Reproductive toxicity in mink (Mustela vison) chronically exposed to environmentally relevant polychlorinated biphenyl concentrations

Article abstract of DOI:10.1002/etc.5620201026

Female mink were exposed to a technical polychlorinated biphenyl (PCB) preparation (Clophen A50 [A50]; 0.1 or 0.3 mg/animal/d), one fraction of A50 containing the non- and mono-ortho-chlorinated congeners (0-1-ortho-chlorobiphenyls [CBs]), another fractio

Full text of DOI:10.1002/etc.5620201026

Environmental Toxicology and Chemistry, Vol. 20, No. 10, pp. 2318–2327, 2001
2001 SETAC
Printed in the USA
0730-7268/01 $9.00 .00
᭧
ϩ
REPRODUCTIVE TOXICITY IN MINK (MUSTELA VISON) CHRONICALLY
EXPOSED TO ENVIRONMENTALLY RELEVANT POLYCHLORINATED
BIPHENYL CONCENTRATIONS
¨
JORN
¨
B
B
RUNSTROM,*† BERT-OVE
L
UND,† ANDERS
B
ERGMAN,‡ LILLEMOR
A
SPLUND,§ IOANNIS
A
THANASSIADIS,§
¨
¨
M
ARIA
A
THANASIADOU,§ SOREN
J
ENSEN,§ and JAN
O
RBERG
†
†Department of Environmental Toxicology, Centre for Reproductive Biology in Uppsala, Uppsala University, Norbyva¨gen 18A,
SE-752 36 Uppsala, Sweden
‡Swedish Museum of Natural History, P.O. Box 50007, SE-104 05 Stockholm, Sweden
§Department of Environmental Chemistry, Wallenberg Laboratory, Stockholm University, SE-106 91 Stockholm, Sweden
(
Received 13 September 2000; Accepted 6 March 2001)
Abstract—Female mink were exposed to a technical polychlorinated biphenyl (PCB) preparation (Clophen A50 [A50]; 0.1 or 0.3
mg/animal/d), one fraction of A50 containing the non- and mono-ortho-chlorinated congeners (0–1-ortho-chlorobiphenyls [CBs]),
another fraction of A50 containing the congeners with two to four ortho-chlorines (2–4-ortho-CBs), or an organic extract from
Baltic gray seal blubber. The animals were exposed for 18 months, including two reproduction seasons. Among the animals given
the highest dose of A50, the whelping frequency was reduced in the second reproductive season, and all kits died within 24 h of
birth. Reproduction was also impaired by the lower dose of A50. Daily exposure to the 0–1-ortho-CBs separated from 0.3 mg A50
severely reduced kit survival. Reproduction was not significantly impaired by daily exposure to the 2–4-ortho-CBs separated from
0.3 mg A50 or by exposure to the blubber extract. We conclude that the reproductive toxicity in chronically PCB-exposed mink
is caused by the aryl hydrocarbon (Ah) receptor agonists. The lowest-observed-effect level for reproductive impairment was 2.4
ng 2,3,7,8-tetrachlorodibenzo-
p
-dioxin (TCDD) equivalents (TEQs) per kilogram body weight and day (22 pg TEQs/g feed).
-dealkylase by the 2–
Ethoxyresorufin- -dealkylase (EROD) was strongly induced by the 0–1-ortho-CBs and pentoxyresorufin-
O
O
4-ortho-CBs. High EROD activity was correlated with low kit production, and consequently EROD may serve as a marker for
reproductive toxicity by Ah receptor agonists in mink.
Keywords—Mink
Reproductive toxicity
Ethoxyresorufin-
O
-dealkylase
Polychlorinated biphenyls
2,3,7,8-Tetra-
chlorodibenzo- -dioxin equivalents
p
INTRODUCTION
Piscivorous birds and mammals living in contaminated en-
lowed by increases since the late 1970s [1]. In feeding studies,
fish taken from the Great Lakes in North America have had
adverse effects on reproduction in mink [6–8]. A hazard as-
sessment was conducted by Giesy et al. [9] to determine the
potential for adverse effects on mink consuming fish from three
rivers that flow into the Great Lakes. It was concluded that
consumption of fish below dams in the rivers would pose a
risk to mink, mainly because of the high PCB levels in the
fish.
It has been known for many years that mink are susceptible
to PCB and that effects on their reproduction occur at relatively
low exposure levels [6,10–12]. In most mink studies, the an-
imals were treated with technical PCB for a few months, in-
cluding a single reproduction season. Kihlstro¨m et al. [13]
exposed 10 female mink to a daily dose of 2 mg Clophen A50
(A50) for three months during the reproduction season, giving
a total dose of about 170 mg PCB/animal at parturition. The
10 females produced all together only one kit. Whether a lower
dose over a substantially longer period would cause similar
dramatic effects on reproduction has scarcely been examined.
