163
Chapter 6
Biological Effects
ญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญ
Biological effects can be measured at different levels of
to very different PCB compositions, expressed as
biological organization, from the molecular to the
PCBs, than is seen in a PCB technical product. As an
ecosystem level. Biomarkers measurable at a molecular
illustration, a weathered mixture of PCB in contami-
level respond early but are not readily interpreted eco-
nated fish was found to be more toxic to mink than a
logically, while measures with established ecological
comparable PCB concentration of a technical product
relevance, such as population declines or reduced re-
(Aroclor 1254) (Giesy and Kannan, 1998).
productive rates, respond too late to have diagnostic or
Differences in species sensitivities to the effects of
preventive value.
POPs make it difficult to know which of the tested
In the previous AMAP assessment, there was rela-
species best represents those in the Arctic. For example,
tively little knowledge of the biological effects of POPs
several Arctic species have delayed implantation (mink,
in Arctic species, but concern was raised that the con-
otter, other mustelids, seal, walrus, and polar bear),
centrations in some species put them at risk for effects
which may make them more sensitive to the reproduc-
on reproduction, the immune system, and on subtle
tive effects of POPs than tested laboratory animals
neurobehavioral functions. Newer studies of biomark-
without delayed implantation (Sandell, 1990). Arctic
ers linked to POP exposures are beginning to show evi-
species also differ from laboratory animals because of
dence that chemical contaminants may be present in
their fat dynamics, differences in life styles and life
sufficient quantities to have biologically significant ef-
strategies, and differences in toxicokinetics. For exam-
fects on some species.
ple, there are wide differences in metabolic capacity in
Two approaches have generally been taken in iden-
different Arctic species compared to laboratory ani-
tifying and estimating the risk for possible effects. The
mals (see Section 3.2.2). Very little is known about the
first involves comparison and extrapolation. Researchers
sensitivity of Arctic species, particularly marine mam-
assess the risk for possible effects by comparing levels
mals, to the effects POPs.
of POPs in Arctic species of interest to known detri-
The second approach studies biological effects by
mental levels, with this knowledge coming from labo-
examining subtle indicators of biological responses
ratory studies, semi-field studies or from observations
(biomarkers) to contaminants. Examination of the ani-
on affected animals in the wild. Extrapolation is rou-
mals for responses known to be associated with the
tinely used in toxicology and is based on the conserved
contaminants found in their tissues is perhaps the only
nature of many of the endpoints being measured across
way to make a convincing case either for or against the
taxa. These include commonalities in endocrinology,
hypothesis that trace contaminants are acting biologi-
immunology, and to a lesser extent, anatomy and re-
cally on the animals. Almost any biological change,
productive biology. The extrapolation is based on
from molecular to ecological, can serve as a biomarker;
similarities in mechanisms of action (e.g., mediation
however, the term most often refers to changes at sub-
through the Ah-receptor, hormone receptors, CYP1A
cellular levels (McCarthy and Shugart, 1990; Huggett
induction, OH-PCB disruption of vitamin A/thyroid,
et al., 1992; Peakall, 1992). Biomarkers, typically, are
etc.). The difficulties in extrapolation relate generally
measures of normal processes that take on abnormal
to differences in sensitivity, where the same types of ef-
values as a result of exposure to chemicals of interest.
fects are seen but at different doses, or to differences in
Most of the biomarkers studied have established sensi-
structure and function that are fairly obvious among
tivities (in laboratory animals) to some of the same
species, especially those that are more distantly related
contaminants measured in Arctic marine mammals and
to each other (i.e. fish vs. birds vs. mammals) (Kim and
fish, notably several PCB congeners, PCDD/Fs, and
Hahn, 2002).
PAHs.
Comparison and extrapolation have some inherent
The MFO cytochrome P450 system, a ubiquitous
weaknesses, however. Laboratory animals are most
enzyme system common to mammals, birds, fish, and
often exposed to single POPs or technical products at
microorganisms, has probably been one of the most
high doses for short periods of time, and it is difficult
widely used biomarkers to date, with numerous labora-
to extrapolate the toxic effects seen at high acute doses
tory and field cases of responses established (Payne et
to possible adverse effects at lower but chronic expo-
al., 1987; Rattner et al., 1989; Goks๘yr and F๖rlin,
sures. Wild animals are generally exposed to lower
1992; Haasch et al., 1993; Beyer et al., 1996; Hylland
concentrations of OCs than laboratory rodents in ex-
et al., 1996). The preferred field study design has been
perimental studies, but they are exposed to mixtures of
the comparison of an exposed group of individuals
POPs and other stressors, and they are exposed over
with similar groups not exposed to the same source.
their entire lifetime.
Another design, less commonly encountered, is the
They are also exposed to weathered mixtures due to
comparison of individuals within a group to search for
the change in composition of many POP mixtures
linkages between biomarker values and exposure as in-
caused by abiotic degradation, metabolism and subse-
dicated by chemical residues. This is a valuable ap-
quent filtering up through the food web. For example,
proach for those contaminants stable enough to remain
marine mammals at high trophic levels will be exposed
identifiable as residues.

---

164
AMAP Assessment 2002: Persistent Organic Pollutants in the Arctic
There are also limitations with these types of stud-
whereas others prefer comparisons on a lipid weight
ies, however. It is not possible to determine causality,
basis. Wet weight comparisons are not always possible
only that a statistical association has been found be-
because different tissues with different lipid contents
tween a biomarker and the contaminant in question.
have been analyzed. Therefore, the values in this report
Most POPs covary, and thus, it is not possible to state
have been normalized to lipid content to enable com-
equivocally that the biomarker response has been
parison of POP concentrations in different tissues. This
caused by a particular contaminant. There may be
conversion assumes that POP concentrations are evenly
other contaminants not analyzed that are just as impor-
distributed in the lipid stores of all organs in an organ-
tant, or the response may be the result of synergistic,
ism and are not fluctuating and, therefore, that the
additive or antagonistic effects of contaminant mix-
lipid-normalized concentrations in all organs will be
tures. Biological variables such as age, sex, body condi-
the same. This is also known to not always be the case.
tion, presence of disease or other stresses also may act
In seals, however, lipid-normalized levels in blubber,
as confounders, as they can cause similar biological ef-
liver and blood in the same individual were within a
fects as those seen from POPs. This means that for
factor of 2 (Boon et al., 1994) and in other marine
most reported biological effects in wildlife, the evi-
mammals, they were usually within a factor of 2 and at
dence for a causal link with a specific chemical contam-
most, a factor of 5 (Aguilar, 1985; Boon et al., 1987;
inant is weak or non-existent. This is mainly due to the
Boon et al., 1992; Kannan et al., 1993; Bernhoft and
complexity of contaminant mixtures, the lack of chem-
Skaare, 1994; Kannan et al., 1994; Jenssen et al.,
ical exposure data, lack of data on the sensitivity of the
1996; Nakata et al., 1998b).
species concerned, and knowledge of mechanisms of
Lipid conversion also adds more uncertainty into the
action. Understanding the linkages between contami-
values since there are errors introduced from the lipid
nants and health effects (e.g., on reproduction or im-
determination. In some cases, literature threshold values
munosuppression) is most likely to come from studies
have been given on a lipid weight basis with no informa-
in laboratory animals. Crucial in establishing causal ev-
tion available for converting back to wet weight. The
idence for chemical-induced wildlife effects are semi-
comparisons for birds are somewhat less problematic,
field or laboratory studies using the wildlife species of
as thresholds have been determined in eggs and are
concern. Semi-field studies represent a useful approach
expressed on a wet weight basis, and most data avail-
to bridge the gap between the controlled conditions of
able for Arctic birds are for eggs on a wet weight basis
the laboratory experiments and the uncontrolled expo-
as well.
sure conditions in the field. Based on these different
Another weakness in this approach is that methods
types of studies, a weight-of-evidence argument can be
for quantifying PCBs used in the thresholds studies
established. For example, based on the combination of
and in the analyses of Arctic species are not identical,
field, semi-field, and laboratory studies, it was con-
and use varying numbers of congeners or are based on
cluded that dioxin-like PCBs that accumulated through
quantification using a technical product (total PCB).
the marine food chain aggravated the severity and ex-
Similarly, TEQs may have been calculated using
tent of the 1988 morbillivirus-related epizootic in har-
PCDD/F, nPCB and/or mono-ortho PCB concentra-
bour seals in northwestern Europe (Ross et al., 1996c;
tions combined with toxic equivalency factors (TEF)
Vos et al., 2003).
from different schemes. Previously, TEF schemes were
In the following assessment, results from studies
based on mammalian models, but fish- and bird-spe-
of biomarkers in Arctic biota are presented first if these
cific TEFs are now available (van den Berg et al.,
have been performed. Where possible, levels of specific
1998). Thus, the terms PCBs and TEQ may not be
POPs in Arctic biota are also compared to no-ob-
completely comparable, and this may lead to under- or
served-adverse-effect-levels (NOAEL) or no-observed-
overestimations of exposure and risk. Mammalian
effect-levels (NOEL), and lowest-observed-adverse-
TEFs are derived from studies in laboratory rodents
effect-levels (LOAEL) or lowest-observed-effect-levels
and may not be appropriate for risk assessment in ma-
(LOEL) known to cause subtle effects in sensitive spe-
rine mammals. Similarly, bird TEFs are based on stud-
cies. The purpose of these comparisons is to assess
ies in domestic chickens, which are more sensitive to
the likelihood that some Arctic species may be at risk
dioxin-like substances than fish-eating bird species
for the effects of some POPs, and to identify these
(Sanderson et al., 1998).
species, the types of effects they may be at risk for, and
Many of the mammalian thresholds for PCBs are
the contaminants that might be associated with these
based on studies in mink and otter. Mink, in particular,
effects.
have been used as a surrogate for seals in many PCB
There are some problems in making such compar-
studies, since mink also have delayed implantation.
isons, particularly for mammals, and some caution is
Mink and otter are extremely sensitive to the effects of
advised. Where possible, effects thresholds derived
PCBs and dioxin-like substances, and thresholds based
from wild species have been used, but the problems
on effects may overestimate the risk if Arctic species
with comparing different species has already been dis-
are less sensitive.
cussed above. In the case of mammals, the PCB
In a few cases, a threshold value is available only
thresholds are based on concentrations found in differ-
for laboratory species such as rats, mice, rabbits or
ent types of tissues (blood, blubber, muscle) with very
dogs (e.g., toxaphene, PFOS, TBT and DBT). Because
different lipid contents.
of the presence of these `new' substances in the Arctic,
Some scientists consider it better to compare thresh-
these thresholds have been used to indicate whether
olds based on the same tissue on a wet weight basis in
any species is at potential risk for these substances,
both the threshold species and the Arctic species,
rather than not making any statement at all. For some

---

Chapter 6 ท Biological Effects
165
Table 6.1. Selected criteria, action levels or guidelines for critical pollutants in the Great Lakes. Parts of the table are modified from De Vault
et al. (1995). All values for fish/aquatic organisms are based on wet weight, and for sediment on dry weight.
ญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญ
US FDA1
IJC2
GLI4
OMEE5
IJC3
TRG6
USEPA7
ERL8
ERM8
EQG6
EQG6
Fish
Fish
Fish
Sedi-
Sedi-
Sedi-
Contaminant
Fish
Water
Water
Water
tissue
tissue
tissue
ment
ment
Water
ment
ญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญ
2,3,7,8 TCDD/
25
0.0096
0.71
0.5
0.02
0.091
other PCDD/Fs
pg/g
pg/L
pg/g b,c
pg/g
pg/L
pg/g
as TEQs
4.75
pg/g b,d
--------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
DDT
5
0.003
0.00087
0.003
1.0
0.014
0.039
1.6
46
g/g
g/L
g/L
g/L
g/g a
g/gb
g/g
ng/g
ng/g
--------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
Total PCBs
17
0.001
0.1
0.015
0.16
23
180
pg/L
g/L
g/ga
g/gb,c
g/g
ng/g
ng/g
(0.79
pg/g TEQ )b,c
0.048
g/gb,d
(2.4
pg/g TEQ )b,d
-------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
Mirex
2
< detection
0.001
g/g
g/L
--------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
Toxaphene
5
0.008
0.008
0.0063
g/g
g/L
g/L
g/g b
--------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
Aldrin/dieldrin
0.3
0.001
0.001
0.3
g/g
g/L
g/L
g/g a
ญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญ
a
Whole fish.
b
Aquatic organism.
c
TEQ refers to dioxin toxic equivalents using toxic equivalency factors (TEFs) for mammals (van den Berg et al., 1998).
d TEQ refers to dioxin toxic equivalents using toxic equivalency factors (TEFs) for birds (van den Berg et al., 1998).
1
USFDA (U.S. Food and Drug Administration) Action Levels in edible portions of fish for regulation of interstate commerce.
2, 3 International Joint Commission Annex 1 ญ objectives for protection of aquatic life and wildlife.
4
USEPA (U.S. Environmental Protection Agency) Great Lakes Water Quality Guidance proposed criteria for protection of wildlife (USEPA,
1995).
5
OMEE (Ontario Ministry of Environment and Energy) guideline for the protection of sediment quality (OMEE, 1993).
6
Canadian Tissue Residue Guidelines for the protection of wildlife consumers of aquatic biota (Environment Canada, 2002). Total PCB
values are calculated assuming that average PCB TEQs are equal to 5 10 ญ5 of PCB concentrations, which is the ratio in Arctic char
(D. Muir, unpublished data).
7 USEPA guideline values for assessment of hazards to fish-eating wildlife (USEPA, 1995).
8 ERL = Effects range low and ERM = effects range median for sediments (Long et al., 1995).
substances such as PBDEs, no threshold data are cur-
Thresholds for effects in birds, mammals, and fish
rently available, and the concentrations in Arctic biota
can not be assessed for possible effects.
Birds
An attempt has also been made to compare POP lev-
In several review articles (Bosveld and van den Berg,
els in the diet of selected Arctic biota to known dietary
1994; Giesy et al., 1994b; Barron et al., 1995), the no-
no-observed-adverse-effect-concentrations (NOAEC) or
effect and low-effect levels for bird eggs and adults as
no-observed-effect-concentrations (NOEC), and lowest-
well as dietary intakes associated with no or low effects
observed-adverse-effect-concentrations (LOAEC) or low-
have been compiled from the literature for PCBs and
est-observed-effect-concentrations (LOEC), or to envi-
dioxin-like compounds. For eggs of fish-eating and
ronmental quality criteria/guideline values for protecting
predatory birds (bald eagle, herring gull, Caspian tern,
aquatic biota/fish-eating wildlife that have been devel-
double-crested cormorant (Phalacrocorax auritus),
oped in various countries (Table 6 ท1). There are consid-
common tern, Forster's tern (Sterna forsteri), great blue
erable limitations in this latter approach, as there is a
heron, black-crowned night heron (Nycticorax nyctico-
general lack of knowledge on the diets of many Arctic
rax)), the following ranges were found. For reproduc-
organisms. It is also assumed that predators eat only one
tive success, the NOEL range for PCBs was 1.3-11
type of food, but even where food preferences are
ตg/g ww in eggs. The LOEL range for various endpoints
known, there may not be analytical data for these par-
of reproductive success (hatching success, egg mortality,
ticular food items.
deformities, and parental attentiveness) ranged from 3.5

---

166
AMAP Assessment 2002: Persistent Organic Pollutants in the Arctic
PCB concentration in eggs (or otherwise indicated tissue),
ตg /g ww
23
25.5
21
13
12
11
10.9 Reproduction*, night heron (1)
10
9
8
8.0 Reproduction**, common tern (1)
7.6 Hatching success**,
common tern (3)
7
6
5.7 Deformities**,
double-crested cormorant (3)
5
5.0 Egg mortality, deformities**,
herring gull (1,2)
4
4.0 Egg mortality**, bald eagle (1,2)
3.5 Egg mortality**,
double-crested cormorant (1,2)
3
2.3 Hatching success*,
2
Forster's tern (3)
1
* No-observed-effect level (NOEL)
** Lowest-observed-effect level
(LOEL) or lowest-observed-
adverse-effects level (LOAEL)
0
)
)
)
)
fin

Shag
Range
Eiders
Puf
Murres
Fulmar
Merlin
Kittiwake
anatumanatum
Gyrfalcon
tundrius
tundrius
Single average
Gray heron
Herring gull
Bald eagle
Glaucous gull
Golden eagle
AMAP Phase II
Eiders (muscle)
Black guillemot
Kola peregrine
Common guillemot
AMAP Phase I
Glaucous gull (liver)
Norwegian peregrine
White-tailed sea eagle
Fennoscandian peregrine
Kola white-tailed sea eagle
Alaskan peregrine (
Canadian peregrine (
Alaskan peregrine (
Canadian peregrine (
Norwegian white-tailed sea eagle
Figure 6ท1. Concentrations of PCB in Arctic bird eggs (or otherwise indicated tissue) compared to thresholds for avian effects (1. Barron et
al., 1995; 2. Giesy et al., 1994b; 3. Bosveld and van den Berg, 1994). Due to numerous limitations in the thresholds data, in quantification
methods for PCBs and problems with extrapolating such data across species, this comparison should be used with caution. See pages 163-165
for more details. AMAP Phase I data are from de March et al., 1998.
to 22 ตg PCBs/g ww in eggs. For adults, PCB concen-
NOEL of 100 pg/g and a LOEL of 210 pg/g TEQ on a
trations in brain tissue higher than 300 ตg/g ww were as-
whole egg (wet weight basis) for induction of cy-
sociated with mortality. Relevant thresholds are pre-
tochrome P450 (CYP) 1A in bald eagle chicks. Bosveld
sented in Figure 6ท1.
et al. (2000) have estimated the LOEL for induction of
PCBs with TCDD-like activity (nPCBs, mono-ortho
CYP1A in the common tern at approximately 25 000 pg
PCBs) have been shown to adversely affect patterns of
TEQ/g lw in liver. These and other relevant thresholds
survival, reproduction, growth, and metabolism, with
are presented in Figure 6ท2.
CB126 being among the most toxic of all PCB congeners
It has been concluded that peregrine falcon eggs with
to birds (Eisler and Belisle, 1996). For dioxin-like sub-
DDE residues of 15 to 20 ตg/g ww would experience re-
stances, the NOAEL range for reproductive effects was
productive failure (20% eggshell thinning) (Peakall et
1.5-200 pg TEQ/g ww in eggs, and the LOAEL range
al., 1990). The adverse effects threshold for dieldrin in
for various reproductive endpoints (deformities, hatch-
peregrine falcon eggs is 1- 4 ตg/g ww, and for heptachlor
ing success, and mortality) ranged from 10 to 2200 pg
epoxide, the adverse effects threshold is 1.5 ตg/g ww
TEQ/g ww in eggs. Elliott et al. (1996) have suggested a
(Peakall et al., 1990). For HCB, results from Vos et al.

