reduced by mercury.
Dlplock (1976) and Grasso et al. (1969) produced toxic symptoms In rats
and chickens fed vitamin E-defldent diets with 0.15% silver acetate In the
drinking water. Rats suffered dystrophlc lesions and necrosis of the liver,
and high mortality; chickens suffered a pro-exudative diathesis effect.
Supplementation with the addition of 1 ppm selenium to the diet resulted 1n
55% protection against the toxic effects of silver. Liver lesions due to
silver toxldty, and vitamin E and selenium toxldty were similar (Grasso et
al., 1969).
Hollo and Sztojcso (1960) demonstrated that death due to thallium
poisoning could be prevented by parenteral administration of selenate.
Ruslecki and Brzezinskl (1966) found that oral administration of selenate
prevented thallium toxlcity and resulted 1n a greater thallium content in
liver, kidney and bone than occurred in nonselenate-treated animals sub-
jected to thallium toxidty. Levander and Argrett (1969) showed that subcu-
taneous Injection of thallium acetate increased the retention of selenium in
the liver and kidney of rats and reduced pulmonary and urinary excretions of
selenium.
Halverson and Monty (1960) demonstrated that dietary sulfate restored by
>40% the growth rate depression 1n rats resulting from excessive dietary
selenium as selenate or selenite. Sulfate, however, appeared to have no
protective effect on liver degeneration associated with selenium.
-19-
-------
Moxon and Dubois (1939) reported that fluoride increased the toxlcity of
selenium In rats. Adding 5 mg fluoride/8, to the drinking water of rats
receiving 11 ppm dietary selenium as seleniferous grain resulted in in-
creased mortality and decreased feed and water intake and rate of body
weight gain. Hadjimarkos (1965, 1969), on the other hand, exposed rats to 3
mg selenium/5, of drinking water with and without the addition of 3 mg
fluoride/2, of drinking water. No additional severity of toxlcity was
observed in the group receiving both selenium and fluoride.
Levander (1982) mentioned interactions between vitamin E and selenium.
Several experiments show that these two nutrients can partially spare one
another. For example, 0.05 ppm of dietary selenium was needed to prevent
exudative diathesis in chicks fed a diet devoid of vitamin E, but only 0.01
ppm was needed when the diet contained 100 ppm vitamin E (Scott et al.,
1967).
The herbicide, paraquat, is thought to be toxic as a result of stimulat-
ing Hpid peroxidation (Bus et al., 1974). Cagen and Gibson (1977) found
that, when exposed to toxic amounts of paraquat, selenium-deficient rats
suffered increased lung damage and selenium-deficient mice suffered in-
creased hepatic injury, compared with control animals that were not defi-
cient in selenium.
Combs and Scott (1975) demonstrated that the addition of 50 ppm of PCB
to the diets of chicks deficient in vitamin E and marginal in selenium
resulted in an Increase in the incidence of exudative diathesis compared
with the incidence in chicks on similar diets without added PCB.
Selenium deficiency appears to protect against toxlcity from certain
xenobiotics. Burk and Lane (1979) observed that selenium deficient rats
suffered less severe hepatotoxicity than selenium-adequate rats when exposed
to acetaminophen or iodipamide.
-20-
-------
Since under certain conditions cyanide Inhibited the activity of GSH-Px
(Kraus and Ganther, 1980), 1t was suggested that cyanide may precipitate
selenium deficiency. Rudert and Lewis (1978) showed that drenching with
potassium cyanide did Increase the Incidence of myopathy, a common manifes-
tation of selenium deficiency in lambs. Selenium toxldty 1n rats has been
alleviated by cyanide (Palmer and Olson, 1979), and Unseed meal, long known
to counteract selenium poisoning, has been shown to contain two cyanogenlc
glycosldes (Palmer et a!., 1980).
-21-
-------
4. CARCINOGENICITY
4.1. HUMAN DATA
Evidence of a carcinogenic role for selenium In humans appears to be
lacking. Although the occupational toxldty of selenium has been recognized
since the early 1900s, no cases of malignancy attributable to exposure to
the element have been reported (Schroeder et a!., 1970). Shamberger and
Frost (1969) cite data suggesting that selenium may help protect the human
population against cancer. In a 31 state area 1n which forages grown for
livestock consumption averaged <0.06 ppm, the death rate In 1965 was lower
than 1n a 17 state area and the District of Columbia, 1n which forages
averaged <0.05 ppm. In a study of the concentration of selenium 1n the
blood and the death rate due to cancer, a nearly perfect negative Pearson
coefficient (r=-0.96, p<0.001) was found, Indicating that protection from
death due to cancer Is associated with higher blood levels of selenium
(Shamberger and Frost, 1969).
In a review of selenium and cancer, Shapiro (1972) cited j_n vitro data
suggesting an antlcancer role for selenium. In experimental therapy 1n
humans, Welsburger et al. (1956) and Welsburger and Suhrlaud (1956a,b)
applied selenium cystelne 1n the treatment of cysteine-dependent leukemlas.
Chemical progress was Impressive, but therapy was terminated due to selenium
toxldty.
4.2. BIOASSAYS
Several Investigators have studied the carclnogenldty of orally admin-
istered selenlferous compounds; however, none of these studies present
reasonably consistent data that suggest a carcinogenic role for selenium.
-22-
-------
After a short-term range finding experiment. Nelson et al. (1943)
exposed seven groups of 18 female Osborne-Mendel rats to 5, 7 or 10 ppm
selenium In the diet using either selenlferous corn or wheat. In addition,
another 10 ppm selenium group was treated with selenlte added to the diet.
A control group of 18 rats was maintained. Mortality was high and tended to
be proportional to the dietary level of selenium. Of 53 treated rats that
survived to 18 months, 11 developed hepatocellular "adenomas or low grade
carcinomas." Rats that died or were killed before 18 months exhibited
cirrhosis starting as early as 3 months, but no tumors were found. The 14
control rats that survived to 18 months showed no evidence of neoplasla.
There was no discussion of the statistical significance of these findings.