In a recent study, female mink were exposed to PCB-contam-
inated carp from Saginaw Bay (Lake Huron, USA) for up to
18 months [14]. Fewer kits survived, and body weights of the
vironments such as the Baltic region accumulate persistent
lipophilic environmental contaminants to concentrations that
can have adverse health effects [1]. The Baltic populations of
gray seal (Halichoerus grypus), ringed seal (Phoca hispida),
and harbor seal (Phoca vitulina) are all under threat by en-
vironmental contaminants [1]. Furthermore, the population
density of the European otter (Lutra lutra) declined in many
regions in Sweden in the 1970s and 1980s, probably partly
because of polychlorinated biphenyl (PCB) contamination of
their diet [2–4]. As endangered species are generally not avail-
able for experimental studies, model species have to be studied.
Mink (Mustela vison) are farmed for their pelt, and ranch mink
are available for toxicological research. Moreover, this mus-
telid is a candidate species for use when monitoring exposure
of piscivorous wildlife to environmental contaminants. Mink
are opportunistic predators, and a large proportion of their diet
consists of fish. It was reported by Erlinge [5] that the diet of
mink along a Swedish river consisted of 60% fish, 24% wa-
terfowl, and 9% mammals. To identify suitable endpoints for
use in biomonitoring, the responses of mink to various envi-
ronmental contaminants have to be studied experimentally.
Mink is not an endangered species in Sweden, but hunting
data indicated diminishing mink populations in the 1960s, fol-
kits were reduced at a concentration of 500
␮g PCB/kg feed
(about 0.1 mg PCB/mink/d). In addition to PCB, the carp also
contained polychlorinated dibenzo-p-dioxins/furans (PCDDs/Fs)
and pesticides.
* To whom correspondence may be addressed
(bjorn.brunstrom@ebc.uu.se).
Various commercial products of technical PCB may differ
2318

Reproductive toxicity in chronically PCB-exposed mink
Environ. Toxicol. Chem. 20, 2001
2319
in chlorine content, and the mix of individual chlorinated bi-
phenyls in technical PCB includes more than 100 different
congeners. Certain of these congeners are aryl hydrocarbon
(Ah) receptor agonists, therefore having toxic effects similar
ml) was passed, followed by 100 ml of toluene. From a first
fraction (110 ml), 288 mg of 2–4-ortho-CBs was isolated and
pooled with the 2.92 g of 2–4-ortho-CBs initially isolated
(compare above). In a second fraction from the second charcoal
column, 221 mg of 0–1-ortho-CBs were isolated. However,
the bulk of the 0–1-ortho-CBs present did not elute from the
first charcoal column but was recovered by Soxhlet extraction
(20 cycles with toluene) of the charcoal [20]. In this way, 620
mg of 0–1-ortho-CBs were obtained, and thus all together 841
mg of 0–1-ortho-CBs were isolated. Of the PCB congeners
obtained from 4.0 g of A50, 80% were 2–4-ortho (3.2 g) and
20% were 0–1-ortho (0.8 g).
to those of 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD) and
other PCDDs/Fs. Highly toxic to adult mink, TCDD has a
median lethal dose (LD50) value of about 47 ng/kg/d in fe-
males exposed for 125 d [15]. Another Ah receptor agonist,
3,3
mink exposed for three months via the feed to a concentration
of 100 g/kg feed, corresponding to a daily dose of 15 g per
animal [16].
Ј,4,4Ј,5,5Ј-hexachlorobiphenyl, caused 50% mortality in
␮
␮
The objective of the present study was to examine the ef-
fects of long-term exposure to relatively low dietary concen-
trations of various PCB preparations on the reproductive per-
formance of mink. An organic extract of blubber from Baltic
gray seal was also studied to ascertain what effects, if any,
weathered PCB and other contaminants accumulated in seal
blubber might have.
Isolation of persistent organic pollutants from gray seal
blubber
Anthropogenic compounds in gray seal blubber were iso-
lated and used for dosing one group of mink. Blubber from
three gray seals, kindly provided by Dr. M. Olsson (Swedish
Museum of Natural History, Stockholm), was used for indus-
trial cold-press isolation of oil (blubber pressed by Sildolje-
and Sildemelindustrins Research Institute, Bergen, Norway).
The isolated oil contained a mixture of lipophilic anthropo-
genic compounds accumulated in the seal blubber. The per-
sistent organic pollutants in the seal blubber oil (4.2 kg) were
isolated by partitioning between the oil and acetonitrile in a
cyclic procedure as described by Jensen et al. [21]. The pro-
cedure was repeated once. The aryl methyl sulfones were re-
moved by dimethyl sulfoxide partitioning [22].
MATERIAL AND METHODS
Chemicals
A commercial PCB brand, Clophen A50, produced by Bay-
er (Leverkusen, Germany), was donated by Dr. Lars Fo¨rlin
(Department of Zoophysiology, University of Gothenburg).