---

Chapter 6 ท Biological Effects
167
TCDD equivalents in eggs (or otherwise indicated tissue),
pg TEQ /g ww
10 000
2200 LD50, pheasant embryo (2); hatching success**, Forster's tern (3)
1000
750 LD50, Caspian tern embryo (4)
550 LD50, double-crested cormorant embryo (5)
200 Reproduction*, Forster's tern; **, bald eagle (3,6)
100
100 Cytochrome P450 1A induction*, bald eagle (6)
20 Reproduction**, wood duck (1)
10
10 Reproduction*, herring gull (1)
4.6 Reproduction*, double-crested cormorant (1)
1
* No-observed-effect level (NOEL)
or no-observed adverse-effects-level (NOAEL)
** Lowest-observed-effect leve (LOEL)
or lowest-observed adverse-effects level (LOAEL)
Range
0.1
Single average
AMAP Phase II
AMAP Phase I
0.01
Figure 6ท2. Concentrations of PCDD/Fs and non- and mono-
ortho PCBs as TCDD equivalents (pg TEQ/g ww) in Arctic bird
eggs compared to thresholds for avian effects (1. Giesy et al.,
1994b; 2. Nosek et al., 1992; 3. Kubiak et al., 1989; 4. Ludwig et
al., 1993; 5. Tillitt et al., 1992; 6. Elliott et al., 1996). Due to nu-
merous limitations in the thresholds data, in calculating TEQs
Fulmar (Canada)
Kittiwake (Canada)
Peregrine (Russia)
Sea eagle (Russia)
and problems with extrapolating such data across species, this
Grey heron (Norway)
Sea eagle (Sweden)
comparison should be used with caution. See pages 163-165 for
more details. AMAP Phase I data are from de March et al., 1998.
Thick-billed murre (Canada)
Ptarmigan (muscle) (Russia)
. waterfowl (muscle) (Russia)
Glaucous gull (liver) (Svalbard)
Div
Black-backed herring gull (Norway)
Common guillemot (Faroe Islands)
(1972) indicate a NOEL in kestrels (Falco tinnunculus)
of 12 000 ng PCB/g lw (muscle) in dams for decreased
of 40 - 49 ตg/g ww in liver. Lindane concentrations of
kit production and reduced kit body weight gain. The
100 - 200 ตg/g ww in eggs of quail and chickens are as-
NOEL and LOEL for vitamin A reduction in otter were
sociated with decreased egg production (Whitehead et
4000 and 11 000 ng/g lw in liver for PCB7 (170 and
al., 1972a; 1972b; 1974). For bald eagle eggs, DDE con-
460 ng/g ww, respectively) (Murk et al., 1998). Captive
centrations below 3 ตg/g ww are associated with near
harbour seals from the Dutch Wadden Sea, exposed to
normal productivity, concentrations above 5.1 ตg/g ww
PCBs via different fish diets, had reduced reproductive
are associated with marked productivity declines, and
success at PCB levels of 25000 ng/g lw in blood (16
concentrations above 15 ตg/g ww are associated with
ng/g ww) (Reijnders, 1986; Boon et al., 1987).
complete reproductive failure (Wiemeyer et al., 1984).
Similarly, the EC50 for dioxin-like compounds was
calculated to be 160 pg TEQ/g ww (5300 - 8000 pg/g lw)
Mammals
in mink muscle for litter size and 200 pg TEQ/g ww
Platonow and Karstad (1973) reported that 1230 ng/g
(6600 -10 000 pg/g lw) for kit survival (Leonards et al.,
ww of Aroclor 1254 in mink liver tissue was associated
1995). In experiments, Heaton (1992) found the
with impaired reproductive success. Reduced growth
LOAEL for mink kit survival to be 490 pg TEQ/g ww in
and survival of mink kits were observed in female mink
liver. In free-living otter, the NOEL for vitamin A reduc-
with 2000 ng/g ww Arcoclor 1254 in liver tissue (Wren
tion was found to be 2000 pg TEQ/g lw in liver or blood
et al., 1987a; 1987b). Olsson and Sandegren (1991a;
(84 pg TEQ/g ww in liver) based on non- and mono-
1991b) proposed an EC50 of 50 000 ng total PCB/g lw,
ortho PCBs (Murk et al., 1998). The EC90 level for vita-
and Kihlstr๖m et al. (1992) proposed an EC50 of 65 000
min A reduction was 5000 pg TEQ/g lw in liver and
ng/g lw and a NOEL of 9000 ng/g lw for litter size in
blood (210 pg TEQ/g ww in liver).
mink, based on muscle concentrations.
Experiments found reduced immune function and
In a reassessment of all reproductive studies of PCBs
disruption of vitamin A physiology in captive harbour
on mink, the EC50 in adult females for litter size was cal-
seals fed herring from the relatively contaminated Baltic
culated to be 40 000 - 60 000 ng total PCB/g lw (approx-
Sea for 2.5 years. Total PCB concentrations of 16 500
imately 1200 ng total PCB/g ww in muscle) and 2400
ng/g lw, corresponding to total TEQ levels of approxi-
ng/g ww in muscle for kit survival (Leonards et al.,
mately 210 pg/g lw, had accumulated in the blubber of
1995). Brunstr๖m and Halldin (2000) found a LOAEL
these seals, suggesting a threshold for these effects below

---

168
AMAP Assessment 2002: Persistent Organic Pollutants in the Arctic
PCB concentration, ng /g lw
500 000
8 0 0 0 0 - 1 2 0 0 0 0
E C 50 kit survival, mink muscle (1)
7 7 0 0 0
Reproductive success, ringed seal (7)
100 000
4 0 0 0 0 - 6 0 0 0 0
E C 50 litter size, mink muscle (1)
2 5 0 0 0
Reproductive success,
harbour seal blood (5)
2 1 0 0 0 Immune effects**,
Rhesus monkey blood (2)
1 6 5 0 0 Immunosuppression,
depressed vitamin A,
10 000
harbor seal blubber (10)
1 2 0 0 0 Decreased kit production**,
kit body weight, mink muscle (9)
1 1 0 0 0 Vitamin A reduction**,
otter liver (8)
9 0 0 0 Mink kit survival*,
muscle (3)
7 5 0 0 Otter reproduction*,
1000
muscle (4)
4 0 0 0 Vitamin A reduction*,
otter liver (8)
1 0 0 0 Visual memory*,
rhesus monkey offspring
blood serum (6)
500-1000 Short-time memory**,
rhesus monkey blood serum (6)
100
* No-observed-effect level (NOEL)
or no-observed adverse-effects-
level (NOAEL)
** Lowest-observed-effect level
(LOEL) or lowest-observed
adverse-effects level (LOAEL)
10
Range
Single average
AMAP Phase II
AMAP Phase I
1
olf
W
alrus
W
Beluga
/reindeer
Narwhal
Mustelids Harp seal Gray seal
Ringed sealHarbor seal
Grey whale
Bearded seal
Minke whale
Mountain hare
Caribou
Lamb (Greenland)
olverine (Canada)
Harbour porpoise
Sea otter (Alaska)
Arctic fox (Alaska)
Muskox (Greenland) W
Polar bear (Alaska)
Polar bear (Canada)
Polar bear (Russia)
Arctic fox (Canada)
Arctic fox (Iceland)
Arctic fox (Svalbard)
Steller sea lion (Alaska)
Polar bear (Svalbard)
Northern fur seal (Alaska)
Bowhead whale (Alaska)
Polar bear (circumpolar)
T E R R E S -
Pilot whale (Faroe Islands)
T R I A L
M A R I N E
Killer whale (resident) (Alaska)
Killer whale (transient) (Alaska)
Figure 6ท3. Concentrations of PCBs in Arctic animals compared to thresholds for mammalian effects (1. Leonards et al., 1995; 2. Tryphonas,
1994; 3. Kihlstr๖m et al., 1992; 4. Roos et al., 2001; 5. Boon et al., 1987; 6. Ahlborg et al., 1992; 7. Helle et al., 1976; 8. Murk et al., 1998;
9. Brunstr๖m and Halldin, 2000; 10. de Swart et al., 1994, 1996). Due to numerous limitations in the thresholds data, in quantification of
PCBs and problems with extrapolating such data across tissues and species, this comparison should be used with caution. See pages 163-165
for more details. AMAP Phase I data are from de March et al., 1998.
these concentrations. Of the TEQs, PCDD/F accounted
11 000 ng/g lw, and the authors suggest the geometric
for 18 pg TEQ, nPCB for 51 pg TEQ, and mono- and di-
mean of 8700 ng/g lw as a threshold value. For TEQs,
ortho PCB for 140 pg TEQ (Ross et al., 1995).
the threshold concentrations in liver and blood for these
Kannan et al. (2000) have reviewed semi-field and
effects were 160-1400 pg/g lw (geometric mean of 520
field studies on seals, mink, and otter to establish mean
pg/g lw).
threshold tissue concentrations of PCBs and TEQs
Assessments based on subtle neurobehavioral effects
linked to effects on hepatic vitamin A stores, thyroid
in offspring of rhesus monkeys treated with PCBs and
hormone concentrations and the immune system. The
human mothers eating PCB-contaminated fish, have re-
purpose of this study was to derive threshold concentra-
sulted in an estimated LOAEL for effects on short-term
tions that could be used for estimating risk in marine
memory of 500 -1000 ng/g lw, and a NOAEL for effects
mammals. For PCBs, the threshold concentrations in
on visual memory of 1000 ng/g lw in offspring or cord
liver or blood for these effects were found to be 6600 -
blood serum (Ahlborg et al., 1992). The LOAEL for im-

---

Chapter 6 ท Biological Effects
169
TCDD equivalents,
2400
pg TEQ /g lw
500
209 (PCDD/F, nPCB, moPCB)
191 (nPCB, moPCB)
140 (moPCB)
100
69 (PCDD/F, nPCB)
51 (nPCB)
18 (PCDD/F)
10
All thresholds:
Immune suppression
in captive harbour seals
(Ross et al., 1995)
1
Range
Single average
AMAP Phase II
AMAP Phase I
0.1
Seals
Canada)
Minke whale
Otter (Sweden)
alrus (Inukjuak)
alrus (Akulivik)
alrus (Inukjuak)
Reindeer (Stilla)
Caribou (Canada)
W
W Beluga (Canada)W
fin Island, Narwhal (Canada)
Reindeer (Russia)
Reindeer (Jarfjord)
Polar bear (Canada)
Reindeer (Sweden)
olar bear (Svalbard)
P
Polar bear (Canada)
Ringed seal (Russia)
Grey whale (Russia)
Bearded seal (Russia)
Spotted seal (Russia)
Ringed seal (Kara Sea)
Mountain hare (Russia)
Harbour porpoise (Norway)Beluga (Kimmirut, Canada)
Arctic fox (Pribilof Islands)
Ringed seal (Holman Island)
Beluga (Baf
Harbour porpoise (Greenland)
Killer whale (residents) (Alaska)
Killer whale (transients) (Alaska)
Northern fur seal (Pribilof Islands)
Ringed seal (Pangnirtung, Canada)
Figure 6ท4. Body concentrations of PCDD/Fs and non- and mono-ortho PCBs as TCDD equivalents (pg/g lw) in Arctic mammals compared to
thresholds for immunosuppression in harbour seal for the same combination of substances analyzed (Ross et al., 1995). Due to numerous lim-
itations in the thresholds data, in calculating TEQs and problems with extrapolating such data across tissues and species, this comparison
should be used with caution. See pages 163-165 for more details. AMAP Phase I data are from de March et al., 1998.
munosuppression is 21 000 ng PCBs/g lw in rhesus
Fish
monkeys (Tryphonas, 1994). PCB concentrations of
The LOEL for EROD induction by PCBs in Arctic char
2000 -5000 ng/g ww in non-human primate brain tissue
is 1000 ng/g ww in liver (J๘rgensen et al., 1999). Labo-
were associated with decreased dopamine concentra-
ratory studies of Arctic char have found that single doses
tions (Seegal et al., 1990). Relevant thresholds for PCBs
of PCB as low as 1000 ng/g body weight affect disease
are presented in Figure 6ท3 and for TEQs and effects on
resistance and stress responses in starved specimens (J๘r-
immunosuppression, in Figure 6ท4.
gensen, 2002). Mayer et al. (1977) found that adult fat-
Several effects thresholds have been determined for
head minnows (Pimephales promelas) and channel cat-
the effects of TBT or its metabolite, DBT. For in vitro
fish exposed to toxaphene had no decreases in hydrox-
inhibition of hepatic cytochrome P450 in Dall's porpoise
yproline levels, but levels in exposed offspring were sig-
and Steller sea lion liver, the threshold was found to be
nificantly decreased. Toxaphene concentrations of 3400
100 ตM TBT (29000 ng/g ww) (Kim et al., 1998b). For
ng/g ww in channel catfish fry tissues were associated
in vivo hepatotoxicity in mice, a threshold concentration
with decreased growth, and 600 ng/g led to altered bone
of 2600 ng DBT/g ww in liver was found (Ueno et al.,
development (Stickel and Hickey, 1977), identifying off-
1994). For in vitro immunotoxicity to rat thymocytes
spring as being more sensitive than adults are to toxic ef-
(Snoeij et al., 1986) and rabbit polymorphonuclear neu-
fects. Mayer et al. (1975) exposed brook trout to
trophils (PMNs) (Elferink et al., 1986), the threshold con-
toxaphene in water and found higher mortality during
centration was found to be 1.0 ตM TBT (290 ng/g ww).
spawning. A 50% mortality was associated with tox-

---

170
AMAP Assessment 2002: Persistent Organic Pollutants in the Arctic
PCB concentration, ng /g ww
20 000
10 000
5 0 0 0 Peregrine falcon, highest
no-effect concentration
in food (1)
1000
160 US EPA guideline (2)
100
100 IJC guideline (3)
72 Mink, highest no-adverse-effect
concentration in food (4)
48 Canadian guideline
for protecting birds (5)
1 5 Canadian guideline
for protecting mammals (5)
10
1
Range
Single average
AMAP Phase II
AMAP Phase I
0.1
skin) skin)
alleye
+ +
๘ya/Jan Mayen) W
Capelin
๘rn
Dab (liver)
fish (liver)
(Bj
Burbot (liver)
Cisco (liver)
olf
Pike (muscle)
Blue mussels
Sculpin (liver)
Perch (muscle)
Navaga (liver)
W
Redfish (liver)
Sturgeon (liver)
Seabirds (eggs)
Seabirds (liver)
Inconnu (muscle)
Arctic cod (liver)
Starry ray (liver)
Grayling (muscle)
Redfish (muscle)
Atlantic cod (liver)
. salmon (muscle)
Lake trout (muscle)
(muscle)
Arctic char (muscle)
Herring (whole fish)
Harp seal (blubber)
Brown trout (muscle)
Navaga (whole fish)
Div Long rough dab (liver) . Greenland fish (liver)
Ringed seal (blubber)
Lake whitefish (muscle)
. Jan Mayen fish (liver)
European carp (muscle)
. Russian fish (muscle)
Greenland shark (liver)
Greenland halibut (liver)
Div
Arctic char (muscle
Checkered eelpout (liver)
Div
Div
Arctic char
F R E S H W A T E R
Broad whitefish (muscle
M A R I N
Greenland cod (whole body)
E
Figure 6ท5. PCBs in Arctic animals as food items compared to NOAECs and environmental quality guidelines for protecting fish-
eating/aquatic wildlife (1. Baril et al., 1990; 2. USEPA, 1995; 3. De Vault et al., 1995; 4. Giesy et al., 1994a; 5. Environment Canada, 2002).
Due to limitations in quantification methods for PCBs and problems with extrapolating such data across tissues and species, this comparison
should be used with caution. See pages 163-165 for more details. AMAP Phase I data are from de March et al., 1998.
aphene concentrations of 870 ng/g ww in muscle and
of 5 ตg/g ww for PCBs, 5 ตg/g ww for DDE and 0.1
2400 ng/g ww in whole body. Decreased survival was
ตg/g ww for dieldrin. For HCB, results from Vos et al.
found in lake trout dosed with 7000 ng/g body weight,
(1972) indicate a dietary NOEC of 100 ตg/g ww.
which was extrapolated to be equivalent to a retained
whole body concentration of 4800 ng/g ww (Delorme et
Mammals
al., 1993; Delorme, 1995).
Giesy et al. (1994c) have estimated the dietary NOAEC
of PCBs, dioxin-like compounds, dieldrin, and DDTs
for reproductive effects in mink based on a diet of con-
Dietary intake thresholds for birds and mammals
taminated fish. These were found to be 72 ng PCBs/g
Birds
ww food, 2 pg TEQ/g ww food, 5000 ng dieldrin/g ww
Dietary LOECs for PCBs range from 2 to 50 ตg/g ww
food, and 100 000 ng DDTs/g ww food, respectively.
food for reproductive endpoints in a number of different
Brunstr๖m and Halldin (2000) found a LOAEC for ad-
bird species. Baril et al. (1990) concluded that only p,p'-
verse effects on mink reproduction of 22.3 pg TEQ/g
DDE, PCBs, and dieldrin intake are likely to affect re-
ww food and a NOAEC of 3 pg TEQ/g ww food. The
production in peregrine falcons and gave dietary LOECs
NOEC for vitamin A reduction in otter from non- and

---

Chapter 6 ท Biological Effects
171
DDT, ng /g ww
20 000
10 000
5 0 0 0 N O E C , peregrine falcon (1)
1000
1 0 0 0 I J C guideline (2)
100
3 9 . 0 US EPA guideline (3)
1 4 . 0 Canadian guideline (4)
10
1
Range
Single average
AMAP Phase II
AMAP Phase I
0.1
skin) skin)
alleye
+ +
๘ya/Jan Mayen) W
Capelin
๘rn
Dab (liver)
fish (liver)
(Bj
Burbot (liver)
Cisco (liver)
olf
Pike (muscle)
Blue mussels
Sculpin (liver)
Perch (muscle)
Navaga (liver)
W
Redfish (liver)
Sturgeon (liver)
Seabirds (eggs)
Seabirds (liver)
Inconnu (muscle)
Arctic cod (liver)
Starry ray (liver)
Grayling (muscle)
Redfish (muscle)
Atlantic cod (liver)
. salmon (muscle)
Lake trout (muscle)
(muscle)
Arctic char (muscle)
Herring (whole fish)
Navaga (whole fish)
Harp seal (blubber)
Div Long rough dab (liver) . Greenland fish (liver)
Ringed seal (blubber)
Lake whitefish (muscle)
. Jan Mayen fish (liver)
European carp (muscle)
. Russian fish (muscle)
Greenland shark (liver)
Greenland halibut (liver)
Div
Arctic char (muscle
Checkered eelpout (liver)
Div
Div
Arctic char
Greenland cod (whole body)
F R E S H W A T E R
Broad whitefish (muscle
M A R I N E
Figure 6ท6. DDTs in Arctic animals as food items compared to NOAECs and environmental quality guidelines for protecting fish-
eating/aquatic wildlife (1. Baril et al., 1990; 2. USEPA, 1995; 3. De Vault et al., 1995; 4. Environment Canada, 2002). Due to problems with
extrapolating data across tissues and species, this comparison should be used with caution. See pages 163-165 for more details. AMAP Phase I
data are from de March et al., 1998.
mono-ortho PCB TEQs was 11 pg TEQ/g lw in fish, and
mission, U.S. Environmental Protection Agency, Envi-
the LOEC was 29 pg TEQ/g lw in fish (Murk et al.,
ronment Canada) for the protection of aquatic life and
1998). For heptachlor epoxide, the dietary NOAEC for
wildlife that consume aquatic biota. The guidelines have
adult mink was 50 000 ng/g ww food (Aulerich et al.,
been derived using contaminant concentrations in prey
1990), and the LOAEC for kit growth was 6250 ng/g
such as fish, known thresholds for effects in sensitive
ww (Crum et al., 1993). No information on the sensi-
fish-eating wildlife species, and research results on bio-
tivity of mink to toxaphene is available; however, a
accumulation and biomagnification rates of particular
NOAEC of 4000 ng toxaphene/g ww food for thyroid
substances. These data have then been used to back-cal-
effects has been estimated from studies in rats and dogs
culate fish tissue concentrations that should be without
(Chu et al., 1986b).
effects in wildlife, and these have been designated as en-
Relevant thresholds and guidelines for intakes of
vironmental quality guidelines/objectives.
PCBs are given in Figure 6ท5, and for DDTs in Figure
In the discussion that follows, only new information
6ท6. The environmental quality guidelines/objectives
(from 1996 or later) on biological effects studies, or
have been taken from Table 6 ท1. These have been devel-
where new concentration data warrant a discussion of
oped by various organizations (International Joint Com-
the potential for biological effects, will be addressed.

---

172
AMAP Assessment 2002: Persistent Organic Pollutants in the Arctic
6.1. Terrestrial environment
6.1.1.4. Lamb
6.1.1. Terrestrial herbivores
Concentrations of PCBs, DDTs and other POPs are
6.1.1.1. Arctic hare/mountain hare
available from West Greenland and Faroe Island lamb
No biological effects studies have been conducted on
(Annex Table 5). The mean PCB levels in different tis-
Arctic or mountain hares. Concentrations of PCBs,
sues (muscle, kidney, liver) range from 7.9 to 38 ng/g lw
DDTs, and other POPs are available from West Green-
and are similar to the levels found in muskox. These
land, several Russian sites and the Faroe Islands (Annex
concentrations do not exceed any effects thresholds for
Table 5). The mean PCB levels in different tissues (mus-
PCBs (Figure 6ท3).
cle, kidney, liver) range from 0.7 to 41 ng/g lw. These
concentrations do not exceed any effects thresholds for

6.1.2. Terrestrial birds
PCBs (Figure 6ท3). TEQ values based on PCDD/F con-
centrations are also available for Russian mountain hares,
6.1.2.1. Ptarmigan/willow grouse
and these varied from 1.4 to 30 pg TEQ/g lw (0.03 to
No biological effects studies have been conducted on
0.6 pg/g ww) in muscle (Annex Table 16). The highest
ptarmigan or willow grouse. Concentrations of PCBs,
TEQs were found in samples from hares from the Kola
DDTs and other POPs are available from West Green-
Peninsula (Lovozero) and are higher than levels associ-
land and several Russian sites (Annex Table 5). PCB
ated with immunosuppressive effects in harbour seal
concentrations ranged between 0.6 and 18.3 ng/g ww
(Figure 6ท4).
(24-270 ng/g lw). The levels are difficult to assess since
the tissues analyzed are liver and muscle, but if the lipid
levels in these tissues are assumed to be similar to that
6.1.1.2. Caribou and reindeer
of eggs, then the liver and muscle concentrations do
In the previous AMAP assessment (de March et al.,
not exceed any thresholds for PCBs (Figure 6ท1). TEQ
1998), the concentrations of -HCH, HCB, PCBs, chlor-
values based on PCDD/F concentrations for ptarmi-
danes, DDTs, PCDD/F, and dieldrin in caribou and
gan/willow grouse from several Russian sites varied
reindeer from across the Yukon, the Northwest Territo-
from 0.02 to 0.2 pg TEQ/g ww (1.2-3.1 pg/g lw) in
ries, Norway, Sweden, and Russia were very low. The
muscle (Annex Table 16). These are below concentra-
Russian data were very limited, and concentrations from
tions expected to result in reproductive effects, assum-
two consecutive years varied considerably. The POP levels
ing similar lipid levels in muscle compared to eggs (Fig-
were several orders of magnitude lower than those ex-
ure 6ท2).
pected to lead to subtle biological effects. No effects stud-
ies have been carried out since the previous assessment.
6.1.3. Waterfowl
Levels and intake assessment
In the previous AMAP assessment (de March et al.,
A much better data set for PCBs, DDTs and other
1998), several waterfowl groups and species (mollusci-
POPs is available from several sites across Russia (An-
vores, piscivores, semipalmated plover, pintail, and
nex Table 5). The mean PCB levels in different tissues
oldsquaw) from Canada had levels of PCBs above
(muscle, kidney, liver) range from 0.2 to 2.4 ng/g ww.
NOELs and LOELs for reproductive effects. No ef-
Based on an estimated lipid content of 4.5% in liver,
fects studies have been carried out since the previous
PCB concentrations are estimated to be 5.6-53 ng/g lw
assessment.
in the Russian reindeer. PCB levels in reindeer from
Finland were even lower (3.5 ng/g lw). These concentra-
Levels and intake assessment
tions do not exceed any effects thresholds for PCBs
PCB and TEQ concentrations based on PCDD/Fs are
(Figure 6ท3). TEQ values based on PCDD/F concentra-
available for several species of waterfowl from several
tions are also available for Russian reindeer, and these
Russian sites (Annex Tables 11 and 16). For PCBs, re-
varied from 0.75 to 20 pg TEQ/g lw (0.05 to 0.98 pg/g
sults are available for liver and muscle from oldsquaw
ww) in muscle and 4.2 to 105 pg TEQ/g lw (0.24 to 6.5
(Clangula hyemalis), goldeneye (Bucephala clangula),
pg/g ww) in liver (Annex Table 16). The highest TEQs
pintail (Anas acuta), wigeon (Anas penelope), green-
were found in muscle and liver samples from reindeer
winged teal (Anas crecca), greater white-footed goose
from the Kola Peninsula (Lovozero) and are higher than
(Anas albifrons), greater scaup (Aythya marila), scoters
levels associated with immunosuppressive effects in har-
(Melanitta spp.), and bean goose (Anser fabalis). PCB
bour seal (Figure 6ท4). TEQs in liver from reindeer from
concentrations in liver range from 0.001 to 0.042 ตg/g
Pechora and Taymir Dudinka were also above this
ww, and for muscle from 0.001 to 0.017 ตg/g ww. The
threshold, but TEQs in muscle were not.
levels are difficult to assess since the tissues analyzed
are liver and muscle, but if the lipid levels in these tis-
sues are assumed to be similar to that of eggs, then the
6.1.1.3. Muskox
liver and muscle concentrations do not exceed any
No biological effects studies have been conducted on
thresholds for PCBs (Figure 6ท1). For TEQs, results
muskox. Concentrations of PCBs, DDTs and other
are available for oldsquaw, goldeneye, and pintail.
POPs are available from West Greenland (Annex Table
TEQs range from 0.02 to 0.52 pg TEQ/g ww (0.5 to
5). The mean PCB levels in different tissues (muscle,
6.4 pg/g lw) in muscle. These are below concentrations
kidney, liver, fat) range from 4.1 to 34 ng/g lw. These
expected to result in reproductive effects, assuming
concentrations do not exceed any effects thresholds for
similar lipid levels in muscle compared to eggs (Figure
PCBs (Figure 6ท3).
6ท2).