Harr et al. (1967) and Tlnsley et al. (1967) assigned 1437 Wlstar rats
to 34 different dietary groups. Dietary selenium levels ranged from 0.5-16
ppm (sodium selenlte and selenate) 1n rats maintained on high (22% casein)
or low (12% casein with or without 0.3% DL-methlonine) protein diets. A
known carcinogen, FAA, was fed at dietary levels of 50 or 100 ppm to some
groups as a positive control. Mortality claimed many rats 1n the 8 and 16
ppm groups and caused early termination of these groups. Most rats lived
<100 days; 175 lived >2 years. Necropsies were performed on 1126 of the
original 1437 rats; 63 neoplasms were found, 43 In FAA-exposed rats and the
other 20 randomly distributed among selenium-treated animals. No hepatomas
were found. Although lifetime exposures to toxic levels of selenium were
shown to produce serious pathological changes in the liver and other organs,
no hepatic cancers were observed among selenium-exposed animals. The total
number of cancers observed and their distribution in treated rats appeared
similar to those observed 1n controls.
-23-
-------
Volgarev and Tscherkes (1967) reported the results of three experiments
1n which rats were fed diets containing variable levels of protein (12-30%
casein) and selenium from selenate at 4.3 or 8.6 ppm. In the first experi-
ment, 23/40 rats lived >18 months, 4 developed sarcomas, 3 hepatocellular
carcinomas and 3 hepatocellular adenomas. Two of the rats with hepatocellu-
lar carcinomas had metastases. In the second experiment Involving 60 rats,
one hepatocellular carcinoma, one hepatocellular adenoma and three sarcomas
were reported. Preneoplastlc lesions or carcinomas were not reported 1n the
third experiment with 100 rats. Negative control rats were not Included 1n
these experiments; hence, no conclusions can be made regarding the Incidence
of these observed tumors.
Schroeder (1967) and Schroeder and MHchener (1971a) exposed 418 wean-
ling Long-Evans rats to 2-3 mg of sodium selenlte or sodium selenate, or 2
mg tellurite/2. of drinking water for lifetime. At 21 months of age, a
virulent epidemic of pneumonia caused considerable mortality. Necropsy of
-75% of the animals was performed at -28 months. H1stolog1cal examinations
were performed only on selected Individuals. Control animals exhibited 11
malignant tumors and 20 were reported in the selenate-exposed rats. No
statistical analyses of these data were reported. Few data regarding malig-
nancies in the selenlte-exposed groups were given. Due to the toxlcity of
selenlte, male rats were switched to selenate and, along with the females on
selenlte, were terminated before the age of high tumor incidence.
Schroeder and Mitchener (1972) repeated these studies in mice. Treat-
ment of 3 mg selenium/1 of drinking water did not have a significant
effect on the incidence of spontaneous tumor formations in mice.
The U.S. EPA (1980a) reported the unpublished results of an NCI (1978)
bioassay. B6C3F1 mice were divided into groups of 50 of each sex and were
designated control, vehicle control, low-dose and high-dose groups.
-24-
-------
Low- and high-dose groups received 11.05 mg selenium/kg and 55.24 mg sele-
nium/kg, respectively, as selenium dlsulflde 1n carboxymethyl cellulose by
gavage for lifetime. It was apparent that this strain of mice had a high
rate of spontaneous tumors of multiple anatomic sites. Among groups of male
mice, exposure to selenium did not result In any statistically significant
Increased Incidence of tumors. High-dose females, however, experienced a
significantly higher (21/49) Incidence of hepatocellular carcinoma compared
with controls (2/50) or vehicle controls (0/49).
These data suggest that, at least 1n female mice, selenium disulflde 1s
a carcinogen. Selenium dlsulflde, however, 1s not just another salt of
selenium but a distinct compound which Is unlikely to be found 1n the
environment as a pollutant. It cannot be assumed that these results suggest
that Inorganic selenium (selenlte or selenate) would be carcinogenic 1n
humans or animals (U.S. EPA, 1980a).
4.3. OTHER RELEVANT DATA
Recent data concerning the mutagenlcHy of selenium or Its compounds
could not be located 1n the available literature. The U.S. EPA (1980a) has
summarized the data to 1980, which 1s presented here.
Selenium 1n selenoamlno acid has been shown to reduce genetic crossing
over 1n DrosophUa melanoqaster. Selenocystine at 2 mmol had a significant
effect on crossing-over 1n the x-chromosome of I), melanogaster (T1ng and
Walker, 1969; Walker and Bradley, 1969).
Senteln (1967) found that selenates, selenltes and selenium dioxide
caused similar effects on segmentation mitoses: polar dissociation with
conserved dominance of the principal pole, stickiness and chromosomal clump-
Ing. In another study, sodium selenlte was found to cause degenerative cell
changes and decreased mltotlc activity In rabbit kidney tissue cultures
(exposure unspecified) (Foklna and Kudryavtserla, 1969).
-25-
-------
No chromosomal aberrations were observed 1n human leukocyte or flbro-
blast cultures exposed to at least two concentrations of sodium selenate or
sodium selenlte for 24, 48 or 72 hours (Paton and Allison, 1972).
4.4. WEIGHT OF EVIDENCE
From the studies reviewed In Section 4.2., 1t 1s apparent that the
cardnogenldty of Inorganic selenium has not been unequivocally established
and the question of the carclnogenlclty of Inorganic selenium will not be
resolved without further Investigation. There 1s limited evidence, however,
that selenium dlsulflde 1s a carcinogen In female mice, but none of the
studies to date provide a weight of evidence or sufficient dose-response
data for risk estimation. The NAS (1976) mentioned that selenium salts have
been used both prophylactlcally and therapeutlcally 1n livestock husbandry
for a number of years. Human exposure to selenium has Included Its Incorpo-
ration Into shampoos and Us presence 1n Industrial plants. Epldemlologlc
and demographic evidence from Us widespread use has failed to Implicate
selenium as a carcinogen; 1n fact, H has been associated with a reduction
1n the Incidence of human ovarian cancer (Frost, 1971; Schroeder and
MHchener, 1972; Anonymous, 1970; Shamberger and Rudolph, 1966; Shamberger
et al., 1972, 1973; Shamberger and WHHs, 1971; Wedderburn, 1972).
Several studies have demonstrated an Inhibitory effect on tumor Inci-
dences 1n laboratory animals, complicating assessment of the carcinogenic
role of selenium. Harr et al. (1972) noted an apparent decreased Incidence
of 2-acetylam1nofluorene-1nduced mammary carcinomas and hepatomas 1n mice
related to content of selenium In the diet. A 50% reduction In the Inci-
dence of dlmethylamlnobenzene-lnduced liver tumors In rats was noted as the
result of adding 5 ppm sodium selenlte to the diet (Clayton and Baumann,
1949).