2,3,3
ner numbers according to Ballschmiter et al. [17]),
2,3,3 ,4,5,5 -hexachlorobiphenyl (CB-159), and 2,2 ,3,6 -tetra-
chlorbiphenyl (CB-46) were synthesized as described by Sund-
stro¨m [18]. 2,2-Bis(phenyl)-1,1-dichloroethene ( -DDE) was
prepared via sodium hydroxide–promoted dehydrochlorination
of 2,2-bis(phenyl)-1,1,1-trichloroethane ( -DDT), prepared by
Ј,4,4Ј,5,5Ј-Heptachlorobiphenyl (CB-189) (PCB conge-
Ј
Ј
Ј
Ј
Animals and treatments
␾
Female mink (Mustela vison, standard breed, six months
old) were divided into six groups of 20 individuals each. The
body weights of the females on commencing the study were
␾
a method similar to the synthesis of 2,2-bis(4-fluorophenyl)-
similar (mean
ϭ 1.2–1.3 kg) in the various groups. The females
13
1,1,1-trichloroethane [19]. A C12-labeled standard mixture of
were housed individually outdoors in sheltered cages and had
free access to tap water. All animals were tested for plasma-
cytosis and proved negative. Twice a day (morning and af-
ternoon) they were fed a basic diet containing cereals (10%),
fish heads and backbones (40%), slaughter offal (50%), and a
mixture of vitamins and trace elements (Rovemix, Hoffman
3,3
Ј
,4,4
Ј
-tetrachloro-[13C12]-biphenyl, 3,3
Ј
,4,4
Ј,5-pentachloro-
[13C12]-biphenyl, and 3,3
Ј
,4,4Ј,5,5Ј
-hexachloro-[13C12]-biphenyl
(5.0 ppm of each CB in nonane, 99% purity) was obtained from
Cambridge Isotope Laboratories (Andover, MA, USA).
All solvents used were of pro analysi quality or higher. The
sodium chloride (Baker reagent) was obtained from J.T. Baker
Chemicals B.V. (Deventer, The Netherlands). The silica gel
(Silica gel 60, 70–230 mesh ASTM) was purchased from
Macherey-Nagel (Du¨ren, Germany) and heated overnight at
LaRoche, Stockholm, Sweden,
was about 130 g.
Ͻ1%). The daily feed ration
Pollutant exposure started on November 24, 1992, when
the females were six months old, and ended just over 18
months later (May/June 1994). The compounds studied were
dissolved in oil prepared from capelin (Mallotus villosus; Nor-
sildmel, Minde, Norway) captured in the north Atlantic. The
concentrations of various pollutants in the oil were low (0.15
280ЊC before use.
Fractionation of A50
Using a slightly modified version of a previously described
procedure [20], the A50 (4.0 g) was separated into one fraction
containing the CBs that lack or have only one chlorine atom
in an ortho position (0–1-ortho-CBs) and one fraction con-
taining CBs with two to four chlorine atoms in ortho positions
of the biphenyl structure (2–4-ortho-CBs). The PCB mixture
was dissolved in heptane:benzene (95:5, v/v, 200 ml) and slow-
ly added (10 ml/h) to a charcoal column (56 g). Heptane:
benzene (95:5, v/v, 2,250 ml) was passed through the column,
followed by toluene (180 ml). The 1,200 ml of solvent eluting
first contained only 2–4-ortho-CBs (2.92 g). The material that
eluted thereafter was pooled and the solvent evaporated. The
PCB residue (510 mg), containing a mixture of PCB congeners
with zero to four ortho-chlorine atoms, was dissolved in hep-
tane:benzene (95:5, v/v, 30 ml) and added to a fresh charcoal
column (5.6 g), through which heptane:benzene (95:5, v/v, 200
␮
g PCB/g oil, 0.06
chlorophenyl]ethane
yl]ethene 1,1-dichloro-2,2-bis[4-chlorophenyl]ethane]/g oil,
0.03 g hexachlorobenzene/g oil, and 0.03 g lindane/g oil,
␮
ϩ
g total DDT [1,1,1-trichloro-2,2-bis[4-
1,1-dichloro-2,2-bis[4-chlorophen-
ϩ
␮
␮
lipid weight). Three times a week, 1 ml of capelin oil solution
was added to the morning feed ration. The capelin oil solution
was palatable to the animals. Designation of the treatment
groups and exposure of the mink were as follows:
A50 high: Females exposed to 0.7 mg A50 three times a
week, corresponding to 0.3 mg/d.
A50 low: Females exposed to 0.24 mg A50 three times a
week, corresponding to 0.1 mg/d.
0–1-ortho: Females exposed to a fraction of A50 containing
0–1-ortho-CBs and any dicyclic aromatic contaminants such
as PCDFs; the daily dose corresponded to the amounts of these

2320
Environ. Toxicol. Chem. 20, 2001
B. Brunstro¨m et al.
congeners separated from 0.3 mg of A50 (the daily dose of
A50 in A50 high group).
2–4-ortho: Females exposed to a fraction of A50 containing
2–4-ortho-CBs; the daily dose corresponded to the amounts
of these congeners separated from 0.3 mg of A50 (the daily
dose of A50 in A50 high group).
Blubber extract: Females exposed to an extract of Baltic
gray seal blubber containing organochlorine pollutants but
without aryl methyl sulfones; the daily dose of the extract
contained 0.064 mg total PCB and 0.021 mg total DDT.
Control: Females given 1 ml of the fish oil vehicle three
times a week.
