21
Chapter 3
Toxicology
ญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญ
crease in PCB concentrations in the carcass but brain
3.1. Toxicokinetics
PCB concentrations increased six-fold and liver concen-
3.1.1. Distribution
trations doubled over the same time period. These re-
The majority of the substances dealt with in this report
sults show a net redistribution of PCBs from the carcass
are lipophilic, stable, and persistent. They are taken up
lipids to the brain and liver during periods of stored
by aquatic living organisms via diffusion over the gills
lipid utilization.
and from food in the gastrointestinal tract. POPs, partic-
Significant differences were seen in PCB concentra-
ularly organohalogen substances, cross the gill/gut mem-
tions with higher concentrations in blubber of molting
brane and enter the blood where they are quickly dis-
(lean) harp seals (Phoca groenlandica) than in pre-wean-
tributed to high lipid tissues such as the liver and adi-
ing (fat) individuals (Kleivane et al., 2003). When the
pose tissue. Metabolism and elimination are often slow,
seals were compared on a total blubber burden basis,
leading to a net increase of these substances in the or-
however, there were no significant differences (i.e., the
ganism over time (bioaccumulation).
same amount of PCB was present in the blubber but was
There are species differences in the tissue distribution
more diluted in the greater blubber mass of the obese
of POPs, partly due to differences in lipid distribution.
seals). In another study on harp seals, POP concentra-
Several examples of this were given in the first AMAP
tions were measured in blood and blubber of seals be-
assessment (de March et al., 1998). Lipid dynamics can
fore and after a four-week fast, and in wild seals sam-
also affect the distribution of POPs. For example, many
pled before the breeding season (fat) and during the molt
Arctic animals go through dramatic periods of fat accu-
(lean) (Lydersen et al., 2002). In the fasting experiment,
mulation followed by long periods of fasting. Lipophilic
POP concentrations in blubber did not change but blood
POPs will be sequestered in the fat. As the fat is utilized
POP levels increased during the fast. In the wild seals,
for energy during fasting, POP concentrations in the re-
POP concentrations in both blubber and blood were
maining fat will increase, driving new equilibria to be
higher in the lean seals than in the fat seals.
established between fat, blood lipids, and the lipids of
Polar bears show seasonal dynamics in OC concen-
other organs, effectively redistributing POPs to other
trations related to fasting (Polischuk, 1999; Norstrom,
compartments in the organism. This implies that differ-
2000). The concentrations of CBz, PCB, and chlor-
ent tissues in different species will be targets for possible
dane (CHL) in fat increased and DDT decreased dur-
effects from POPs.
ing 47- 68 days of fasting and this was entirely due to
For example, lipid dynamics and resultant redistribu-
lipid utilization. For a more thorough presentation of
tion of OCS and several PCBs have been studied in wild
this material, see Section 4.4.7.
anadromous Arctic char. Lipid composition was studied
in descending (May) fish, ascending (mid-July) fish and
3.1.2. Metabolism and elimination
in fish caught in mid-July but held in captivity until late
September. From May to July, lipid stores increased five-
Metabolism of xenobiotics occurs mainly in the liver via
fold with 50% of the lipid content being in the carcass
a two-phase process. These processes are catalyzed by
and 35 -50% in muscle (J๘rgensen et al., 1997a; Jobling
liver enzymes such as the cytochrome P450 containing
et al., 1998). Most of this was triacylglycerols (TAG).
monooxygenases (Nebert and Gonzalez, 1987). Lipo-
Body lipids decreased from mid-July to September in
philic substances that are resistant to metabolism will be
maturing char by 30 - 40% from all lipid depots, but the
selectively accumulated in living organisms. In addition
major mobilization was of TAG from the carcass and
to detoxification, the enzymatic processes can also cre-
muscle depots. For mature females, the ovaries contained
ate reactive intermediates that may be mutagenic and/or
more than 25% of the remaining lipids but for males,
carcinogenic, or metabolites that are stable to further
the testes only contained 3% of the remaining lipids. Fe-
metabolism, that are lipophilic and retained, or that are
males lost approximately 80% of their body lipids dur-
biologically active, with the ability to bind selectively to
ing spawning and overwintering, whereas males only
proteins and accumulate in the organism.
lost 50-55% of body lipids.
Many POPs form metabolites that are biologically
The distribution of OCS was studied in lean and fat
active. DDT is metabolized in living organisms to DDE,
char. A higher proportion of the body burden was
which is lipophilic and toxic, and accumulates in biota
found in extra-adipose organs such as the liver (2 times)
(WHO, 1989a). In some cases, a methyl sulfone (MeSO2)
and brain (4 times) of lean char than of fat char (J๘r-
group is added during metabolism and a number of
gensen et al., 1997b). Similar results were obtained for
MeSO2-DDE and MeSO2-PCB congeners have been
PCB distribution in a study where wild anadromous
identified in animals including several Arctic species
char were captured when ascending and treated with
(Jensen and Jansson, 1976; Lund et al., 1988; Haraguchi
PCBs in September (J๘rgensen et al. 2002a). The fish
et al., 1990; Bergman et al., 1992b;1994b; Brandt et al.,
were not fed during the winter and PCB and lipid analy-
1992; Haraguchi et al., 1992; Letcher et al., 1995a;
ses were done at three time periods during the winter
1995b; 1998; 2000b). MeSO2-DDE has a high binding
and spring. From October to May, there was a 20% de-
affinity for the adrenal cortex and is highly toxic to this

22
AMAP Assessment 2002: Persistent Organic Pollutants in the Arctic
tissue in mice (Lund et al., 1988; J๖nsson et al., 1991;
via their eggs, and female mammals via placental trans-
1992; Brandt et al., 1992; Lindhe et al., 2001). This
fer to the fetus and in breast milk. A particular charac-
DDT metabolite is suspected of being one possible cause
teristic of Arctic marine mammals is that most have very
of adrenal hyperplasia seen in Baltic Sea grey seals (Hali-
high fat contents in breast milk in order to facilitate
choerus grypus) (Jensen and Jansson, 1976; Bergman
rapid growth in the young during the short growing sea-
and Olsson, 1985). MeSO2-PCBs may also play a role in
son. For example, polar bear milk has a fat content of
hyperadrenocorticism in Baltic Sea seals (Bergman et al.,
20 - 46% (Derocher et al., 1993; Oehme et al., 1995a;
1992a; Mortensen et al., 1992).
Polischuk et al., 1995; Bernhoft et al., 1997) and various
Previously, research has shown that several PCB
seal species have milk fat contents of 30 - 60% (Addison
congeners also form hydroxylated metabolites (Jansson
and Brodie, 1977; 1987; Bacon et al., 1992; Pomeroy et
et al., 1975). This type of metabolite has been found to
al., 1996; Beckmen et al., 1999). Therefore, excretion of
bind selectively to transthyretin (TTR), one of the major
POPs via milk is more important than placental transfer
transport proteins for thyroid hormones in the blood
for adult females of marine mammal species. This in
(Brouwer et al., 1988; 1990; 1998; Bergman et al., 1994a;
turn enhances POP exposure of young, particularly for
Letcher et al., 2000a). TTR is complexed to retinol-
polar bears, Arctic foxes (Alopex lagopus), whales, and
binding protein (RBP), which transports vitamin A
seals.
(retinol). Other POPs have also been found to form hy-
For example, in Alaska, northern fur seals (Callorhi-
droxylated metabolites, and are suspected of being bio-
nus ursinus), pups have significantly higher blood con-
logically active in a manner similar to the hydroxylated
centrations of several POPs when compared to their
PCBs. For example, studies have found that some PBDE
dam's blood and milk (Beckman et al., 1999). The pups
congeners form hydroxylated metabolites (Klasson Weh-
of younger dams (primaparous) have much higher con-
ler et al., 1996; ึrn, 1997; ึrn and Klasson Wehler,
centrations than those of older dams (multiparous).
1998; Meerts et al., 2000; Klasson Wehler et al., 2001;
Young harp and hooded seals (Cystophora cristata) have
M๖rck et al., 2003; Hakk et al., 2002). 4-Hydroxy-
as high levels of some POPs as their mothers at the end
heptachlorostyrene has been identified as a metabolite of
of the lactation period (Espeland et al., 1997). In Steller
OCS in polar bear and ringed seal plasma and has been
sea lions (Eumetopias jubatus), 80% of the POP burden
found to bind to human TTR in vitro (Sandau et al.,
may be transferred from a female to her first offspring
2000). Pentachlorophenol, a metabolite of HCB, also
via lactation (Lee et al., 1996). Young polar bears (1- 2
binds to TTR (van den Berg, 1990). Toxaphene con-
years) have PCB levels similar to adult females with high
geners Parlar 32 and 62 are metabolized to hydroxy-
PCB levels (Bernhoft et al., 1997) and polar bear cubs-
lated metabolites by seal liver microsomes (van Hezik et
of-the-year have higher concentrations of many POPs
al., 2001).
than their mothers (Polischuk et al., 1995). This is of
TBT is metabolized by the cytochrome P450 system
concern, as young animals may be more sensitive to the
in mammals and fish to DBT and MBT, both of which
effects of POPs than adults.
are biologically active (Fish et al., 1976; Kimmel et al.,
Polischuk et al. (2002) also found clear evidence of
1977; Lee, 1991; Martin et al., 1989; Fent and Stege-
sex-specific metabolism in the polar bear. Adult male
man, 1993).
polar bears continued to lose body burdens of chlordane
PAHs are metabolized by the cytochrome P450 sys-
compounds during a three-month fast, indicating metab-
tem to reactive intermediates that covalently bind to
olism, while non-lactating females did not lose any of
macromolecules such as DNA and proteins, creating
their body burden. On the other hand, lactating females
adducts.
lost PCBs faster than males. This results in higher PCB
Chlorinated paraffins (CPs) have been shown to be
levels, but lower chlordane levels, in male than female
biotransformed in fish (Fisk et al., 2000), birds (Darne-
polar bears.
rud and Brandt, 1982), and mammals (Darnerud, 1984),
A special case is excretion of TBT, which distributes
with the susceptibility decreasing with increasing carbon
not only to internal organs but also to bird feathers and
chain length and chlorine content (Tomy et al., 1998).
seal fur (Tanabe, 1999). The yearly molts of birds and
It would appear that there are a number of metabolic
seals may be an important excretion pathway for bu-
pathways that can degrade CPs, but chlorine content
tyltins. Guruge et al. (1996) estimated that up to 25%
and carbon chain length can influence which pathway is
of the body burden could be excreted in cormorants
utilized. There is little information, however, on the en-
during a complete molting cycle. Comparable butyltin
zymes involved in the degradation. Although older re-
concentrations are seen in male and female marine
ports suggested that CPs induce phase I (mixed-func-
mammals, indicating that these are not transferred
tion-oxygenase enyzmes, e.g., CYP450) and phase II en-
from mother to fetus/pup to the same degree as other
zymes (conjugation reactions, e.g., mercapturic acid syn-
POPs (Tanabe, 1999).
thesis) (Haux et al., 1982), more recent work has failed
The net result of uptake, distribution, metabolism,
to find CYP1A induction in fish despite high CP expo-
and excretion will determine the POP levels found in
sures (Fisk et al., 1996).
an organism. This is in turn affected by other factors.
The major excretion route of POPs and their me-
Studies carried out to determine the uptake, distribu-
tabolites is via feces and to some extent, urine. Some of
tion, metabolism, and excretion of POPs usually inves-
this is passive diffusion over the gut membrane and
tigate one substance at a time. Wildlife and humans,
some from bile excretion of metabolites. In invertebrates
however, are exposed to complex mixtures of POPs.
and fish, excretion of low log Kow compounds may also
Very little is known about how different POPs affect
occur by diffusion through the gill membranes. Female
each other's toxicokinetics. POPs that induce the hepatic
fish and birds excrete lipophilic, organohalogen POPs
cytochrome P450 system will affect the metabolism of