-26-
-------
IARC (1975) concluded that the available animal data are insufficient
for evaluation of the cardnogenldty of selenium compounds. The available
human data do not suggest that selenium 1s carcinogenic in man, and "the
evidence for a negative correlation between regional cancer deaths and sele-
nium 1s not convincing" (IARC, 1975). Applying the criteria for evaluating
the overall weight of evidence for the carcinogenicity of selenium (and
compounds) for humans proposed by the Carcinogen Assessment Group of the
U.S. EPA (Federal Register, 1984), selenium Is most appropriately designated
a Group D - Not Classified chemical.
-27-
-------
5. REGULATORY STANDARDS AND CRITERIA
The ACGIH (1980) recommended a TWA-TLV for selenium 1n the workroom
atmosphere of 0.2 mg/m3, primarily to protect against the Irritative
effects of selenium. No STEL has been recommended. The ACGIH (1983) recom-
mended a TWA-TLV of 0.2 mg/m3 selenium for selenium hexafluoMde. The
OSHA standard for selenium has been set at 0.2 mg/m3 (Code of Federal
Regulations, 1981).
The U.S. EPA (1980a) concluded that the cardnogenlcHy of selenium has
not been clearly demonstrated, and based the derivation of an ambient water
quality criterion on the collective data of several short-term toxlcity
studies in rats. Based on an estimated LOEL in rats, occurring at 0.5 ppm
selenium in the diet, the U.S. EPA (1980a) recommended an ambient water
quality criterion of 10 ng/fc. This value was chosen with the considera-
tion that only 5-10% of the dally exposure to selenium should come from
ingestion of water.
The FDA ruled that sodium selenlte or sodium selenate may be added to
the complete feed for swine and chickens up to 16 weeks of age not to exceed
0.1 ppm, and for turkeys, not to exceed 0.2 ppm (U.S. EPA, 1980a).
-28-
-------
6. RISK ASSESSMENT
6.1. ACCEPTABLE INTAKE SUBCHRONIC (AIS)
6.1.1. Oral. Reports of toxlclty In humans with subchronic oral exposure
to selenium could not be located in the available literature. Animal data
concerning subchronic oral selenium toxicity are difficult to interpret.
Harr et al. (1967) reported hepatomegaly in rats associated with 0.5 ppm
selenium in the diet. Harr and Muth (1972) suggested that a dietary level
of 0.25 ppm was the threshold associated with liver lesions and that 0.75
ppm was related to reduced longevity; however, artificially contrived
purified diets were used that circumvented the normal interactions of trace
elements and resulted in unusual apparent toxicity associated with selenium,
and these data must be interpreted with caution. Anspaugh and Robison
(1971) demonstrated liver atrophy in rats and dogs, and Vesce (1974) found
"endocrine changes" in guinea pigs exposed to diets containing 5 ppm
selenium. Halverson et al. (1966) fed diets containing 1.6 (intrinsic),
3.2, 4.8, 6.4, 8.0, 9.6 or 11.2 ppm. selenium to rats for 6 weeks. Dietary
levels >6.4 ppm resulted in mortality, anemia and pathologic changes in
liver, pancreas and spleen. A dietary level of 4.8 ppm was associated with
reduced growth rate in some rats. In this study, 3.2 ppm selenium corre-
sponding to an intake of 0.16 mg/kg bw/day, assuming rats eat food equiva-
lent to 5% of their body weight/day, was a NOEL. From this animal dose, an
AIS can be calculated by multiplying the animal dose (0.16 mg/kg bw/day) by
70 kg (body weight of man) and dividing by an uncertainty factor. The fact
that selenium is a required nutrient for animals and presumably for humans
(NAS, 1980) enters into the selection of an uncertainty factor. An uncer-
tainty factor of 50 is chosen, a factor of 5 for interspecies extrapolation,
because it seems reasonable that both rats and humans may have similar
-29-
-------
dietary requirements for selenium (U.S. EPA, 1980a) and a factor of 10 to
provide additional protection for especially sensitive populations, such as
people living 1n known selenlferous areas. The resultant AIS = 0.224 mg/day
for subchronlc oral exposure. The NAS (1980) Food and Nutrition Board
recommended a human dietary requirement of 10-200 yg/day for selenium and
estimated that the average adult Ingests -130-150 yg of selenium/day from
food.
6.1.2. Inhalation. Pertinent data regarding subchronlc Inhalation
exposure of humans or laboratory animals to selenium could not be located 1n
the available literature; hence, no Interim AIS for subchronlc Inhalation
exposure can be calculated.
6.2. ACCEPTABLE INTAKE CHRONIC (AIC)
6.2.1. Oral. Beath (1962) reported symptoms of lassitude, alopecia, dis-
coloration of skin, and loss of fingernails 1n people exposed to well water
containing 9 mg selenium/8.. Assuming an Intake of 2 i of water/day, an
average of -18 mg selen1um/man/day was Ingested. Smith et al. (1936)
examined the diets of people farming selenlferous soils and reported that
average dally Intakes of -7-14 mg of selenium were associated with symptoms
similar to those described by Beath (1962).
Several Investigators (Franke and Potter, 1935; Nelson et al., 1943;
Fitzhugh et al., 1944) chronically exposed laboratory animals to selenium by
the oral route, but the exposures used resulted 1n mortality and did not
define NOELs, NOAELs or LOAELs that are useful to risk assessment. More
recently, however, Bowen (1972) exposed young monkeys for 5 years to drink-
Ing water containing 1 ppm selenium to study the Impact of a low level of
selenium on the formation of dental carles. Carious lesions developed twice
as frequently and 3 times faster in selenium-treated monkeys than 1n control
-30-
-------
monkeys. No other signs of selenium toxldty were reported. Assuming that
monkeys weigh 3.5 kg and drink 450 ms. water/day, an Intake of 0.13 mg
selenium/kg bw/day Is calculated. That the dose of 0.13 mg/kg bw/day
associated with Increase in dental carles 1s below the threshold for other
effects In monkeys Is supported by the data reported by Frost and Llsh
(1975). In this study, no manifestations of toxldty were observed In
monkeys given 200 yg/kg/day by capsule. Dental carles were not reported
1n this study at this dosage level but It Is not clear whether an oral
examination was performed and this study ran for a much shorter period of
time than the 5-year study of Bowen (1972).