---

Chapter 6 ท Biological Effects
173
PCB levels exceed most NOELs and LOELs found for
6.1.4. Birds of prey
hatching success, egg mortality and deformities in a
6.1.4.1. Peregrine falcon
number of wild bird species (Figure 6ท1).
In the previous AMAP assessment, it was concluded that
The p,p'-DDE concentrations in the American sub-
Arctic populations of Canadian peregrine falcons were
species ranged between 0.48 and 14.1 ตg/g ww (geomet-
still at risk for reproductive effects from DDTs and
ric mean of 3.4 ตg/g ww ), and in the Arctic subspecies
PCBs in their eggs and in their food. For Fennoscan-
between 1.2 and 13.3 ตg/g ww (geometric mean of 3.0
dian peregrines, only high levels of PCBs in eggs ex-
ตg/g ww), indicating that some individuals have concen-
ceeded NOELs and LOELs for reproductive effects in
trations just below the critical threshold for reproductive
other bird species. Peregrine falcons from the Kola
failure (15-20 ตg/g ww).
Peninsula, Russia, had PCB levels in eggs that exceeded
Geometric mean dieldrin concentrations were below
most NOEL and LOEL levels for reproductive effects in
the adverse effects threshold of 1-4 ตg/g ww for both sub-
other bird species. TEQs based on PCDD/F and dioxin-
species and in 1991-1995, no individual eggs exceeded
like PCB concentrations exceeded most NOAELs and
these levels. Geometric mean heptachlor epoxide con-
LOAELs for reproductive effects in other bird species
centrations never exceeded the adverse effects threshold
and several LD50 values (the dose that causes the death
of 1.5 ตg/g ww.
of 50% of a group of test animals). No new effects or
PCB concentrations in Norwegian peregrine falcon
levels studies have been carried out in these populations
eggs collected in 1991-1997 ranged from 2.3 to 25.5
since the previous assessment.
ตg/g ww (mean of 10.6 ตg/g ww) (Annex Table 5). These
concentrations exceed most thresholds for reproductive
Reproductive effects
effects in other bird species (Figure 6ท1).
Peregrine falcon populations in North America and Eu-
rope declined drastically in most parts of the species'
6.1.4.2. Merlin
range in the 1960s and 1970s, primarily due to the ef-
fects of OC pesticides, especially DDT and dieldrin. Sub-
In the previous AMAP assessment, Fennoscandian mer-
species found in Alaska are the Arctic peregrine, which
lin were suffering from 10% shell thinning, but the pop-
nests in northern tundra, the American peregrine, which
ulation was recovering. Merlin seemed to be less sensi-
nests in the forested interior and the Peale's peregrine,
tive to the effects of DDT, since DDT levels exceeded
which nests along the southern coast from the Aleutian
those expected to cause reproductive failure in peregrine
Islands to southeast Alaska.
falcons (20% eggshell thinning results in crushed shells).
Persistent OC contaminants were measured in Amer-
PCB levels were in the range of some NOELs and LOELs
ican and Arctic peregrine falcon eggs from Alaska from
for effects in white leghorn chicken, an extremely sensi-
1979 to 1995 (Ambrose et al., 2000). Dieldrin, p,p'-
tive species. No effects studies have been carried out since
DDE, heptachlor epoxide, oxychlordane, and total Aro-
the previous assessment.
clor PCBs were consistently measured and detected, and
were tested multivariately for relationships with time
Levels and intake assessment
and productivity. Eggshell thickness was significantly
PCB concentrations in Norwegian merlin eggs col-
negatively correlated with p,p'-DDE concentrations, and
lected from 1991 to 1997 ranged between 1.4 and 2.0
mean eggshell thicknesses in 1991-1995 were 12.0 and
ตg/g ww (Annex Table 5). These concentrations are
10.6% thinner in American and Arctic subspecies, re-
below thresholds for reproductive effects in other bird
spectively, than pre-DDT era peregrine falcon eggs from
species (Figure 6ท1).
Alaska (Anderson and Hickey, 1972).
Significant multivariate analyses indicated that some
6.1.4.3. White-tailed sea eagle
POPs were associated with decreased reproduction. Dif-
ferences were greatest early in the study, but over the en-
In the previous AMAP assessment, effects studies
tire time span, dieldrin, oxychlordane and total PCBs
showed improvement in reproduction and population
were significantly greater in eggs from unsuccessful nests
numbers for white-tailed sea eagles in Sweden and
(no young at expected age of 1-3 wk) compared to suc-
Norway, with only a small amount of eggshell thinning
cessful nests (1 young) for the American subspecies.
(2-5%). However, levels of TEQs in Swedish eagles,
There were no significant differences in p,p'-DDE or
measured as PCDD/F and several dioxin-like PCBs, ex-
heptachlor epoxide, and no significant differences for
ceeded most NOAELs and LOAELs for reproductive
any of the five contaminants in the Arctic subspecies.
effects in other bird species and approached the LD50
Eggs from unsuccessful nests also had higher mercury
for white leghorn chicken embryo. PCB levels in the
concentrations, which may also have affected reproduc-
northerly Norwegian eagles exceeded or overlapped
tion (AMAP, 2004).
most NOELs and LOELs for subtle reproductive ef-
fects in other bird species. PCB and TEQ levels in one
Levels and intake assessment
Russian white-tailed sea eagle egg from the Kola Penin-
Concentrations of PCBs in Alaskan peregrine eggs col-
sula exceeded some thresholds for effects in other bird
lected in 1991-1995 ranged between 0.4 and 8.5 ตg/g
species. The intake assessment for Norwegian sea ea-
ww (geometric mean of 1.6 ตg/g ww) for the American
gles, which prey mainly on deep-water marine fish
subspecies and between 0.6 and 6.0 ตg/g ww (geometric
species, concluded that dietary intakes of DDTs and
mean of 1.3 ตg/g ww) for the Arctic subspecies (Am-
PCBs may be high enough to lead to effects. No ef-
brose et al., 2000) (Annex Table 5). These mean PCB
fects studies have been carried out since the previous
levels are below thresholds for effects, but the maximum
assessment.

---

174
AMAP Assessment 2002: Persistent Organic Pollutants in the Arctic
Levels and intake assessment
Levels and intake assessment
PCB concentrations in Norwegian white-tailed sea eagle
PCB concentrations in Norwegian gyrfalcon eggs col-
eggs collected in 1991-1997 ranged from 3.1 to 13.0
lected in 1991-1997 ranged from 0.29 to 5.3 ตg/g ww
ตg/g ww (mean of 9.0 ตg/g ww) (Annex Table 5). These
(mean of 2.0 ตg/g ww) (Annex Table 5). These concen-
concentrations exceed most thresholds for reproductive
trations exceed a few thresholds for reproductive effects
effects in other bird species (Figure 6ท1).
in other bird species (Figure 6ท1).
6.1.4.4. Bald eagle
6.1.5. Carnivores
Reproductive effects
6.1.5.1. Mink
The relationship between OC concentrations, diet, and
In the previous AMAP assessment, Canadian mink from
productivity was studied in nesting bald eagles from
Quebec (Grand Baleine) and Fort Providence, NWT, had
Adak, Tanaga, Amchitka, and Kiska Islands in the Ale-
PCB levels that exceeded the NOAEL and LOAEL for
utian archipelago (Anthony et al., 1999). Productivity
effects on subtle neurobehavioral effects in offspring of
on Kiska Island (the most westerly island) was depressed
rhesus monkeys, and the Quebec mink levels were close
(lower breeding success and low clutch sizes), averaging
to the NOEL for mink kit survival. The intake assess-
0.67 young per occupied site. This was associated with
ment for Canadian mink indicated that dietary intakes
higher DDE and OC pesticide concentrations. Mercury
of dioxin-like compounds and PCBs from some species
concentrations were also higher. Productivity on the
of fish at some lakes may be high enough to lead to ef-
other three islands was comparable to healthy popula-
fects. No effects studies have been carried out since the
tions in other parts of the U.S. (0.88-1.24 young per
previous assessment. No new information on current
occupied site), and DDE concentrations were lower.
POP levels is available.
Eggshell thickness was not significantly different among
the islands and was not correlated to DDE concentra-
6.1.5.2. Otter
tions, but was significantly negatively correlated to PCB
concentrations. PCB concentrations were not correlated
In the previous assessment, otter populations in north-
to DDE concentrations, but instead, reflected previous
ern Sweden were beginning to recover in the 1990s after
military activity.
having existed only as isolated groups (de March et al.,
1998). The PCB levels were considered to be at the NOEL
Levels and intake assessment
for otter reproduction. These levels exceeded the NOAEL
Mean PCB concentrations in bald eagle eggs from the
and LOAEL for effects on subtle neurobehavioral effects in
above four islands in the Aleutian chain were between
offspring of rhesus monkeys, however. The intake assess-
0.7 and 2.1 ตg/g ww (range 0.1-9.9 ตg/g ww) (Annex
ment for Swedish otter indicated that dietary intakes of
Table 11). The mean concentrations are just below the
dioxin-like compounds from some species of fish at some
NOEL for hatching success in Forster's tern (Figure
lakes may be high enough to lead to effects. No effects
6ท1). However, the maximum PCB concentrations at
studies have been carried out since the previous assessment.
some sites exceed most of the thresholds for reproduc-
No new information on current POP levels is available.
tive effects in other bird species including the LOAEL
for egg mortality in bald eagles. The mean p,p'-DDE
6.1.5.3. Wolverine
concentrations were between 0.7 and 2.75 ตg/g ww
(range 0.3 to 4.1 ตg/g ww) with the highest concentra-
No biological effects studies have been conducted on
tions on Kiska Island (range 1.5-4.1 ตg/g ww). These
wolverines. Concentrations of PCBs, DDTs, and
concentrations are within the range known to cause re-
other POPs are available from the Northwest Territo-
productive impairment in bald eagles (Wiemeyer et al.,
ries, Canada (Annex Table 5). The mean PCB levels in
1984).
liver are 1960 ng/g lw. These concentrations exceed the
NOAEL and LOAEL for subtle neurobehavioral effects
if wolverines are as sensitive as offspring of rhesus mon-
6.1.4.5. Golden eagle
keys and humans (Figure 6ท3).
Levels and intake assessment
PCB concentrations in Norwegian golden eagle eggs
6.1.5.4. Wolf
collected in 1991-1997 ranged from 0.26 to 4.7 ตg/g
ww (mean of 1.4 ตg/g ww) (Annex Table 5). These con-
In the previous AMAP assessment, concentrations of
centrations exceed a few thresholds for reproductive ef-
PCB levels in wolves from the Canadian Arctic were
fects in other bird species (Figure 6ท1).
several orders of magnitude lower than those expected
to result in effects on reproduction. The levels in cari-
bou/reindeer were also much lower than the dietary con-
6.1.4.6. Gyrfalcon
centrations expected to cause such effects. No biological
In the previous AMAP assessment, DDE and PCB con-
effects studies have been conducted on wolf since the
centrations in Canadian gyrfalcon eggs were several or-
previous assessment.
ders of magnitude lower than those considered to cause
reproductive effects. Based on dietary LOECs, OC con-
Levels and intake assessment
centrations in gyrfalcon prey from Canada were also
Mean concentrations of PCBs in wolves (liver) from
below those expected to cause effects. No effects studies
the Canadian Yukon, and median concentrations from
have been carried out since the previous assessment.
northwest Russia were between 17 ng/g lw and 450 ng/g

---

Chapter 6 ท Biological Effects
175
lw (Gamberg and Braune, 1999; Shore et al., 2001).
pled in 1993-1995 along the Norwegian coast (F๘lsvik
These concentrations are below those expected to result
et al., 1999). Some degree of imposex occurred in almost
in effects (Figure 6ท3). The dietary assessment is the
all populations of dogwhelks studied, except in four
same as for the previous AMAP assessment.
from northern Norway. The concentration of organotin
compounds in the gastropods from the unaffected popu-
lations was below the detection limit (7 ng Sn/g dw). The
6.2. Freshwater environment
concentration of TBT in dogwhelks from affected popu-
6.2.1. Fish
lations was in the range of 48-1096 ng Sn/g dw. A posi-
In the previous AMAP assessment, freshwater fish did
tive relationship between the concentration of TBT in
not exceed effect levels for toxaphene, but fish from
dogwhelks and the degree of imposex was found.
some lakes had levels close to the threshold for bone de-
In Iceland, the status of imposex was evaluated in the
velopment effects and mortality during spawning, indi-
dogwhelk in 1998 and compared to the levels of impo-
cating that some individuals in Canada may exceed
sex evaluated in 1992/1993 (Svavarsson, 2000). The
these levels. Besides Lake Laberge, possibly affected lo-
level of imposex has decreased considerably since 1992/
cations included several lakes in the Yukon (Bennett,
1993, two years after implementation of restrictions on
Tagish, and Marsh Lakes), Atlin Lake in northern
the use of TBT-based anti-fouling paint. VDSI and RPSI
British Columbia, and Great Slave Lake in the North-
levels (see definition in Table 6ท2) have declined con-
west Territories.
siderably, both near large and small harbors. The impact
area of a large harbor complex has decreased consider-
Cytochrome P450 activities
ably, while lesser changes were seen in the impact areas
Arctic char from two lakes on Bj๘rn๘ya, Norway, have
near smaller harbors. This study shows that, at least in
been studied for OC contamination and biomagnifica-
Icelandic waters, the situation has improved considerably.
tion: Ellasj๘en, which is strongly affected by seabird
Imposex has been observed in the subtidal common
guano from nearby colonies and has high PCB levels
whelk near Svalbard and in Icelandic waters (Brick and
and ุyangen, an oligotrophic lake unaffected by the
Bolte, 1994; Svavarsson et al., 2001). The levels ob-
bird colonies and with lower PCB levels. Arctic char
served in Breiafj๖rur, southwestern Iceland, were low,
have also been studied for liver enzyme activity includ-
and only 26.4% of the females observed had imposex,
ing testosterone hydroxylation enzymes. When compar-
and the penises were small (0.7 mm ฑ 0.6 mm SD) (Sva-
ing char from both lakes, correlations were seen be-
varsson et al., 2001). Near harbors (Reykjavํk and
tween CYP enzyme activities and high PCB levels. Lev-
Straumsvik harbors), the frequency was higher, and the
els of testosterone hydroxylation enzymes were similar
female penises were considerably longer (mean 3.8 and
at both lakes and not correlated to PCB levels (Skot-
6.8 mm, respectively).
vold et al., 1999).
Studies of the occurrence of imposex in dogwhelks in
the Faroe Islands were done in 1996 and in 2001 (FEA,
Levels and intake assessment
2002). The results do not indicate any major changes in
Toxaphene levels in some burbot from the East Arm of
the five-year period from the first to the second sampling
Great Slave Lake, NWT and Lake Laberge and Kusawa
period (Table 6ท2). The occurrence of imposex in the
Lake, Yukon (Canada), exceed levels associated with ef-
Faroe Islands is still widespread, but there are sites
fects on bone development in channel catfish (600 ng/g
where the phenomenon is hardly seen.
ww) (Annex Table 7). Toxaphene levels in freshwater
fish from all other sites measured in the Arctic are below
Table 6ท2. Imposex in dogwhelks, Nucella lapillus, in the Faroe Islands.
ญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญ
threshold levels for effects. PCB concentrations in bur-
VDSI a
VDSI
RPSI b
RPSI
bot liver from Fairbanks and Yukon Flats (Alaska), Lake
Feb.-Mar.
March
Feb.-Mar. March
Laberge (Canada) and in some Arctic char from Bj๘rn-
Station
1996
2001
1996
2001
๘ya (Ellasj๘en) are close to or exceed the LOEL for in-
ญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญ
duction of EROD in Arctic char (Annex Table 7). PCB
T๓rshavn
Argir
4.4
4.0
32
48
levels in other Arctic freshwater fish do not exceed this
Kirkjub๘ur
0.1
0.6
0
0
threshold.
Vestmanna
4.3
4.1
13
17
Trongisvแgsfj๘rur
Kolatoftir
4.4
31
6.3. Marine environment
Hvํtanes (April '01)
4.2
36
6.3.1. Invertebrates
Klaksvํk
Kunoy
4.1
4.0
13
10
In the previous AMAP assessment, imposex, the induc-
Skแlafj๘rur
4.0
53
tion of male sex characteristics in females, was found in
N๓lsoy
females of the common whelk in Kongsfjorden, Sval-
Vํkin c
4.1
20
bard, dogwhelks in Norway and Iceland, and a marine
Kirkjutangi
4.3
1.3 d
8
0 d
ญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญ
snail (Nucella lima) in Alaska exposed in harbors with
a VDSI = Vas deferens sequence index. The VDSI scale ranges from
significant boat mooring.
0 (normal) to 6 (last stages of development of vas deferens and
penis in females leading to sterility).
Reproduction
b RPSI = Relative penis size index (female mean penis length3/ male
There have been only modest studies on TBT and its ef-
mean penis length 3) ท100.
c Too few females in 1996 sample for analyses.
fects on gastropods recently in Arctic or subarctic wa-
d Imposex in all stages were found, a sample with unusually large
ters. In Norway, TBT has been measured, and imposex
individual variation.
has been evaluated in dogwhelks in 41 populations sam-
Source: FEA (2002).