Chapter 3 ท Toxicology
23
other xenobiotics, for example. This may lead to an in-
and gap junction intercellular communication. Almost
crease in xenobiotic metabolism, thus increasing excre-
all POPs considered in this report also cause visible
tion. An increase in xenobiotic metabolism may also
changes in the liver, including hypertrophy, lesions, and
lead to an increase in the formation of reactive interme-
in some cases, tumors.
diates, with increased toxicity and tissue damage (Boon
POPs can cause short-term acute effects when ad-
et al., 1992). An example of this is the bioactivation of
ministered in high doses as well as long-term chronic ef-
benzo[a]pyrene (B[a]P) and 7,12-dimethylbenz[a]anthra-
fects at lower doses. In the Arctic, the major concern is
cene (DMBA) in animals after exposure to the CYP1A
long-term chronic exposures as organisms are exposed
inducer, CB126. Bioactivated B[a]P and DMBA metabo-
to POPs over their entire lifetimes. In this context, the
lites were found to bind irreversibly to endothelial cells
major effects of concern are those that may affect repro-
in certain arteries, veins and capillaries of mice, rats and
duction and survival at the individual and population
chicken embryos (Granberg, 2001; Granberg et al.,
level. Effects at the individual, population or ecosystem
2000; Annas et al., 1999; 2000).
level, however, come at a late stage of exposure. It would
TBT exposure has been shown to lead to the degra-
be more useful to have earlier warning of exposure to
dation of cytochrome P450 proteins (CYP1A, CYP2B
POPs and, therefore, biological marker systems based on
and CYP3A) and concomitant inhibition of enzyme ac-
subtle, low-dose effects are being used or developed.
tivities in the liver. This in turn would inhibit the metab-
Most biological markers measure effects at the molecu-
olism of organohalogen compounds if an organism was
lar, cellular or organ level; however, it is still not estab-
exposed to both substance groups simultaneously, lead-
lished what these changes might mean at the individual
ing to higher accumulation of organohalogens, and pos-
or population level.
sible risk of toxic effects.
Thus, it is very difficult to evaluate the toxicokinetics
3.2.1. Reproduction and development
of environmental exposures to mixtures of POPs. The
interactions that have been seen indicate that the relative
POPs have a number of effects on the ability of organ-
amounts and the composition of various contaminants
isms to reproduce and develop normally (for reviews,
in animals may partly be the result of selective effects on
see Peterson et al., 1993; Bosveld and van den Berg,
the organism's uptake, metabolism, and excretion, and
1994; Barron et al., 1995; Brouwer et al., 1995). Expo-
not solely a result of the specific pollution burden of any
sure to some POPs may cause embryo- and fetotoxicity,
single contaminant in the area.
decreased offspring survival, abnormalities in the estrus
cycle and sex hormone levels, reduced sperm produc-
tion, reduced litter sizes, and even total reproductive
3.2. Types of effects
failure in mammals. In birds, some POPs cause de-
In most laboratory experiments studying the toxicolog-
creased egg production, retarded egg production, in-
ical effects of POPs, animals are exposed to single sub-
creased embryo mortality, eggshell thinning, embryonic
stances or to technical products, often at acutely toxic
deformities, growth retardation, and reduced egg hatch-
doses. In a few studies, combinations of a few sub-
ability, as well as detrimental effects on parental behav-
stances have been used. It should be remembered that
ior. In fish, some POPs cause decreased egg and larval
wildlife are exposed to complex mixtures of POPs,
survival, reduced sexual maturation, and reduced gonad
most often at low doses, and these mixtures may not
size. Other effects of POPs on organisms may include
resemble the original technical products released into
structural malformations, neurotoxic effects, and neuro-
the environment because of weathering processes.
logical and behavioral changes in offspring. Behavioral
There are often considerable species differences in sen-
changes also occur in adult animals, including changes
sitivity to specific POPs as well as differences in re-
in mating behavior.
sponse. It is, therefore, often difficult to generalize re-
Some POPs act as hormones or interfere with en-
sults found in one species to other species. This is par-
docrine systems and are therefore called endocrine or
ticularly difficult when extrapolating effects seen in
hormone disruptors (for reviews, see Vos et al., 2000;
controlled studies in laboratory species such as rodents
Damstra et al., 2002). The reproductive effects of em-
to effects seen in marine mammals in the wild. Other
bryonic or fetal exposure to these disruptive com-
factors such as fat dynamics, delayed implantation, dif-
pounds may only become obvious at later developmen-
ferences in physiology, and toxicokinetics may make
tal stages or at sexual maturity. The estrogenic and
wildlife more or less sensitive to the effects of POPs.
antiestrogenic effects of POPs are the best studied of
The following is a short, general summary of results
these effects. Endocrine disruption is also implicated in
from laboratory studies as well as results from field
thyroid and immune system effects, which are treated
studies where POPs have been quantified and concen-
in Sections 3.2.3 and 3.2.4. POPs may also function
trations have been correlated with biological and toxi-
as androgens or antiandrogens. Estrogens and andro-
cological effects.
gens are important in the normal sexual differentiation
A wide range of effects is seen after exposure to
of developing organisms. A number of biomarkers
POPs. Some of these types of effects are currently being
have now been developed for testing the estrogenicity
used as biological markers for POP exposure. These in-
of POPs.
clude, among other things, effects on reproduction, de-
Studies have shown that there is a critical phase in
velopment (including the brain), cytochrome P450-de-
neonatal mouse brain development when the brain is
pendent enzymes, the immune system, the adrenals, the
particularly susceptible to effects of low-dose exposure
thyroid gland, thyroid hormone levels, vitamin A levels,
to toxic substances such as PCB, DDT, pyrethroids,
formation of DNA adducts, peroxisome proliferation,
organophosphates, paraquat, and nicotine (Eriksson,

24
AMAP Assessment 2002: Persistent Organic Pollutants in the Arctic
1997). This critical phase is known as the `brain growth
that have an activity profile very similar to CYP2B in
spurt' (BGS) and disruption leads to persistent disrup-
mammals, especially as judged by the ability to metabo-
tion in adult brain function. The BGS occurs at different
lize specific PCB structures (Norstrom, 1988). This en-
time points in different mammalian species (Davison
zyme may be in the CYP2 family, but there is no im-
and Dobbing, 1968). In rats and mice, it occurs in the
munochemically recognizable CYP2B in birds. Sub-
first 3- 4 weeks of life (neonatal period) whereas in hu-
stances that induce CYP2B are DDT, chlordane, aldrin,
mans, it occurs during the third trimester of pregnancy
endrin, di- to tetra-ortho PCBs, and 3-MeSO2-PCB.
and throughout the first two years of life. Some POPs,
Mono-ortho PCB congeners and technical PCBs are
most notably the non-dioxin-like PCBs, may act as neu-
mixed type inducers inducing both CYP1A and 2B.
rotoxins by decreasing dopamine content in the brain
There is strong correlative evidence that CYP2B is in-
and altering calcium homeostasis (for a review, see Til-
duced by a combination of PCBs and chlordanes in po-
son and Kodavanti, 1998).
lar bears (Letcher et al., 1996).
CYP3A forms are among the most versatile iso-en-
zymes with low substrate specificity. Substances that in-
3.2.2. Cytochrome P450 system
duce CYP3A include carcinogens, pesticides, some
and other xenobiotic metabolizing
drugs, and steroid hormones such as testosterone. Induc-
enzyme systems
tion is often measured as the formation of hydroxylated
The most developed of the biological markers is the
metabolites (e.g., 6-
hydroxylation) of testosterone.
study of cytochrome P450-dependent liver enzymes
Recently, CYP3A has been found to be involved in the
(e.g., F๖rlin et al., 1994; Jensen and Hahn, 2001; Kim
metabolism of several toxaphene congeners in harbour
and Hahn, 2002). Exposure to OCs and some PAHs in-
(Phoca vitulina) and grey seals (van Hezik et al., 2001)
duces liver cytochrome P450-dependent enzymes (CYP)
and implicated in toxaphene metabolism in harp and
known as mixed function oxidases (MFO), which me-
ringed seals (Wolkers et al., 1998b; 2000). A CYP3A
tabolize xenobiotics and endogenous substances (Ne-
form that is immunologically cross-reactive with anti-rat
bert and Gonzalez, 1987). Exposure to high concentra-
CYP3A1 was induced in chickens by PB (Ourlin et al.,
tions of MFO-inducing POPs can affect the metabolism
2000). CYP3A forms may be involved in the metabolism
of endogenous substrates, such as steroid hormones,
of some POPs in birds.
leading to disturbances in critical biological functions
In the previous AMAP assessment report, de March
(Kupfer and Bulger, 1976). Other methods that can
et al. (1998) reviewed species differences in the cyto-
be used to study CYP forms include measuring CYP
chrome P450 system, particularly with reference to the
messenger ribonucleic acid (mRNA), DNA, and pro-
Arctic situation. At that time, the P450 system had been
tein levels. Protein levels are determined using elec-
characterized in a number of Arctic species including
trophoresis combined with immunoblotting, which re-
harp and hooded seals, polar bears, harbour porpoise
quires the availability of antibodies for each CYP form
(Phocoena phocoena), beluga (Delphinapterus leucas),
to be identified.
short-finned pilot whale (Globicephala macrorhynchus),
There are several gene families of cytochrome P450
and minke whale (Balaenoptera acutorostrata). For ter-
in vertebrates (Nelson et al., 1993; 1996) and those
restrial mammals, they generally have two CYP1A iso-
most relevant for the metabolism of POPs are the CYP
forms known as 1A1 and 1A2 as well as functioning
1A, 2B, and 3A gene families. CYP1A forms are induced
CYP2B forms and thus, have a high capacity for metab-
primarily by planar aromatic hydrocarbons such as PAHs,
olizing both groups of POPs.
PCDD/Fs, non-ortho and some mono-ortho PCBs.
Harp, grey and hooded seals, and harbour porpoise,
Induction of CYP1A is mediated by the aryl hydrocar-
like other mammals, seem to have functional CYP1A1
bon (Ah) receptor, a cytosolic receptor found in all ver-
and 1A2 and thus, a higher capacity to metabolize pla-
tebrates studied so far, and for which the most potent
nar compounds, but have weak CYP2B activity and
ligand is 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD).
thus, a reduced ability to metabolize xenobiotics that are
The activated ligand-receptor complex triggers the ge-
substrates for these particular MFOs (Boon et al., 1992;
netic expression and production of a number of proteins
Goks๘yr et al., 1992; Goks๘yr, 1995a). Seals seem to have
including CYP1A. Induction is often measured as in-
more CYP2B activity than whales (Goks๘yr, 1995a).
creases in activity of several enzymes including ethoxy-
Polar bears have functional CYP1A1, 2B1, 3A1, and
resorufin-O-deethylase (EROD) and aryl hydrocarbon
epoxide hydrolase (Bandiera et al., 1995), and have a
hydroxylase (AHH) as well as by caffeine demethyl-
high metabolic capacity, particularly for PCB and DDT
ation. The efficacy of CYP1A to metabolize POPs that
(Norstrom and Muir, 1994).
induce these enzymes appears to vary considerably
Previous studies have shown that fish seem to lack
among taxa.
CYP2B and have only one version of CYP1A (Nebert et
CYP2B forms are induced in mammals by another
al., 1989; Stegeman, 1989; Stegeman and Hahn, 1994;
class of substances, typified by phenobarbital (PB) and
Goks๘yr, 1995b), and thus have a low POP metaboliz-
measured in laboratory rodents as aminopyrine N-
ing capacity. However, one exception to this has been
demethylase (APND) activity, aldrin epoxidase (AE),
found recently. Deepwater sculpin (Myxocephalus thomp-
and pentoxyresorufin-O-dealkylase (PROD), for exam-
soni) from the Great Lakes, have been found to metabo-
ple. Considerable caution is needed in interpretation of
lize PCBs and form 3- and 4-MeSO2-PCB metabolites,
these activities in wild mammals. For example, PROD
indicating that they have CYP2B-like activity (Stapleton
activity was shown to be more highly correlated with
et al., 2001). With regard to birds, cytochromes P450
CYP1A than CYP2B induction in polar bears (Letcher
1A1 and 1A2 are found in this group of animals (Living-
et al., 1996). Birds possess one or more CYP enzymes
stone and Stegeman, 1989).

Chapter 3 ท Toxicology
25
Since 1996, new studies have further characterized
tion or cancer. Immunosuppressive effects of POPs have
the cytochrome P450 system in Arctic species. CYP1A is
often been studied as: reduced antibody production
present in Arctic char and is active even at low tempera-
when exposed to a foreign antigen (e.g., suppression of
tures (Wolkers et al., 1996; 1998c; J๘rgensen and Wolk-
the anti-sheep red blood cell plaque-forming response);
ers, 1999). Recent studies in glaucous gull (Larus hyper-
decreased delayed-type hypersensitivity; decreased natu-
boreus) have shown some EROD activity implying the
ral killer cell activity; and, decreased resistance to
presence of CYP1A (Henriksen et al., 1998a; 2000).
pathogens (e.g., viral infections) (Vos and Luster, 1989;
Ringed seals have been found to have CYP1A and 3A
Tryphonas, 1994; Wong et al., 1992). All of these have
activities, and possibly CYP2B activity (Mattson et al.,
been used as indicators of immunosuppressive effects in
1998; Wolkers et al., 1998a; 1998b; Hyyti et al., 2001;
laboratory and wild animals.
Nyman et al., 2001). Besides CYP1A, both grey and
Immunosuppressive effects may be one of the most
ringed seals may have active CYP1B1 (Nyman et al.,
sensitive and environmentally relevant effects of POPs
2000), which has previously only been found in human,
(Vos and Luster, 1989). For example, immunosuppres-
rat and mouse tissues (Nelson et al., 1996). CYP1B1 me-
sion has been measured in harbour seals fed Baltic fish in
tabolizes PAHs, often initiating carcinogenesis (Shimada
semi-field experiments and was found to correlate with
et al., 1996; Baron et al., 1998; Kim et al., 1998a). Harp
levels of PCDD/F and planar PCBs expressed as toxic
seals have been found to have active CYP3A (Wolkers et
equivalents (TEQs) (de Swart et al., 1995; Ross et al.,
al., 1999; 2000). Steller sea lions have been found to
1995; 1996a). Immunosuppression is also suspected to
have CYP1A1 and 1A2, whereas minke whales only
be the cause of an increasing prevalence of moderate to
have CYP1A1 (Teramitsu et al., 2000). Beluga and pilot
severe intestinal ulcers in Baltic grey seals (Bergman,
whales (Globicephala melas) both have CYP1A1 activ-
1999). Indications of immunosuppression have also
ity and show the presence of CYP2B, though it is not
been found in some Arctic species including polar bears,
clear if this is active (White et al., 2000).
northern fur seals, and glaucous gulls (Chapter 6).
Organisms lacking functional CYP1A, 2B or 3A will
not be able to eliminate the POPs metabolized by these
3.2.4. Thyroid and retinol effects
enzymes, leading to their bioaccumulation. This is par-
ticularly the case for fish, making them amplifiers of
Thyroid hormones control metabolism and growth, and
many POPs in food webs. The presence of functional cy-
are essential for normal reproduction. They are also im-
tochrome P450 enzymes means that POPs may be me-
portant for the development of normal brain functions
tabolized and eliminated; metabolized to lipophilic and
during fetal development (Morse et al., 1993; Morse,
toxic metabolites; metabolized to hydrophilic metabo-
1995). The thyroid gland produces predominantly T4
lites that bind to proteins and are retained; and/or, that
(thyroxine), which is transported in plasma to target tis-
POP exposure may lead to cytochrome P450 enzyme in-
sues by the transport protein TTR. Once delivered, T4 is
duction, increasing the amounts of metabolic enzymes
deiodinated by T4-monodeiodinase to triiodothyronine
present. Those organisms that cannot eliminate POPs
(T3), which is the active hormone. Some POP effects
may accumulate parent compound concentrations high
on the thyroid may be related to the ability of some
enough to cause effects. Those organisms that metabo-
POP phenolic metabolites, such as hydroxylated PCBs,
lize POPs to lipophilic and toxic metabolites, and/or hy-
hydroxylated PBDEs, pentachlorophenol and 4-hy-
drophilic and toxic metabolites that are retained, may
droxy-heptachlorostyrene, to attach to the binding
instead accumulate these metabolites to high enough
sites on the transthyretin-retinol-binding protein com-
concentrations to cause effects.
plex (TTR-RBP) in plasma, thereby disrupting the nor-
mal transport of thyroid hormones T3 and T4 as well
as vitamin A (retinol) to their target tissues (Rolland,
3.2.3. Immunological effects
2000; Simms and Ross, 2000). POPs may also interfere
Many POPs disrupt both humoral and cell-mediated im-
with the enzymes that control thyroid hormone metab-
mune responses of the specific (acquired) arm of the im-
olism such as uridine diphosphate glucoronosyl trans-
mune system, as well as causing effects on the non-spe-
ferase (UDPGT), which is involved in glucuronidation
cific (innate) arm. As a result, the resistance to infectious
and subsequent excretion of T4 (Brouwer et al., 1998).
agents may be reduced. Humoral-mediated immunity in-
Recently, TTR has been identified in polar bear plasma
volves the body's ability to recognize foreign substances
(Sandau et al., 2000) and in harbour seals (Simms and
(helper T-cells) and mount a response by stimulating the
Ross, 2000).
production of antibodies (B-cells). Cell-mediated immu-
The structural requirements for binding of a POP
nity is involved in delayed hypersensitivity reactions
metabolite to TTR are hydroxy-substitution in para- or
(e.g., skin reactions to allergens) and the production of
meta-positions of one or both phenyl rings with adjacent
cytotoxic T-cells against tumors and viruses. Natural kil-
halogen unsubstituted sites (Lans, 1995). X-ray diffrac-
ler cells are involved in the non-specific response (i.e.,
tion studies have shown that several phenolic organo-
absence of memory) and are a first line of defense against
halogen compounds bind with the hydroxy group in the
virus-infected cells and tumors. Most POPs cause multi-
central channel of the TTR molecule (Ghosh et al.,
ple effects on the immune system.
2000). Disruption of normal thyroid hormone transport
Some OCs (PCDD/Fs, some PCBs) as well as TBT
leads to lowered plasma levels of T3 and T4, which in
have direct effects on the thymus, causing atrophy of
turn may initiate an increased release of thyroid stimu-
this lymphoid organ responsible for the maturation of T-
lating hormone (TSH) to stimulate the thyroid gland
cells. The most insidious effect of POPs on the immune
to secrete more T3 and T4. This disruption of the feed-
system is to decrease an organism's resistance to infec-
back system for thyroid hormones may lead to thyroid