The epidemiology study of Yang et al. (1983) Is chosen rather than the
monkey study of Bowen (1972) for derivation of an ADI because the LOAEL from
the human study, 3.20 mg/day or 46 vg/kg bw/day, 1s lower than the NOEL of
200 yg/kg bw/day from the monkey study. Furthermore, the effects were
carefully catalogued and care was taken In estimating the range of Intakes
associated with selenosls. The uncertainties associated with animal to
human extrapolation are also eliminated. The U.S. EPA (1985) derived an ADI
for selenium based on the LOAEL of 3.20 mg/day associated with selenosls
from the human ep1dem1olog1cal data (Yang et al., 1983). An uncertainty
factor of 15 was applied to convert the LOAEL of 3.20 mg/day to an ADI. A
factor of 15 rather than 10 was applied to reflect more efficient absorption
from water than from the diet. An ADI of 0.21 mg/day was calculated for
selenium. It Is suggested that this ADI be adopted as the oral AIC. This
ADI Is similar to the ADI of 0.224 mg/day calculated for subchronlc exposure
from the Halverson et al. (1966) study 1n rats.
-31-
-------
U.S. EPA (1983a) reviewed the data base for selenium and Its compounds
and selected the multlgeneration study In mice (Schroeder and MHchener,
1971b) as being most appropriate for calculating a CS to most stringently
represent the toxlcity of selenium compounds. Assuming that mice consume
water equivalent to 17% of their body weight, an Intake of 0.51 mg/kg bw/day
was calculated. The animal dose was multiplied by the cube root of the
ratio of the body weight of mice (assumed to be 0.03 kg) to that of humans
(assumed to be 70 kg) and multiplied by 70 kg to derive a human MED of 2.69
mg/day. This MED corresponds to an RV of 4.9 and the effect observed,
Increased neonatal mortality, was assigned an RV of 10. The resulting
CS, 49, was calculated as the product of RV. and RV .
6.2.2. Inhalation. Glover (1967) measured urinary concentrations of
selenium 1n workers exposed to atmospheric selenium and found that a urinary
concentration of 0.1 mg/a. corresponded to an air concentration of 0.1
mg/m.3, which he recommended as a maximum allowable concentration 1n work-
room air. Lacking more definitive data on the effects of selenium by
Inhalation, an AIC can be derived from these data as from a TLV. Assuming a
worker Inhales 10 m3 of air during the workday, this exposure corresponds
to a daily Intake of 1 mg/day during a 5-day workweek. An AIC Is calculated
by multiplying by 5/7 to expand the exposure to a 7-day week and by dividing
by an uncertainty factor of 10, intended to provide greater protection to
members of the population unusually sensitive to selenium inhalation. The
resultant AIC is 0.07 mg/day.
6.3. CARCINOGENIC POTENCY (q^)
6.3.1. Oral. Although the NCI (1978) bioassay indicated a possible
carcinogenic role for selenium disulfide in female 86C3F1 mice, as was
discussed in Section 4.4., the carclnogenicity of other selenium compounds
-32-
-------
1s uncertain at the present time. The U.S. EPA (1980a), suggesting that the
currently available Information does not warrant classification of selenium
or Us compounds as carcinogenic and reviewing the conclusion of the MAS
(1976) that selenium intake may be correlated with a reduction in the
incidence of cancer, concluded that insufficient evidence was available from
which to calculate a q *
6.3.2. Inhalation. Since no reports of carcinogenicity associated with
inhalation exposure to selenium or Us compounds have been located in the
available literature, no q,* for inhalation exposure can be calculated.
-33-
-------
7. REFERENCES
AC6IH (American Conference of Governmental Industrial Hygienists). 1980.
Documentation of the Threshold Limit Values for Substances in Workroom Air,
4th ed. with supplements through 1981. Cincinnati, OH. p. 361-362. (Cited
in U.S. EPA, 1983a)
ACGIH (American Conference of Governmental Industrial Hygienists}. 1983.
Threshold Limit Values for Chemical Substances and Physical Agents in the
Workroom Environment with Intended Changes for 1984. Cincinnati, OH.
Anonymous. 1970. Selenium and cancer. Nutr. Rev. 28: 75-80. (Cited in
U.S. EPA, 1980a)
Anspaugh, L.R. and W.L. Robison. 1971. Trace elements in biology and
medicine. Prog. Atomic Med. 3: 63. (Cited in U.S. EPA, 1980a) .
Beath, O.A. 1962. The story of selenium in Wyoming. University of
Wyoming, Laramie. (Cited in U.S. EPA, 1980a)
Bellies, R.P. 1981. Phosphorus, selenium and tellurium. In.: Patty's
Industrial Hygiene and Toxicology, 3rd ed., Vol. IIA, G.D. Clayton and F.E.
Clayton, Ed. John Wiley and Sons, Inc., NY. p. 2130-2140.
Bowen, W.H. 1972. No title provided. J. Ir. Dent. Assoc. 18: 83. (Cited
in Underwood, 1977; U.S. EPA, 1983a)
-34-
-------
Burk, R.F. and J.M. Lane. 1979. Ethane production and liver necrosis 1n
rats after administration of drugs and other chemicals. Toxlcol. Appl.
Pharmacol. 50: 467. {Cited In Levander, 1982)
Bus, J.S., S.D. Aust and J.E. Gibson. 1974. Superoxlde- and singlet
oxygen-catalyzed llpld peroxldatlon as a possible mechanism for paraquat
(methyl vlologen) toxldty. Blochem. Blophys. Res. Commun. 58: 749.
(Cited 1n Levander, 1982)
Buttner, W. 1963. Action of trace elements on the metabolism of fluoride.
J. Dent. Res. 42: 453. (Cited In U.S. EPA, 1980a)
Cagen, S.Z. and J.E. Gibson. 1977. Liver damage following paraquat In
selenium-deficient and dlethyl maleate-pretreated mice. Toxlcol. Appl.
Pharmacol. 40: 193. (Cited 1n Levander, 1982)
Callahan, M.A., H.W. SUmak, N.W. Gabel, et al. 1979. Water-related
environmental fate of 129 priority pollutants. Vol. I. U.S. EPA, Office of
Water Planning and Standards, Office of Water and Waste Management, Wash-
ington, DC. EPA 440/4-79-0293.