---

176
AMAP Assessment 2002: Persistent Organic Pollutants in the Arctic
In Greenland, TBT concentrations in blue mussels
dian, Norwegian, Greenlandic, Icelandic, and Russian
and the occurrence of imposex in several whelk species
waters did not exceed any of the guidelines for protect-
was studied in and around five harbors (Nuuk, Qaqor-
ing fish-eating wildlife. However, PCB and DDT lev-
toq, Manitsoq, Qeqertarssuak, and Uummannaq), and
els in a range of fish species from many Arctic sites ex-
at an uninhabited reference site (Strand and Asmund,
ceeded some, and in a few cases for PCBs, all of the en-
2003). Some degree of imposex was found in several
vironmental quality guidelines, implying that dietary in-
neogastropod species (Buccinum spp.) in all the harbors
takes of DDTs and PCBs in fish-eating seabirds might
but not outside the harbors or at the reference site. Ef-
be high enough to lead to effects if seabirds prey on these
fects of contaminants on blue mussels have been evalu-
fish species. A few fish species in Canada and Norway
ated at sites in southwestern Icelandic waters by study-
had PCDD/F or nPCB levels that exceeded the guide-
ing Scope for Growth (SFG) (Widdows, 1998). This
lines.
method allows evaluation of combined effects of con-
Based on POP levels found in eggs from guillemots,
taminants on the physiology of the mussel. The studies
puffins, cormorants, and fulmars from the Canadian
show that SFG is low near Reykjavํk harbor amongst
and Norwegian Arctic, dieldrin, HCB, DDT, and PCB
mussels with high levels of TBT and PAHs, while among
levels were below dietary LOECs for reproductive ef-
mussels living near small harbors (low TBT levels; inter-
fects in peregrine falcon. Mean dieldrin levels in these
mediate PAH levels), the SFG is fairly high (Halld๓rsson,
seabird species' eggs also did not exceed any environ-
2002).
mental guidelines for protecting aquatic wildlife. How-
ever, DDT levels in these species from these sites ex-
ceeded Canadian Environmental Quality Guidelines for
6.3.2. Fish
protection of animals that consume aquatic biota and
In the previous AMAP assessment, only Greenland hali-
U.S. EPA guideline values for assessment of hazards to
but (turbot) from the Canadian Arctic had toxaphene
fish-eating wildlife. PCB levels exceeded all guideline
levels close to the effect levels for bone development and
levels, implying that dietary intakes of DDTs and
mortality during spawning. No effects studies have been
PCBs in some seabirds that prey on seabird eggs and
carried out since the previous assessment.
chicks might be high enough to lead to effects.
Levels and intake assessment
6.3.3.1. Eiders
Mean PCB levels in marine and anadromous fish are
below the LOEL for EROD induction in Arctic char
In the previous AMAP assessment, common and Steller's
except for Greenland shark (liver) from Cumberland
eiders (Polysticta stelleri) from Canada, Russia, and
Sound and for Greenland halibut (liver) from West Green-
Norway had PCB levels below effects thresholds. A re-
land (Annex Table 10). Based on mean toxaphene levels,
cent study in spectacled eiders (Somateria fischeri) from
no anadromous or marine fish exceed effects levels.
western Alaska found no biological effects that could be
attributed to the low concentrations of POPs present
(Trust et al., 2000).
6.3.3. Seabirds
In the previous AMAP assessment, PCB levels in
Levels and intake assessment
seabirds from the Canadian and Norwegian Arctic ap-
Mean PCB levels in common eider muscle from Green-
proached or exceeded reproductive NOELs and LOELs
land, Iceland, and western Russia (Chukotka), and King
at the low range of the scale compared to peregrine fal-
eider (Somateria spectabilis) from Greenland, are below
cons. Mean PCB levels in eider eggs were below those
those expected to result in reproductive effects, assum-
expected to result in reproductive effects, levels in shag
ing similar lipid levels in muscle compared to eggs
(a species of cormorant) and fulmar were at or exceeded
(Figure 6ท1).
the LOAEL for embryo deformities and the lower NOEL
Based on dietary LOECs for PCBs (5000 ng/g ww),
for hatching success in white leghorn chicken, an ex-
HCB (100 000 ng/g ww), DDE (5000 ng/g ww), and
tremely sensitive species. In addition, mean PCB levels
dieldrin (100 ng/g ww) in peregrine falcon prey, levels
in puffin, murre, common guillemot, black guillemot,
in blue mussels and anadromous and marine fish are
and kittiwake exceeded the upper NOEL for hatching
several orders of magnitude below those expected to
success in white leghorn chicken.
cause effects in animals that eat them (Annex Table 10).
Mean PCB levels in glaucous gull and herring gull
Mean dieldrin levels in blue mussels and Arctic anadro-
eggs were somewhat higher and approached or exceeded
mous and marine fish from Canadian, Norwegian, and
the LOEL for hatching success in white leghorn chicken
Greenlandic waters (Annex Table 10) do not exceed any
and the NOEL for hatching success found for Forster's
of the guidelines for protecting aquatic biota-eating
tern. None of the Arctic seabirds studied appear to have
wildlife (Table 6.1). PCB and DDT levels in a range of
PCB levels high enough to be associated with thresh-
fish species from many Arctic sites exceed some, and in a
olds for egg mortality. Mean DDT, HCH, dieldrin, and
few cases for PCBs, all of the environmental quality
HCB levels in all seabirds studied were several orders of
guidelines for protecting aquatic biota consumers/fish-
magnitude below those expected to cause effects.
eating wildlife given in Table 6.1 (Figures 6.5 and 6.6).
Based on dietary LOECs for PCBs, HCB, DDE, and
Toxaphene levels in several fish species also exceed the
dieldrin in peregrine falcon prey, levels in anadromous
Canadian Tissue Residue Guideline for protecting wild-
and marine fish were several orders of magnitude below
life consumers of aquatic biota (Table 6.1). This implies
those expected to cause effects in falcons. Mean dieldrin
that dietary intakes of DDTs, PCBs, and toxaphene in
levels in Arctic anadromous and marine fish from Cana-
some fish-eating seabirds may be high enough to lead to

---

Chapter 6 ท Biological Effects
177
effects if seabirds prey on these species. No data are
ilar PCB and TEQ (based on eight mono-ortho CBs)
available for levels of dioxin-like compounds in anadro-
levels. No effects of either PCBs or TEQs on plasma
mous and marine fish so the dietary intake of TEQs can-
retinol levels were found at either site (Jenssen et al.,
not be assessed.
2001).
For DDTs, the fish species that exceed both Cana-
A borderline significant positive correlation was
dian and U.S. EPA guidelines are spotted wolffish
found between lipid weight PCB concentrations in yolk
(liver) from southwest Greenland; Atlantic cod (liver)
and plasma retinol levels in one-day-old shag hatchlings
from Greenland, Iceland, and the Faroe Islands; short-
from the central Norwegian coast (Murvoll et al., 1999).
horn sculpin (liver) from northeast and southwest
Greenland and the Faroe Islands; starry ray (liver) from
Levels and intake assessment
Greenland; Greenland shark (liver) from Canada; stur-
Mean PCB concentrations in Norwegian shag hatch-
geon (liver) from the Kara Sea; and, Greenland halibut
lings were 1.2 ตg/g ww and for grey heron hatchling
(liver) from Greenland. The fish species that exceeded
yolk sacs, 2.1-2.5 ตg/g ww (range 0.57-4.9 ตg/g ww)
the International Joint Commission objectives for pro-
(Annex Table 11). Grey heron levels exceed the NOEL
tection of aquatic life and wildlife of 1000 ng/g ww
for hatching success in Forster's tern and several LOELs
(Table 6.1) were Greenland shark (liver) from Canada
for egg mortality and embryo deformities in other bird
and Greenland halibut (liver) from West Greenland
species (Figure 6ท1). For grey heron, the mean TEQ con-
(Figure 6ท6).
centrations based on eight mono-ortho PCBs were 50-79
For PCBs, the species that exceed Canadian guide-
pg/g ww (range 21-179 pg TEQ/g ww) (1020-1170 pg/g
lines for avian predators are polar cod (liver) from Jan
lw), exceeding some NOAELs and LOAELs for repro-
Mayen (Norway); Atlantic cod (liver) from Greenland
ductive effects in other bird species (Figure 6ท2) (Annex
and Iceland; shorthorn sculpin (liver) from Greenland
Table 16).
and the Faroe Islands; starry ray (liver) from Greenland;
The assessment for dietary intake of dieldrin, HCB,
Greenland shark (liver) from Canada; sturgeon (liver)
PCB, DDT, and toxaphene is the same as given for eiders
from the Kara Sea; Greenland halibut (liver) from
(Section 6.3.3.1).
Greenland; Arctic char (muscle) from Nain, Canada,
and a number of other marine species (liver) from
6.3.3.3. Alcids
around Jan Mayen (coalfish, long rough dab, checkered
eelpout, and Atlantic poacher). The fish species that ex-
Cytochrome P450 activities/retinol effects
ceed International Joint Commission objectives are At-
Three groups of black guillemot nestlings in Saglek Bay,
lantic cod (liver) from Greenland; shorthorn sculpin
Canada, with low, medium, and high PCB exposures,
(liver) from southwest Greenland and the Faroe Islands;
were studied for several liver biomarkers (Kuzyk et al.,
sturgeon (liver) from the Kara Sea; Greenland shark
2003). This bay has PCB-contaminated marine sedi-
(liver) from Canada; Greenland halibut (liver) from
ments due to a former military site. There was a signifi-
Greenland; and, long rough dab (liver) and checkered
cant dose-dependent increase in liver size, EROD activ-
eelpout from Jan Mayen. Only liver from sturgeon from
ity up to a threshold of 100 ng/g ww liver, and reduced
the Kara Sea, Greenland halibut and Atlantic cod from
liver retinol and retinyl palmitate (females only) with in-
Greenland, Greenland shark from Canada, and long
creasing liver PCB concentrations. Many effects were
rough dab and checkered eelpout from Jan Mayen ex-
more pronounced in female nestlings than in males.
ceed the U.S. EPA guideline values for assessment of haz-
ards to fish-eating wildlife (Figure 6ท5).
Levels and intake assessment
For toxaphene, the species that exceed the Canadian
In many cases, tissues other than eggs (muscle, liver and
Tissue Residue Guideline are spotted wolffish (liver)
fat) have been analyzed in alcids (little auk, thick-billed
from southwest Greenland; polar cod (liver) from Bar-
murres, black guillemots and common guillemots) at
row, Alaska; Atlantic cod (liver) from southwest Green-
various sites. This makes comparisons difficult because
land and Iceland; herring (whole fish) from the White
concentrations in other tissues may not be reflective of
Sea, Russia; shorthorn sculpin (liver) from southwest
those found in the eggs (Braune and Norstrom, 1989;
Greenland and the Faroe Islands; dab (liver) from Ice-
Drouillard and Norstrom, 2001). If only PCB concen-
land; starry ray (liver) from southwest Greenland and
trations in eggs are used, values range from 0.087 to
Arctic char (fillet) from Barrow (Annex Table 10).
0.27 ตg/g ww for thick-billed murres from the Pribilof
Islands, Alaska (Bering Sea), Lancaster Sound, Canada
and Bj๘rn๘ya, and 0.11 to 0.29 ตg/g ww in black guille-
6.3.3.2. Grey heron, shag
mots from Qeqertarsuaq and Ittoqqortoormiit, Green-
Reproductive effects
land and Bj๘rn๘ya (Annex Table 11). The PCB concen-
A significant negative correlation was found between
trations based on eggs are below thresholds for effects in
wet weight PCB concentrations in yolk and egg vol-
other fish-eating birds (Figure 6ท1).
ume, yolk mass, and hatchling mass in one-day-old shag
PCB concentrations in liver of black guillemots
hatchlings from the central Norwegian coast (Murvoll et
from Baffin Bay and two sites at Saglek Bay (Canada),
al., 1999).
Qeqertarsuaq and Ittoqqortoormiit (Greenland), the
Faroe Islands, and Svalbard are similar to or lower than
Retinol effects
levels seen in eggs from Greenland and Bj๘rn๘ya (eggs
Grey heron hatchlings from two rookeries in Norway,
have lipid contents of approximately 10%, liver has a
one at Fr๘ya on the central Norwegian coast and one at
lipid content of 4-6%). However, PCB concentrations
Finnfjord๘y on the north coast, were found to have sim-
in black guillemot livers from Jan Mayen (1.2 ตg/g ww)

---

178
AMAP Assessment 2002: Persistent Organic Pollutants in the Arctic
and at the more contaminated site at Saglek Bay (0.17-
individuals with different levels, controlling for a set
6.5 ตg/g ww) are higher, and indicate that these popula-
of potential confounding variables (body condition,
tions probably have elevated levels in eggs as well. The
breeding areas, laying dates and others) (Bustnes et al.,
levels at Jan Mayen are probably below thresholds for
2003). Females with high circulating levels of OCs, in-
effects, but at Saglek Bay, they indicate that some indi-
cluding HCB, oxychlordane, DDE, and PCBs, were
viduals probably have levels that may be above thresh-
more likely to have non-viable eggs in the nest than fe-
olds for effects (Figure 6ท1). PCB concentrations in
males with low PCB levels. Moreover, the body condi-
liver of thick-billed murres from Baffin Bay (Canada),
tion at hatching was poorer for the first chick in the
Nuuk (Greenland), and Svalbard are lower than in eggs
clutch in females with high concentrations of all per-
from Alaska and Lancaster Sound (Canada), indicating
sistent OCs. For the second chick in the clutch, only
that eggs from these populations probably have levels
HCB, -HCH, and CB28 showed such a negative rela-
below thresholds for effects.
tionship. Apart from a negative association between
TEQ concentrations based on PCDDs, PCDFs and
concentration of some OCs and laying date, no other
non-ortho PCBs were 24 pg/g ww (719 pg/g lw) in liver
reproductive parameters, such as clutch size, egg size,
and 26 pg/g ww in eggs for Canadian thick-billed mur-
incubation time, nest predation or early chick survival,
res collected in 1993 (Braune and Simon, 2002), 25 pg/g
showed any association with OCs. Hence, reproduc-
ww (640 pg/g lw) in black guillemot liver from Saglek
tion seems moderately affected by OCs. Adult survival
Bay collected in 1999 and 9.8 pg/g ww (66 pg/g lw) for
was significantly negatively related to four different
common guillemot eggs collected from the Faroe Islands
OCs: HCB, DDE, CB153, and, in particular, oxychlor-
in 2000 (Annex Table 16). The value for Canadian
dane. In long-lived birds, such as glaucous gulls, adult
thick-billed murre eggs exceeds the LOAEL for repro-
survival probability is the key parameter to which the
ductive effects in wood duck, a sensitive species, but for
population growth rate is most sensitive, suggesting
common guillemots, the levels are below thresholds (Fig-
that OCs may have considerable effect on growth in
ure 6ท2). Lipid weight TEQ levels in liver in these species
glaucous gull populations. However, the effects on spe-
are well below the LOEL for induction of CYP1A in
cific populations will depend on the proportion of the
common terns (25000 pg TEQ/g lw).
population exposed to high intake of contaminants via
The assessment for dietary intake of dieldrin, HCB,
the diet.
PCB, DDT, and toxaphene is the same as given for eiders
It was also demonstrated that glaucous gulls with
(Section 6.3.3.1).
high levels of various persistent OCs had asymmetry in
wing feathers (i.e. feathers on the left and right wing
were of different length). Feather asymmetry is a well-
6.3.3.4. Gulls
known indication of developmental stress, suggesting
At Bj๘rn๘ya, Bourne and Bogan (1972) were the first to
that OCs are an extra stress for the birds during molt. In
observe aberrant behaviors in glaucous gulls with high
this study, it was shown that the effect of HCB was
PCB levels. In the Svalbard archipelago, sick and dying
much stronger than for the other OCs, such as PCBs and
glaucous gulls were found to have high PCB levels
DDE (Bustnes et al., 2002).
(Daelemans, 1994; Gabrielsen et al., 1995).
Cytochrome P450 activities
Reproductive effects
Only a weak association was found between PCB (nine
Recently, a study was carried out to examine whether
congeners) concentrations and CYP activity measured as
behavior at the nesting stage was negatively influenced
EROD activity in Bj๘rn๘ya glaucous gulls (Henriksen et
by PCBs. Individual patterns of incubation and nest-site
al., 2000). No correlations were seen between OC levels
attentiveness were studied in relation to PCB concentra-
and testosterone hydroxylation activity.
tions in the blood of twenty-seven glaucous gulls in two
breeding areas. PCB concentrations in the blood ranged
Immunosuppression
from 52 ng/g ww to 1079 ng/g ww. There were signifi-
Numbers of intestinal macroparasites were compared to
cant differences between the two breeding areas, and fe-
hepatic concentrations of several POPs in 40 glaucous
males had significantly lower concentrations than males
gulls from Bj๘rn๘ya (Sagerup et al., 2000). After con-
(Bustnes et al., 2001).
trolling for nutritional condition, no single parasite
Plasma PCB concentrations were positively related to
species was significantly correlated with PCB or pesti-
the proportion of time the birds were absent from the
cide concentrations. However, the intensity of all nema-
nest, both overall and when not incubating, and to the
todes grouped together was positively correlated with
number of absences, when controlling for possible con-
concentrations of p,p'-DDT, mirex, and PCBs (nine
founding variables (area, sex, body condition, and oth-
congeners). This indicates that high POP concentrations
ers). Increased absence from the nest site in individual
may affect immune function in the glaucous gull.
glaucous gulls with high blood concentrations of OCs
suggests that they need more time to gather food be-
Thyroid and retinol effects
cause of either endocrine disruption or neurological dis-
No significant correlations were seen between PCB
orders. This probably led to increased energetic costs
concentrations and hepatic retinol and retinyl palmitate
during incubation and reduced reproductive output
concentrations in forty glaucous gulls from Bj๘rn๘ya
(Bustnes et al., 2001).
(Henriksen et al., 2000). A significant negative correla-
To examine the effects of OCs on individual fitness,
tion was seen between HCB, p,p'-DDE, and PCB blood
blood concentrations of various compounds were com-
concentrations and plasma T4 levels in male glaucous
pared to reproductive parameters and adult survival in
gulls from Bj๘rn๘ya (Verreault et al., 2002).

---

Chapter 6 ท Biological Effects
179
Levels and intake assessment
that consume aquatic biota (Figure 6ท6). Most PCB
PCB concentrations in the brains of glaucous gulls found
levels exceed all guideline levels (Figure 6ท5) and where
dead on Svalbard ranged from 0.9 to 29.5 ตg/g ww
measured, toxaphene levels exceed Canadian guide-
(Gabrielsen et al., 1995) and in living glaucous gulls,
lines. PCDD/F and nPCB levels given as TEQs exceed
from 0.5 to 9.5 ตg/g ww (Henriksen et al., 1998b).
Canadian (avian) and U.S. EPA guideline values (Table
Tissues other than eggs (muscle, liver, and fat) have
6.1) in kittiwake, fulmar and murre eggs and liver,
been analyzed in glaucous gulls at various sites (Annex
black guillemot liver and common guillemot eggs (An-
Table 11). As stated previously, this makes comparisons
nex Table 16). Thus, seabirds that prey on seabird eggs,
difficult. Mean PCB concentrations in liver range from
chicks and adults may have dietary intakes of TEQs,
0.45 ตg/g ww to 22.7 ตg/g ww and indicate that several
DDTs, PCBs and toxaphene high enough to lead to
populations probably have elevated levels in eggs as
effects.
well, some of which may be above thresholds for effects
(Figure 6ท1). The populations with high PCB concen-
6.3.3.5. Black-legged kittiwakes
trations in liver that might be associated with egg levels
above thresholds for effects are from Bj๘rn๘ya, Sval-
On a Swedish Arctic research expedition in 1996 in the
bard, Novaya Zemlya, Jan Mayen, and Franz Josef
Barents Sea, several juvenile kittiwakes with crossed bills
Land.
and clump feet were observed (Kylin, 1997a; 1997b).
TEQ concentrations were calculated to be 2500 pg/g
ww (approximately 60 000 pg/g lw) based on the nPCB
Levels and intake assessment
and mono-ortho PCB concentrations in Svalbard glau-
In many cases, tissues other than eggs (muscle, liver and
cous gull liver (Daelemans et al., 1992) (Annex Table
fat) have been analyzed in black-legged kittiwakes at
16). These TEQ levels exceed all NOAELs and LOAELs
various sites. As stated previously, this makes compar-
for reproductive effects and LD50s in a range of other
isons difficult. If only values for PCB concentrations in
bird species (Figure 6ท2). Lipid weight TEQ levels in
eggs are used, the mean is 0.28 ตg/g ww at Lancaster
liver also exceed the LOEL for induction of CYP1A in
Sound, Canada (Annex Table 11). The PCB concentra-
common terns.
tions based on eggs do not exceed any thresholds for ef-
For ivory gull from north Baffin Bay, liver and fat
fects in fish-eating birds (Figure 6ท1). PCB concentra-
tissues were analyzed (Annex Table 11). PCB concen-
tions in liver range from 0.12 to 1.4 ตg/g ww in kitti-
trations in liver are 0.3 ตg/g ww, and indicate that levels
wakes from sites in Baffin Bay, Nuuk (Greenland)
in eggs are also low and probably below thresholds for
Bj๘rn๘ya, Jan Mayen, Svalbard, Franz Josef Land, No-
effects. For great black-backed gulls from Jan Mayen,
vaya Zemlya, and along the Russian Barents Sea coast.
liver was analyzed. PCB concentrations in liver are
This indicates that several populations probably have el-
9.6 ตg/g ww, and indicate that levels in eggs may also be
evated levels in eggs as well, but these are probably
high and probably exceed some thresholds for effects.
below thresholds for effects.
TEQ concentrations based on PCDD/Fs, nPCBs and
TEQ concentrations based on PCDD/Fs and nPCBs
mono-ortho PCBs were determined in herring gulls and
were 47 pg/g ww (1117 pg TEQ/g lw) in liver and 78
great black-backed gulls from four sites in northern
pg/g ww in eggs for Canadian black-legged kittiwakes
Norway (Gabrielsen, 2002). These concentrations ranged
collected in 1993 (Annex Table 16) (Braune and Simon,
from 37 to 72 pg TEQ/g ww. These TEQ levels exceed
2002). The value for eggs exceeds the LOAEL for repro-
the LOAEL for reproductive effects in wood duck, a sen-
ductive effects in wood duck, a sensitive species (Figure
sitive species (Figure 6ท2).
6ท2). Lipid weight levels in liver are well below the
Levels of DBT (n.d.-51 ng/g ww) and MBT ( n.d.-14
LOEL for induction of CYP1A in common terns.
ng/g ww) in glaucous gull livers from Bj๘rn๘ya (Berge et
The assessment for dietary intake of dieldrin, HCB,
al., 2002) are well below the thresholds associated with
PCB, DDT, and toxaphene is the same as given for eiders
hepatic and immune effects. TBT levels were below the
(Section 6.3.3.1).
detection limits.
Ivory, herring, and great black-backed gulls feed
6.3.3.6. Fulmar
primarily on fish but also scavenge. Glaucous gulls
prey on eggs, chicks, and even adult seabirds as well as
No effects studies have been carried out on fulmars in
fish. The assessment for dietary intake of dieldrin,
the Arctic.
HCB, PCB, DDT, and toxaphene from fish is the same
as given for eiders (Section 6.3.3.1). Based on contami-
Levels and intake assessment
nant levels found in eggs from thick-billed murres,
Besides eggs, other tissues (liver, fat) have been analyzed
black guillemots, black-legged kittiwakes, and fulmars
in northern fulmars at various sites (Annex Table 11). As
from the Alaskan, Canadian, Greenlandic, and Norwe-
stated previously, this makes comparisons difficult. If
gian Arctic (Annex Table 11), dieldrin, DDT, and
only values for PCB concentrations in eggs are used,
PCB levels are below dietary LOECs for reproductive
the mean is 0.27 ตg/g ww at Lancaster Sound, Canada
effects found in peregrine falcons. Mean dieldrin levels
(Annex Table 11). The mean PCB concentrations in
in these seabird species' eggs also do not exceed any en-
eggs from Lancaster Sound are below any effects thresh-
vironmental guidelines for protecting aquatic wildlife.
olds (Figure 6ท1). PCB concentrations in liver range
However, DDT levels in eggs and liver from these
from 0.16 ตg/g ww to 0.69 ตg/g ww in fulmars from
species from many of these sites, as well as kittiwakes
northern Baffin Bay and Jan Mayen. These PCB con-
from Russian sites in the Barents Sea, exceed Canadian
centrations indicate that levels in eggs are probably low
and U.S. EPA guideline levels for protecting wildlife
and below thresholds for effects.