26
AMAP Assessment 2002: Persistent Organic Pollutants in the Arctic
hyperplasia (goiter), hypertrophy, hypothyroidism, dis-
Methods for measuring mutagenicity include the
ruptions of metabolism and possibly a tendency to de-
measurement of DNA adducts since there is a positive
velop thyroid tumors. Another effect may also be re-
correlation between a chemical's carcinogenic potency
lated to the ability of some POPs, such as PCBs, to in-
and the extent that it binds to DNA (Kriek et al., 1998).
duce the production of liver enzymes involved in the
Correlations have also been shown between the level of
breakdown of thyroid hormones. This results in re-
exposure to PAHs and the amount of adducts formed
duced amounts of thyroid hormones circulating in the
(French et al., 1996; Shugart and Theodorakis, 1998;
plasma.
Wirgin and Waldman, 1998). Other standard methods
Imbalance in vitamin A (retinol and its esters) status
include the formation of micronuclei, sister chromatid
can cause immunosuppression, susceptibility to cancers,
exchange, chromosome aberration assays in peripheral
skin lesions, as well as disruption of reproduction,
blood lymphocytes, as well as the Ames test for single
growth, and development. Some POPs (particularly
nucleotide mutations.
PCDDs and PCBs) affect vitamin A metabolism and
transport. Biomarkers used for thyroid and retinol ef-
3.2.6. Effects of mixtures
fects include measuring plasma levels of free and bound
T3 and T4, TSH as well as vitamin A levels.
In the previous AMAP assessment report, the additive,
antagonistic and synergistic effects of mixtures were dis-
cussed (de March et al., 1998). Some recent studies indi-
3.2.5. Mutagenic and carcinogenic effects
cate that certain POPs singly have little or no endocrine
Current research supports a two-stage cancer model
disrupting effects but when tested as mixtures, endocrine
characterized by a primary mutagenic event (initiation)
disruption occurs, indicating synergism. For example,
followed by a long latency period and second event (pro-
chlordane, dieldrin and toxaphene singly showed no
motion) that leads to tumor growth. Peroxisomes are
ability to compete with estradiol when tested against al-
cellular organelles mainly found in the liver and kidney.
ligator and human estrogen receptors (Arnold et al.,
They contain peroxisomal enzymes and are essential for
1997; Vonier et al., 1996). However, when tested as a
lipid metabolism, cellular respiration and gluconeogene-
mixture, the combination inhibited estradiol binding by
sis among other things (for a review, see Youssef and
20-40%. A mixture of 15 POPs mimicking the composi-
Badr, 1998). A number of POPs, including perfluori-
tion in ringed seal blubber was found to disturb in vitro
nated compounds such as PFOS and PFOA, are potent
porcine oocyte maturation and development at concen-
peroxisome proliferators. Increased oxidative stress by
trations comparable to highly exposed humans or mam-
peroxisome proliferation is thought to be one possible
mals in the Arctic (Campagna et al., 2001).
mechanism for tumor promotion (Cattley and Preston,
1995). Another mechanism may be down-regulation of
3.3. Effects of specific POPs
gap junctional intercellular communication (GJIC),
which has been linked to tumor-promoting properties of
Because the previous AMAP assessment report con-
many carcinogens (Trosko and Ruch, 1998). Gap junc-
tained thorough reviews of the toxicology of legacy OCs
tions are protein channels that allow for the transport of
and their metabolites (PCDDs, PCDFs, PCBs, MeSO2-
substances between cells. This communication is neces-
PCBs, aldrin, dieldrin, chlordanes, DDT, MeSO2-DDE,
sary for normal growth and function and, if inhibited,
HCB, HCHs, mirex, toxaphene), only those compounds
may lead to tumor promotion. Various PCBs, DDT, diel-
where substantial new information has become available
drin, toxaphene and brominated biphenyls have been
since 1996 are covered in this report. Emphasis has been
shown to inhibit GJIC in human breast epithelial cells
placed on updates of these few compounds, plus the tox-
(Kang et al., 1996) and methyl sulfone PCBs seem par-
icology of new compounds that have been found in the
ticularly potent (Kato et al., 1998). GJIC is measured
Arctic. The following descriptions of the toxicology
using a bioassay where a monolayer of cells is scraped
of different POPs are short reviews and not meant to
after exposure to a contaminant, then exposed to a
be comprehensive. They mainly cover chronic effects
dye and dye migration into the cells is quantified (Up-
and effects that are relevant to the Arctic discussion.
ham et al., 1997).
Results for controlled studies in laboratory animal
High doses of planar aromatic hydrocarbons induce
species are discussed first. Where done, controlled stud-
oxidative stress possibly through induction of CYP en-
ies using wild animal species under laboratory condi-
zymes (Toborek et al., 1995; Park et al., 1996; Hennig et
tions are then presented. If field studies outside of the
al., 1999; Schlezinger et al., 1999; Slezak et al., 1999;
Arctic have been performed, these are then presented.
Slim et al., 1999). This is manifested by increased pro-
Finally, studies correlating specific effects with contami-
duction of reactive oxygen species, lipid peroxidation
nant concentrations found from field studies of wild
and DNA damage.
species, in areas with known high burdens of contami-
Several PAHs, such as B[a]P and DMBA, are well
nants such as the Baltic Sea and the Great Lakes, are dis-
known mutagens and both initiate and promote tumors.
cussed. This last type of study has inherent problems, as
Most of the other POPs dealt with in this assessment are
it is never possible to state that the contaminant meas-
not mutagenic, but many are strong tumor promoters.
ured is the cause of the effect, since there may be other
Several POPs are associated with increased tumor preva-
contaminants not measured that co-vary with the one
lence found in highly exposed wildlife from areas out-
measured. Field studies and correlation studies of Arctic
side of the Arctic, including fish (Myers et al., 1998) and
species are discussed under Section 6. An overview of
beluga from the St. Lawrence Estuary (De Guise et al.,
toxic effects of the POPs discussed in this report is given
1994; Martineau et al., 1985).
in Table 3.1.

Chapter 3 ท Toxicology
27
Table 3ท1. Overview of toxic properties of various POPs. = suppression or decrease,
= induction or increase.
ญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญ
Reproductive/
Cytochrome
Thyroid/
developmental
Neurotoxic
P450
Immune
retinol
effects
effects
effects
effects
effects
Cancer
Other
ญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญ
Aldrin and
Reproduction
Induces cyto-
Suppresses
Non-mutagenic.
dieldrin
chrome P450 2B immune system
Increased
liver tumors
ญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญ
Chlordanes
Reproduction
Induces cyto-
Suppresses
Non-mutagenic
chrome P450 2B immune system
tumor promoter
ญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญ
DDT and
Egg-shell thinning
Induces cyto-
Suppresses
Thyroid
Adrenal cortex
metabolites
in bird eggs.
chrome P450 2B immune
weight
hyperplasia
Reproduction
system
ญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญ
HCBz
Fetotoxic.
Induces
Suppresses
T3 and T4
Non-mutagenic
Porphyria
Teratogenic.
cytochrome
immune
tumor
Reproduction
P450 1A
system
TSH
promoter
and 2B
Thyroid weight
ญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญ
-HCH
No information
Induces cyto-
Non-mutagenic
chrome P450 2B
tumor promoter
ญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญ
-HCH
Estrogenic
Induces cyto-
Suppresses
Thyroid
Non-mutagenic
chrome P450 2B immune weight
tumor
promoter
system
ญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญ
-HCH
Estrogenic and
Induces cyto-
Non-mutagenic
(lindane)
antiestrogenic.
chrome P450
tumor promoter
Reproduction
1A and 2B
ญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญ
Mirex
Reproduction
Induces cyto-
Suppresses
Non-mutagenic.
chrome P450 2B immune system
Induces tumors
ญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญ
Toxaphenes Fetotoxic.
Induces
Suppresses
Thyroid-
Mutagenic,
Bone brittle-
Reproduction
cytochrome
immune
weight
potent carci-
ness in fish.
P450 1A, 2B
system
TSH nogen.
Adrenal
and 3A
Inhibits GJIC
hypertrophy
ญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญ
Endosulfan
Fetotoxic.
Induces cyto-
Suppresses
Non-mutagenic
Reproduction
chrome P450
immune system
1A and 2B
ญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญ
PCDD/Fs Fetotoxic.
Permanent
Induces
Thymic
T3 and T4
Non-mutagenic
Porphyria
and nPCBs
Deformities.
changes
cytochrome
atrophy.
tumor pro-
and meta-
Reproduction in learning,
P450 1A
Suppresses
Vitamin A
moters.
bolites
behavior,
immune Affects
GJIC
memory
system
(nPCB)
ญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญ
Other PCBs Fetotoxic.
Permanent
Induces
Suppresses
T3 and T4
Non-mutagenic
Porphyria
Deformities.
changes
cytochrome
immune tumor
pro-
Hyperadreno-
Reproduction in learning,
P450 2B
system
Vitamin A
moters.
cortism
behavior,
Affects GJIC
memory.
Decreased
dopamine
ญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญ
SCCPs
Fetotoxic
Motor
Induces No
T4
Non-mutagenic.
Deformities.
perform-
cytochrome
information
Peroxisome
Reproduction
ance
P450 1A
TSH
proliferation.
Inhibits GJIC
ญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญ
PCNs
Embryotoxic. Induces
cyto-
Reproduction
chrome P450 1A
ญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญ
Octachloro-
Induces
Binds to TTR
styrene and
cytochrome
in vitro
metabolites
P450 1A and 2B
ญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญ
PBDEs
Estrogenic and
Permanent
Induces
Suppresses
T4
Non-mutagenic
antiestrogenic
changes
cytochrome
immune
in learning,
P450 1A
system
Vitamin A
behavior,
and 2B
memory
ญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญ
PFOS/PFOA
Reproduction
Non-mutagenic,
tumor promoter
Peroxisome
proliferation.
Inhibits GJIC
ญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญ
TBT
Imposex in
Inhibits liver
Suppresses
May be carcinogenic
and
invertebrates.
cytochrome
immune system
metabolites
Deformities.
P450 1A,
Reproduction
2B, and 3A
ญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญ

28
AMAP Assessment 2002: Persistent Organic Pollutants in the Arctic
3.3.1. Halogenated industrial chemicals
cant behavioral alterations in rat offspring after pre- and
and by-products
post-natal exposure (including lactation exposure in
males) to several specific CB congeners (CBs 118, 126,
3.3.1.1. Update on PCDDs, PCDFs, and PCBs,
153). In a recent study (Holene et al., 1999), female rats
including PCB metabolites
were exposed to CB153 through mother's milk. The
The 2,3,7,8-PCDD/Fs, as well as PCBs substituted in the
females showed a significant sex-specific behavioral re-
3,3',4,4'-positions with no (non-ortho PCB (nPCB)) or one
sponse, being less sensitive than males studied previ-
ortho chlorine (mono-ortho PCB) are among the most
ously, since only deficient acquisition of time discrimina-
toxic POPs. The most toxic and best studied of these pla-
tion was seen.
nar compounds is TCDD. These substances exert their
Repeated exposure to Aroclors caused decreases in
toxic effects via a common mechanism that requires bind-
brain dopamine and the bioaccumulation of several
ing to the Ah receptor. They produce essentially the same
ortho-substituted CBs in the brain of rats and non-hu-
spectrum of toxic effects in treated animals as TCDD,
man primates (Seegal et al., 1986; 1991a; 1991b; Shain
differing only in their potencies. Except in some bird
et al., 1986; 1991; Seegal and Schantz, 1994). Koda-
species, all are less potent than TCDD. The non-ortho
vanti et al. (1995; 1996) have since been able to show
PCBs are more toxic than the mono-ortho congeners. A
that the ortho-substituted, non-dioxin-like PCB con-
thorough review of the effects of PCDDs, PCDFs, and
geners are associated with neurotoxicity. The ortho-
PCBs was presented in the previous AMAP assessment
PCBs were found to disrupt intracellular signal trans-
report (de March et al., 1998). The following is an up-
duction in cells from the cerebellum. Mariussen et al.
date on relevant effects studies published since 1996.
(1999) tested 14 PCBs in vitro for their ability to com-
petitively inhibit dopamine uptake into synaptic vesicles
Reproductive and developmental effects
and found that only ortho-PCB congeners were active.
Competitive-binding studies using estrogen receptors from
EC50s (the concentration affecting 50% of the animals)
humans, green anole (Anolis carolinensis), and rainbow
ranged from 2.3 to 5.6 ตg/g (7-30 ตM) for single con-
trout (Salmo gairdnerii) showed that chlorobiphenyls
geners. In non-human primates, brain concentrations of
(CBs) 104, 184, and 188 were significant competitors in
2 to 5 ตg/g of PCB were associated with decreased
all three systems and thus have estrogen-like activity
dopamine concentrations (Seegal et al., 1990). In a study
(Matthews and Zacharewski, 2000). In the rainbow
of 20 PCB congeners, inhibition of dopamine uptake in
trout assay, CBs 41, 51, 91, 115, 143, and 173 were also
synaptic vesicles was found to be most potent for CBs
found to compete to some extent. None of these con-
41, 91, 112, and 143 (all ortho-substituted) and no inhi-
geners are, however, major components in commercial
bition was seen with the non-ortho CBs tested (Anders-
PCB mixtures, and some are not present at all. The find-
son, 2000).
ings are therefore of small relevance to the environment.
Aroclor 1254 has been found to affect bone devel-
Hydroxy-PCBs that are found in blood have been shown
opment in rats (Andrews, 1989) and recently, CB126
to be weakly anti-estrogenic in the MCF-7 human breast
(3,3',4,4',5-PeCB) has been shown to cause extensive al-
cancer cell line and in HeLa cells (Kramer et al., 1997;
terations in the long bones of rats, including decreases in
Moore et al., 1997). Andersson et al. (1999) found that
density and strength (Lind et al., 1999; Lind, 2000).
four hydroxy-PCBs significantly induced vitellogenin
TCDD also affects bone structure in rats in a similar
synthesis in rainbow trout hepatocytes and induced pro-
manner (Jไmsไ et al., 2001) and has also been found to
liferation of MCF-7 cells, both indicative of the capabil-
impair their molar tooth development (Kattainen et al.,
ities of these substances to bind to the estrogen receptor
2001; Lukinmaa et al., 2001). A synthetic mixture of 16
and elicit a response. Also shown, was the induced pro-
MeSO2-PCBs, prepared in the same relative concentra-
liferation of MCF-7 cells by CBs 104 and 188.
tions as found in the blubber of Baltic grey seals, was fed
Experiments have been performed exposing neonatal
to female mink (Mustela vison) for one year and repro-
mice to a single oral dose of specific CB congeners at a
ductive outcome was studied (Lund et al., 1999). Com-
critical time point during the brain growth spurt and
pared to controls, the treated mink had significantly in-
studying a range of neurobehavioral effects (Eriksson
creased litter size, but lower birth weights and reduced
et al., 1991; Eriksson and Fredriksson, 1996a; 1996b;
kit survival. In vitro studies of liver showed increased
1998). Permanent changes in spontaneous behavior were
breakdown of progesterone in the treated mink. The
seen after neonatal exposure to ortho-substituted PCBs
muscle concentrations found in the exposed females and
(2,4,4'-TrCB (CB28) at 14 ตmol (3.6 mg)/kg body
their kits were 18 000 ng/g lw and 21 000 ng/g lw, indi-
weight; 2,2',5,5'-TeCB (CB52) at 14 ตmol (4.1 mg)/kg
cating considerable carry-over from dams to kits.
body weight; 2,2',4,4',5,5'-HxCB (CB153) at 14 ตmol
In mink, continuous exposure to 250 ng PCB/g food
(5.1 mg)/kg body weight); non-ortho PCBs (3,3',4,4-
caused delayed onset of estrus and lowered whelping
TeCB (CB77) at 14 ตmol (4.1 mg)/kg body weight;
rates (Restum et al., 1998). PCB exposure was from a
3,3',4,4',5-PeCB (CB126) at 0.14 ตmol (0.046 mg)/kg
diet made using Saginaw Bay carp as the PCB source.
body weight; and, 3,3',4,4',5,5'-HxCB (CB169) at 1.4
Exposure to 500 ng PCB/g food led to increased litter
ตmol (0.51 mg)/kg body weight). The effects worsened
mortality and reduced kit body weights. Short-term pa-
with age. As well, learning and memory in adult mice
rental exposure reduced kit survival of subsequent gene-
were affected by CB52 and CB153 at the same doses,
rations of mink conceived months after the parents were
CB126 at 1.4 ตmol (0.46 mg)/kg body weight, and
placed on PCB-free feed.
CB169 at 14 ตmol (5.1 mg)/kg body weight.
Kestrels (Falco sparverius) were exposed to PCBs in
In previous studies reviewed in the AMAP assess-
ovo via parents that were fed a diet containing a mixture
ment report, Holene et al. (1995; 1998) showed signifi-
of technical PCBs (7 mg/kg body weight/day) through

Chapter 3 ท Toxicology
29
breeding and hatching (Fernie et al., 2001). This diet led
and starved, non-exposed and fed, and non-exposed and
to an exposure of 34 ตg PCB/g whole egg ww. These sec-
starved (J๘rgensen et al., 1999). The PCB-exposed fish
ond-generation kestrels were paired with unexposed, ex-
were fed a single oral dose of 1 ตg Aroclor 1260/g body
perienced kestrels and compared with non-exposed con-
weight. Increased EROD activity was found in the PCB-
trols. In ovo PCB exposure was found to suppress egg
exposed and starved group, which also had the highest
laying in 25% of exposed females, delay clutch initiation
PCB concentrations and exhibited fin erosion. The
and lead to smaller clutch sizes for both exposed sexes.
threshold for EROD effects in Arctic char was found to
Exposed kestrels also had reduced fledgling success and
be 1 ตg PCB/g ww in liver.
higher incidence of complete brood mortality. Greater
Induction of hepatic CYP1A1 activity was found to
effects were seen in PCB-exposed females than males.
be correlated with PCDD/F concentrations in common
A mixture of 20 PCB congeners administered to ze-
tern chicks (Murk et al., 1996). EROD induction was
brafish (Brachydanio rerio) via food caused reproduc-
found to correlate with PCB and DDT levels in a com-
tive disturbances (ึrn et al., 1998). Egg production and
parison of ringed and grey seals from the Baltic Sea and
offspring survival time were reduced. CBs 60, 104, and
from reference sites (Nyman, 2000).
a hydroxy-PCB were highly embryotoxic in zebrafish
oocytes exposed via maternal transfer (Westerlund et al.,
Immunosuppression
2000). A significant negative correlation was found be-
Beluga leukocytes and splenocytes were exposed to sev-
tween PCB levels above 20 ng/g ww (as Aroclor 1260
eral POPs in vitro, and CB138 was found to significantly
in liver) and baculum (penis bone) length in wild juve-
reduce splenocyte proliferative responses (De Guise et
nile mink (Harding et al., 1999).
al., 1998). No effects were seen with CBs 153, 180, 169
Aroclor 1242 caused significant sex reversal in red-
or TCDD.
eared sliders (Trachemys scripta elegans), a species of tur-
Previously, a diet of Baltic Sea herring (Clupea haren-
tle, overriding male-producing temperature levels to re-
gus) fed to captive harbour seals was found to cause im-
sult in female hatchlings, indicating that this PCB mix-
munosuppression, which was associated with the con-
ture has estrogenic effects (Willingham and Crews, 1999).
tent of PCDD/PCDF and dioxin-like PCBs (de Swart et
Later egg laying, prolonged incubation, and smaller
al., 1994; 1995; Ross et al., 1995; 1996a). In a labora-
eggs and chicks were correlated with higher yolk sac
tory study, using rats as surrogates for harbour seals,
PCDD/F concentrations in common tern chicks (Sterna
pregnant rats were administered extracts from the same
hirundo) (Murk et al., 1996). Toxic effects in colonial
diets of Atlantic or Baltic Sea herring as were given to
fish-eating birds in the Great Lakes area of North Amer-
harbour seals, or a positive control of Atlantic herring
ica were correlated with the concentrations of PCDDs/
oil spiked with TCDD, on a daily basis for 41 days (Ross
PCDFs and non-ortho PCBs found in the different bird
et al., 1997). Immune function was assessed in the off-
species studied (reviewed in Gilbertson et al., 1991;
spring and found to be most suppressed in the TCDD-
Giesy et al., 1994a; 1994b). The effects in birds include
spiked group (reduced thymus weight, thymocyte and
reduced egg hatching, embryotoxicity, deformities such
splenocyte proliferative response, natural killer cell and
as crossed bills and clump feet, and impaired parental
specific antibody responses). Similar types of immuno-
behavior (Hoffman et al., 1987; Kubiak et al., 1989;
suppression were seen in the Baltic herring group but
Tillitt et al., 1989; 1991; 1992; 1993; Yamashita et al.,
were less pronounced. The daily intakes for the pregnant
1993).
rats were 0.3 pg TEQ/g body weight for the cleaner At-
lantic herring oil, 2.1 pg TEQ/g body weight for the
Cytochrome P450-dependent monooxygenases
Baltic Sea herring oil and 134 pg TEQ/g body weight for
Rats and mice exposed to either Aroclor 1254 or CB105
the spiked oil.
had induced cytochrome P450 activity measured as
Two juvenile harp seals were exposed experimentally
EROD, methoxy-O-resorufin deethylase (MROD) and
to increasing doses of selected PCB congeners (0.4, 2.0,
PROD activity, with rats being more sensitive than mice
4.0, 20.0, and 40.0 mg/day, one week per dose), while
(Hallgren et al., 2001).
two more seals acted as controls (Lohman et al., 2002a).
In the study of 16 MeSO2-PCBs fed to mink, dis-
The in vitro release of tumor necrosis factor alpha
cussed previously, an 11-fold induction of PROD activ-
(TNF- ) from isolated monocytes stimulated with Esche-
ity was seen in the exposed adult females and a five-fold
richia coli (E. coli) lipopolysaccharides and -1,3-glucan
increase was seen in their five-week-old kits (Lund et al.,
were assayed during the treatment period and during a
1999). This is in accordance with previous research
subsequent 30-day fasting period. Changes in the adre-
showing high CYP2B induction for MeSO2-PCBs in rats
nal gland were also studied (Lohman et al., 2002b). No
(Kato et al., 1995a; 1995b; 1997; 1999).
differences were seen between the two groups until after
Mink fed a diet with varying PCB doses made from
30 days of fasting. Release of TNF-
was elevated in
clean fish or contaminated carp from Saginaw Bay, Lake
both groups, with higher levels in the controls than in
Huron, had dose-dependent induction of cytochrome
the PCB-treated seals, indicating possible decrease in
P450 activity (Shipp et al., 1998). Young mink were
monocyte reaction to lipopolysaccharides. Elevated ba-
more sensitive than older mink.
sal levels of serum cortisol and aldosterone were seen in
Adult male mallard ducks (Anas platyrhynchos)
the PCB-exposed seals, compared to the control seals.
dosed orally with Aroclor 1254 had induced cytochrome
After 30 days of fasting, both groups had increased lev-
P450 activity as measured by elevated EROD and
els of cortisol and aldosterone compared to the treat-
PROD activity (Fowles et al., 1997).
ment period, with levels in the PCB-treated seals exceed-
In an experimental study, Arctic char were divided
ing those in the controls. Although this study is limited
into four groups, PCB-exposed and fed, PCB-exposed
because of small sample sizes, the results may indicate