Cantor, A.H., H.L. Scott and T. Noguchl. 1975a. Biological availability of
selenium 1n feedstuffs and selenium compounds for restoring blood plasma
glutathlone peroxldase activity 1n selenium-depleted chicks. J. Nutr. 105:
96-105. (Cited 1n Young et al.. 1982)
-35-
-------
Cantor, A.M., M.L. Langevln, T. Noguchi and M.L. Scott. 1975b. Efficiency
of selenium In selenium compounds and feedstuffs for prevention of pan-
creatic flbrosls 1n chicks. J. Nutr. 105: 106-111. (Cited In Young et
al., 1982)
Carlson, C.W., et al. 1951. Some effects of selenium, arsenlcals and
vitamin B12 on chick growth. Poultry Sd. 33: 768. (Cited In U.S. EPA,
1980a)
Clayton, C.C. and C.A. Baumann. 1949. Diet and azo diet tumors; effect of
diet during a period when the dye 1s not fed. Cancer Res. 9: 575. (Cited
1n U.S. EPA, 1980a)
Clinton, M., Jr.. 1947. Selenium fume exposure. J. Ind. Hyg. Toxlcol.
29: 225.
Code of Federal Regulations. 1981. OSHA Safety and Health Standards. 29
CFR 1910.1000.
Combs, G.F., Jr. and M.L. Scott. 1975. Polychlorlnated biphenyl-stlmulated
selenium deficiency 1n the chick. Poultry Sc1. 54: 1152. (Cited In
Levander, 1982)
Dlnkel, et al. 1963. Effects of season of breeding on reproductive and
weaning performance of beef cattle grazing selenlferous range. J. An1m.
Sd. 22: 1043. (Cited in U.S. EPA, 1980a)
-36-
-------
Dlplock, A.T. 1967. Vitamin E and stress 3. The metabolism of 0
tocopherol 1n the rat under dietary stress with silver. Br. J. Nutr. 21:
115-125. (Cited in U.S. EPA, 1980a}
Federal Register. 1984. Environmental Protection Agency. Proposed guide-
lines for carcinogenic risk assessment. 49 FR 46294-46299.
Ferm, V.H. 1972. The teratogenlc effects of metals on mammalian embryos.
Adv. Teratol. 5:51.
FHzhugh, O.G., A.A. Nelson and C.I. Bliss. 1944. The chronic oral toxic-
Hy of selenium. J. Pharmacol. Exp. Therap. 80: 289-299. (Cited in U.S.
EPA, 1983a)
Flohe, L., et al. 1973. Glutathione peroxidase: A selenioenzyme. Fed.
Eur. Biochem. Soc. Lett. 32: 132. (Cited in U.S. EPA, 1980a)
Foklna, A.M. and T.P. KudryavtseMa. 1969. Mikroelem med. Zhivotnovod, M.
Kruming, Ed. USSR:Elm Baku. (CHed In U.S. EPA, 1980a)
Franke, K.W. and V.R. Potter. 1935. New toxicant occurring naturally 1n
certain samples of plant foodstuffs; toxic effects of orally injected
selenium. J. Nutr. 10: 213. (Cited in U.S. EPA, 1980a)
Franke, K.W. and W.E. Tully. 1935. A new toxicant occurring naturally in
certain samples of plant foodstuffs. V. Low hatchability due to deformities
in chicks. Poultry Sci. 14: 273. (Cited in U.S. EPA, 1980a)
-37-
-------
Franke, K.W., et al. 1936. Monstrosities produced by Injection of selenium
salts into hens' eggs. Ant. Rec. 65: 15. (Cited 1n U.S. EPA, 1980a)
Frost, O.V. 1971. The case for selenlte as a feed additive. Feedstuffs.
43: 12. (Cited In U.S. EPA, 1980a)
Frost, D.V. and P.M. Llsh. 1975. Selenium in biology. Ann. Rev.
Pharmacol. 15: 259.
Gabrielsen, 8.0. and J. Opstvedt. 1980. Availability of selenium in fish
meal 1n comparison with soybean meal, corn gluten meal and selenomethlonine
relative to selenium In sodium selenlte for restoring glutathlone peroxidase
activity in selenium-depleted chicks. J. Nutr. 110: 1096-1100. (Cited 1n
Young et al., 1982)
Ganther, H.E. and C.A. Baumann. 1962. Selenium metabolism. I. Effects of
diet, arsenic, and cadmium. J. Nutr. 77: 210. (Cited in U.S. EPA, 1980a)
Glover, J.R. 1967. Selenium 1n human urine: A tentative maximum allowable
concentration for Industrial and rural populations. Ann. Occup, Hyg. 10:
3-10. (Cited in U.S. EPA, 1983a)
Glover, J.R. 1970. Selenium and its industrial toxocology. Ind. Med.
Surg. 39:50.
-38-
-------
Grasso, P., R. Abraham, R. Hendy, A.T. Dlplock, L. Golberg and J. Green.
1969. The role of dietary silver 1n the production of liver necrosis In
vitamin E-defldent rats. Exp. Mol. Pathol. 11: 186-199. (Cited In U.S.
EPA, 1980a)
Gruenwald, P. 1958. Malformation caused by necrosis 1n the embryo Illus-
trated by the effects of selenium compounds on chick embryo. Am. J. Pathol.
34: 77. (Cited 1n U.S. EPA, 1980a)
Gunn, S.A., T.C. Gould and W.A.D. Anderson. 1968. Mechanisms of zinc,
cystelne and selenium protection against cadmium-Induced vascular Injury to
mouse testls. J. Reprod. Pert. 15: 65-70. (Cited 1n U.S. EPA, 1980a)
Hadjlmarkos, D.M. 1965. Effect of selenium on dental carles. Arch.
Environ. Health. 10: 893-899. (Cited 1n U.S. EPA, 1980a)
Hadjlmarkos, D.M. 1969. Selenium: A carles enhancing trace element.
Carles Res. 3: 14. (Cited 1n U.S. EPA, 1980a)
Hadjlmarkos, D.M. and C.W. Bonhorst. 1961. The selenium content of eggs,
milk, and water In relation to dental carles In children. J. Pedlatr. 59:
256. (Cited 1n U.S. EPA, 1980a)
Halverson, A.M. and K.J. Monty. 1960. An effect of dietary sulfate on
selenium poisoning 1n the rat. J. Nutr. 70: 100. (Cited In U.S. EPA,
1980a)
-39-
-------
Halverson, A.W., et al. 1966. Tox1c1ty of selenium to postweanllng rats.