---

180
AMAP Assessment 2002: Persistent Organic Pollutants in the Arctic
TEQ concentrations based on PCDD/Fs and non-
Pribilof stock abundance is less than half of historical
ortho PCBs were 357 pg/g ww (8192 pg TEQ/g lw) in
levels and is listed as depleted under the Marine Mam-
liver and 83 pg/g ww in eggs for Canadian Arctic ful-
mal Protection Act (Loughlin et al., 1994; York et al.,
mars collected in 1993 (Annex Table 16) (Braune and
1997). The St. George subpopulation underwent an un-
Simon, 2002). The value for eggs exceeds the LOAEL
explained decline of 4-6% per year for more than a
for reproductive effects in wood duck, a sensitive species
decade prior to the mid-1990s study (York et al., 1997).
(Figure 6ท2). Lipid weight levels in liver are well below
Long-term monitoring of population trends suggest that
the LOEL for induction of CYP1A in common terns.
the decline was due, at least in part, to increased post-
The assessment for dietary intake of dieldrin, HCB,
weaning mortality at sea (Trites and Larkin, 1989;
PCB, DDT, and toxaphene is the same as given for eiders
Trites, 1992). Cause(s) of the increased mortality is un-
(Section 6.3.3.1).
known, but is thought to be associated with shifts in the
abundance and composition of their primary prey
species.
6.3.3.7. Great skua
Steller sea lion populations have been declining in
No effects studies have been carried out on great skuas
western Alaska including the Aleutian Islands, and the
in the Arctic.
western stocks are considered endangered.
Levels and intake assessment
Cytochrome P450 effects
Only liver has been analyzed in great skuas from Jan
Studies have been done on liver enzyme induction in
Mayen. The mean PCB concentration in liver was 15.9
ringed and harp seals around Svalbard, correlated to lev-
ตg/g ww. This indicates that great skuas at Jan Mayen
els of OCs such as PCB and toxaphene. Elevated hepatic
probably have elevated levels in eggs as well, and these
EROD activities were found in subadult harp seals sam-
are probably above thresholds for effects (Figure 6ท1).
pled from the northwest Barents Sea, east of Svalbard
The dietary assessment is the same as for glaucous gulls
(Wolkers et al., 2000). No correlation was found be-
(Section 6.3.3.4).
tween these activities and the PCB concentrations in the
seals. A highly positive correlation was, however, found
between toxaphene levels and testosterone 6- hydroxy-
6.3.4. Pinnipeds
lation activities (CYP3A). A positive relationship was
In the previous AMAP assessment, increased CYP1A en-
found between CYP enzyme activity and PCB levels in
zyme activities were correlated to PCB and dieldrin lev-
ringed seals from Svalbard (Wolkers et al., 1998b).
els in Canadian ringed seals and to PCB levels in West
Ringed seals from Svalbard and grey seals from Sable Is-
Ice (Jan Mayen area) hooded seals. PCB levels in Arctic
land (Canadian Arctic) had lower EROD and PROD ac-
harp, ringed, harbour and grey seals from all sites stud-
tivities than grey and ringed seals from the Baltic Sea
ied exceeded the NOAEL and LOAEL for subtle neu-
(Nyman et al., 2000).
robehavioral effects if they were as sensitive as the off-
spring of rhesus monkeys and humans, but were below
Thyroid and retinol effects
the NOEL for otter reproduction and mink kit survival
Retinol levels in northern fur seal neonates were nega-
(Figure 6ท3). PCDD/F and nPCB levels expressed as
tively correlated to two recalcitrant PCB congeners in
TEQs in ringed and harp seals from Svalbard, the
whole blood: CB138 and CB153/87 (rank correlation
Greenland Sea, and several sites in Canada were some-
coefficients of ญ 0.403 and ญ 0.452, respectively, p < 0.05)
what lower than levels associated with immunosuppres-
(Beckmen, 1999). Serum retinol levels in neonates were
sive effects in harbour seal (Figure 6ท4).
also negatively correlated to the TEQs in the perinatal
Concentrations of dieldrin, DDTs, and chlordanes
milk (correlation coefficient of ญ 0.475, p = 0.029, n = 21).
in marine crustaceans and fish were several orders of
Total T4 was negatively correlated with CB101/99/
magnitude below those expected to result in effects on
149/196, CB118, CB138, and TEQs (Spearman rank cor-
seal reproduction, and toxaphene levels in fish were
relation coefficients = ญ 0.313, ญ 0.519, ญ 0.457, ญ 0.354,
below those associated with thyroid effects. PCB levels
respectively, p < 0.05). The negative correlation of retinol
were also below effect levels in crustaceans, but levels
and thyroid hormones with PCBs in northern fur seal
exceeded the dietary NOAEC for reproduction in sev-
blood suggests that blood levels of these may be reduced
eral fish species. However, PCB and DDT levels in a
because of PCB exposure in young pups. Thus, PCB ex-
range of fish species from many Arctic sites exceeded
posure in young pups has the potential to affect immune
some, and in a few cases for PCBs, all of the environ-
function, and therefore, health, both through direct im-
mental quality guidelines for protecting fish-eating
munosuppression and indirectly, by lowering circulating
wildlife. This indicates that dietary intakes of DDTs
levels of retinol and thyroxine.
and PCBs in fish-eating marine mammals may be high
enough to lead to effects.
Immune system effects
The native Aleut populations of the villages on St.
George and St. Paul are dependent on an annual subsis-
6.3.4.1. Seals and sea lions
tence harvest of subadult (2-4 years of age) male fur
Reproductive effects
seals both culturally and as a major source of protein.
Breeding rookeries for more than 72% of the world's
Aleut concerns over the unexplained decline in the pop-
population of northern fur seals are located on the two
ulation of northern fur seals prompted a study to eval-
largest Pribilof Islands, St. Paul and St. George, Alaska
uate the potential effects of OC contaminant exposure
in the Bering Sea (Loughlin et al., 1994). The current
on immune function in a cohort of free-ranging pups.

---

Chapter 6 ท Biological Effects
181
In 1996, 50 perinatal pups were captured for blood
proximately monthly intervals from birth to weaning,
sample collection (during the ten-day perinatal period
further substantiated the results of the original study.
and referred to as neonates) and forty-three were re-
The pups were given both primary and secondary vac-
sampled (`pups') 29 to 51 days later. Groups of pups
cinations with tetanus toxoid and examined for pri-
were compared based on the relative age of the dam
mary and secondary (memory) antibody response. When
to study immunologic effects of exposure via milk to
converted on a molecular weight basis, the effect of
high (young dam, presumably primiparous) or low
PCB blood levels at the time the initial and booster vac-
(old dam, multiparous) doses of OCs (Beckmen et al.,
cinations with tetanus toxoid were given had a highly
1999). There were no significant differences in the
significant (p = 0.008 and 0.026, respectively) negative
mass or length (when adjusted for sex) of pups during
effect on the increase in antibody titer. All congeners
the perinatal period between pups born to young and
examined, except for p,p'-DDE, were significant indi-
old dams. Likewise, growth rates during the study were
vidually. In these first-born pups, using set correlation
similar, and the survival rates to the middle of the nurs-
analysis, it was determined that the total immunoglob-
ing period were not significantly different although the
ulin concentrations were related to both TEQs (nega-
survival rates of pups of young dams were higher (88%
tive) and the age (in days) of the pup (p < 0.0001,
vs. 84%, p = 0.858). The only significant difference was
p = 0.0512, respectively). Additionally, total immuno-
a slightly higher body condition index score in pups of
globulins were related to PCB concentrations and the
old dams (p = 0.047). Mean blood OC levels were
age of the pup. Perinatal TEQs had a significant nega-
higher in neonates than at recapture, and neonates of
tive correlation to the total immunoglobulin levels at
young dams had higher mean blood OC levels than
that time and at the next recapture. TEQs at the first
neonates of older dams (Beckmen, 1999; Beckmen et
recapture were also significantly negatively correlated
al., 1999).
to total immunoglobulins at that time and nearly sig-
In the same pups, humoral immune function was as-
nificantly negatively correlated subsequently (r = ญ 0.65
sessed by antibody responses to tetanus toxoid vaccina-
and ญ 0.633; p = 0.0501 and 0.0583, respectively). Set
tion and total immunoglobulin levels. Cellular immune
correlation analysis confirms this effect is correlated to
function was assessed using mitogen-induced lympho-
TEQs by taking into consideration the effect of age on
cyte proliferation assays. Additional indicators of health
immunoglobulin levels. The strong correlation with
status included complete blood cell counts and hapto-
PCB exposure expressed as TEQs suggests that the ef-
globin levels. A higher proportion of pups born to old
fect on humoral immunity may be mediated through,
dams developed a two-fold or greater increase in tetanus
or by a similar mechanism as, the Ah-receptor. PCBs,
antibodies compared to the pups of young dams. Forty-
TEQs, and p,p'-DDE were all negatively correlated
one percent (9 of 22) of the pups of old dams responded
(Spearman Rank Order Correlation) to lymphoprolif-
with the expected 2-fold or greater increase, whereas
erative assays. However, when a set correlation analy-
only 5% of the pups of young dams responded to the
sis was applied, age explained most of the effect. Thus,
antigen (1 of 21). The difference in the proportion of
PCB exposure appears to have a negative impact on hu-
pups responding by the dam's age was significant
moral immunity (immunoglobulin levels and tetanus
(Z = 2.443, p = 0.015). When hemograms of neonates of
antibody responses), but the effects on cellular immune
young and old dams were compared, the mean hemato-
function were not substantiated in this cohort (Beck-
crit, total plasma protein and absolute eosinophil counts
men, 2002).
were significantly greater in the neonates of old dams. In
Blood and blubber samples obtained from 24 free-
neonates, elevated immature (band) neutrophil counts
ranging Steller sea lion pups and juveniles during the
(indicative of inflammation) were correlated to higher
course of live-capture studies in the ranges of both east-
TEQs (r = 0.347, p = 0.016, n = 48). In recaptured pups,
ern and western stocks, have recently been subjected to
the total leukocyte counts were positively correlated to
similar OC contaminant analysis and immune function
higher total PCB concentrations in blood (r = 0.359,
assays as the northern fur seals. In this species as well,
p = 0.023, n = 41).
there were significant negative correlations between
Levels of serum haptoglobin were used as a further
lymphocyte proliferative responses to both T cell and
measure of subclinical inflammation. There were no dif-
mixed B and T cell mitogens and increased blubber
ferences between neonatal and recapture levels for the
TEQs (concanavalin A: r = ญ0.440; pokeweed mitogen:
pups of old dams, but haptoglobin levels increased sig-
r = ญ0.448, p < 0.05). In a subset of eight pups in the
nificantly from neonatal to recapture for pups of young
eastern, endangered stock, the total immunoglobulin
dams (Mann-Whitney Rank Sum test, p < 0.001). No di-
serum concentrations were strongly negatively corre-
rect correlations of serum haptoglobin concentration
lated with blubber concentrations of PCBs, TCDD
with blood OC concentrations were detected. Neonates
TEQs and p,p'- DDE (r = ญ0.933, p = 0.005; r = ญ0.881,
of young dams had significantly (p < 0.001) lower mean
p < 0.001; r = ญ0.893, p < 0.001, respectively) (Beckmen,
immunoglobulin levels than neonates of old dams, 3.14
2002).
ฑ 1.2 ตg/ml versus 4.6 ฑ1.1 ตg/ml, respectively. Lympho-
proliferative responses, when combined with the results
Levels and intake assessment
of the previous year's cohort of neonatal fur seal pups,
The mean PCB levels in Arctic ringed, harbour, grey,
were negatively correlated with blood OC levels, indi-
spotted, and bearded seals, Steller sea lions (Gulf of
cating a possible suppression of cellular immune func-
Alaska and southeast Alaskan populations) and north-
tion (Beckman, 1999).
ern fur seals range from 72 to 8400 ng/g lw in blubber
A recent study examining a cohort of 12, known-age,
(Annex Table 12). These levels exceed the NOAEL and
female pups born to young dams and sampled at ap-
LOAEL for subtle neurobehavioral effects in offspring

---

182
AMAP Assessment 2002: Persistent Organic Pollutants in the Arctic
of rhesus monkeys and humans, the NOEL for vitamin
rine mammal reproduction. Based on the NOAEC for
A reduction in otter, and in some species, are above the
rats and dogs, toxaphene levels in fish are below those
NOEL for otter reproduction, but below that for mink
associated with thyroid effects. Assuming that marine
kit survival (Figure 6ท3). Highest PCB levels were
mammals are as sensitive as mink, mean levels of PCBs
found for ringed seal from Saglek Bay, Labrador (9400
in tissues from several fish species exceed the dietary
ng/g lw) and from the Kara Sea (1900-11000 ng/g lw).
NOAEC for reproduction of 72 ng/g ww (Figure 6ท5).
Both exceed the NOEL for mink kit survival, and the
These include shorthorn sculpin (liver) from several sites
levels in the Kara Sea seals are just at the LOEL for vita-
on Greenland and the Faroe Islands; Greenland halibut
min A reduction in otter. These levels are considerably
(liver) and Atlantic cod (liver) from Greenland and Ice-
lower than those associated with poor reproductive suc-
land; starry ray (liver) from Greenland; sturgeon (liver)
cess in harbour and ringed seal (Figure 6ท3). No biopsy
from the Kara Sea; Greenland shark (liver) from Davis
data are available for Steller sea lions from endangered
Strait and Cumberland Sound; polar cod (liver) from Jan
populations in the Aleutian Islands. The PCB levels are
Mayen; and, several species (liver) from around Jan
much higher in scat (up to 7000 ng/g lw) from these po-
Mayen (long rough dab, Atlantic poacher, checkered
pulations when compared to those from the other two
eelpout) (Annex Table 10).
areas, indicating higher body burdens, but no assess-
Mean dieldrin levels in Arctic anadromous and ma-
ment of these endangered populations can be made. The
rine fish from Canadian, Norwegian, and Greenlandic
PCB levels in harp seal from Svalbard range from 247
waters (Annex Table 10) do not exceed any of the guide-
to 20 400 ng/g lw in blubber. These levels exceed the
lines for protecting fish-eating wildlife (Table 6.1). How-
thresholds for subtle neurobehavioral effects, the NOEL
ever, PCB and DDT levels in a range of fish species
and LOEL for vitamin A reduction in otter, the NOEL
from many Arctic sites do exceed some, and in a few
for mink kit survival, the LOAEL for decreased kit pro-
cases for PCBs, all of the environmental quality guide-
duction and kit body weight gain in mink (Figure 6ท3).
lines for protecting fish-eating wildlife given in Table 6.1
As well, these levels exceed the threshold associated with
(Figures 6.5 and 6.6). Toxaphene levels in several fish
immunosuppression in harbour seals and are just below
species also exceed the Canadian Tissue Residue Guide-
the threshold for immune effects in rhesus monkeys
line for protecting wildlife consumers of aquatic biota.
(Figure 6ท3).
This implies that dietary intakes of DDTs, PCBs, and
Levels of nPCBs and mono-ortho PCBs given as
toxaphene in some fish-eating mammals may be high
TEQs range from 27 to 90 pg/g lw in blubber and blood
enough to lead to effects if mammals prey on these fish
from various age groups of Alaskan northern fur seals
species. No data are available for levels of dioxin-like
(Annex Table 16). Highest TEQs are found in pups. The
compounds in anadromous and marine fish so the di-
TEQ levels for pups are below the combined nPCB and
etary intake of TEQs cannot be assessed.
mono-ortho TEQs (190 pg/g lw) associated with im-
For DDTs, the fish species that exceed both Cana-
munosuppressive effects in harbour seals (Figure 6ท4).
dian and U.S. EPA guidelines are spotted wolffish (liver)
For ringed seals from Pangnirtung, nPCB and mono-
from southwest Greenland; Atlantic cod (liver) from
ortho PCB concentrations in blubber given as TEQs
Greenland, Iceland and the Faroe Islands; shorthorn
were 0.51 to 0.85 pg/g lw. For ringed seals from Holman
sculpin (liver) from northeast and southwest Greenland
Island, TEQs based on PCDD/Fs, nPCBs and mono-
and the Faroe Islands; starry ray (liver) from Greenland;
ortho PCBs ranged from 4.8 to 97 pg/g lw and for ringed
Greenland shark (liver) from Canada; sturgeon (liver)
seals from the Kara Sea, nPCB and mono-ortho PCB
from the Kara Sea; and, Greenland halibut (liver) from
concentrations in blubber given as TEQs were 160 pg/g
Greenland. The fish species that exceeded the Interna-
lw. For ringed seals, bearded seals and spotted seals from
tional Joint Commission objectives for protection of
eastern Russia (Chukotka), TEQs based on PCDD/Fs
aquatic life and wildlife of 1000 ng/g ww were Green-
were 1.0-1.5 pg/g lw. These TEQ levels are all below the
land shark (liver) from Canada and Greenland halibut
threshold for immunosuppression in harbour seals (Fig-
(liver) from West Greenland (Figure 6ท6).
ure 6ท4).
For PCBs, the species that exceed Canadian guide-
PFOS levels in grey, ringed, and northern fur seal
lines for mammalian predators are spotted wolffish
from the Arctic as well as Steller sea lions from the
(liver) from southwest Greenland; polar cod from Jan
Alaskan coast are lower than in marine mammals from
Mayen (Norway); Atlantic cod (liver) from Greenland,
more southerly latitudes, ranging from <3 to 120 ng/g
the Faroe Islands, Iceland and Svalbard; shorthorn
ww in liver or plasma (Giesy and Kannan, 2001; Kan-
sculpin (liver) from Greenland and the Faroe Islands;
nan et al., 2001a). These levels are well below the
dab (liver) from Iceland; starry ray (liver) from Green-
NOAEL (15 000 ng/g ww in liver) and LOAEL (58 000
land; Greenland shark (liver) from Canada; sturgeon
ng/g ww in liver) for second generation effects in rats.
(liver) from the Kara Sea; Greenland halibut (liver and
Levels of TBT (1.9-5.6 ng/g ww) and DBT (n.d.-20
muscle) from Greenland; Arctic char (muscle) from
ng/g ww) in Alaskan Steller sea lions, in Svalbard ringed
several sites in Canada and liver from some Russian
seals (DBT-3.1 ng/g ww, MBT-1.5 ng/g ww, TBT not de-
sites; Atlantic salmon (Salmo salar) (muscle) from West
tected), and Canadian ringed seals (below detection) are
Greenland; and, a number of other marine species
well below the thresholds associated with hepatic and
(liver) from around Jan Mayen (coalfish, long rough
immune effects in laboratory rodents.
dab, checkered eelpout, daubed shanny, Atlantic poacher,
Based on the dietary NOAECs and LOAECs given
grey gurnard) and the White Sea, Russia (Gadus sp,
for mink, mean levels of dieldrin, DDTs (Figure 6ท6),
herring, navaga (Eliginius navaga), sculpin). The fish
and chlordanes in marine fish are several orders of mag-
species that exceed International Joint Commission ob-
nitude below those expected to result in effects on ma-
jectives are Atlantic cod (liver) from Greenland; short-