The level of PCB in the basic diet was analyzed in aliquots
sampled on two different occasions during the experiment. The
highest total PCB and total DDT concentrations in the feed
Results from analysis of part of the samples collected are
presented in this paper. The studies were approved by the Local
Ethics Committee for Research on Animals.
Enzyme assays
Frozen pieces (ϳ3 g) of the maternal livers were thawed
in ice-cold buffer (0.15 M KCl in 0.05 M Tris-HCl, pH 7.5)
and homogenized in a Potter-Elvehjem homogenizer (Kontes
Glass, Vineland, NJ, USA). Microsomes were prepared by
differential centrifugation as previously described [23]. Ethox-
yresorufin-O-dealkylase (EROD) activity in the microsomal
fractions was determined using a slightly modified version of
the method presented by Pohl and Fouts [24] as previously
described [25]. Pentoxyresorufin-O-dealkylase (PROD) was
determined using essentially the method described by Lubet
et al. [26]. Microsomal protein concentration was determined
by the method of Lowry et al. [27] using bovine serum albumin
as a standard.
were 0.011 and 0.004 ␮g/g (wet-wt basis), respectively.
During the mating periods (last three weeks in March), 30
nonexposed males were used. In general, each female was
given two opportunities to mate, and the mating day(s) was
noted. The frequency of successful mating (number of mated
females/total number of females) was high (0.8–1.0) both years
in all groups.
Instruments for chemical analysis
All tissue samples were homogenized with a DISP 25 ho-
mogenizer (InterMed, Roskilde, Denmark) equipped with a
scissors rod (8-mm i.d.). A vacuum centrifuge (Genevac SF
50, Sales Development, Ipswich, England) was used for sol-
vent evaporations. Gas chromatography was performed on a
Varian 3400 gas chromatograph, equipped with a Varian 8200
autosampler and an electron-capture detector. The column used
Measurements and sampling
At parturition the first reproductive season (April–May
1993), the total number of kits (dead and alive) in each litter
was recorded. The kits were removed from the dams within
24 h after birth, sexed, and inspected for gross malformations.
The live kits were weighed (
Ϯ
0.01 g) and killed by decapi-
was a DB-5 fused silica capillary column (30-m
ϫ 0.25-mm
tation. Many of the dead kits were dehydrated, and therefore
only live kits were weighed. Immediately after decapitation,
blood was collected in heparinized glass capillaries from blood
vessels in the neck. After centrifugation, the plasma samples
were pooled separately for each sex and litter and frozen
i.d., 0.25- m film thickness) manufactured by J&W Scientific
(Folsom, CA, USA). Hydrogen was used as a carrier gas and
nitrogen as a makeup gas. The column temperature was held
␮
at 80
which was held for 10 min. The split/splitless injector in split-
less mode was held at 250 C and the electron-capture detector
at 360 C. The data system EldsPro from Chromatography Data
ЊC for 2 min and then increased by 10ЊC/min to 300ЊC,
(
Ϫ
20
Њ
C). Thymus was excised and weighed (
Ϯ
0.0001 g). Var-
Њ
ious organs were excised and frozen in liquid nitrogen or on
dry ice.
At parturition the second reproductive season (April–May
1994), the total number of kits (dead and alive) in each litter
was recorded. Within 24 h of birth, the kits were sexed and
inspected for gross malformations, and the live kits were
Њ
Systems AB (Kungsho¨g, Sweden) was used to collect and
process the data.
High-performance liquid chromatography was performed
on a module consisting of a Varian 9012 programmable solvent
delivery system, a Varian 9050 variable wavelength ultravi-
olet-visible light detector (Varian, Walnut Creek, CA, USA),
a Valco injector (Vici AG, Schenkon, Switzerland) equipped
weighed (Ϯ0.01 g). The kits were weighed at two and five
weeks of age, and immediately after the last weighing they
were anesthetized by electric shock and killed by decapitation.
During the anesthesia, the jugular veins and carotid arteries
were opened by an incision in the neck. Blood from these
vessels was collected in heparinized glass tubes, and after
centrifugation the plasma was transferred to plastic vials. Var-
with a 100-
PYE columns (2-[1-pyrenyl]ethyldimethylsilylated silica gel,
particle size 5 m, Nacalai Tesque, Kyoto, Japan) coupled in
␮l loop, and two 150 ϫ 4.6-mm-i.d. comsosil 5-
␮
series and equipped with a column cooler (Jones Chromato-
graphy, Lakewood, CO, USA).
ious organs were excised and weighed (
Ϯ
0.001 g). Samples
Gas chromatography/mass spectrometry (GC/MS) was per-
formed on a TSQ 700 (Finnigan MAT, Bremen, Germany)
were frozen in liquid nitrogen or on dry ice or were fixed in
formalin or glutaric aldehyde for subsequent histopathological
examination.
equipped with a DB-5 column (30-m
ϫ 0.25-mm i.d., 0.25-
␮
m film thickness). Helium was used as carrier gas at a head
The adult females were weighed (
Ϯ
0.1 g) at the start of
pressure of 12 psi. The mass spectrometer was operated in
electron ionization mode with an electron energy of 70 eV.