Toxlcol. Appl. Pharmacol. 9: 477. (Cited 1n U.S. EPA, 1980a)
Hamilton, A. 1974. Industrial Toxicology. Harper, NY.
Harr, J.R. 1978. Biological Effects of Selenium, in: Toxlclty of Heavy
Metals 1n the Environment, F.W. Oehme, Ed. Dekker, New York. p. 393.
Harr, J.R. and O.H. Muth. 1972. Selenium poisoning In domestic animals and
Us relationship to man. Clln. Toxlcol. 5: 175-186. (Cited In U.S. EPA,
1980a)
Harr, J.R. J.F. Bone, I.J. Tlnsley, P.H. Weswlg and R.S. Yamamoto. 1967.
Selenium toxldty 1n rats. II. Hlstopathology. In: Selenium 1n B1omed1-
dne, O.H. Muth, Ed. Proc. 1st Intl. Symp., Oregon State University, 1966.
AVI Publishing Co., Westport, CT. p. 153-178. (Cited 1n U.S. EPA, 1980a)
Harr, J.R., et al. 1972. Effect of dietary selenium on N-2-fluorenyl-
acetamlde (FAA)-lnduced cancer 1n vitamin E supplemented, selenium depleted
rats. Clln. Toxlcol. 5: 187. (Cited In U.S. EPA, 1980a)
Hollo, M.Z. and Z. Sztojcso. 1960. Prevention of death by means of
selenium salts administered before roentgen-irradlatlon. Borgyogy Vener
Szenle. 36: 203. (Cited 1n U.S. EPA, 1980a)
-40-
-------
Holmberg, R.E., Jr. and V.H. Perm. 1969. Interrelationships of selenium,
cadmium, and arsenic in mammalian teratogenesis. Arch. Environ. Health.
18: 873-877. (Cited in U.S. EPA, 1980a)
IARC (International Agency for Research on Cancer). 1975. Evaluation of
Carcinogenic Risk of Chemicals to Man. Monograph No. 9. Lyon, France.
(Cited in U.S. EPA, 1980a)
Kar, A.B., et al. 1959. Ovarian changes in prepubertal rats after treat-
ment with cadmium chloride. Acta Biol. Med. Ger. 3. (Cited in U.S. EPA,
1980a)
Kar, A.B., et al. 1960. Prevention of cadmium-induced changes in the
gonads of the rat by zinc and selenium: A study in the antagonism between
metals in the biological system. Proc. Natl. Inst. Sci. India Part B,
Suppl. 26. (Cited in U.S. EPA, 1980a)
Kraus, R.J. and H.E. Ganther. 1980. Reaction of cyanide with glutathione
peroxidase. Biochem. Biophys. Res. Commun. 96: 1116. (Cited in Levander,
1982}
Lantzy, R.J. and F.T. MacKenzie. 1979. Atmospheric trace metals: Global
cycles and assessment of man's impact. Geochim. Cosmochim. Acta. 43:
511-525.
Levander, O.A. 1982. Selenium: Biochemical actions, Interactions, and some
human health implications. in: Clinical, Biochemical and Nutritional
Aspects of Trace Elements. Alan R. Liss, Inc., NY. p. 345-368.
-41-
-------
Levander, O.A. and L.C. Argrett. 1969. Effects of arsenic, mercury, thal-
lium, and lead on selenium metabolism in rats. Toxlcol. Appl. Pharmacol.
14: 308-314. (Cited In U.S. EPA, 1980a)
Levander, O.A. and C.A. Baumann. 1966. Selenium metabolism. VI. Effect of
arsenic on the excretion of selenium 1n the bile. Toxlcol. Appl. Pharmacol.
9: 106. (Cited 1n U.S. EPA, 1980a)
Ludwlg, T.G. and B.C. Biddy. 1969. Geographic variations In the prevalence
of dental carles 1n the United States of America. Carles Res. 3: 32.
(CUed 1n U.S. EPA, 1980a)
Mason, K.E. and J.O. Young. 1967. Effectiveness of selenium and zinc In
protecting against cadmium-Induced Injury of the rat tests. In: Selenium 1n
Blomedldne, O.H. Muth, Ed. Proc. 1st Intl. Symp., Oregon State University,
1966. AVI Publishing Co., Westport, CT. p. 383. (Cited 1n U.S. EPA, 1980a)
Moxon, A.L. 1938. The effect of arsenic on the toxlclty of selenlferous
grains. Science. 88: 81. (Cited 1n U.S. EPA, 1980a)
Moxon, A.L. and K.P. Dubols. 1939. Influence of arsenic and certain other
eleme'nts on the toxldty of selenlferous grains. J. Nutr. 18: 447. (Cited
1n U.S. EPA, 1980a)
Muhleman, N. and K. Konlg. 1964. Carles activity of male and female
Osborne-Mendel rats in different environments. Helv. Odontol. Acta. 8: 79.
(Cited in U.S. EPA, 1980a)
-42-
-------
NAS (National Academy of Sciences). 1976. Selenium. Comm. Med. Blol.
Effects, Environ. Pollut. Subcomm. Selenium, Washington, DC. (Cited In U.S.
EPA, 1980a)
NAS (National Academy of Sciences, Food and Nutrition Board). 1980. Recom-
mended dietary allowances, 9th rev. ed. NRC, NAS, Washington, DC. p. 162.
(Cited 1n U.S. EPA, 1980a)
Natusch, D.F.S., J.R. Wallace and C.A. Evans, Jr. 1974. Toxic trace
elements: Preferential concentration resplrable particles. Science. 183:
202-204.
NCI (National Cancer Institute). 1978. NCI carcinogenesls bloassay experi-
mental design status report for selenium sulflde (preliminary data). NCI
Contract No. 702209. (Cited in U.S. EPA, 1980a)
Nelson, A.A., et al. 1943. Liver tumors following cirrhosis caused by
selenium In rats. Cancer Res. 3: 230. (Cited In U.S. EPA, 1980a)
Page, G.W. 1981. Comparison of groundwater and surface water for patterns
and levels of contamination by toxic substances. Environ. Sci. Technol.