30
AMAP Assessment 2002: Persistent Organic Pollutants in the Arctic
that short-term exposure to PCB may induce hyperadre-
Herring gull (Larus argentatus) and Caspian tern
nocorticism without noticeable morphological changes
(Sterna caspia) chicks from five Great Lakes sites were
in the adrenal gland of seals.
assessed for immune function using the PHA skin test, a
No immunosuppression was seen in rats fed diets
sensitive indicator of T-cell-mediated immunity in birds.
containing beluga blubber from the St. Lawrence Estu-
Suppression of T-cell-mediated immunity was found to
ary (highly PCB contaminated), the Arctic or combina-
correlate most strongly to higher PCB concentrations
tions of the two (Lapierre et al., 1999). Mice fed diets
(Grasman et al., 1996; Luebke et al., 1997; Grasman
based on varying amounts of beluga blubber from the
and Fox, 2001).
St. Lawrence Estuary and the Arctic, providing different
Significant correlations have been found between
PCB exposures, showed some indications of immuno-
high PCB concentrations, suppressed T-cell function
suppression but no difference was seen between the
and increased antibody titers after immunization in
treatment groups (Fournier et al., 2000).
Caspian terns from Lake Huron (Grasman and Fox,
Chicken embryos exposed to CB126 in ovo had de-
2001). Significant negative correlations were found be-
creased thymus mass, decreased live T-cell numbers in
tween yolk sac levels of PCDD/F, PCB, non-ortho PCBs
the thymus, declines in some types of thymocytes, and
and TEQs, and plasma corticosterone levels in herring
decreased viable B-cell numbers in the bursa of Fabricius
gull embryos from the Great Lakes (Lorenzen et al.,
(Grasman and Whitacre, 2001).
1999). The activities of phosphoenolpyruvate carboxy-
Glaucous gull chicks with high and low dietary PCB
kinase and malic enzyme, two metabolic enzymes regu-
exposure were immunized with antigen. Antibody pro-
lated in part by corticosteroids, were also negatively cor-
duction, lymphocyte response, mitogen-induced lympho-
related to yolk sac PCDD/F concentrations.
cyte proliferation and hematology were measured after
A significant association was found between PCB
56 days of age (Larsen et al., 2002d). A significantly
concentrations and mortality due to infectious disease in
higher lymphocyte response to phytohemagglutinin
harbour porpoises from England and Wales indicating a
(PHA) in the high PCB group indicated general immune
possible causal relationship between PCB exposure and
system stimulation. Significantly lower antibody titers in
immunosuppression (Jepson et al., 1999).
response to influenza virus immunization in the high
PCB group indicated impaired ability to produce anti-
Thyroid and retinol effects
bodies and may be associated with decreased resistance
Until 1996, only a few hydroxy-PCB metabolites had
to infections.
been identified and studied. Since then, up to 30 hy-
In the study on Arctic char by J๘rgensen et al.
droxy-PCBs have been detected in plasma of various
(1999), previously described in the section on P450-de-
species and 13 such metabolites have been identified
pendent monooxygenases, increased plasma cortisol lev-
(Letcher et al., 2000a). Most of the identified hydroxy-
els were seen in the PCB-exposed and fed group. In a
PCB metabolites have the hydroxy-group attached to
similar study, J๘rgensen et al. (2002b; 2002c) exposed
one of the two para-positions with chlorine atoms at-
Arctic char to varying, single doses of Aroclor 1254 and
tached ortho to the hydroxy-group. This structural con-
held them either with or without food for five months
figuration mimics that of T4, the thyroid hormone,
before subjecting them to a ten-minute handling distur-
except that T4 has iodine atoms instead of chlorines.
bance. Starved fish were given doses of 0, 1, 10 or 100
The affinities of several hydroxy-PCBs for TTR, the
ตg/g body weight and fed fish were given doses of 0 or
plasma protein that transports T4, are up to ten times
100 ตg/g body weight. Starved control fish had elevated
greater than for the natural hormone (Brouwer et al.,
plasma cortisol levels compared with fed fish before
1998; Lans et al., 1993; 1994). 4-OH-3,3',4',5-TeCB
handling. These basal cortisol levels were suppressed by
(CB77 metabolite) and 4-OH-CB107 reduced total
PCBs in starved fish and were elevated in fed fish. The
plasma T4 levels in pregnant mice and in the fetuses
cortisol response to handling was suppressed by PCBs in
(Sinjari et al., 1998a) but the effect was less when CB77
a dose-dependent way in starved fish. Plasma glucose
was administered and allowed to metabolize in vivo
levels were affected in the same manner as cortisol lev-
(Darnerud et al., 1996). 4-OH-2',3,3',4',5-PeCB (CB105
els. The findings indicate that stress responses in Arctic
metabolite) was also found to reduce T4 levels but no ef-
char are compromised by PCBs and the effect of fasting
fect was seen with 4-OH-2,3,3',4',5-PeCB (CB105 me-
makes char sensitive to the effects of PCBs.
tabolite) (Sinjari and Darnerud, 1998). Hydroxy-PCBs
The char were also subjected to tests of disease resist-
also influence T4 metabolism by inhibiting sulfation, a
ance (Maule et al., 2002). Mortality after exposure to a
major regulation pathway in the fetus. Inhibition of sul-
disease agent was dose-related (0, 1, 10, and 100 ตg/g body
fation has been shown to occur in vitro (Schuur et al.,
weight) among the starved fish, with a significant trend
1996; 1998; 1999). If this occurs in vivo, it has implica-
toward higher disease susceptibility with increasing PCB
tions for fetal brain development (Brouwer et al., 1998).
dose. No differences in mortality were seen in the two
Aroclor 1254 or CB105 treatment significantly reduced
fed groups (0 and 100 ตg/g body weight). However, total
free and total plasma T4 levels and hepatic vitamin A in
mortality was higher in the fed groups compared to the
rats and mice but had no effect on levels of TSH (Hall-
starved groups. The results indicate that PCB reduces
gren et al., 2001).
immunocompetence in starved Arctic char in a dose-de-
In the study of 16 MeSO2-PCBs fed to mink, dis-
pendent manner, but that lean fish are also more disease-
cussed previously, plasma concentrations of total T3 and
resistant than fed fish. Similarly, juvenile chinook sal-
T4 were reduced in the exposed dams (Lund et al.,
mon (Oncorhynchus tshawytscha) exposed to Aroclor
1999). The authors speculate that the MeSO2-PCBs de-
1254 had higher mortality than controls when exposed
crease the total T4 concentration via enzyme induction
to a bacterial pathogen (Arkoosh et al., 2001).
of UDPGT, which is involved in glucuronidation of T4.

Chapter 3 ท Toxicology
31
Total and free plasma T4 concentrations increased
and yolk sac retinyl palmitate were significantly cor-
with increasing TEQ levels in mink exposed via a diet of
related to higher yolk sac PCDD/F concentrations in
carp from Saginaw Bay, Lake Huron, Michigan, but
common tern hatchlings from Belgium and the Nether-
total and free T3 decreased, indicating a reduction in
lands (Murk et al., 1996). Lower plasma retinol levels in
T4-monodeiodinase activity (Heaton et al., 1995). In a
herring gull and Caspian tern chicks from several Great
similar study, T4 concentrations were found to be signif-
Lakes colonies were associated with high PCB con-
icantly higher in mink fed a fish diet containing PCB
centrations in eggs from the same colony (Grasman et
than in those not exposed to PCB (Nieminen et al.,
al., 1996).
2000). Rats fed a diet containing Baltic Sea herring that
Higher PCB concentrations were found to be associ-
had previously been used in a semi-field study of im-
ated with increased metabolism of retinoic acid by
munosuppression in harbour seal (Ross et al., 1995),
cytochrome P450 enzymes, decreases in hepatic reti-
were found to have lower plasma T4 levels compared to
noid stores and an increase in developmental deformities
the control group (Ross et al., 1996b).
in lake sturgeon (Acipenser fulvescens) from the St.
Chick embryos exposed in ovo to Aroclor 1242 or
Lawrence River compared to a reference site (Doyon et
1254 had reduced plasma T4 concentrations and re-
al., 1999).
duced hepatic monodeiodinase activity (Gould et al.,
1999). A correlation was also seen between femur length
Cancer
and plasma concentrations of T3 and T4, indicating a
The coplanar PCBs, CB77 and 169, as well as TCDD,
decrease in skeletal growth due to reduced thyroid hor-
inhibit GJIC in mouse Hepa1c1c7 cells (de Haan et al.,
mone levels. Adult male mallards dosed with Aroclor
1994). Of 20 tested PCB congeners, 14 were found to in-
1254 had significantly increased thyroid weights and de-
hibit GJIC in rat liver white blood cell culture (Anders-
creased plasma total T3 concentrations (Fowles et al.,
son, 2000). Most potent were CBs 51, 143, 173, and
1997). Adult great blue herons (Ardea herodias) dosed
184 (tri- and tetra-ortho congeners). Other active con-
with 2,3,7,8-TCDD had a significant increase in plasma
geners included CBs 41, 60, 91, 104, 112, 115, 153,
T4 levels, but no effects were seen on plasma total T3 or
188, 190, and 193. Inactive congeners were CBs 58, 68,
the T3:T4 ratio (Janz and Bellward, 1997). White
78, 99, 126, and 169.
leghorn hens were fed a diet with different PCB levels
Glaucous gull chicks from Svalbard were fed a clean
derived from carp from Saginaw Bay, Lake Huron (Zile
(Arctic cod, hen eggs) or a contaminated (Arctic cod, gull
et al., 1997). The high PCB intake group had decreased
eggs) diet (Kr๘kje et al., 2002). Chromosome aberra-
molar ratios of retinol to retinyl palmitate in eggs.
tions and DNA adducts were measured and higher fre-
CB77 injected intraperitoneally into brook trout
quencies of aberrations were seen in both males and fe-
(Salvelinus fontinalis) caused decreased plasma retinol in
males of the exposed group. DNA adduct levels were
males (Ndayibagira et al., 1994) and exposure with
also higher in males of the exposed group.
CB77 in rainbow trout caused increased hydroxylation
of retinoic acid due to induction of CYP1A enzymes
3.3.1.2. SCCPs/C
(Gilbert et al., 1995). Oral dosing of lake trout (Salveli-
10-C13 polychlorinated n-alkanes
nus namaycush) with CB126 caused decreased liver
For reviews of the toxicology of SCCPs, see Environ-
retinoids (Palace and Brown, 1994).
ment Canada (1993), Willis et al. (1994), WHO (1996)
In free-living European otter (Lutra lutra), a strong
and Tomy et al. (1998). There are difficulties in assessing
negative correlation was found between hepatic vitamin
the toxicity of SCCPs as most data have been generated
A levels and TEQs calculated from non- and mono-
using technical SCCP products, which can contain thou-
ortho PCBs (Murk et al., 1998), and animals with the
sands of chemical compounds. This makes it difficult to
higher concentrations had a higher incidence of infec-
study the toxicity of individual components, and effects
tious diseases. Immature northern elephant seals (Miro-
may be due to stabilizers and impurities in the products.
unga angustirostris) with northern elephant seal skin
As well, degradation, bioaccumulation, and metabolism
disease had depressed total T3 and T4 levels, depressed
change the relative amounts of SCCP components found
retinol levels, and higher concentrations of PCB (and
in organisms in the environment, making it difficult to
p,p'-DDE) compared to unaffected controls (Beckmen et
assess exposure. Recently, studies examining the effects
al., 1997). The ratio of T3 to T4 was significantly corre-
of SCCPs on fish have concluded that they have low
lated to the concentration of CB169 in grey seals from
acute toxicity and a narcotic mode-of-action (Fisk et
the U.K. (Hall et al., 1998). Plasma thyroid hormone
al., 1999b; Cooley et al., 2001), although histopatholog-
concentrations and PCBs were measured in a group of
ical lesions were observed in the livers of exposed rain-
grey seal pups from the Norwegian west coast (Froan,
bow trout (Cooley et al., 2001). In general, SCCPs ap-
Trondheimfjord) and compared to a more highly exposed
pear to be much less toxic than other persistent organic
group in the Baltic Sea (S๘rmo et al., 2002). No rela-
pollutants.
tionship was found for T4, but T3 levels were lower in
the more contaminated Baltic pups as compared to the
Reproductive and developmental effects
Norwegian pups. Disruption of vitamin A and/or thy-
Pregnant rats were dosed orally with 0, 100, 500, and
roid hormones related to high PCB concentrations has
2000 mg/kg body weight/day of a SCCP (58% Cl) on
also been observed in other captive and free-ranging seal
days 6 through 19 of gestation to study teratogenic ef-
species (Brouwer et al., 1989; de Swart et al., 1994; Rol-
fects (WHO, 1996). The high-dose group of dams had
land, 2000; Simms et al., 2000; Simms and Ross, 2000).
32% mortality and decreased body weight, and there
Decreased yolk sac retinoids and plasma thyroid lev-
were increased incidences of post-implantation loss, fe-
els, and increased ratios between plasma retinol levels
tal malformations, and decreases in viable fetuses. The