the study, after 12 months of exposure, and finally at the end
of the 18-month exposure period. One to two weeks after
removing the kits, the dams were killed by cervical dislocation
during anesthesia provoked by electric shock. While the ani-
mals were anesthetized, blood was collected in heparinized
vacutainers by heart puncture. After centrifugation, the plasma
The injector and transfer line temperature was 280
temperature in the ion source was 150 C. The column tem-
perature was kept at 60 C for 4 min, followed by a linear
temperature increase of 30 C/min up to 180 C and then a linear
temperature increase of 3 C/min up to a final temperature of
300 C that was held for 3 min. The instrument was run in
ЊC, and the
Њ
Њ
Њ
Њ
Њ
was transferred to plastic vials and frozen (
Ϫ
20
Њ
C). The non-
Њ
reproducing dams were killed one to two weeks after expected
parturition. Tissues from all dams were excised and treated in
the same way as tissues from the five-week-old kits.
selective ion monitoring detection mode. The mass of the two
most intense ions in each molecular ion cluster was measured
by the isotope dilution technique.

Reproductive toxicity in chronically PCB-exposed mink
Environ. Toxicol. Chem. 20, 2001
2321
Table 1. Concentrations of major non- and mono-ortho-chlorobiphenyls (CBs) assigned a toxic equivalency factor (TEF) value [28] in the various
capelin oil solutions of PCB used to dose the minka
PCB congener concn. (ng/ml)
Treatment
group
CB-77
CB-126
CB-169
CB-105
4.6
CB-118
7.4
CB-156
1.6
CB-157
0.73
Controlb
A50 low
A50 high
0–1-ortho
2–4-ortho
Blubber extract
0.87
69
Յ
0.36
Յ
0.91
1.3
20
61
17
9.2
25
24
ϫ
ϫ
ϫ
ϫ
103
22
61
ϫ
ϫ
ϫ
ϫ
ϫ
103
103
103
103
103
2.2
ϫ
ϫ
ϫ
ϫ
ϫ
103
0.57
1.8
1.8
ϫ
ϫ
ϫ
103
0.20
ϫ
103
2.0
1.2
0.81
1.0
103
103
103
6.5
6.0
103
103
103
103
103
103
69
52
Յ
3.0
1.1
Յ
0.86
0.85
0.63
6.0
0.17
0.35
0.23
49
43
89
a CB-123 and CB-167 were detected in the oils A50 low, A50 high, and 0–1-ortho, but values for them are not given in the table, as they
coeluted with other congeners. Their concentrations were estimated on the basis of relative congener concentrations in Clophen A50 [30].
Calculations from the estimated concentrations and the congener-specific TEFs showed that their contribution to total 2,3,7,8-tetrachlorodibenzo-
p
-dioxin equivalent concentrations was of a minor degree (
Ͻ4%).
b The concentrations found in the capelin oil.
Analysis of PCB in oil preparations used in the experiment
(2:5, v/v). Following centrifugation, the supernatant was trans-
ferred to a test tube containing sodium chloride (0.9%) and
phosphoric acid (0.1 M) for extraction. The tissue residue was
extracted twice with methyl tert-butyl ether:hexane (1:9, v/v),
and after centrifugation the solvent was transferred to the ex-
traction tube containing sodium chloride/phosphoric acid so-
lution. The phases were mixed by rocking the tube for a few
minutes. Following centrifugation, the organic phase was
transferred to a preweighed tube for subsequent gravimetric
determination of lipid weight. The aqueous phase was re-ex-
tracted with hexane, and after centrifugation the hexane was
transferred to the preweighed tube. The bulk of organic solvent
was removed by means of the vapor vacuum pump, and any
solvent residue was evaporated in a heating block under a
gentle stream of nitrogen. The lipid weight of each sample
was determined gravimetrically. The residue was dissolved in
a small amount of hexane, and the lipids were removed on a
sulfuric acid–impregnated silica gel column (1:2, w/w, 1.0 g).
The CBs were eluted with hexane as the mobile phase and
were collected in 10 ml.
Two samples (1 ml) of each oil preparation were analyzed
for 0-ortho-CBs and 1-ortho-CBs. Each oil sample was dis-
solved in hexane (10 ml) and spiked with 25.5
labeled internal standard mixture for the 0-ortho-CBs and with
0.01 (control) to 14 g (A50 high) CB-159 for the 1-ortho
␮
g of the 13C12-
␮
-
CBs. The samples were then treated with concentrated sulfuric
acid (5 g). The tubes were carefully inverted 20 times and
then centrifuged, and the hexane was removed. The procedure
was repeated once. The volume of the pooled hexane was
reduced to 0.5 ml under a gentle stream of nitrogen. This
solution was applied to a sulfuric acid:silica gel column (1:2,
w/w, 1.2 g), and hexane (10 ml) was used as the mobile phase.