15: 1475-1481.
Palmer, I.S., R.C. Arnold, C.W. Carlson. 1973. Toxlcity of various
selenium derivatives to chick embryos. Poult. Sci. 52: 1841.
-43-
-------
Palmer, I.S. and O.E. Olson. 1979. Partial prevention by cyanide of sele-
nium poisoning In rats. Blochem. Blophys. Res. Commun. 90: 1379. (CHed
1n Levander, 1982)
Palmer, I.S., O.E. Olson, A.W. Halverson, R. Miller and C. Smith. 1980.
Isolation of factors 1n linseed oil meal protective against chronic seleno-
s1s 1n rats. 3. Nutr. 110: 145. (CHed In Levander, 1982)
PaMzek, J. 1964. Vascular changes at sites of oestrogen biosynthesis
produced by parenteral Injection of cadmium salts: The destruction of the
placenta by cadmium salts. J. Reprod. Fert. 7: 263-265. (Cited In U.S.
EPA, 1980a)
PaMzek, J., I. Ostadalova, I. Benes and A. Bablcky. 1968. Pregnancy and
trace elements: The protective effect of compounds of an essential trace
element - selenium - against the peculiar toxic effects of cadmium during
pregnancy. J. Reprod. Fert. 16: 507-509. (CHed In U.S. EPA, 1980a)
PaMzek, J., et al. 1971. The detoxifying effects of selenium. Inter-
relations between compounds of selenium and certain metals. In.: Newer Trace
Elements In Nutrition, W. Mertz and W.E. Cornatzer, Ed. Marcel Dekker, NY.
p. 85.
Paton, 6.R. and A.C. Allison. 1972. Chromosome damage In human cell
cultures Induced by metal salts. Mutat. Res. 16: 332-336. (Cited In U.S.
EPA, 1980a)
-44-
-------
Robertson, D.S.F. 1970. Selenium, a possible teratogen? Lancet. 1:
518-519. (Cited 1n U.S. EPA, 1980a)
Rosenfeld, I. and O.A. Beath. 1964. Selenium. Geobotany, Biochemistry,
Toxldty, and Nutrition. Academic Press, NY. (Cited 1n U.S. EPA, 1980a)
Rudert, C.P. and A.R. Lewis. 1978. The effect of potassium cyanide on the
occurrence of nutritional myopathy 1n lambs. Rhod. J. Agrlc. Res. 16: 109.
(Cited 1n Levander, 1982)
Rusleckl, W. and J. Brzezlnskl. 1966. Influence of sodium selenate on
acute thallium poisonings. Acta. Pol. Pharmacol. 23: 69. (Cited 1n U.S.
EPA, 1980a)
Schroeder, H.A. 1967. Effects of selenate, selenlte and tellurlte on the
growth and early survival of mice and rats. J. Nutr. 92: 334-338. (Cited
In U.S. EPA, 1980a)
Schroeder, H.A. and M. MHchener. 1971a. Selenium and tellurium 1n rats:
Effect on growth, survival, and tumors. J. Nutr. 101: 1531. (Cited 1n
U.S. EPA, 1980a)
Schroeder, H.A., M. MHchener. 1971b. Toxic effects of trace elements on
the reproduction of rats and mice. Arch. Environ. Health. 23: 102.
Schroeder, H.A. and M. Mltchener. 1972. Selenium and tellurium 1n mice:
Effects on growth, survival, and tumors. Arch. Environ. Health. 24: 66.
(Cited In U.S. EPA, 1980a)
-45-
-------
Schroeder, H.A., O.V. Frost, J.J. Balassa. 1970. Essential trace metals In
man: Selenium. J. Chron. D1s. 23: 227.
Scott, M.L., G. Olson, L. Krook and W.R. Brown. 1967. Selenium-responsive
myopathles of myocardium and of smooth muscle 1n the young poult. J. Nutr.
91: 573. (Cited 1n Levander, 1982)
Senteln, P. 1967. Action de derives du selenium (selenlates et selenHes)
comparee a celle d'addes divers (crotonlque, acetlque, dtrlque et phos-
phorlque) sur les mitoses de segmentation. Chromasoma (Berl.) 23: 95-136.
(Cited 1n U.S. EPA, 1980a).
Shamberger, R.J., O.V. Frost. 1969. Possible protective effect of selenium
against human cancer. Can. Med. Assoc. J. 100: 682.
Shamberger, R.J. and 6. Rudolph. 1966. Protection against cocardnogenesls
by antloxldants. Experlentla. 22: 116.
Shamberger, R.J. and C.E. Willis. 1971. Selenium distribution and human
cancer mortality. Crit. Rev. Cl1n. Lab. Sci. 2: 211. (Cited in U.S. EPA,
1980a)
Shamberger, R.J., et al. 1972. Antloxldants 1n cereals and in food pre-
servatives and declining gastric cancer mortality. Cleveland Clin. 39:
119. (Cited in U.S. EPA, 1980a)
-46-
-------
Shamberger, R.J., et al. 1973. Antloxldants and cancer. I. Selenium 1n
the blood of normal and cancer patients. J. Natl. Cancer Inst. 50: 863.
(Cited 1n U.S. EPA, 1980a)
Shapiro, J.R. 1972. Selenium and cardnogenesls: A review. Ann. N.Y.
Acad. Sc1. 192: 215.
Shapiro, O.R. 1973. Selenium compounds 1n nature and medicine: Selenium
and human biology. In: Organic Selenium Compounds: Their Chemistry and
Biology, D.L. Klayman and W.H.H. Guenther, Ed. John Wiley and Sons, Inc.,
NY. (Cited 1n U.S. EPA, 1980a)
Smith, M.I. 1939. Public Health Report 54, 1441. (Cited 1n Underwood,
1977; U.S. EPA, 1983a)
Smith, M.I., et al. 1936. The selenium problem 1n relation to public
health. A preliminary survey to determine the possibility of selenium
Intoxication 1n the rural population living on selenlferous soil. U.S. Pub.
Health Rep. 51: 1496. (Cited In U.S. EPA, 1980a)
Tank, G. and C.A. Storvlck. 1960. Title not provided. J. Dent. Res. 39:
473. {Cited 1n Glover, 1967; U.S. EPA, 1983a)
Thomson, C.D. and R.D.H. Stewart. 1973. Metabolic studies of (75Se)
selenomethlonlne and (7SSe) selenlte In the rat. Br. J. Nutr. 30: 139.