32
AMAP Assessment 2002: Persistent Organic Pollutants in the Arctic
no-observed-effect-level (NOEL) for teratogenic effects
plasma TSH levels, a 30-40% decrease in total and free
was set at 500 mg/kg body weight/day. In pregnant rab-
T4, a two-fold increase of UDPGT activity but no effect
bits exposed orally to 0, 10, 30, and 100 mg/kg body
on T3 levels (Wyatt et al., 1993). Elcombe et al. (1994)
weight/day on gestation days 6 through 27, no effects
observed a similar increase in UDPGT activity and TSH
were seen on dams or fetuses, and the no-observed-ad-
level and a decrease in T4 levels in male and female rats
verse-effect level (NOAEL) was set at 100 mg/kg body
exposed to a SCCP (58% Cl) at similar doses to the
weight/day.
above study. They also noted thyroid follicular cell hy-
Mallard ducks were exposed to dietary concentra-
pertrophy.
tions of 0, 28, 166, and 1000 mg/kg of a SCCP (58% Cl)
in a one-generation test (cited in Willis et al., 1994). No
Cancer
effects were seen on the adults but some eggshell thin-
A C10-C23 SCCP (70% Cl) was not found to be muta-
ning was seen although the authors questioned the bio-
genic in the Ames test using three different strains of -
logical significance of this. Hatchlings were fed the same
Salmonella typhimurium (Meijer et al., 1981). A SCCP
diets for 14 days and those in the high-dose group
(60% Cl) was found to increase hepatocellular neo-
showed 10% mortality.
plasms in both sexes of mice and rats; kidney tubular
cell adenomas; adenocarcinomas and mononuclear cell
Neurological effects
leukemia in male rats; and, thyroid follicular cell neo-
Mice exposed intravenously to single doses of 30-300
plasms in female rats and mice (Bucher et al., 1987).
mg/kg body weight of two SCCPs (49% and 70% Cl)
The rats were given repeated oral doses of 312 and 625
were studied for effects on motor performance and ther-
mg/kg body weight/day and the mice, 125 and 250
moregulation (Eriksson and Kihlstr๖m, 1985). Dose-de-
mg/kg body weight/day and their responses were fol-
pendent decreases in motor performance and thermoreg-
lowed for two years.
ulation were seen with increasing doses of both SCCPs.
Peroxisomal proliferation, as measured by signifi-
Mice exposed to 300 mg/kg body weight of the lower
cantly increased peroxisomal fatty acid oxidation, was
chlorinated SCCP showed significantly decreased motor
seen in rats and mice exposed to two SCCPs (56% and
performance 15 minutes after injection. As well, signifi-
58% Cl) (Wyatt et al., 1993). Peroxisome proliferation
cantly decreased rectal temperature was seen in mice in-
was confirmed using microscopic methods in rats ex-
jected with 300 mg/kg body weight of both SCCPs after
posed to the same two SCCPs (Elcombe et al., 1994).
60 minutes.
SCCPs (50% and 60% Cl) have been found to be po-
tent inhibitors of GJIC in rat liver epithelial cells, indi-
Cytochrome P450-dependent monooxygenases
cating that they may be tumor promoters (Kato and
Cytochrome P450 concentrations increased in rats ex-
Kenne, 1996).
posed intraperitoneally to 1000 mg/kg body weight/day
of two short-chain SCCPs (49% and 71% Cl) for four
3.3.1.3. PCNs
days (Nilsen and Toftgๅrd, 1981). The increase was
greater for the more highly chlorinated SCCP. Cyto-
PCNs are planar molecules, like PCDDs, PCDFs, and
chrome P450 induction potential was studied in rats
non-ortho PCBs, and also seem to exert their effects via
using five different SCCP formulations dosed orally at
the Ah receptor. Acute and chronic exposure to PCNs
1000 mg/kg body weight/day for four days (Nilsen et al.,
leads to effects similar to those seen for PCDDs, PCDFs
1981). Only the higher chlorinated short-chain formula-
and non-ortho PCBs (for reviews, see Kover, 1975;
tions (59% and 71% Cl) led to increased P450 concen-
Brinkman and Reymer, 1976; Crookes and Howe, 1993;
trations and EROD activity. A lower chlorinated short-
Jakobsson and Asplund, 2000). The most toxic con-
chain (49% Cl), a medium-chain (50% Cl), and a long-
geners are the PeCNs and HxCNs. Based on in vitro
chain formulation (49% Cl) had no effect.
studies, several PCN congeners have been assigned
High single oral doses (1000 mg/kg body weight) of
TCDD toxic equivalency factors (TEFs) (Hanberg et al.,
C10-C13 SCCP (49% Cl) caused an increase in benzo-
1990; 1991; Blankenship et al., 2000; Villeneuve et al.,
[a]pyrene hydroxylase activity in female flounder held in
2000).
brackish water (Haux et al., 1982). No effects were seen
in females kept in seawater, males kept in brackish or
Reproductive and developmental effects
seawater, or any fish exposed to a more highly chlori-
A commercial PCN product (Halowax 1014) as well as
nated product (70% Cl). Rainbow trout exposed to two
a mixture of 1,2,3,5,6,7- and 1,2,3,4,6,7-hexachloro-
C12 SCCPs (56% and 69% Cl) showed no induction of
naphthalenes are both toxic to chick embryos (Engwall
P450 CYP1A as measured by EROD activity (Fisk et al.,
et al., 1993; 1994). Male and female chickens fed differ-
1996).
ent doses of Halowax 1014 have been mated and egg
production studied. At higher doses (20 mg/kg), egg
Immunosuppression
hatchability was reduced and no eggs were produced in
There is no information on the immunotoxicity of
chickens fed the highest dose of 100 mg/kg (Pudelkie-
SCCPs.
wicz et al., 1959).
In another study, Halowax 1014, 1013 or 1051 were
Thyroid effects
nanoinjected into fertilized medaka (Oryzias latipes)
UDPGT is produced in the liver and decreases plasma
eggs at various dose levels (0.3-30 ng/egg) and the em-
T4 levels, stimulating TSH release by the pituitary
bryos allowed to develop to adulthood and sexual matu-
gland. Rats exposed to 1000 mg/kg body weight of two
rity (Villalobos et al., 2000). Early life stage and early
SCCPs (56% and 58% Cl) showed two-fold increases in
adult life stage assessments were carried out. Halowax

Chapter 3 ท Toxicology
33
1014 was found to be more toxic than Halowax 1013
Thyroid and retinol effects
and 1051. The 16- day LD50 (dose that kills 50% of the
Sandau et al. (2000) found that 4-hydroxy-heptachlo-
exposed animals) for Halowax 1014 in embryos was
rostyrene (4-OH-HpCS), a metabolite of OCS, had a
4.2 ng/egg and death was caused by cardiovascular
similar affinity for human TTR as T4, the native hor-
abnormalities. The lowest-observed-adverse-effect level
mone. The potential of 4-OH-HpCS to bind to TTR
(LOAEL) was 3.0 ng/egg with hemorrhage and yolk sac
makes it capable of disrupting thyroid hormone trans-
edema as the major effects. Halowax 1014 decreased the
port and potentially affecting circulating retinol concen-
gonadosomatic index in adult females. Halowax 1013
trations.
caused high mortality at 10 ng/egg and premature hatch-
ing of embryos at all doses. Halowax 1051 was the least
3.3.1.5. Update on PBDEs
toxic PCN.
PBDEs are numbered according to the same system as
Cytochrome P450-dependent monooxygenases
PCBs, based on chlorination degree and placement of
Three-spined sticklebacks (Gasterosteus aculeatus) fed
the chlorines on the two aromatic rings. Long-term ex-
the commercial PCN mixture Halowax 1014 have
posure to DeBDE has been found to induce thyroid hy-
shown a dose-related increase in EROD activity as well
perplasia, hepatocellular and thyroid adenomas, and
as lipid accumulation in the liver (Holm et al., 1993).
carcinomas in mice (Great Lakes Chemical Corporation,
Rainbow trout fry show dose-related increases in EROD
undated; 1987).
activity after microinjection of Halowax 1014 at the em-
bryo stage (Norrgren et al., 1993). A commercial mix-
Reproductive and developmental effects
ture of tetra-, penta- and hexachlorinated PCN (Ha-
The estrogenic potency of several BDEs was tested using
lowax 1014) as well as a mixture of 1,2,3,5,6,7- and
the estrogen receptor (ER)-CALUX bioassay, and the
1,2,3,4,6,7-HxCNs both caused liver enzyme induction
most potent congeners in descending order were BDEs
in chick and eider duck embryos (Engwall et al., 1993;
100, 75, 51, 30, and 119 (Brouwer et al., 2001; Meerts
1994). A HpCN congener had much lower EROD in-
et al., 2001). Potency was much less than for estrogen.
duction potency.
BDEs 166 and 190 were antiestrogenic.
Three in vitro bioassays were used to test the ability
Neonatal exposure to 2,2',4,4'-TeBDE (BDE47)
of 18 PCN congeners to induce CYP1A activity (Vil-
(10.5 ตg/g body weight), 2,2',4,4',5-PeBDE (BDE99)
leneuve et al., 2000). The PLHC-1 fish hepatoma cell
(0.8 or 12 ตg/g body weight) or 2,2',4,4',5,5'-HxBDE
bioassay was fairly insensitive to PCNs but the EROD
(BDE153) (0.9 or 9 ตg/g body weight), administered
and luciferase assays using recombinant H-4-II E rat
orally to mice on day 10, induced permanent aberra-
hepatoma cells were more sensitive. The HxCN con-
tions in spontaneous motor behavior which worsened
geners tested were most potent, followed by the PeCNs.
with age (Eriksson et al., 2001; Viberg et al., 2001a;
The TeCNs, TrCNs, DiCNs, and MoCNs tested were
2002). Neonatal exposure to BDE99 (12 ตg/g body
less active.
weight) or BDE153 (0.9 or 9 ตg/g body weight) also af-
Rainbow trout sac fry were treated with Halowax
fected learning and memory functions in the adult ani-
1014, a mixture of 1,2,3,4,6,7- and 1,2,3,5,6,7-HxCN,
mal. BDE209 (2.22 or 20.1 ตg/g body weight) adminis-
or 1,2,3,4,5,6,7-HpCN injected into the yolk sac (Peso-
tered orally to neonatal mice on day 3 induced perma-
nen et al., 2000). After two weeks, immunohistochemi-
nent aberrations in spontaneous motor behavior, but
cal analysis was performed for CYP1A expression and
when administered on day 10, had no effect (Viberg et
was most pronounced in the hepatocytes. Exposure of a
al., 2001b). BDE99 exposure (0.6, 6 and 30 ตg/g body
primary cell culture of trout hepatocytes to these PCNs
weight/day) during pregnancy to post-natal day 21 in
led to increased EROD activity and CYP1A mRNA con-
mice caused increased hyperactivity in offspring (Bran-
tent, with the HxCN mix being most potent followed by
chi et al., 2002).
HpCN and then Halowax 1014.
In a follow-up study to Eriksson et al. (2001), Eriks-
son et al. (2002) investigated whether there is a critical
time in neonatal mouse brain development for induction
3.3.1.4. OCS
of the neurotoxic effects of BDE99. One single oral dose
Long-term dietary exposure of rats to OCS causes ele-
of 8 ตg/g body weight (14 ตmol/kg body weight) was
vated serum cholesterol and histological changes in the
administered to 3-day, 10-day and 19-day-old mice. The
thyroid, liver, and kidney of rats (Chu et al., 1986a).
mice exposed to BDE99 on day 10 showed significant
behavior aberrations, as was previously seen in Eriksson
Reproductive and developmental effects
et al. (2001), and mice exposed on day 3 showed similar
OCS was found to have binding affinity for both the an-
aberrations but to a lesser degree. The mice exposed on
drogen and estrogen receptor in an in vitro assay (Satoh
day 19 showed no significant change from the controls.
et al., 2001).
Uptake and retention of BDE99 in the brain was also
studied by administering 14C-labelled BDE99 to 3-day,
Cytochrome P450-dependent monooxygenases
10-day and 19-day-old mice (Eriksson et al., 2002). The
OCS significantly induced CYP1A activity in mice
retention of BDE99 in mice exposed on day 3 indicates
(Smith et al., 1994). Long-term dietary exposure to high
that the effects observed may be due to the amount of
doses of OCS induced the CYP2B enzymes aniline hy-
BDE99 still present in the brain on day 10. The neuro-
droxylase and aminopyrine N-demethylase (APND) in
toxic effects seem to involve changes in the cholinergic
male and female rats, with males being more sensitive
system as mice given BDE99 on day 10 and then chal-
(Chu et al., 1986a).
lenged as adults with a low dose of nicotine behaved