The hexane was evaporated to leave a residue of about 50 ␮l,
and the sample was then analyzed in the high-performance
liquid chromatography system described previously. The two
serially coupled PYE columns were kept at 5ЊC, and the mobile
phase was kept at 0 C. The fraction borders were chosen ac-
Њ
cording to the results obtained from a separation of a standard
mixture of CB-77, CB-118, CB-157, and CB-169. Three frac-
tions (fractions 1–3) were collected: fraction 1: 4.0 to 7.3 min
(2–4-ortho-CBs); fraction 2: 7.3 to 9.8 min (1-ortho-CBs);
and fraction 3: 9.8 to 16.3 min (0-ortho-CBs). The column
was retroflushed for 40 min to avoid any contamination of
subsequent samples. The solvent volume in fraction 3 (0-or-
tho-CBs) was reduced by evaporation, and the samples were
then transferred to GC vials and the volume adjusted to 100
To facilitate recovery evaluation, an injection standard (CB-
46) was added to each sample prior to GC analysis. Recovery
was 98%
Ϯ 5%. The individual CBs were quantified using
A50 as a standard for PCB. The relative amounts were cal-
culated according to Schultz et al. [30].
Statistics
Mean values of the treated mink were compared with the
␮
l prior to analysis by GC/MS. ⌽-DDE (18.9 ng) was added
control values using Dunnett’s
were used when variances differed significantly (Bartlett’s test,
0.05). Medians were compared using the Mann–Whitney
t test. Log-transformed values
as injection standard. The 0-ortho-CBs were quantified by us-
ing a multilevel calibration curve and the internal standard
method. For analysis of the 1-ortho-CBs, CB-189 was added
as injection standard. The analyses were carried out by GC/
electron-capture detection. The recovery of the internal stan-
p
Ͻ
test, and frequencies were compared by using Fisher’s exact
probability test. In the statistical treatment of the kit data, the
mean value for each litter was used as the representative value
for the corresponding female.
dard (CB-159) was 93%
Ϯ 3%. The 2–4-ortho-CBs were an-
alyzed as described in the following for the analysis of these
congeners in muscle.
RESULTS
The TCDD-equivalent (TEQ) exposure of the mink in the
various groups was calculated using the toxic equivalency fac-
tors (TEFs) recently suggested by a World Health Organization
(Paris, France) working group [28] for CB-77, -105, -118, -123,
-126, -156, -157, -167, and -169.
The PCB analysis of the oil preparations used to dose the
mink in the groups A50 high and A50 low confirmed that the
animals in these groups were exposed to amounts correspond-
ing to 0.3 and 0.1 mg PCB/d, respectively. Congener-specific
analysis of the oil preparations (Table 1) showed that the frac-
tionation was not entirely successful, especially with regard
to the 0-ortho-CBs. The concentrations of CB-77 and CB-126
in the 0–1-ortho-CB fraction were only about 30% of those
in A50 high, and the remainder was not recovered in the 2–
Analysis of PCB in mink muscle
All tissue samples were extracted using a slightly modified
version of a general method for PCB analysis [29]. The tissue
was cut in small pieces and homogenized in hexane:acetone

2322
Environ. Toxicol. Chem. 20, 2001
B. Brunstro¨m et al.
Table 2. The various treatments of female mink and body weights during the exposure period of 18 months
Body weight (g)b
Average
daily exposure
of each animal
TEQ
exposure
(ng/d)a
Treatment
group
After 18
months
At start
After one year
Control
Ͻ
1.5
␮
g PCB
0.02
2.9
1,198
Ϯ
ϭ
Ϯ
ϭ
Ϯ
ϭ
Ϯ
ϭ
Ϯ
ϭ
Ϯ
ϭ
188
1,297
Ϯ
ϭ
Ϯ
ϭ
Ϯ
ϭ
Ϯ
ϭ
Ϯ
ϭ
Ϯ
ϭ
230
18)
161
19)
167
20)
151
19)
167
20)
121
19)
1,064
Ϯ
ϭ
Ϯ
ϭ
178
(
n
20)
194
20)
148
20)
155
20)
170
20)
153
20)
(
n
(
n
18)
189
19)
170
19)
174
19)
172
20)
160
19)
A50 low
0.1 mg Clophen A50
0.3 mg Clophen A50
1,265
1,295
1,133
(n
995 Ϯ
(
n
(
n
A50 high
0–1-ortho
2–4-ortho
Blubber extract
8.4
1,197
1,178
(
n
(
n
(
n
ϭ
Ϯ
ϭ
Ϯ
ϭ
Ϯ
ϭ
Amounts in 0.3 mg
Clophen A50c
Amounts in 0.3 mg
Clophen A50
0.064 mg PCB
ers
5.8
1,300
1,303
1,131
(n
1,105
(n
1,107
(
n
(
n
0.41
0.11
1,209
1,165
(
n
(
n
ϩ
oth-
1,276
1,227
(
n
(
n
(n
a 2,3,7,8-Tetrachlorodibenzo-
p-dioxin equivalent (TEQ) exposure was calculated from congener-specific analysis of the various oils and the toxic
equivalency factors suggested by van den Berg et al. [28]. The TEQ concentration in the feed is not included but was deemed to be a minor
contributor to total TEQ exposure, except in the control group.
b Values are presented as mean
Ϯ standard deviation.
c Only about 30% of the non-ortho-chlorinated congeners was recovered.