(Cited In U.S. EPA, 1980a)
-47-
-------
Thomson, C.D. and R.D.H. Stewart. 1974. The metabolism of (75Se)
selenlte In young women. Br. J. Nutr. 32: 47. (Cited In U.S. EPA, 1980a)
Thomson, C.D., et al. 1975. Metabolic studies of (75Se) selenocystlne
and (7SSe) selenomethlonlne 1n the rat. Br. J. Nutr. 34: 501. (Cited In
U.S. EPA, 1980a)
Ting, K.P. and G.W.R. Walker. 1969. The distributive effect of selenoamlno
add treatment on crossing-over In Drosophlla melanogaster. Genetics. 61:
141-151. (Cited In U.S. EPA, 1980a)
Tlnsley, I.J., et al. 1967. Selenium toxlclty 1n rats. I. Growth and
longevity. In.: Selenium 1n B1omed1c1ne, O.H. Muth, Ed. Proc.. 1st Intl.
Symp., Oregon State University, 1966. AVI Publishing Co., Westport, CT.
p. 141. (Cited 1n U.S. EPA, 1980a)
Underwood, E.O. 1977. Trace Elements 1n Human and Animal Nutrition, 4th
ed. Academic Press, New York. p. 302-346.
U.S. EPA. 1980a. Ambient Water Quality Criteria for Selenium. Environ-
mental Criteria and Assessment Office, Cincinnati, OH. EPA 440/5-80-070.
NTIS PB 81-117814.
U.S. EPA. 1980b. Guidelines and Methodology Used In the Preparation of
Health Effects Assessment Chapters of the Consent Decree Water Quality
Criteria. Federal Register. 45: 79347-79357.
-48-
-------
U.S. EPA. 1983a. Reportable Quantity for Selenium (and Compounds). Pre-
pared by the Environmental Criteria and Assessment Office, Cincinnati, OH,
OHEA for the Office of Solid Waste and Emergency Response, Washington, DC.
U.S. EPA. 19835. Methodology and Guidelines for Reportable Quantity Deter-
minations Based on Chronic Toxldty Data. Prepared by the Environmental
Criteria and Assessment Office, Cincinnati, OH, OHEA for the Office of Solid
Waste and Emergency Response, Washington, DC.
U.S. EPA. 1985. Acceptable Dally Intakes for Volatile Organic Chemicals.
Inorganic Chemicals and Synthetic Organic Chemicals. Criteria and Standards
Division, Office of Drinking Water. (Federal Register draft)
Vesce, C.A. 1974. Intosslcazlone spermentale da selenlo. Intoss1caz1one
Sperlmentale da Selenlo, Folia Med. (Ital.) 33: 209. (Cited 1n U.S. EPA,
1980a)
Volgarev, M.N., L.A. Tscherkes. 1967. Further studies 1n tissue changes
associated with sodium selenate. in: Selenium In B1omed1c1ne, O.H. Muth,
Ed. Proc. 1st Intl. Symp., Oregon State University, 1966. AVI Publishing
Co., Westport, CT. p. 179-184. (Cited 1n Harr, 1978)
Wahlstrom, R.C. and O.E. Olson. 1958. The relation of prenatal and pre-
weaning treatment to the effect of arsanlUc acid on Se poisoning. J. Anim.
Sci. 18: 578. (Cited in U.S. EPA, 1980a)
-49-
-------
Walker, G.W.R. and A.M. Bradley. 1969. Interacting effects of sodium mono-
hydrogenarsenate and selenocystlne on crossing over 1n Drosophila mela.no-
gaster. Can. J. Genet. Cytol. 11: 677-688. (Cited In U.S. EPA, 1980a)
Weast, R.C. 1980. CRC Handbook of Chemistry and Physics, 61st ed. CRC
Press, Boca Raton, FL.
Wedderburn, J.F. 1972. Selenium and cancer. New Zealand Vet. J. 20: 56.
(Cited 1n U.S. EPA, 1980a)
Weisberger, A.S. and L.G. Suhrlaud. 1956a. Studies on anologles of
L-cyste1ne and L-cyst1ne. II. The effect of selenium cystlne on Murphy
lymphosarcoma tumor cells 1n the rat. Blood. 11: 11-18. (CHed 1n
Shapiro, 1972)
Weisberger, A.S. and L.G. Suhrlaud. 1956b. Studies on anologles of
L-cyste1ne and L-cystlne. III. The effect of selenium cystlne on leukemia
In the rat. Blood. 11: 19. (CHed 1n Shapiro, 1972)
Weisberger, A.S., L.G. Suhrlaud, J. Selffer. 1956. Studies on analogies of
L-cyste1ne and L-cystine. I. Some structural requirements for inhibiting
the Incorporation of radioactive L-cyst1ne by leukocytes. Blood. 11: 1-10.
(CHed in Shapiro, 1972)
Yang, G.t W. Shuzhen, R. Zhou, S. Sun. 1983. Endemic selenium Intoxication
of humans 1n China. Am. J. CUn. Nutr. 37: 872-888.
-50-
-------
Young, V.R., A. Nahapetlan and M. Janghorbanl. 1982. Selenium bloavail-
abllity with reference to human nutrition. Am. J. CUn. Nutr. 35(5)
1076-1088.
-51-
-------
APPENDIX
Summary Table for Selenium and Compounds
Species
Experimental
Dose/Exposure
Effect
Acceptable Intake
(AIS or AIC)
Reference
Inhalation
AIS
AIC
human
0.1 mg/m3
none
ND
0.07 mg/day
Glover, 1967
Oral
AIS
AIC
rat
human
Maximum mouse
composite
score
3.2 ppm diet
(0.16 mg/kg/day)
3.2 mg/day from
diet of selenlf-
erous foodstuffs
3 ppm In drinking
water (0.51 mg/kg/
day) for 3 genera-
tions (RV,j-4.9)
none
loss of hair,
nails, derma-
titis nemo-
muscular
dysfunction
Increased neo
natal mortal-
ity (RVe=10)
0.224 mg/day
0.21 mg/day
49
Halverson et al.,
1966
Yang et al.
1983; U.S. EPA,
1905
Schroeder and
MHchener, 1971b;
U.S. EPA, 1983a
ND = Not derived
------- |