34
AMAP Assessment 2002: Persistent Organic Pollutants in the Arctic
completely the opposite of controls. From these studies,
Studies in whole cultured chick embryo liver showed
it was concluded that the window for permanent effects
induction of EROD activity after exposure to BDEs 47,
of BDE99 and BDE47 is day 10 in neonatal mice
99, and 153 as well as the commercial mixture Bromkal
(Eriksson et al., 2001; 2002).
70-5 DE (Pettersson et al., 2001). BDE99 was most po-
In Sinjari et al. (1998b), female rats given BDE47
tent but was much less potent than TCDD (TEF of
orally for a period of two weeks were then killed and the
0.000004).
choroid plexus of the brain removed, homogenized and
Microsomal enzyme activities were studied in rats
incubated with 125I-T4. Compared to controls, there was
and mice, and results from Bromkal 70-5 DE, and BDE
a dose-dependent reduction in the binding of 125I-T4
47 showed induction of EROD, MROD, and PROD in
to the choroid plexus. In contrast, in vitro incubation of
both species (Hallgren and Darnerud, 2002; Hallgren et
rat choroid plexus with BDE47 revealed no competitive
al., 2001). Rats exposed to Bromkal 70-5 DE orally for
inhibition of labeled T4 binding. This indicates that
28 days had dose-dependent increases in EROD and
BDE47 metabolites can cross the bloodญbrain barrier
PROD activities (Fattore et al., 2001).
and bind to the choroid plexus T4-binding sites. This in
Rainbow trout dosed orally with BDE47 or BD99
turn could cause interference of T4 transport to the
for 6 and 22 days had significantly inhibited EROD ac-
brain, with risks for effects on neural development.
tivity in the liver with BDE47 being most powerful in
BDE99 at 10 ตM induced death of 23% of cerebellar
this effect (Tjไrnlund et al., 1998).
neurons in an in vitro neurotoxicity test (Llansola et
al., 2001). Higher concentrations induced more neu-
Immunosuppression
ronal death.
Mitogen-induced DNA synthesis and immunoglobulin
Microinjection of BDE47, 2,2',3,4,4'-PeBDE (BDE
synthesis by human lymphocytes in vitro were examined
85) or BDE99 into newly fertilized rainbow trout eggs
after exposure to purified BDEs 47 and 85. No effects
in an early life stage mortality bioassay showed no ef-
on mitogen-induced proliferation or immunoglobulin syn-
fects compared to TCDD (Hornung et al., 1996). Expo-
thesis were observed (Fernl๖f et al., 1997). The results
sure to low levels of BDE47 affected developmental
indicate that proliferation and immunoglobulin synthe-
rates in the invertebrate Acartia tonsa, and juveniles
sis are insensitive to the direct action of PBDEs.
were more sensitive than adults (Breitholtz et al., 2001).
Immunotoxicity was studied after oral treatment
This may implicate disruption of juvenile hormones or
with Bromkal 70-5 DE or BDE47 in rats and mice (Dar-
ecdysteroids.
nerud and Thuvander, 1998). In mice, BDE47 caused re-
duced splenocyte number as reflected in decreased
Cytochrome P450-dependent monooxygenases
numbers of CD45R+, CD4+, and CD8+ cells in spleens.
Using a recombinant H- 4-II E rat hepatoma cell line
In mice treated with Bromkal 70-5 DE, absolute num-
having Ah receptor mediated expression of a luciferase
bers of double negative thymocytes were significantly
reporter gene (the dioxin receptor (DR)-CALUX assay)
lower than in controls, and mice also showed reduced
(Aarts et al., 1995; Murk et al., 1998), a number of indi-
production of IgG. No effects were seen in rats. Thus,
vidual PBDE congeners have been tested for their po-
BDE47 and Bromkal 70-5 DE, which contains BDE47,
tency to activate/deactivate the Ah receptor (Meerts et
both seem to be immunotoxic in mice.
al., 1998). In order to study antagonism, the same PBDE
congeners were also tested in the presence of TCDD. Of
Thyroid and retinol effects
the 17 PBDE congeners tested, seven (BDEs 32, 85, 99,
Seventeen PBDE congeners (BDEs 15, 28, 30, 32, 47,
119, 153, 166, 190) showed ability to activate the Ah re-
51, 71, 75, 77, 85, 99, 100, 119, 138, 153, 166, and
ceptor. Potencies could only be determined for BDEs
190) were incubated individually with rat hepatic micro-
166 and BDE190 and are in the same range as mono-
somes from rats treated with -naphthaflavone, pheno-
ortho PCB congeners 105 and 118 (Sanderson et al.,
barbital or clofibrate (Meerts et al., 2000). The parent
1996). Some congeners such as BDEs 85, 99, and 119
congeners and the metabolites formed were then tested
showed both agonist and antagonist activities depending
for their ability to compete with T4 for binding to
on the concentration tested. Nine congeners, including
human TTR in vitro. Results showed no competition
BDEs 15, 28, 47, 77, and 138, showed antagonist activ-
with the parent compounds, but considerable potency
ities against TCDD. The observed antagonism may be
for several of the metabolites, indicating the metabolism
due to competition between PBDEs and TCDD at the
of PBDE to hydroxylated PBDE. The results indicate
Ah receptor level. In a recent study, more BDE congeners
that hydroxylated metabolites of PBDE may be potent
have been tested in the DR-CALUX assay and BDEs 30,
competitors of T4 and could disrupt normal thyroid
47, 51, 71, 75, and 100 have also been found to activate
hormone function in wildlife and humans if present. No
the Ah receptor, but their potencies could not be deter-
binding competition was seen for several of the higher
mined (Brouwer et al., 2001).
brominated PBDEs such as BDEs 138, 153, 166 and 190
EROD induction was studied in chick and rat hepa-
after incubation with the microsomes. This may indicate
tocytes, liver cell lines from rainbow trout, rats and hu-
that these congeners are not readily metabolized.
mans, and in a human intestinal cell line (Chen et al.,
The Bromkal 70-5DE product causes decreased thy-
2001). BDEs 77, 100, 119 and 126 induced the greatest
mus weight, increased liver/body weight ratios in mice
EROD activity in all cell types, but were less potent than
and decreased T4 in rats and mice (Fowles et al., 1994;
TCDD. BDEs 153 and 183 were weaker inducers. BDEs
Darnerud and Sinjari, 1996; Hallgren et al., 2001). De-
47 and 99 were not inducers in any cell line. The con-
creases in T4 were also seen when rats and mice were
geners that did not induce EROD also failed to bind to
treated with the single congener BDE47, but no effects
the Ah receptor.
on TSH were seen for BDE47 or Bromkal 70 (Darnerud

Chapter 3 ท Toxicology
35
and Sinjari, 1996; Hallgren et al., 2001). Bromkal 70-5
birth, and close to one-third of first generation pups in
DE and BDE47 also caused significant reductions in he-
the 1.6 mg/kg/day group died within four days of birth.
patic vitamin A concentrations in rats, and Bromkal 70-
Only pups in the 0, 0.1 and 0.4 mg/kg/day groups were
5 DE caused reductions of hepatic vitamin A concentra-
carried to the second generation. For second-generation
tions in mice at high doses (Hallgren et al., 2001). Simi-
offspring, reductions in pup weight and reversible delays
larly, Bromkal 70-5 DE caused dose-dependent reduc-
in reflex and physical development were seen in the
tion in hepatic vitamin A in rats dosed orally for 28 days
high-dose groups. For the second-generation offspring,
(Fattore et al., 2001).
the NOAEL for reduced pup weight was determined to
In subsequent studies, the interactive effects of differ-
be 0.1 mg/kg/day and the LOAEL, 0.4 mg/kg/day. These
ent organohalogen compounds (PCB, PBDE, and chlo-
doses corresponded to PFOS liver concentrations of 15
rinated paraffins (CP)) on T4 levels and microsomal
ตg/g and 58 ตg/g ww, respectively, in the rats.
enzyme activities were tested (Hallgren and Darnerud,
In rabbits, PFOS caused maternal toxicity (decreased
2002). Female rats were orally exposed to single com-
body weight gain) at a dose of 1.0 mg/kg/day or higher
pounds or combinations at isomolar concentrations dai-
(Case et al., 2001). Levels causing maternal toxicity also
ly over 14 days. The results show that PCBs (Aroclor
led to increased abortions and reduced fetal weights.
1254) and PBDEs (BDE47) significantly reduce the T4
levels in rats, in the actual exposure interval (6 -18 mg/
Cancer
kg body weight/day), with Aroclor 1254 resulting in the
PFOA and PFOS are not mutagenic but are known to be
strongest effect. EROD and MROD, but to a lesser ex-
liver tumor promoters in rats. Perfluorinated fatty acids
tent PROD and UDPGT, activities were negatively corre-
such as PFOA and PFOS increase peroxisome levels and
lated to T4 effects. Regarding the mixed BDE47 + CP
inhibit GJIC (Upham et al., 1998). PFOS is almost as po-
group, a synergistic decrease in free T4 levels and in-
tent as PFOA in causing increased peroxisome prolifera-
crease in EROD activity was observed.
tion (Sohlenius et al., 1993). Several other perfluori-
nated compounds (perfluorooctanoic sulfonamide, per-
Cancer
fluorohexane sulfonate) also affect GJIC (Hu et al.,
BDE47 was shown to induce a statistically significant in-
2001).
crease in intragenic recombination when studied in one
of two tested in vitro assays using mammalian cells
3.3.2. Persistent organic pesticides
(Helleday et al., 1999). This may indicate that BDE47
can induce cancer via a non-mutagenic mechanism.
3.3.2.1. Update on toxaphene
A review of the effects of technical toxaphene was given
in the previous AMAP assessment report. The toxicity
3.3.1.6. PFOS and PFOA
data available at that time were rather limited. The fol-
PFOS is a surfactant with both lipophobic and hy-
lowing is an update on relevant effects studies that have
drophobic properties. Therefore, it does not accumulate
been published recently. For a recent review on toxa-
in lipids, but instead accumulates in the liver, gall blad-
phene, including toxicology, see de Geus et al. (1999).
der and the blood (where it binds to proteins). It is spec-
ulated that the body treats PFOS as a bile acid. The liver
Reproductive and developmental effects
makes bile acids from cholesterol, which are excreted
Technical toxaphene, T2 (Parlar 26), and T12 (Parlar
from the gall bladder into the intestine to facilitate the
50) were tested for their estrogenicity using the MCF7-
emulsification and uptake of fats in the gut. The bile
E3 human breast cancer cell model and were found to
acids are then recycled back into the liver via enterohep-
have weak estrogenic activity (Stelzer and Chan, 1999).
atic circulation. PFOS may also weaken cell membranes
T2 and T12 had lower proliferative effects on the cells
(Hu et al., 2000). PFOA is used as a lubricant, detergent,
than technical toxaphene, and T2 was more potent than
and wetting agent (Guethner and Vietor, 1962).
T12. Effects of mixtures of the three indicated additive
Liver enlargement and reduced serum cholesterol
effects, and none of the compounds had effects on estro-
levels are early responses to exposure to PFOS. Acute
gen receptor or progesterone receptor levels. In another
toxicity with 100% mortality was seen in rhesus mon-
study using MCF-7 cells, technical toxaphene, T2, and
keys fed 10 mg/kg/day for three weeks in one experi-
T12 were tested for their effects on estrogen receptor
ment and in rhesus monkeys fed 4.5 mg/kg/day for seven
function (J๘rgensen et al., 1997c). The results indicated
weeks in another (Seed, 2000). Even doses of 0.75 mg/
that toxaphene and T12 are antiestrogens, and that the
kg/day led to changes in cynomolgus monkey (Macaca
effects occur at the gene transcription level. In support
fascicularis) livers, reductions in blood cholesterol, dis-
of this, toxaphene tested in a battery of assays for estro-
interest in food, and death. PFOA also causes liver en-
genic activity (MCF-7 cells, competitive receptor bind-
largement and increased liver lipid levels in rats and
ing) was found to be weakly antiestrogenic (Arcaro et
mice (Kawashima et al., 1995; Kudo and Kawashima,
al., 2000). In a battery of estrogenic screening methods,
1997; Kudo et al., 1999).
including mouse uterus, MCF-7 human breast cancer
cells, and yeast-based reporter gene assays, toxaphene
Reproductive and developmental effects
was found to exert minimal estrogenic effects (Rama-
In a two-generation reproductive toxicity study of PFOS
moorthy et al., 1997).
in rats, pup-survival in the first generation was signifi-
Technical toxaphene and congeners T2 and T12 were
cantly decreased in the two highest dose groups receiv-
tested for their effects on cultured rat embryos during
ing 1.6 and 3.2 mg/kg/day (Seed, 2000). All first-genera-
the period corresponding to a critical period of morpho-
tion pups in the high-dose group died within one day of
genesis and organogenesis (gestational days 10-12).

36
AMAP Assessment 2002: Persistent Organic Pollutants in the Arctic
Technical toxaphene and both single congeners caused
Immunosuppression
significant changes in total morphology, somite number,
Cynomolgus monkeys were treated with toxaphene at
head and crown-rump length, and central nervous sys-
1 mg/kg body weight/day for 52 weeks to study immune
tem scores of the embryos, including a high incidence of
effects (Tryphonas et al., 2000). Effects were seen that
central nervous system defects (Calciu et al., 1997). The
were not statistically significant, but which indicated
altered total morphology, including decreases in head
possible negative effects of long-term exposure to toxa-
and crown-rump lengths indicate that toxaphene and
phene on humoral immunity.
the two single congeners retard growth and morpholog-
ical development. There were differences in potency and
Thyroid and retinol effects
type of toxicity, indicating that specific congeners can
Toxaphene caused thyroid follicular cell hypertrophy,
produce effects not predicted by the technical mixture.
intrafollicular hyperplasia and increased production of
Technical toxaphene fed to female zebrafish for two
TSH in rats given 100 mg/kg/day for three days (Waritz
weeks at doses of 20, 230, and 2200 ng/g body weight/
et al., 1996). No changes were seen for T3 or T4. The
day was found to have no effects on total number of
mechanism proposed for this effect was that toxaphene
eggs spawned, percentage of fertilized eggs, percentage
induced cytochrome P450 enzymes (CYP2B type) which
of embryo mortality or percentage of hatching (Fๅhr-
led to increased excretion of T4, thereby stimulating the
aeus -Van Ree and Payne, 1997). Toxaphene, however,
pituitary gland to produce and excrete more TSH. This,
did cause a dose-related decrease in the percentage of
in turn, led to thyroid hypertrophy and hyperplasia.
oviposition.
The ability of toxaphene to displace native ligands
Cancer
from the estradiol receptor, testosterone receptor and
Toxaphene is a potent carcinogen in rats and mice (re-
cortisol receptor was tested using rainbow trout liver
viewed in Reuber, 1979; Saleh, 1991; de Geus et al.,
and brain tissues (Knudsen and Pottinger, 1999). Toxa-
1999). Toxaphene induces malignant liver tumors, retic-
phene did not bind to any of the receptors and was con-
ulum cell sarcomas, uterine sarcomas, reproductive sys-
cluded not to be estrogenic.
tem tumors, mammary gland tumors, and tumors in the
In red-eared slider turtle, incubation temperature de-
pituitary, adrenal and thyroid glands.
termines the sex of hatchlings, but male-producing tem-
Toxaphene is mutagenic in the Ames test (Hooper et
peratures can be overridden if the eggs are exposed to
al., 1979; Mortelmans et al., 1986). When toxaphene
estrogenic compounds. Toxaphene was tested for its es-
and four single congeners (Parlars 26, 50, 62, and 32)
trogenic activity by application on eggs set to become
were tested for mutagenicity using two different bacter-
males (Willingham and Crews, 1999). No estrogenic
ial (Salmonella typhimurium) strains, toxaphene was
effects were seen for toxaphene. Similarly, toxaphene
mutagenic but the single congeners were not (Steinberg
showed no estrogenic effects when tested for its ability
et al., 1998). Schrader et al. (1998) found that technical
to displace native estradiol from alligator or human es-
toxaphene was mutagenic in all five S. typhimurium
trogen receptors (Vonier et al., 1996; Arnold et al.,
strains tested but high concentrations were required.
1997). However, Palmer et al. (1998) found that water
However, no mutagenesis was seen for toxaphene when
exposure to toxaphene induced significant vitellogenin
tested in Chinese hamster V79 lung fibroblasts. Toxa-
production in the male African clawed frog (Xenopus
phene and Parlar 32 were found to be genotoxic using
laevia), which is an estrogenic effect.
the Mutatox assay, but Parlars 26, 50, and 62 were not
genotoxic (Boon et al., 1998).
Cytochrome P450-dependent monooxygenases
Low concentrations of toxaphene induce micronuclei
Subacute levels of toxaphene given to guinea pigs led to
formation in vitro in beluga skin fibroblasts (Gauthier et
induced cytochrome P450 and increased aniline hydrox-
al., 1999).
ylase in the liver and kidney (Chandra and Durairaj,
Non-cytotoxic concentrations of toxaphene inhib-
1993). In mice, toxaphene exposure led to increases in
ited GJIC in normal human breast epithelial cells, in a
CYP2B levels (Hedli et al., 1998). Rats and Japanese quail
dose-dependent manner (Kang et al., 1996), indicating it
exposed to single doses of technical toxaphene ranging
may be a tumor promoter. No DNA adducts were found
from 0.012 to 40 mg/kg body weight showed induced
in the livers of mice treated with toxaphene (Hedli et al.,
P450 systems only at the highest dose (Drenth et al.,
1998). Toxaphene has been found to down-regulate the
2000). These included increased PROD, formation of
retinoblastoma gene, a tumor suppressor gene, indicat-
15 -hydroxytestosterone and 2-hydroxyestradiol in the
ing that toxaphene could promote tumors by turning off
rat, and increased formation of 6 -, 15 - and 16 -hy-
tumor suppression (Rought et al., 1999).
droxytestosterone in the quail. The doses required to in-
duce the P450 system were close to those known to cause
3.3.3. Other pesticides
mortality, so it was concluded that P450 activity induction
is probably unlikely in wildlife exposed to toxaphene.
3.3.3.1. TBT and its metabolites (DBT, MBT)
In cynomolgous monkeys, toxaphene treatment in-
TBT is one of the most toxic substances deliberately in-
duced aminopyrene, MROD, and EROD activities, indi-
troduced into natural waters (Goldberg, 1986). TBT is
cating that toxaphene is a mixed-type inducer that in-
moderately lipophilic and bioconcentrates and bioaccu-
duces both CYP1A and 2B (Bryce et al., 2001). Liver en-
mulates in the marine environment (Tanabe, 1999;
zyme induction occurred at doses that did not cause any
Maguire, 2000; Hoch, 2001), and is a classic endocrine
toxic effects in the monkeys, indicating that they may be
disrupter (Matthiessen and Gibbs, 1998). For recent re-
less sensitive to the toxic effects of toxaphene than labo-
views on TBT toxicology, see Fent (1996a), Maguire
ratory rodents.
(2000) and WHO (1999). Chronic effects are observed