4-ortho-CB fraction. The recoveries of the 1-ortho-CBs on
fractionation were high. Eighty-six percent of CB-118 was
obtained in the 0–1-ortho-CB fraction, representing the lowest
recovery of the 1-ortho-CBs analyzed. The small portion of
1-ortho-CBs not recovered in the 0–1-ortho-CB fraction was
present in the 2–4-ortho-CB fraction. The TEQ exposures of
the mink in the various groups, as calculated from the chem-
ically determined concentrations of Ah receptor agonists, are
shown in Table 2.
In the PCB-exposed adult females, no signs of toxicity were
observed, judging by the general appearance of the animals
and the lack of effects on body weight (Table 2). In the first
reproductive season, after six months of exposure of the dams,
a decreased birth weight (18% lower than in controls) was
observed in group A50 high (0.3 mg A50/animal/d), whereas
whelping frequency and litter size were not significantly af-
fected (Table 3). Thymus weights were decreased in the kits
in groups A50 high and 0–1-ortho (Table 4). The kits in A50
high also appeared to have a reduced viability, judging by their
feeble efforts to search for milk and their weak vocal calls.
In the second season, the effects on reproduction were more
pronounced and clearly dose dependent (Table 5). In group
A50 high, only 39% of the dams produced offspring, compared
with 93% in the control group. The newborn kits in group
A50 high weighed 25% less than the control kits, and they all
died within 24 h. Reproduction was impaired also in group
A50 low (0.1 mg A50/mink/d). Kit production, measured as
the number of two-week-old kits per mated female, was re-
duced (Fig. 1), and the two-week survival frequency of the
kits born (including stillborn kits) was 36%, whereas survival
was 73% in the control group. The body weight of kits in
group A50 low was lower than in the control kits, at both two
and five weeks of age (Fig. 2). In both seasons, the frequency
of abnormalities (mainly edemas) among the stillborn kits in
group A50 high (29% the first season and 24% the second
season) exceeded that in the control group (0 and 10%).
Treatment with the 0–1-ortho-CB fraction strongly reduced
(by two postnatal weeks) the number of kits per mated female
(Fig. 1), and the two-week survival frequency in this group
was only 7%. The growth rate of the kits was substantially
Table 3. Reproductive performance of female mink after six-month exposure to PCBs in the feed
% Mated
females
Litter size,
Treatment
group
Litter size,
live kitsb,c
dead and live
Birth weight
(g)b,d
whelpinga
kitsb,c
Control
93
89
90
78
79
89
4.0
Ϯ
ϭ
Ϯ
ϭ
Ϯ
ϭ
Ϯ
ϭ
Ϯ
ϭ
Ϯ
ϭ
2.7
25)
2.1
17)
2.3
17)
2.3
12)
2.2
15)
2.2
17)
4.9
Ϯ
ϭ
Ϯ
ϭ
Ϯ
ϭ
Ϯ
ϭ
Ϯ
ϭ
Ϯ
ϭ
2.4
25)
2.1
17)
2.3
17)
2.4
12)
2.1
15)
1.8
17)
9.6
(n ϭ
9.5 Ϯ
(n ϭ
7.9 Ϯ
(n ϭ
9.1 Ϯ
(n ϭ
9.9 Ϯ
(n ϭ
9.2 Ϯ
(n ϭ
Ϯ
2.0
22)
1.9
16)
1.2*
13)
1.7
12)
2.1
14)
1.6
16)
(
n
(
n
A50 low
5.2
5.9
(
n
(
n
A50 high
0–1-ortho
2–4-ortho
Blubber extract
3.8
5.1
(
n
(
n
3.6
4.8
(
n
(
n
4.2
5.3
(
n
(
n
4.9
5.4
(
n
(
n
a There were 27 mated females in the control group and 18 to 20 in the other groups.
b Values are presented as mean
Ϯ standard deviation.
c Nonwhelping females not included.
d Only live kits were weighed; values given are means of the litter means in each group.
* Significantly different from control value (Dunnett’s t test, p Ͻ 0.05).

Reproductive toxicity in chronically PCB-exposed mink
Environ. Toxicol. Chem. 20, 2001
2323
Table 4. Kit organ weights (as % or ‰ of body wt) at autopsy; values given are means of the litter means in each groupa
Newborn kits,
first season
Five-week-old kits,
second season
Treatment
group
Thymus
(‰ of body wt)
Thymus
(% of body wt)
Liver
(% of body wt)
Adrenal
(‰ of body wt)
Control
0.97
Ϯ
ϭ
0.25
19)
0.35
Ϯ
ϭ
0.11
12)
3.98
Ϯ
ϭ
0.25
12)
0.102
Ϯ
0.01
(
n
(
n
(
n
(n ϭ 12)
A50 low
0.91
Ϯ
ϭ
0.20
15)
0.34
Ϯ
ϭ
0.05
10)
3.68
Ϯ
ϭ
0.44
10)
0.112
Ϯ
0.018
(
n
(
n
(
n
(n ϭ 10)
A50 high
0–1-ortho
2–4-ortho
Blubber extract
0.56
(n ϭ 13)
0.78
Ϯ
0.18**
—
—
—
Ϯ
ϭ
Ϯ
ϭ
Ϯ
ϭ
0.13*
12)
0.18
14)
0.14
16)
0.27
Ϯ
ϭ
Ϯ
ϭ
Ϯ
ϭ
0.11
4)
0.09
15)
0.09
12)
3.72
Ϯ
ϭ
Ϯ
ϭ
Ϯ
ϭ
0.66
4)
0.40
15)
0.50
12)
0.117
(n
0.109
(n
0.096
Ϯ
ϭ
Ϯ
ϭ
Ϯ
ϭ
0.015
4)
0.035
15)
0.014
12)
(
n
(
n
(
n
0.93
0.35
3.64
(
n
(
n
(
n
0.85
0.33
3.72
(
n
(
n
(
n
(n
a Values are presented as mean
Ϯ standard deviation.