---

Chapter 6 ท Biological Effects
183
horn sculpin (liver) from southwest Greenland and the
Levels of DDTs, PCBs, and PCDD/Fs in invertebrates
Faroe Islands; sturgeon (liver) from the Kara Sea;
are also below the various environmental guidelines
Greenland shark (liver) from Canada; Greenland hal-
(Table 6.1). Mean dieldrin levels in blue mussels from
ibut (liver) from Greenland; and, long rough dab (liver)
Canadian and Greenlandic waters (Annex Table 10) do
and checkered eelpout from Jan Mayen. Only liver
not exceed any of the guidelines for protecting fish-eat-
from sturgeon from the Kara Sea, Greenland halibut,
ing wildlife (Table 6.1). Toxaphene levels in blue mussels
and Atlantic cod from Greenland, Greenland shark
from one Greenland site and from several sites around
from Canada as well as long rough dab and checkered
Iceland are below the Canadian guideline. However,
eelpout from Jan Mayen exceed the U.S. EPA guideline
PCB levels in blue mussels from some sites in northern
values for assessment of hazards to fish-eating wildlife
Quebec, Canada, exceed the Canadian Tissue Residue
(Figure 6ท5).
Guideline for protecting wildlife consumers of aquatic
For toxaphene, the species that exceed the Canadian
biota (Figure 6ท5).
Tissue Residue Guideline are spotted wolffish (liver)
from southwest Greenland; polar cod (liver) from Bar-
6.3.5. Cetaceans
row, Alaska; Atlantic cod (liver) from southwest Green-
land and Iceland; herring (whole fish) from the White
In the previous AMAP assessment, increased CYP1A
Sea, Russia; shorthorn sculpin (liver) from southwest
enzyme activities were correlated to mono-ortho and
Greenland, and the Faroe Islands; dab (liver) from Ice-
nPCB concentrations in beluga from the Canadian Arc-
land; starry ray (liver) from southwest Greenland and
tic. Minke whale, beluga, and narwhal had PCB levels
Arctic char (fillet) from Barrow (Annex Table 10).
which exceeded the NOAEL and LOAEL for subtle
neurobehavioral effects if they are as sensitive as off-
spring of rhesus monkeys and humans, but were below
6.3.4.2. Walrus
the NOEL for otter reproduction and mink kit survival
In the previous AMAP assessment, PCB levels for the
(Figure 6ท3). For harbour porpoise from the southern
NOAEL and LOAEL for subtle neurobehavioral effects
Barents Sea, PCB levels also exceeded the NOEL for
in offspring of rhesus monkeys and humans were ex-
otter reproduction, mink kit survival, and the levels as-
ceeded in walrus from eastern Baffin Island, eastern
sociated with immunosuppression, and approached the
Hudson Bay, northeastern Hudson Bay, and Svalbard.
levels associated with poor reproductive success in har-
PCB levels in walrus from eastern Hudson Bay (Inuk-
bour seal (Figure 6ท3). Canadian beluga and narwhal
juak) and Svalbard also exceeded the NOEL for otter re-
had TEQ levels that were considerably lower than
production and mink kit survival (Figure 6ท3).
those associated with immunosuppressive effects (Fig-
Walrus from Inukjuak in eastern Hudson Bay had
ure 6ท4).
PCDD/F and/or nPCB TEQ levels that exceeded TEQ
Concentrations of dieldrin, DDTs, and chlordanes
levels associated with immunosuppression in harbour
in cetacean food items, such as marine crustaceans and
seals (Figure 6ท4).
fish, were several orders of magnitude below those ex-
For walrus feeding on marine invertebrates, no risks
pected to result in effects on reproduction, and dietary
for reproductive effects were indicated based on levels of
toxaphene levels in fish were below those associated
dieldrin, DDTs, chlordanes, PCBs, and PCDD/Fs, or
with thyroid effects. PCB levels were also below di-
based on environmental guidelines.
etary effect levels in crustaceans, but levels exceeded
Some walrus prey on ringed seal, and PCBs, DDTs,
the dietary NOAEC for reproduction in several fish
and TEQs in ringed seal blubber exceeded dietary
species. However, PCB and DDT levels in a range of
NOAECs and a range of guidelines for protecting aqua-
fish species from many Arctic sites exceeded some, and
tic wildlife, indicating exposure to these substances at
in a few cases for PCBs, all of the environmental qual-
levels that could be expected to lead to effects.
ity guidelines for protecting fish-eating wildlife, indi-
No effects studies have been carried out since the
cating that dietary intakes of DDTs and PCBs in
previous assessment.
fish-eating mammals could be high enough to lead to
effects.
Levels and intake assessment
No biological effects studies have been carried out on
Mean PCB levels in walrus from Alaska (Bering Sea),
cetaceans since the previous assessment.
Canada, and northeast and northwest Greenland
ranged from 33 ng/g lw to 3412 ng/g lw (Annex Table
6.3.5.1. Mysticetes
12). PCB levels in walrus from Alaska, Canada, and
6.3.5.1.1. Minke whales
northwest Greenland (Avanersuaq) are below all thresh-
olds for effects (Figure 6ท3). PCB levels in northeast
Levels and intake assessment
Greenland walrus from Ittoqqortoormiit exceed the
PCB levels in minke whales ranged from 230 ng/g lw to
NOAEL and LOAEL for subtle neurobehavioral effects
20760 ng/g lw. Mean PCB levels in all minke whales
in offspring of rhesus monkeys and humans, and are
exceed the NOAEL and LOAEL for subtle neurobehav-
close to the NOEL for vitamin A reduction in otter
ioral effects if they are as sensitive as the offspring of
(Figure 6ท3).
rhesus monkeys and humans. Some minke whales from
Based on the dietary NOAECs and LOAECs given
several sites (Jan Mayen, North Sea, Svalbard, north-
for mink, mean levels of dieldrin, DDTs, and chlor-
west Kola Peninsula, northern Norway/northwest Rus-
danes in marine invertebrates are below those expected
sia) have PCB levels that exceed the NOEL and LOEL
to result in effects on walrus reproduction. PCB levels
for vitamin A reduction in otter. Some minke whales
in the diet are also below effect levels in invertebrates.
from northern Norway/northwest Russia also exceed

---

184
AMAP Assessment 2002: Persistent Organic Pollutants in the Arctic
the NOEL for otter reproduction and mink kit survival,
6.3.5.2.2. Killer whales
the threshold for decreased kit production and kit body
weight gain in mink, the threshold associated with im-
Levels and intake assessment
munosuppression and vitamin A disruption in harbour
The mean PCB levels in killer whales from Alaska
seals, but are just below the LOAEL for immune effects
(Prince William Sound) are 14 400 ng/g lw in residents
in rhesus monkeys (Figure 6ท3).
(range: 1100-65 000 ng/g lw) and 240 000 ng/g lw in
TEQ concentrations based on PCDD/Fs, nPCB, and
transients (range: 66 000-550 000 ng/g lw) (Annex
mono-ortho PCBs were 15-74 pg TEQ/g lw in Svalbard
Table 13). The mean PCB levels in resident killer
(Spitsbergen) minke whales (Annex Table 16). These
whales exceed the NOAEL and LOAEL for subtle neu-
concentrations are below the threshold associated with
robehavioral effects in offspring of rhesus monkeys and
immunosuppression in harbour seals (Figure 6ท4).
humans, the NOEL and LOEL for vitamin A reduction
The dietary intake assessment for PCBs, DDTs,
in otter, the NOEL for otter reproduction and mink kit
dieldrin, chlordanes, and toxaphene is the same as for
survival, and the threshold for decreased kit production
seals and sea lions (Section 6.3.4.1).
and kit body weight gain in mink. Based on the maxi-
mum levels, PCB concentrations exceed the threshold
associated with immunosuppression and vitamin A dis-
6.3.5.1.2. Grey whales
ruption in harbour seals, the LOAEL for immune effects
Levels and intake assessment
in rhesus monkeys, the threshold for poor reproductive
PCB levels in Bering Sea and Chukotka grey whales
success in harbour seal, and the EC50 for decreased litter
ranged from 230 to 2700 ng/g lw, with means of 460-
size in mink (Figure 6ท3). For the transient killer whales,
1400 ng/g lw. The levels in the Bering Sea whales exceed
mean and maximum PCB levels are above all thresh-
the NOAEL and LOAEL for subtle neurobehavioral ef-
olds (Figure 6ท3).
fects if they are as sensitive as the offspring of rhesus
Mean TEQ concentrations based on mono-ortho
monkeys and humans, but are below the NOELs for
PCBs were 100 pg/g lw (range: 5.9-470 pg/g lw) for res-
otter reproduction and mink kit survival (Figure 6ท3).
idents and 860 pg/g lw (range: 190-2400 pg/g lw) for
The TEQ concentration based on PCDD/F levels in the
transients (Annex Table 16). The mean TEQ for resi-
Chukotka whales was 1.8 pg/g lw, which is below
dents is below the threshold associated with immuno-
thresholds for effects (Figure 6ท4).
suppression in harbour seals, but individuals with maxi-
mum values exceed this threshold (Figure 6ท4). All TEQs
for transients exceed this threshold.
6.3.5.1.3. Bowhead whales
The resident killer whales feed primarily on fish.
Levels and intake assessment
Therefore, the dietary intake assessment for PCBs,
Mean PCB levels in Alaskan bowhead whales are
DDTs, dieldrin, chlordanes, and toxaphene for resi-
below thresholds for effects (Figure 6ท3).
dent killer whales is the same as for seals and sea lions
(Section 6.3.4.1). The transient killer whales feed on
6.3.5.2. Odontocetes
other marine mammals. Based on mean POP levels in
ringed, bearded and northern fur seals as well as Steller
6.3.5.2.1. Beluga
sea lion blubber from various sites in Alaska (Annex
Levels and intake assessment
Table 12), dieldrin, chlordanes (heptachlor epoxide),
The mean PCB levels in beluga from Alaska, Canada,
and DDT levels are below the dietary NOAECs for re-
Greenland, and Svalbard range from 700 to 9000 ng/g
productive effects in mink. Assuming that marine mam-
lw with the highest concentrations at Point Lay (Ala-
mal-eating transient killer whales are as sensitive as
ska), Hendriksen Island (western Canada) and Kim-
mink, mean levels of PCBs in these seal species exceed
mirut (eastern Canada) (Annex Table 13). These PCB
the dietary NOAEC for reproduction of 72 ng/g ww. As
levels exceed the NOAEL and LOAEL for subtle neu-
prey items, mean DDT levels in all Alaskan seal
robehavioral effects in offspring of rhesus monkeys and
species exceed Canadian and U.S. EPA guidelines for
humans, the NOEL and LOEL for vitamin A reduction
protecting aquatic wildlife, and for northern fur seal,
in otter, and the NOEL for otter reproduction and mink
International Joint Commission objectives for protec-
kit survival (Figure 6ท3). The PCB levels in beluga brain
tion of aquatic life and wildlife are also exceeded (Fig-
from Alaska are below those associated with decreased
ure 6ท6). For PCBs, all Alaskan seal species exceed all
dopamine concentrations in non-human primate brain
environmental guidelines for protecting aquatic wildlife
tissue.
(Figure 6ท5) (Table 6.1). For nPCB TEQs, northern fur
TEQ concentrations based on nPCBs and mono-
seal blubber exceeds both the Canadian mammalian tis-
ortho PCBs in beluga from Kimmirut were 0.3-2.4 pg/g
sue residue guideline and the U.S. EPA guideline for
lw (Annex Table 16). If PCNs are included, the TEQs
protecting wildlife. No data for toxaphene are available
were 0.33-2.9 pg/g lw. TEQ concentrations based on
in Alaskan seal species and the dietary intake cannot be
nPCBs in beluga from Cumberland Sound (Baffin Is-
assessed.
land) were 6.1 pg/g lw. The TEQ levels in beluga are
well below the threshold associated with immunosup-
6.3.5.2.3. Long-finned pilot whales
pression in harbour seals (Figure 6ท4).
Levels of TBT and DBT in Canadian beluga were
Levels and intake assessment
below detection limits. The dietary intake assessment for
The mean PCB concentrations in long-finned pilot
PCBs, DDTs, dieldrin, chlordanes, and toxaphene is
whales from the Faroe Islands ranged from 16 000 ng/g
the same as for seals and sea lions (Section 6.3.4.1).
lw to 38 000 ng/g lw (Annex Table 13). The mean PCB

---

Chapter 6 ท Biological Effects
185
concentrations exceed the NOAEL and LOAEL for sub-
disruption in harbour seals, the LOAEL for immune ef-
tle neurobehavioral effects in offspring of rhesus mon-
fects in rhesus monkeys, and the threshold for poor re-
keys and humans, the NOEL and LOEL for vitamin A
productive success in harbour seals. The dietary intake
reduction in otter, the NOEL for otter reproduction and
assessment for PCBs, DDTs, dieldrin, chlordanes, and
mink kit survival, and the threshold for decreased kit
toxaphene is the same as for seals and sea lions (Section
production and kit body weight gain in mink (Figure
6.3.4.1).
6ท3). Based on the maximum levels, PCB concentra-
Levels of PCDD/Fs given as TEQs in Greenland har-
tions exceed the threshold associated with immunosup-
bour porpoise range from 0.2 to 0.9 pg/g lw, and the
pression and vitamin A disruption in harbour seals, the
mean TEQ concentration for PCDD/Fs, nPCBs, and
LOAEL for immune effects in rhesus monkeys, and the
mono-ortho PCBs in the Norwegian harbour porpoises
threshold for poor reproductive success in harbour seal
was 111 pg/g lw (Annex Table 16). These TEQ levels do
(Figure 6ท3). Low concentrations of DBT and TBT were
not exceed the threshold associated with immunosup-
found in a few adults and one fetus, but levels were well
pression in harbour seals (Figure 6ท4).
below thresholds for effects.
Levels of TBT (55-151 ng/g ww), DBT (127-490 ng/g
The dietary intake assessment for PCBs, DDTs,
ww), and MBT (11-58 ng/g ww) in Norwegian harbour
dieldrin, chlordanes, and toxaphene is the same as for
porpoise from the coast of the Barents Sea are below the
seals and sea lions (Section 6.3.4.1).
thresholds associated with hepatic and immune effects in
laboratory rodents.
6.3.5.2.4. Narwhal
Levels and intake assessment
6.3.6. Polar bear
Many measurements of PCB concentrations in narwhal
In the previous AMAP assessment, high cub mortality
blubber are based on only ten congeners (PCB10). The
was found in Svalbard polar bears. A significant nega-
mean PCB10 concentrations in narwhal range from 260
tive correlation was found between retinol and PCB
to 4750 ng/g lw (Annex Table 13). In general, the
concentrations in Svalbard polar bears. Hepatic CYP1A1
PCB10 levels are approximately 30% of PCB levels in
and 1A2 content in Canadian male polar bears were
beluga, and in one set of narwhal data where both val-
found to be correlated with levels of mono-ortho- and
ues were calculated. Assuming this is the case for the
nPCBs, and CYP2B content was correlated with concen-
other narwhal data, then the range of PCB concentra-
trations of total chlordane (mainly oxychlordane and
tions in narwhal from Canada, Greenland, and Svalbard
nonachlor), and total ortho-substituted PCBs. A spatial
is estimated to be 900 to 16 000 ng/g lw. These PCB
study of POP concentrations in fat was carried out
levels exceed the NOAEL and LOAEL for subtle neu-
which included polar bear populations from 16 areas
robehavioral effects in offspring of rhesus monkeys and
across the Arctic; from Canada, Greenland, and Sval-
humans, the NOEL and LOEL for vitamin A reduction
bard. All PCB levels exceeded the NOAEL and LOAEL
in otter, the NOEL for otter reproduction and mink kit
levels found for subtle neurobehavioral effects in off-
survival, and the threshold for decreased kit production
spring, if polar bear are as sensitive as offspring of rhe-
and kit body weight gain in mink, but are just below the
sus monkeys and humans. The PCB levels also ex-
threshold associated with immunosuppression and vita-
ceeded the NOEL for kit survival in mink in four of the
min A disruption in harbour seals (Figure 6ท3).
areas: Svalbard, East Greenland, M'Clure Strait, and
Narwhal eat squid, Arctic cod, shrimp, and Green-
eastern Hudson Bay. PCB levels were close to the kit
land halibut. Based on POP concentrations in Arctic
survival NOEL in several other areas. PCB levels in
marine fish, the dietary intake assessment for PCBs,
polar bear from three areas were at or above the LOAEL
DDTs, dieldrin, chlordanes, and toxaphene is the same
for immunosuppression in rhesus monkeys: Svalbard,
as for seals and sea lions (Section 6.3.4.1).
East Greenland, and M'Clure Strait.
Highest PCB levels were found in the population at
Svalbard. The PCB levels for different groups of Sval-
6.3.5.2.5. Harbour porpoise
bard polar bears exceeded the NOAEL and LOAEL for
Levels and intake assessment
offspring neurobehavioral effects as well as the NOEL
Mean PCB levels in harbour porpoise from northern
for mink kit survival. Some individuals in all groups
Norway are 15 000 ng/g lw (range: 7200-33 000 ng/g lw)
over three years of age exceeded the LOAEL for im-
and from southwest Greenland, 1300 ng/g lw (range:
munosuppression. Some individuals also exceeded the
880-1530 ng/g lw) (Annex Table 13). The mean PCB
levels known to be correlated with poor reproductive
levels from both sites exceed the NOAEL and LOAEL
success in harbour seals, as well as those correlated with
for subtle neurobehavioral effects if harbour porpoise
poor reproductive success in ringed seal and the EC50
are as sensitive as offspring of rhesus monkeys and hu-
for reduced litter size in mink.
mans. The mean PCB levels for the porpoises from
Some polar bear from the Canadian and Svalbard
Norway also exceed the NOEL and LOEL for vitamin A
populations had TEQ levels based on PCDD/F and/or
reduction in otter, the NOEL for otter reproduction and
nPCB levels that exceeded levels associated with im-
mink kit survival, and the threshold for decreased kit
munosuppressive effects in harbour seal.
production and kit body weight gain in mink, but are
Concentrations of PCBs, DDT, and TEQs in ringed
below the threshold associated with immunosuppression
seal blubber exceeded dietary NOAEC levels and a
and decreased vitamin A in harbour seals (Figure 6ท3).
range of guidelines for protecting aquatic wildlife, indi-
The maximum PCB levels, however, exceed the thresh-
cating exposure to these substances at levels that could
old associated with immunosuppression and vitamin A
be expected to lead to effects.