Chapter 3 ท Toxicology
37
at exposure levels of 1000 ng/L or less for oysters, mus-
liver weight, and liver cell hyperplasia (Seinen et al.,
sels and crustaceans (Rexrode, 1987), while the most
1981). A NOAEL of 40-50 ng/L for rainbow trout yolk
sensitive species (dogwhelk snails, e.g., Nucella lapis)
sac fry has been proposed (De Vries et al., 1991). TBT
show sublethal effects at concentrations of only a few
exposure in medaka embryos on day 0 led to concentra-
ng/L or less (Bryan and Gibbs, 1991; Gibbs, 1993; Stew-
tion-related lethality, reductions in hatching success, in-
art and Thompson, 1994).
creases in gross abnormalities (bent, curled and/or short-
Recently (since the mid-1990s), studies have been
ened tails caused by reduced number of somites) and
published showing the occurrence of TBT and other
slowed developmental rates (Bentivegna and Piatkow-
butyltins (MBT, DBT) in fish, birds, and terrestrial and
ski, 1998). The LOAEL for TBT combined chronic ef-
marine mammals from the Pacific Ocean (Japan, Aus-
fects in medaka embryos was 12 500 ng/L.
tralia, Taiwan, India, Bangladesh, Thailand, Vietnam,
TBT-oxide (TBTO) exposure (0, 2.7 and 9 ng/ml) for
Indonesia, Alaska, U.S., and open ocean areas) (Iwata et
four days led to significant changes in spatial position,
al., 1995; Kannan et al., 1995a; 1995b; Guruge et al.,
response to predator attack, recovery time and latency
1996;1997; Kim et al., 1996a; 1996b; 1996c; Takahashi
time in three-spined stickleback (Wibe et al., 2001).
et al., 1997; 1999; Tanabe et al., 1998), in the Baltic Sea
Dietary exposure of Japanese quail (Coturnix co-
(Kannan and Falandysz, 1997), on the U.S. Atlantic and
turnix japonica) to 60 000 and 150 000 ng TBTO/g
Gulf coasts (Kannan et al., 1997), in a freshwater lake in
food led to reduced egg hatchability and an increase in
the Netherlands (Stไb et al., 1996), and along the coast
the percent of chicks found dead in the shell (Coenen et
of Italy (Kannan et al., 1996).
al., 1992; Schlatterer et al., 1993). The no-observed-ef-
Butyltins concentrate in the liver, blubber, and mus-
fect concentration (NOEC) for reduced egg weight and
cle of vertebrates. Higher relative amounts of MBT and
hatchability was 60 000 ng/g food (Schlatterer et al.,
DBT are found in birds and mammals as compared to
1993).
fish, due to metabolism of TBT.
Cytochrome P450-dependent monooxygenases
Reproductive and developmental effects
TBT exposure to 3.3, 8.1 and 16.3 mg/kg (intraperito-
TBT has been found to have high binding affinity to the
neal injection) led to the concentration-dependent degra-
androgen receptor in an in vitro assay but shows no
dation of CYP1A and loss of EROD activity in vitro and
affinity for the estrogen receptor (Satoh et al., 2001).
in vivo in scup (Stenotomus chrysops) (Fent and Stege-
At concentrations of 1-2 ng/L, dogwhelk snails ex-
man, 1991; 1993; Fent et al., 1998). The lowest dose
hibit imposex. Imposex is caused by TBT interference
level resulted in liver TBT concentrations of 8000 ng/g.
with the biosynthesis of steroid hormones (i.e., the syn-
At the high dose levels used, CYP2B (responsible for
thesis of 17 -estradiol from testosterone and the synthe-
testosterone 15 -hydroxylase activity) and CYP3A (re-
sis of estrone from androstenedione) (Bettin et al.,
sponsible for testosterone 6 -hydroxylation) proteins
1996). High levels of testosterone result in the develop-
were also destroyed, which could lead to effects on
ment of a penis and vas deferens in female neogastro-
steroid metabolism and possible endocrine disruption.
pods. At levels of 7-10 ng/L, the vas deferens can over-
TBT exposure in vitro in rainbow trout, bullhead
grow the genital opening of the female, resulting in re-
(Cottos gobio), and eel (Anguilla anguilla) hepatic mi-
productive failure of the species (Gibbs et al., 1987).
crosomes led to strong inhibition of EROD activity and
Thresholds for other species are poorly known. Studies
reduction in total cytochrome P450 protein levels (Fent
with cod (Gadus morhua) (Granmo et al., 2002) indi-
and Bucheli, 1994), with rainbow trout microsomes
cate that threshold levels for cod embryos are higher
being most sensitive. Rainbow trout hepatocytes also
than found in the Arctic environment.
showed reduced cytochrome P450 levels after exposure
Imposex has been observed most frequently in
to 1 ตM TBT (290 000 ng/L) (Reader et al., 1996). In
coastal areas near obvious TBT sources such as marinas
the fish hepatoma cell line PLHC-1, TBT exposure led to
or harbors, and has been associated with TBT paints on
inhibition of EROD, and decreased levels of CYP1A and
both pleasure boats and commercial shipping. Sedi-
DBT exposure also led to inhibition of EROD, but was
ments in particular seem to be reservoirs for TBT espe-
less potent than TBT (Br�schweiler et al., 1996). These
cially after use has stopped, leading to continued expo-
studies indicate that TBT is metabolized by the P450
sure (Maguire, 2000). Open ocean areas are exposed to
monooxygenase system but that it also inhibits this sys-
TBT from large vessels that are still allowed to use TBT
tem, affecting its own metabolism and that of other sub-
on their hulls, with imposex being found in whelks
stances. TBT and DBT both inhibit carboxylesterases in
(Buccinum undatum) collected from the open North
the tropical marine fish Siganus canaliculatus, with DBT
Sea along shipping routes, indicating that the problem
being most potent (Al-Ghais et al., 2000). The IC50s
is not confined to coastal areas (Ellis and Pattisina,
(concentrations at which 50% inhibition occurs) were
1990; Bryan and Gibbs, 1991; Ten Hallers-Tjabbes et
180-385 ตM (52 200 -112 000 ng/g) for TBT and 17-49
al., 1994).
ตM (4000 -11 400 ng/g) for DBT.
TBT causes developmental effects in the early life
In vitro inhibition of hepatic cytochrome P450 in
stages of fish. Water concentrations of 690 - 820 ng/L
Dall's porpoise (Phocoenoides dalli) and Steller sea lion
cause scoliosis and an inability to swim in minnows
hepatic microsomes by TBT has also been shown (Kim
(Phoxinus phoxinus) (Fent, 1992) and notochord length
et al., 1998b). Total P450 levels decreased and EROD
is significantly reduced in larval striped bass (Morone
and PROD activity decreased with increasing doses. The
saxatilis) at 514 ng/L (Pinkney et al., 1990). Long-term
apparent effect threshold concentration was estimated
exposure of rainbow trout yolk sac fry to 1000 ng/L re-
to be 100 ตM TBT (29 000 ng/g). Comparison of the
sulted in decreased numbers of red blood cells, increased
composition of TBT and its metabolites DBT and MBT

38
AMAP Assessment 2002: Persistent Organic Pollutants in the Arctic
in marine mammals and their prey indicates that pin-
ostasis, formation of reactive oxygen species and DNA
nipeds (seals, sea lions) have a higher metabolic capacity
fragmentation (Gennari et al., 2000).
for organotins than cetaceans (dolphins, porpoises,
TBT is also immunotoxic in rainbow trout immune
whales) (Tanabe et al., 1998; Tanabe, 1999). Both TBT
cells, but DBT has been found to be more potent as an
and DBT are hepatotoxic in mice, with DBT being more
immunotoxin (O'Halloran et al., 1998). Both TBT and
potent than TBT (Ueno et al., 1994). TBT toxicity in this
DBT were found to suppress mitogenic activity in trout
study was attributed to the metabolism of TBT to DBT
head kidney and splenic cells at exposure concentrations
and subsequent accumulation of DBT. The effect thresh-
of 50 ng/g or greater (154 nM). Studies in rainbow trout
old for hepatotoxic effects in mice was 2600 ng DBT/g
in vivo also show similar types of effects to those seen
ww in liver.
in mammals, such as lymphoid depletion and immune
modulation (Schwaiger et al., 1992). In channel catfish
Immunosuppression
(Ictalurus punctatus), TBT affected humoral immunity
Both TBT and DBT have been found to be immunosup-
as measured by response to Edwardsiella ictaluri infec-
pressive in a range of animals, causing thymic and
tion (Rice et al., 1995). The effects were significant for
splenic atrophy, reductions in thymic, circulating and
all doses (10, 100, and 1000 ng/g administered intra-
splenic lymphocytes, suppression of T-cell-dependent
peritoneally), but the strength of the response was dose-
immunity, and suppression of tumoricidal activity (Sei-
dependent. Flounder (Platichthys flesus) exposed to fairly
nen et al., 1977; Krajnc et al., 1984; Vos et al., 1984;
high TBT concentrations (ตg/L) had a significant decrease
Snoeij et al., 1985; 1988). For a review of organotin im-
in thymus volume (Grinwis et al., 1998).
munotoxicity, see Snoeij et al. (1987). DBT and TBT
In Canada, studies have shown effects of TBT, DBT,
also suppressed concanavalin-A-induced mitogenesis in
and MBT on the in vitro phagocytic activity of hemo-
peripheral blood monocytes of marine mammals
cytes from three marine bivalve species, Mytilus edulis,
(Nakata et al., 2002). In vitro immunotoxicity is seen in
Mya arenaria, and Mactromeris polynyma, using flow
rat thymocytes (Snoeij et al., 1986) and rabbit polymor-
cytometry (Bouchard et al., 1999). Phagocytosis was re-
phonuclear leucocytes (Elferink et al., 1986) with an ef-
duced with increasing doses of TBT and DBT, and the
fect threshold less than 1.0 ตM TBT (290 ng/g ww) in
toxicity of butyltins on hemocytes decreased in the order
both studies. A NOEL of 500 ng/g body weight has been
DBT > TBT > MBT. The comparison of the relative sensi-
proposed for TBTO immunotoxicity in rats (Verdier et
tivity of the three species showed that blue mussels (M.
al., 1991). The most sensitive in vivo effect of TBT-in-
edulis) were more tolerant of butyltin compounds than
duced immunosuppression is seen in weanling rats, as
both clam species.
measured by reduced IgE titers and reduced resistance to
parasitism by Trichinella spiralis in muscle (Vos et al.,
Cancer
1990). The NOAEL was determined to be 25 ตg/kg/day
Generally, negative results have been obtained when TBT
(equivalent to 500 ng/g in diet), and the LOAEL was
is tested for mutagenicity in various test systems (WHO,
250 ตg/kg/day (equivalent to 5000 ng/g in diet).
1990). However, Hamasaki et al. (1993) have shown
One possible mechanism that has been proposed for
that MBT, DBT and TBT were mutagens in S. typhimu-
TBT immunotoxicity is the induction of apoptosis, or
rium TA100. Carcinogenicity has not been well investi-
programmed cell death, in the thymus (Pieters et al.,
gated. Since the data on genotoxicity of TBT are not con-
1994). This may be related to alterations of Ca2+ home-
sistent, it is not possible to draw definite conclusions.