---

186
AMAP Assessment 2002: Persistent Organic Pollutants in the Arctic
The elevated levels of PCBs in adipose tissue of polar
reported previously (Wiig, 1998). The proportion of fe-
bears from the Svalbard area (ranging from 4790 to
males with cubs-of-the-year that were older than 16
80 300 ng/g) (Bernhoft et al., 1997) reported in the pre-
years old was found to be significantly lower on Sval-
vious POPs assessment, have prompted further biologi-
bard (12.7%) when compared to western Hudson Bay
cal effects studies linked to contaminant analyses.
(40.3%).
A significant negative relationship was found be-
Reproductive and developmental effects
tween PCBs and testosterone, as well as total pesticides
Some adult female polar bears at Svalbard were equip-
and testosterone in polar bear plasma, after correcting
ped with satellite transmitters programmed to send in-
for possible confounders such as age and condition, in-
formation every six days for two to three years. Infor-
dicating that PCBs and/or pesticides may decrease circu-
mation on the location of the transmitter as well as sen-
lating testosterone levels in male polar bears (Oskam et
sor data on internal transmitter temperature and short-
al., 2001) (Figure 6ท7). Testosterone is the major andro-
and long-term bear activity were recorded. Female polar
genic steroid hormone playing a crucial role in male sex-
bears normally have a three-year reproductive cycle
ual development.
(Ramsay and Stirling, 1988). Only pregnant bears den
Total pesticide concentration
Total PCB concentration
over winter. Reproductive rates were estimated from
in polar bear plasma,
in polar bear plasma,
ng/g ww
ng/g ww
satellite data (Wiig, 1995; 1998).
5000
25 000
From these studies of adult female polar bears at
Svalbard, the reproductive rate was found to be approx-
imately 0.75. This is similar to corresponding values
4000
20 000
found in other polar bear populations (Wiig, 1998). No
difference in the PCB levels between females available
3000
15 000
for mating that became pregnant, and those that did not
become pregnant, was found. The sample sizes for this
comparison were small, however (Bernhoft et al., 1997).
2000
10 000
Relatively low cub survival was found at Svalbard, and
there are indications that the reproduction cycle was less
1000
5000
than three years (Wiig et al., 1992; Wiig, 1998). Thus,
epizoological studies suggest that reproduction and cub
survival in polar bears at Svalbard may be impaired. The
0
0
1
2
3
4
1
2
3
4
high intake of PCBs at a crucial period could adversely
Testosterone quantile
Testosterone quantile
influence the early development of cubs and lead to
Figure 6ท7. Total pesticide and PCB concentrations (mean and 95%
higher mortality. However, other factors such as popula-
C.I.) by testosterone quantile (Oskam et al., 2001).
tion density may play a role, and a causal link with PCBs
cannot be established.
In 1999, 52 experienced polar bear hunters living in
POP concentrations were determined in adipose tis-
East Greenland were interviewed about their observa-
sue, plasma, and milk samples from seven female polar
tions of aberrant bears including pathological changes/
bears and their cubs near Cape Churchill, Hudson Bay,
anomalies in internal organs (Dietz et al., 2001; Sandell
between 1992-1996 (Norstrom, 1999b; Polischuk, 1999).
et al., 2001). Information on approximately 1110 bears
Pregnant females were captured from August 7 to Oc-
that had been shot between 1945 and 1999 was ob-
tober 7, and the same females with cubs were captured
tained. Thirteen anomalous polar bears were reported.
from March 2 to March 17 of the following year, after
The anomalies included supernumerary nipples or
emerging from dens but before they had moved onto
claws, unilateral collapse of lung, abnormal and missing
the ice to begin hunting seals. Mothers that were recap-
claws, partial melanism, missing limbs, and a malformed
tured a third time in the following autumn without
newborn.
cubs were found to have had high OC concentrations in
A study of contaminant-induced changes in polar
their milk when they emerged from the den in the previ-
bear skulls from East Greenland was initiated in 1999.
ous spring (Figure 4ท55). By comparison, mothers re-
The results are still very preliminary as both the POP
captured in autumn and still accompanied by cubs, had
analyses and the age determination of the bears are still
low OC concentrations in their milk the previous
in progress. Approximately 180 skulls (1892-1987)
spring. The differences in concentrations were signifi-
from the National Zoological Museum in Copenhagen
cant (p < 0.05) for all residue classes. For example, PCB
were compared with 100 recent (1999-2001) skulls to
concentrations were approximately three times higher
detect possible macroscopical pathological changes,
(5780 ng/g lw) in females that lost their cubs than in fe-
asymmetrical changes as well as osteopenia (osteoporo-
males that kept their cubs (1830 ng/g lw). It is not
sis and osteomalacia). Both directional and fluctuating
known how much significance can be attached to this
asymmetry were found in several bones of the skulls.
finding in terms of reproductive performance, but it is
Surprisingly, the amount of fluctuating asymmetry
suggestive, at least, that cub survival may be dependent
seemed to have decreased from the period of 1892-1960
on degree of exposure to OCs in milk (Norstrom,
to the period 1960-2001. The results are only prelimi-
1999b).
nary, however, and awaiting the inclusion of age data in
The age structure of polar bears on Svalbard was
the analysis.
compared to that of several other populations from less
The frequency of paradontitis (loose teeth) among
contaminated areas (Derocher et al., 2003) as a low fre-
subadult polar bears was higher, but not statistically sig-
quency of older female polar bears on Svalbard has been
nificant, in the period of 1960-2001 than in the period

---

Chapter 6 ท Biological Effects
187
1892-1960. A clear sex difference was observed in the
Log10 retinol, ตmol / L plasma
bone mineral density measured as calcium-phosphate
0.4
content. There was also a tendency for the males from
Polar bear, Svalbard
the period of 1960-2001 to have lower bone mineral
density as compared to males from the period of 1900-
0.2
1960.
Several female polar bears from Svalbard have been
found to be pseudohermaphrodites, and high PCB levels
have been hypothesized as one possible causative factor
0.0
(Wiig et al., 1998). The link to POP exposure; however,
is not strong because of the occurrence of the same syn-
drome in black bears (Norstrom, 2002). One pseudoher-
ญ 0.2
maphroditic female was also found among more than
100 sampled polar bears in 1999 on East Greenland
(Sonne-Hansen et al., 2002) and the occurrence of this
ญ 0.4
one case is lower than the estimated frequency found at
Svalbard of approximately 3% (Derocher, 2002).
1.2
1.4
1.6
1.8
2.0
2.2
2.4
2.6
Cytochrome P450 activities
Log10 PCBs, ng/g plasma ww
CYP1A1 has been determined in white blood cells of
Figure 6ท8. The association between retinol (corrected for age and
polar bears and the results (n = 13) show a significant
sex) and PCBs in plasma of 79 polar bears at Svalbard by regres-
positive correlation between PCBs (the sum of PCBs
6.3.6.2
sion analysis. The regression line is shown (log residual retinol =
0.40 ญ 0.21 log PCBs). The Pearson correlation coefficient is r =
99, 118, 153, 156 , 180, and 194) and CYP1A1 area in
ญ 0.33 (p = 0.003) (Skaare et al., 2001a).
western blots (p = 0.026). The strongest correlation was
found with CB156 (p = 0.011). These results are promis-
Log10 Total T4 : Free T4 ratio
ing with regard to the potential for CYP1A1 in blood
0.4
cells as a biomarker for PCB exposure in polar bears
Polar bear, Svalbard
(Skaare et al., 2000).
0.2
Thyroid and retinol effects
Normal regulation of vitamin A and thyroid hormones
is important for a wide range of biological functions,
such as growth, cell differentiation, reproduction, be-
0.0
havior, and the immune system. In Svalbard polar bears,
retinol and thyroid hormones (T3 and T4), and concen-
trations of several OC contaminants have been deter-
ญ 0.2
mined in blood plasma of 71 individuals collected from
1991 to 1994. The determination of multivariate associ-
ations between retinol, thyroid hormones, and the ratio
ญ 0.4
of total and free thyroid hormones, respectively, and the
concentrations of various OC components (PCBs, DDE,
HCB and HCHs) revealed significant OC associations
1.2
1.4
1.6
1.8
2.0
2.2
2.4
2.6
Log
for retinol and the ratio of total T4 : free T4, after cor-
10 PCBs, ng / g plasma ww
recting for age and sex (Skaare et al., 2001a). Significant
Figure 6ท9. The association between the ratio of total T4 and free
negative correlations were found between retinol and
T4 (TT4 : FT4) (corrected for age and sex) and PCBs in plasma of
PCBs (Figure 6ท8), as well as retinol and HCB and
78 polar bears at Svalbard by regression analysis. The regression
line is shown (log residual total T4 : free T4 = 0.21 ญ 0.12 log PCBs).
HCHs. Significant negative correlations were also found
The Pearson correlation coefficient is r = ญ 0.28 (p = 0.013) (Skaare
between total T4 : free T4 and PCBs (Figure 6ท9) as
et al., 2001a).
well as HCB.
Concentrations of the thyroid hormones, T3 and T4,
in females from the Svalbard population, and FT3 index
and retinol were determined in plasma of polar bears
was lowest in Resolute females.
from Resolute Bay in Canada and Svalbard, which have
Correlations among thyroid hormone concentra-
among the lowest and highest POP concentrations, re-
tions, FT3 and FT4 indices, retinol concentrations and
spectively, in polar bears (Norstrom, 2000). Free T3
the complete suite of POP concentrations in adult polar
(FT3) and free T4 (FT4) indices were also determined.
bear plasma (n = 60) from Resolute Bay, Canada, and
Resolute bears had significantly higher total T4 and FT4
Svalbard were examined by principal component analy-
index, and lower total T3 and FT3 index than Svalbard
sis. Because hydroxy compounds may affect circulating
bears (p < 0.001). Retinol concentrations were not sig-
levels of free and bound T4 due to competitive binding
nificantly different between regions. None of the biolog-
of hydroxy metabolites and T4 to TTR, all chemical
ical measures were significantly related to age, even
concentrations were converted from mass to molar con-
when separated into region and sex categories. This is in
centrations prior to statistical analysis.
contrast to the above study, which found that total T4,
Using the entire data set (n = 60), retinol concentra-
FT4, total T3, and FT3 are associated with age in male
tions were negatively correlated (r = ญ 0.465, p < 0.001)
polar bears (Skaare et al., 2001a). Total T3 was higher
with persistent PCBs (PC1) and positively correlated

---

188
AMAP Assessment 2002: Persistent Organic Pollutants in the Arctic
Retinol concentration in polar bear plasma, ตg/L ww
Retinol concentration in polar bear plasma, ตg/L ww
600
600
Resolute males
Resolute bears only
Resolute females
500
Svalbard males
500
Svalbard females
400
400
All bears
300
300
All bears
200
200
100
100
Resolute bears only
0
0
ญ 3
ญ 2
ญ1
0
1
2
3
ญ 3
ญ 2
ญ1
0
1
2
3
PC1 (Persistent PCBs)
PC2 (OH-PCBs)
Figure 6ท10. Correlation between retinol concentrations (ตg/L) and the first two principal components from analysis of POP concentrations in
polar bear plasma from Resolute Bay in the Canadian Arctic in April-May 1997, and from Svalbard in April-May 1998.
(r = 0.309, p = 0.02) with OH-PCBs (PC2) (Figure 6ท10).
may influence the levels. Three individual PCB con-
If only Resolute bears (n = 25) were included in the
geners showed significant inverse correlations to IgG:
analysis, retinol was more highly correlated with PC1
CBs 99, 194 and 206 (Figure 6ท11). In addition, HCB
(r = ญ 0.744, p < 0.001) and PC2 (r = 0.692, p < 0.001).
was also inversely correlated to IgG. OCs were found to
These results suggest that plasma retinol concentrations
account for 11% of the variation of IgG levels.
are more likely to be affected by the influence of persist-
In a study of mothers and cubs on Svalbard sampled
ent PCBs on retinol metabolism and storage in liver than
in 1995-1998, a negative correlation was found between
by the interference of OH-PCBs with the transport of
IgG levels in cub plasma and plasma PCB concentra-
retinol via RBP:TTR dimer formation.
tions (Lie et al., 2002). The results of these studies demon-
Total T4 plasma concentrations were negatively as-
strate a possible contaminant-associated suppression of
sociated with both persistent PCBs (PC1), r = ญ 0.337, p =
antibody-mediated immunity in polar bears at Svalbard.
0.01, n = 56) and non-persistent PCBs (PC3) (r = ญ 0.293,
The effects of high PCB exposure on the immune sys-
p = 0.03, n = 56), but not with any other contaminant
tem have been further studied by comparing the immune
group, including OH-PCBs. Since both PC1 and PC3 are
system functions in polar bears with high (Svalbard) and
PCB-related, this correlation suggests a common mecha-
low (Canada) PCB exposure (Skaare et al., 2001b; 2002;
nism of action of all PCB congeners in reduction of
Larsen et al., 2002c). This field study comprised a vacci-
plasma T4 concentrations. Total T4 concentrations were
nation model with recapture after immunization. Thirty
also negatively correlated with concentrations of PCBs
bears at Svalbard and thirty in Canada were immunized
(r = ญ 0.29, p = 0.04, n = 56). Thirty-three female polar
with different herpes-, reo-, and influenza viruses and
bears without cubs and with two-year old cubs were
tetanus toxoid to stimulate the production of protective
sampled at Svalbard and in the Barents Sea region dur-
antibodies such as virus-neutralizing antibodies, virus
ing the period 1995-1998, and blood samples were ana-
hemagglutination inhibition antibodies and toxin-neu-
lyzed for a range of POPs, thyroid hormones, and prog-
tralizing antibodies. The immunization also included
esterone (Braathen et al., 2000). Significant differences
keyhole limpet hemocyanin that, together with tetanus
were seen in progesterone levels between the females
toxoid, would stimulate cell-mediated immune response.
from the two areas, with higher progesterone levels in
Blood was sampled at immunization, and four to six
the Barents Sea bears. However, no relationship could be
IgG concentration in polar bear plasma, mg/mL
shown between progesterone levels and the PCB con-
40
centrations. For thyroid hormones, significant differ-
males
ences were found for the ratio of total T3 : free T3 with
females
higher ratios in the Svalbard bears, but no relationship
could be seen between these and PCB levels.
30
Immune effects
Production of antibodies plays an important role in pro-
20
tection against infections. Antibodies are divided into
different immunoglobulin classes, where IgG is the ma-
jor one in blood. IgG and OC concentrations were de-
10
termined in blood sampled from 56 free-living polar
bears of different ages and both sexes between 1991 and
1994 (Bernhoft et al., 2000). Total IgG concentration in-
creased with age and was significantly higher in males
0 0
2
4
6
8
10
12
14
than in females. A significant decrease of IgG with in-
CB99 concentration in polar bear plasma, ng/g ww
creased PCB level was found (r = ญ0.29, p = 0.03). IgG
Figure 6ท11. Linear regression of IgG level and CB99 in blood plas-
was standardized for sex and age, since both sex and age
ma of polar bears of both sexes at Svalbard (Bernhoft et al., 2000).

---

Chapter 6 ท Biological Effects
189
weeks later, for detection of antibodies. In addition, in
three sites in Russia (Franz Josef Land, Kara Sea, Siber-
vitro lymphocyte stimulation was performed with mito-
ian Sea) overlap the NOEL for kit survival in mink, and
gens and specific antigens (phytohemaglutinin, conca-
several of these also exceed the LOEL for vitamin A re-
navalin A, pokeweed mitogen, Mycobacterium antigen,
duction in otter (Figure 6ท3). The threshold associated
and lipopolysaccharide) (Larsen et al., 2002a; 2002b).
with immunosuppression and vitamin A disruption in
The resistance factor was measured directly as neutral-
harbour seal blubber is 16 500 ng PCB/g lw, and the
ization of virus infections in cell cultures, inhibition of
LOAEL for immunosuppression is 21 000 ng PCBs/g
virus hemaglutination, and toxin neutralization. There-
lw (blood) in rhesus monkeys. PCB levels in polar bears
fore, the effect of PCBs on infection resistance was meas-
from three areas are at or above these two thresholds:
ured without dependence on infection experiments or by
Franz Josef Land, Kara Sea, and Svalbard (cubs and
conducting registration of disease outbreaks.
yearlings). Since Svalbard polar bears with high PCB
Preliminary results from this study demonstrated sig-
levels have been shown to exhibit signs of immunosup-
nificantly higher PCB plasma levels in polar bears at
pression, it is probable that such effects are also occur-
Svalbard as compared to Canada. In the Canadian
ring in Russian polar bears from these sites. PCB levels
bears, significantly higher antibody titers against in-
in some polar bears from Davis Strait, Canada are above
fluenza, reo- and herpes viruses were found following
the threshold for immunosuppression in harbour seals
immunization than at Svalbard. Furthermore, signifi-
and close to the LOAEL for immunosuppression in rhe-
cantly lower antibody titers against influenza virus fol-
sus monkeys. Temporal-trend studies for PCBs in
lowing immunization, were associated with higher PCB
Canadian and Svalbard polar bears indicate that there
levels. Significantly lower lymphocyte responses were
have been no declines during the 1990s. This implies
found to pokeweed mitogen, lipopolysaccharide from
that the assessment of contaminant levels compared to
E. coli and Mycobacterium antigen with high PCB expo-
thresholds for other polar bear populations, given in the
sure levels, and significant negative correlations between
previous AMAP assessment, is still valid.
PCBs and IgG were also found.
Levels of 2,3,7,8-TeCDD and 1,2,3,7,8-PeCDD in
From the present preliminary results on effects of
Canadian polar bears from four sites sampled in 1990
PCBs on the immune system of polar bears, it is reason-
range from 1.2 to 3.8 pg TEQ/g lw in adipose tissue
able to assume that PCBs are associated with decreased
(Annex Table 16). These are well below the threshold
resistance to infections. This is supported by the finding
for immunosuppression in harbour seals (Figure 6ท4).
that polar bears with high plasma PCB have higher inci-
This may, however, be an underestimation since mono-
dence of Pasteurella bacteria, one of the most common
ortho PCBs usually contribute most to the total TEQs
microbes in the environment (Larsen, 2002).
compared to nPCBs and PCDD/Fs (Letcher et al., 1996)
and these were not included in the study.
Levels and intake assessment
PFOS levels in polar bears range from 26 to 52 ng/g
PCB concentrations (22-40 congeners) in polar bear fat
ml in plasma in Beaufort Sea individuals and 180-680
collected between 1996 and 1997 at several sites around
ng/g ww in liver from bears from Barrow, Alaska (Kan-
Alaska (Beaufort Sea, Bering Sea, Chukchi Sea) ranged
nan et al., 2001a; Giesy and Kannan, 2001). These levels
between 910 and 11 000 ng/g lw with means in the range
are well below the NOAEL (15000 ng/g ww in liver) and
of 2600-9640 ng/g lw (Annex Table 14). Mean PCB
LOAEL (58000 ng/g ww in liver) for second-generation
concentrations in polar bear fat collected in 1999-2000
effects in rats.
from East Greenland ranged from 4800 to 7700 ng/g lw.
Based on mean POP levels in ringed seal blubber
In polar bear fat from one individual from the pack ice
from various sites in the Arctic (Annex Table 12), diel-
near Iceland and collected in 1993, PCB concentration
drin, chlordanes (heptachlor epoxide), and DDT levels
(7 congeners) was 7860 ng/g ww (approximately 10 000
are below the dietary NOAECs found for reproductive
ng/g lw) (Annex Table 14). PCB concentrations (1990)
effects in mink. Some harp seals from east of Svalbard
from a recent temporal-trend study of Canadian polar
have chlordane levels that exceed the LOAEC found for
bear from several sites ranged from 720 to 20 200 ng/g lw
kit growth in mink. Based on the NOAEC (4000 ng/g ww)
with geometric means in the range of 2600-7500 ng/g lw
for rats and dogs, toxaphene levels in ringed and harp
(Annex Table 14). Russian polar bears collected from
seals are below those associated with thyroid effects. As-
four areas (Franz Josef Land, Kara Sea, Siberian Sea,
suming that polar bears are as sensitive as mink, mean
Chukchi Sea) have plasma PCB concentrations (5 con-
levels of PCBs in ringed and harp seal blubber from all
geners) ranging from 1100 to 31 000 ng/g lw, and Sval-
sites exceed the dietary NOAEC for reproduction of 72
bard polar bears have plasma PCB concentrations rang-
ng/g ww. Mean DDT levels in ringed and harp seals ex-
ing from 2100-14 000 ng/g lw in the same study (Ander-
ceed Canadian and U.S. EPA guidelines for protecting
sen et al., 2001b). For Svalbard polar bears of different
aquatic wildlife, and in some cases (i.e. Svalbard, Rus-
ages collected in 1995-1998, plasma PCB levels range
sia), International Joint Commission objectives are also
from 1060 to 29 000 ng/g lw, with highest concentrations
exceeded (Table 6.1) (Figure 6ท6). For PCB levels,
found in cubs-of-the-year and yearlings (Annex Table 14).
ringed and harp seal blubber from all sites exceed all en-
Most PCB levels in these polar bears exceed the
vironmental guidelines for protecting aquatic wildlife
NOAELs and LOAELs found for subtle neurobehav-
(Figure 6ท5). Toxaphene levels in ringed and harp seal
ioral effects in offspring if polar bears are as sensitive as
blubber, where measured, exceed Canadian guidelines
offspring of rhesus monkeys and humans, and also, the
for protecting wildlife consumers. Data for dioxin-like
NOEL for vitamin A reduction in otter (Figure 6ท3). The
compounds are very limited but TEQs in ringed seal
PCB levels in polar bears from several sites in Alaska
blubber from Canada and Russia exceed Canadian
and Canada, on the Iceland pack ice, at Svalbard and
(mammalian) and U.S. EPA guidelines (Annex Table 16).

---

190
AMAP Assessment 2002: Persistent Organic Pollutants in the Arctic
ciated with thyroid effects. Mean dieldrin levels in these
6.3.7. Arctic fox
seabird species' eggs also do not exceed any environ-
In the previous AMAP assessment, mean PCB levels in
mental guidelines for protecting aquatic wildlife. How-
Svalbard Arctic fox from 1993 and 1994 were found to
ever, DDT levels in eggs and liver from many of these
exceed the NOAEL and LOAEL for subtle neurobehav-
species from many of these sites exceed Canadian and
ioral effects in offspring of rhesus monkeys and humans,
U.S. EPA guideline levels for protecting wildlife that
the NOEL for kit survival in mink, levels known to
consume aquatic biota (Figure 6ท6). Most PCB levels
cause poor reproductive success in harbour seals, and
exceed all guideline levels (Figure 6ท5) and where meas-
the EC50 for reduced litter size in mink. Some individu-
ured, toxaphene levels exceed Canadian guidelines.
als also exceeded the levels associated with poor repro-
PCDD/F and nPCB levels given as TEQs exceed Cana-
ductive success in ringed seals and the EC50 for kit sur-
dian (mammalian) and U.S. EPA guideline values (Table
vival in mink (80 000-120 000 ng/g lw). The mean PCB
6.1) in kittiwake, fulmar, murre, black guillemot, com-
levels in Arctic fox were also above the LOAEL for im-
mon guillemot, glaucous gull, herring gull, and great
munosuppression in rhesus monkeys. No effects studies
black-backed gull eggs and/or liver (Annex Table 16).
have been carried out since the previous assessment.
Thus, Arctic fox that prey on seabird eggs, chicks, and
adults may have dietary intakes of TEQs, DDTs, PCBs,
Levels and intake assessment
and toxaphene high enough to lead to effects. The di-
Mean concentrations of PCBs in Arctic fox collected in
etary assessment for Arctic fox that feed on seals is the
1999-2000 from Holman Island, NWT, were 860 ng/g
same as for polar bears (Section 6.3.6).
lw in muscle (range: 76-8050 ng/g lw) and 1350 ng/g lw
in liver (range: 110-14 600 ng/g lw). Mean PCB con-
6.3.8. Sea otter
centrations in Arctic fox from Barrow, Alaska and from
inland Iceland were 1000-1600 ng/g lw in liver. Mean
No biological effects studies have been carried out in sea
concentrations of PCBs in Arctic fox fat collected in
otter.
1996-1997 in the Pribilof Islands, Alaska, were somewhat
higher, between 2100 ng/g lw and 5000 ng/g lw (range:
Levels and intake assessment
240-12 000 ng/g lw). Highest mean PCB concentrations
Mean concentrations of PCBs in sea otter from south-
were seen in Arctic fox from the coast of Iceland at 72 500
east Alaska were 270 ng/g lw and for sea otter from the
ng/g lw (Annex Table 14). The PCB levels for Arctic
Aleutian Islands, 10 300 ng/g lw (Annex Table 14). The
fox from all sites exceed the NOAEL and LOAEL for
PCB levels in sea otter from southeast Alaska do not
subtle neurobehavioral effects in offspring of rhesus mon-
exceed any threshold levels. The PCB levels in sea otter
keys and humans. The PCB levels in Canadian, Pribilof
from the Aleutian Islands exceed the NOAEL and
Islands, and coastal Iceland foxes also exceed the
LOAEL for subtle neurobehavioral effects in offspring
NOELs for vitamin A reduction in otter, effects on otter
of rhesus monkeys and humans, the NOELs for vitamin
reproduction and mink kit survival, the LOAELs for vi-
A reduction in otter, and effects on otter reproduction
tamin A reduction in otter, and decreased kit production
and mink kit survival (Figure 6ท3).
and kit body weight gain in mink (Figure 6ท3). The
PCB levels in the coastal Iceland foxes also exceed the
6.4. Summary and conclusions
threshold for immunosuppression and vitamin A disrup-
ญ biological effects
tion in harbour seals, the LOAEL for immune effects in
rhesus monkeys, the threshold for poor reproduction in
It is very difficult to link contaminant levels or biochem-
harbour seals, and the EC50 for litter size in mink. Levels
ical indicators of effects to effects on Arctic animals at
of mono-ortho PCBs given as TEQs in Pribilof Islands
the individual or population level. Such assessments are
Arctic foxes ranged between 9 and 290 pg/g lw (means:
also complicated by the fact that the thresholds for ef-
48 and 151 ng/g lw for females and males, respectively)
fects of many contaminants are not well known, and
and these exceed the mono-ortho TEQs associated with
very little is known about effects of contaminant mix-
immunosuppressive effects in harbour seals (Figure 6ท4)
tures. It is also possible that there may be other causes of
Besides feeding on terrestrial mammals and birds,
the effects seen that are unrelated to POP exposure.
Arctic foxes may also eat marine birds and eggs, seal
As far as OCs are concerned, Arctic marine mam-
pups, and placentas, as well as scavenge on seals, mostly
mals are often regarded as controls for much more con-
from polar bear kills, depending on what is available at
taminated members of their populations or related spe-
the particular site they are located. Based on contami-
cies in temperate regions. Biological effects studies on Arc-
nant levels found in eggs and liver from thick-billed
tic animals do, however, show some subtle responses,
murres, black guillemots, black-legged kittiwakes, glau-
and there are now stronger links between these effects
cous gulls, herring gulls, great black-backed gulls, and
and current levels of some POP contaminants. Based on
fulmars from the Alaskan, Canadian, Greenlandic, Nor-
the results of the biological effects studies that have been
wegian, and western Russian Arctic (Annex Table 11),
carried out, the following conclusions can be drawn.
dieldrin, DDT, and heptachlor levels are below the di-
etary NOECs and LOECs for reproductive effects in
6.4.1. Observed effects
mink. PCB levels are above the dietary NOEC for re-
6.4.1.1. Field studies
productive effects and TEQs, based on PCDD/Fs,
nPCBs, and/or mono-ortho PCBs are above the NOECs
Reproduction
and LOAECs for effects on reproduction and vitamin A
ท Alaskan peregrine falcons (tundrius and anatum sub-
in mink. Dietary toxaphene levels are below those asso-
species) still suffer from eggshell thinning of 10.6%

---

Chapter 6 ท Biological Effects
191
and 12%, respectively, when compared to eggs from
and EROD activity up to a threshold of 100 ng/g ww
the pre-DDT era.
in liver. After this, EROD activity leveled off.
ท Bald eagles showed reduced productivity on Adak Is-
ท High correlations have been found between testos-
land in the Aleutians, which was associated with the
terone 6- hydroxylation (CYP3A) and toxaphene lev-
higher DDE concentrations found there than at three
els in harp seals around Svalbard. Positive correlations
other Aleutian islands. Eggshell thickness was signifi-
have been found between EROD levels and PCBs in
cantly negatively correlated with PCB concentrations
Svalbard ringed seals.
on the four islands.
ท A significant correlation was found between CYP1A1
ท Positive correlations were found between TBT concen-
in polar bear white blood cells and PCB levels.
trations and degree of imposex in dogwhelks from the
Norwegian coast, and imposex is also found along the
Thyroid and retinol effects
Faroe Islands and Greenlandic coasts. Imposex has also
ท A borderline significant correlation was found be-
been seen in dogwhelk from Icelandic harbors but has
tween PCB concentrations in yolk and plasma retinol
decreased after restrictions on the use of TBT as an anti-
levels in one-day-old shag from Norway.
fouling paint on boat hulls were implemented in 1990.
ท A significant negative correlation was found between
ท In glaucous gulls from Bj๘rn๘ya, Norway, increased
liver PCB concentrations and liver retinol and retinyl
absence from nests was correlated with high OC levels.
palmitate levels in black guillemots at Saglek Bay.
Female gulls with high OC levels were more likely to
ท A significant negative correlation was seen between
have non-viable eggs and chick body condition was
HCB, p,p'-DDE, and PCB blood concentrations and
poorer. Adult survival was significantly negatively cor-
plasma T4 levels in male glaucous gulls from Bj๘rn๘ya.
related to OC levels. Glaucous gulls with high OC lev-
ท Retinol levels and total T4 levels in northern fur seal
els have wing feather asymmetry, an indication of de-
pups were found to be negatively correlated to PCB
velopmental stress.
congeners and to TEQs.
ท Egg volume, yolk mass, and hatchling mass in one-
ท In Svalbard polar bears, significant negative correla-
day-old shag were negatively correlated to PCB con-
tions were found between retinol and PCB levels, as
centrations.
well as retinol and HCB and HCHs. Significant nega-
ท A significant negative correlation was found between
tive correlations were also found between the ratio of
testosterone levels and PCB levels, as well as total pes-
total T4 : free T4 and PCB as well as HCB. In a com-
ticide levels in polar bear plasma, indicating that these
parison of polar bears from Svalbard (high PCB levels)
POPs may decrease circulating testosterone levels. Fe-
and Resolute, Canada (low PCB levels), a significant
males from southwest Hudson Bay that had lost their
negative correlation was found between total T4 levels
cubs between emerging from the den and the following
and PCBs but not with OH-PCBs. Retinol levels were
summer/autumn had significantly higher OC concen-
negatively correlated with persistent PCBs and posi-
trations when emerging from the den than females that
tively correlated with OH-PCBs, suggesting that PCB
kept their cubs.
affects retinol metabolism and storage in the liver.
Other observations have been made where POPs are sus-
Immune effects
pected of playing a role, but where there is as yet no cer-
ท In glaucous gulls from Bj๘rn๘ya, nematode density
tain link between the effects seen and POP concentra-
was positively correlated with concentrations of p,p'-
tions. These are the following.
DDT, mirex, and PCBs.
ท Black-legged kittiwakes with crossed bills and clump
ท In northern fur seal and Steller sea lion pups, various
feet have been observed during an expedition in the
measures of normal immune function were negatively
Barents Sea.
correlated to PCB levels, indicating that high PCB ex-
ท Steller sea lion populations have been declining in
posure may be causing immunodysfunction.
western Alaska including the Aleutian Islands and the
ท In Svalbard polar bears, a significant decrease in anti-
western stock is considered endangered. These popula-
bodies (IgG) with increased PCB levels was found. In
tions excrete higher concentrations of POPs than less-
mothers and cubs, a similar negative correlation was
affected populations with lower POP concentrations.
found for IgG levels in cubs and plasma PCB levels as
ท Reproductive rates in Svalbard polar bears are similar
well as with a number of specific congeners (CBs 99,
to those of other polar bear populations, but cub sur-
137, 153, 157, 170, 180, and 194). In a vaccination study
vival is lower. Female pseudohermaphroditic polar
using two polar bear populations, one with higher PCB
bears have been found on Svalbard and Greenland.
levels (Svalbard) and one with lower PCB levels (Can-
The proportion of females with cubs-of-the-year that
ada), polar bears with high PCB levels were found to
were older than 16 years was significantly lower on
exhibit immunosuppression expressed as reduced IgG
Svalbard than in western Hudson Bay.
production and lowered lymphocyte responses. This
may indicate decreased resistance to infections.
Cytochrome P450 activity
ท CYP enzyme activities are correlated to PCB levels in
6.4.1.2. Laboratory studies using Arctic species
Arctic char from two freshwater lakes (one with high
POP levels, one with low POP levels) on Bj๘rn๘ya.
Cytochrome P450 activity
ท A weak association between PCB levels and EROD ac-
ท Increased EROD activity was found in Arctic char fed
tivity was seen in glaucous gulls at Bj๘rn๘ya.
a single oral dose of Aroclor 1260 and then starved,
ท A positive correlation was found in black guillemot
compared to PCB-exposed and fed char, non-exposed
from Saglek Bay, Canada, between PCB concentrations
and fed char, and non-exposed and starved char.

---

192
AMAP Assessment 2002: Persistent Organic Pollutants in the Arctic
Immune effects
PCB levels in Alaskan and Canadian burbot liver
ท Increased cortisol levels were found in Arctic char fed
from some sites and in some Arctic char from Bj๘rn๘ya
a single oral dose of Aroclor 1260 and then given food,
are close to or exceed the LOEL for induction of liver
compared to PCB-exposed and starved char, non-ex-
enzymes found in Arctic char. Toxaphene levels in some
posed and fed char, and non-exposed and starved char.
Canadian burbot exceed levels associated with effects on
In another study, basal cortisol levels were suppressed
bone development found in channel catfish. In marine
by PCBs in starved fish but were elevated in fed fish
and anadromous fish, PCB levels in Greenland shark
after handling, indicating that stress responses are com-
from Cumberland Sound and Greenland halibut from
promised by PCBs, and the effect of fasting makes char
West Greenland exceed the LOEL for induction of liver
sensitive to the effects of PCBs. Fasting and PCB expo-
enzymes found in Arctic char.
sure were also studied in relation to disease susceptibil-
Piscivorous seabirds such as grey herons, alcids, and
ity. Disease susceptibility was highest in the fed char
kittiwakes have lower PCB levels than predatory sea-
with no difference due to PCB exposure, while disease
birds, and only grey herons had PCB levels that ex-
susceptibility increased with PCB exposure in the
ceeded some reproductive thresholds for effects. Black
starved group. The results indicate that PCBs reduce
guillemots from the most contaminated site at Saglek
immunocompetence in starved Arctic char, but that
Bay, Canada and predatory seabirds, such as glaucous
starved fish are more disease resistant than fed fish.
and great black-backed gulls and great skuas, have
ท Glaucous gull chicks with high dietary PCB exposure
PCB levels in liver that indicate that egg levels might
show an impaired ability to produce antibodies when
exceed some threshold levels for reproductive effects.
challenged with an antigen.
For grey heron, mono-ortho PCB levels, expressed as
ท Two juvenile harp seals treated with increasing doses
TEQs, exceeded some reproductive thresholds in other
of selected PCB congeners for 40 days and then fasted
bird species. TEQs based on PCDD/F and nPCB levels
for 30 days had increased serum cortisol and aldos-
for Canadian thick-billed murres, black-legged kitti-
terone levels as well as tumor necrosis factor alpha
wakes, and northern fulmars, and TEQs based on
compared to the controls.
PCDD/Fs, nPCBs and mono-ortho PCBs for herring gull
and great black-backed gull eggs from northern Norway
Mutagenic effects
exceed the LOAEL for reproductive effects in wood
ท Higher frequencies of chromosome aberrations and
duck, a sensitive species. Glaucous gulls from Svalbard
DNA adducts were found in glaucous gull chicks fed a
have nPCB and mono-ortho PCB levels expressed as
POPs contaminated diet.
TEQs that exceed all thresholds for reproductive effects
in other bird species and the threshold for induction of
liver enzymes found in common terns.
6.4.2. Assessment of current levels in biota
Ringed seals from Alaska, Canada, and Greenland,
Current concentrations of some POPs in several Arctic
harbour seals from Alaska, Steller sea lions from Alas-
species are at or above the known thresholds associated
kan eastern stocks, harbour porpoises and walrus from
with effects that have been seen in other species studied
Greenland, and grey whales from the Bering Sea, have
either in the laboratory or in the field.
PCB levels that are low but that still exceed levels asso-
Canadian wolverines have PCB levels that exceed
ciated with subtle neurobehavioral effects in offspring of
those associated with subtle neurobehavioral effects in
rhesus monkeys and humans. Higher PCB levels are
offspring of rhesus monkeys and humans. Mountain
found in ringed seal from the Barents Sea/Kara Sea area
hare and reindeer from the Kola Peninsula (Lovozero)
and Saglek Bay, Labrador, harbour, grey, bearded and
have PCDD/F concentrations, expressed as TEQs, which
northern fur seals and some Steller sea lions. These
exceed levels associated with immunosuppression in
PCB levels also exceed the NOEL associated with de-
harbour seals. TEQs in reindeer from Pechora and Tay-
creased vitamin A in otter. Svalbard, Kara Sea, and
mir Dudinka also exceed this threshold.
Saglek Bay ringed seals and northern fur seals also ex-
Eggshell thinning in Alaskan peregrine falcon eggs has
ceed the threshold for decreased reproduction in otter,
improved, but eggshells are still thinner than pre-DDT era
and sea otter from the Aleutian Islands. Some ringed seal
eggs, indicating that present DDE levels are still causing
(Kara Sea) and some beluga exceed the NOEL for mink
effects. The p,p'-DDE concentrations in some Alaskan
kit survival. Harbour porpoises from Norway, long-
peregrine falcon eggs are just below the critical threshold
finned pilot whales, narwhal, killer whales (residents
for reproductive failure. PCB levels in Alaskan pere-
and transients), harp seals, and some minke whales have
grines exceed some NOELs and LOELs for reproductive
even higher PCB levels that are associated with de-
endpoints in a wide range of wild bird species. PCB con-
creased vitamin A in otter, and decreased otter and mink
centrations in Norwegian peregrine falcons and white-
reproduction. Harp seals and some minke whales exceed
tailed sea eagle exceed most or all thresholds for repro-
PCB levels associated with immunosuppression and vi-
ductive effects in other bird species. Norwegian gyrfalcon
tamin A disruption in harbour seals but are below the
and golden eagles have lower PCB concentrations, but
threshold for immunosuppression in rhesus monkeys. In
these still exceed some thresholds for reproductive ef-
addition, some harbour porpoises from northern Nor-
fects. Bald eagles from one of four studied Aleutian Is-
way, some resident killer whales and all transient killer
lands have p,p'-DDE concentrations in the range known
whales from Alaska, and some long-finned pilot whales
to cause reproductive impairment. Maximum PCB con-
from the Faroe Islands have PCB levels that exceed
centrations in some Aleutian bald eagles exceed most of
thresholds for immunosuppression in rhesus monkeys
the thresholds for reproductive effects in other bird spe-
and poor reproductive success in harbour seal. Some res-
cies including the LOAEL for egg mortality in bald eagles.
ident and many transient killer whales exceed the EC50

---

Chapter 6 ท Biological Effects
193
for litter size in mink, and many transient killer whales
rine fish. Killer whales, walrus, polar bear, and Arctic
also exceed the threshold for poor reproductive success
fox that prey on seals have dietary intakes of PCBs that
in ringed seal and the EC50 for kit survival in mink.
may cause reproductive effects. When environmental
Levels of mono-ortho PCBs, given as TEQs, in Arctic
quality guidelines for protecting aquatic wildlife are used,
fox from the Pribilof Islands, and in resident and tran-
dietary intakes of DDTs, PCBs, and dioxin-like sub-
sient killer whales, exceed the mono-ortho TEQs associ-
stances are problematic for many marine and freshwater
ated with immunosuppressive effects in harbour seals.
piscivorous species and, in some cases, for molluscivores.
Current PCB levels in polar bear from Alaska,
The same is true for predatory seabirds feeding on sea-
Canada, Greenland, the Iceland ice pack, Svalbard, and
bird eggs and chicks, and for killer whale, walrus, polar
Russia indicate that these populations have levels that
bear, and Arctic fox that consume marine mammals.
exceed those associated with subtle neurobehavioral ef-
fects in offspring of rhesus monkeys and humans, de-
6.4.3. Conclusions
creased vitamin A in otter, and NOELs for otter and
mink reproduction. PCB levels are highest in the Sval-
Assessment of contaminant levels in the previous AMAP
bard and Russian populations, and these levels exceed
assessment suggested that several species were at risk for
those associated with immunosuppression in harbour
neurobehavioral, reproductive, and immune system ef-
seals and rhesus monkeys and poor reproductive success
fects, particularly polar bears, predatory birds, seabirds,
in harbour seals. The results from biological marker
and seals. This has been borne out in biological effects
studies, particularly immune system effects in Svalbard
studies done on polar bears, northern fur seals, glaucous
polar bears and observations of decreased cub survival,
gulls and possibly also Steller sea lions. The implica-
support this assessment.
tions of these findings are that there are other popula-
Arctic fox from Barrow, Alaska, and inland Iceland
tions of these species, and other highly contaminated
have PCB levels associated with subtle neurobehavioral
species, that are being affected by current levels of some
effects in offspring of rhesus monkeys and humans. Arc-
POPs.
tic fox from the Pribilof Islands, Alaska, Canada, and
Effects that are biologically significant are those that
coastal Iceland have PCB levels associated with subtle
affect resistance to infection, reproduction, and behav-
neurobehavioral effects in offspring of rhesus monkeys
ior. Anything that affects these negatively reduces the
and humans, decreased vitamin A in otter and reproduc-
margin of safety for the affected species, putting them at
tive effects in otter and mink. PCB levels in Arctic fox
higher risk. Biomarkers for POP effects measure changes
from coastal Iceland also exceed those associated with
at the cellular or individual level, and are warning sig-
immunosuppression in harbour seals and rhesus mon-
nals. The results from biomarker studies in the Arctic
keys, poor reproductive success in harbour seals and the
have shown that there are associations between several
EC50 for litter size in mink. Levels of mono-ortho PCBs,
biomarkers and concentrations of some POPs. Results
given as TEQs, in Pribilof Islands Arctic foxes exceed
from field experiments and laboratory studies give
the mono-ortho TEQs associated with immunosuppres-
added weight to the possible link between some POPs
sive effects in harbour seals.
and specific effects. Therefore, based on the present evi-
Consequently, based on known thresholds for ef-
dence, it is believed that effects of biological significance
fects, several Arctic species appear to be at risk for, pri-
are occurring in some Arctic species related to POPs ex-
marily, reproductive and/or immunosuppressive effects
posure. These effects of biological significance are:
from current levels of DDTs, PCBs, and/or dioxin-like
ท polar bears are at higher risk for infections due to im-
substances. Those at greatest risk include peregrine fal-
mune effects of POPs;
cons, bald eagles, white-tailed sea eagles, glaucous and
ท glaucous gulls with high POP levels are at higher risk
great black-backed gulls, great skuas, some alcids, har-
of immune, behavioral and reproductive effects, and
bour porpoises, seals, northern fur seals, Steller sea
effects on adult survival;
lions, belugas, long-finned pilot whales, narwhal, minke
ท northern fur seals are at higher risk for infections due
whales, killer whales, sea otters, polar bears, and Arctic
to immune dysfunction correlated to POP exposure;
fox. If the risk for subtle neurobehavioral effects from

ท peregrine falcons continue to exhibit egg shell thinning
PCBs in exposed offspring of mammals is included,
and reproductive effects of POPs;
then some reindeer and mountain hares, wolverines,
ท Arctic char exhibit immune effects of PCB; and
walrus, and grey whales are also potentially at risk. Bur-
ท dogwhelks exhibit the reproductive effects of TBT.
bot from some sites have toxaphene levels high enough
to affect bone development in fry. Burbot, land-locked
Knowledge gaps, such as understanding the influence
char (Bj๘rn๘ya), Greenland shark, and Greenland hali-
of confounding factors, indicate that other biomarkers
but have PCB levels associated with increased liver en-
studied in Arctic biota (thyroid hormones, vitamin A
zyme production in Arctic char. Some invertebrates are
and cytochrome P450 activity) should be considered in-
at risk for the reproductive effects of TBT, particularly
dicators of increased exposure. It is not yet possible to
those that are exposed in harbors.
conclude that any documented changes in these bio-
An assessment of risks from dietary intake has been
markers imply increased risk. Studies on the effects of
attempted based on results from laboratory feeding ex-
other stressors, such as long periods of fasting, on PCB
periments, POP levels in prey items, and information on
effects in Arctic char show that PCB exposure increases
different species' food preferences. Reproduction in pis-
susceptibility.
civorous marine mammals, such as minke whale, beluga,
The lack of experimental dosage/response data con-
narwhal, long-finned pilot whale, harbour porpoise, and
tinues to limit the ability to interpret residues in Arctic
seals, may be affected by dietary levels of PCBs in ma-
animals.

---