* Significantly different from control value (Dunnett’s
** Significantly different from control value (Dunnett’s
t
test;
p
p
Ͻ
Ͻ
0.05).
0.01).
t
test,
reduced by the 0–1-ortho-CBs (Fig. 2), whereas the 2–4-or-
tho-CB fraction affected neither kit production (Fig. 1) nor
their growth rate (Fig. 2). However, a numerically lower two-
week survival frequency (50%) than in the control group (73%)
was observed also for the kits in group 2–4-ortho. In the group
exposed to the seal blubber extract, no adverse effects on re-
production were noted.
No significant effects on thymus, liver, or adrenal weights
(relative to the body wt) were observed in any group in the
five-week-old kits at autopsy (Table 4).
PROD (Fig. 3), whereas EROD was only slightly induced in
this group (Fig. 4).
Reduced kit production occurred concomitantly with in-
creased hepatic EROD activity in the adult females. The neg-
ative correlation between TEQ exposure and kit production
and the positive correlation between TEQ exposure and EROD
activity are shown in Figure 5. Exposure to TEQs was cal-
culated from the chemically analyzed levels of Ah receptor
agonists in the oils. The TEQ concentration in the basal feed
was not included in this calculation.
The hepatic cytochrome P450–dependent enzyme activities
PROD and EROD were induced in a dose-dependent manner
in the female adults exposed to A50 (Figs. 3 and 4). We ob-
served a similar PROD induction by the highest dose of A50
and the 2–4-ortho-CB fraction and a lack of PROD induction
in animals exposed to the 0–1-ortho-CB fraction (Fig. 3). Both
PCB fractions induced EROD, although the 0–1-ortho-CB
fraction was much more potent (Fig. 4). The contaminants
extracted from Baltic gray seal blubber strongly induced
The concentration of total PCB in muscle of the adult fe-
males after the experiment is shown in Figure 6. The mean
concentration (n ϭ 7) of total PCB in A50 high was 54 ␮g/g
(lipid wt), whereas the concentrations in the other exposed
groups were much lower and in the same range (10.6–14.4
␮
g/g), even though different types of PCB were given. The
lipid content in muscle was 2.4 0.4%. The bioaccumulation
Ϯ
factors (BAFs), defined as the muscle concentrations (lipid wt)
of the various CB congeners, divided by feed concentrations
Table 5. Reproductive performance of female mink after 18-month exposure to PCBs in the feed
% Mated
females
Litter size,
Treatment
group
Litter size,
live kitsb,c
dead and live
Birth weight
(g)b,d
whelpinga
kitsb,c
Control
93
4.4
Ϯ
ϭ
Ϯ
ϭ
Ϯ
ϭ
Ϯ
ϭ
Ϯ
ϭ
Ϯ
ϭ
2.5
14)
2.4
14)
1.7*e
7)
2.3
13)
2.1
18)
1.8
14)
5.1
Ϯ
ϭ
Ϯ
ϭ
Ϯ
ϭ
Ϯ
ϭ
Ϯ
ϭ
Ϯ
ϭ
2.4
14)
2.3
14)
1.8
7)
2.2
13)
2.3
18)
1.9
14)
8.9
(n ϭ
8.0 Ϯ
(n ϭ
6.7 Ϯ
(n ϭ
8.1
(n ϭ
9.2 Ϯ
(n ϭ
8.9 Ϯ
(n ϭ
Ϯ
1.8
14)
1.0
14)
2.2*
5)
1.2
12)
1.5
18)
1.7
14)
(
n
(
n
A50 low
88
5.9
6.2
(
n
(
n
A50 high
0–1-ortho
2–4-ortho
Blubber extract
39**
72
2.0
4.1
(
n
(
n
4.0
5.0
Ϯ
(
n
(
n
95
4.9
6.0
(
n
(
n
93
5.1
5.7
(
n
(
n
a The numbers of mated females were 15 to 19.
b Values are presented as mean
standard deviation.
c Nonwhelping females not included.
Ϯ
d Only live kits were weighed; values given are means of the litter means in each group.
e All kits died within 24 h.
* Significantly different from control value (
** Significantly different from control value (Fisher’s exact probability test,
p
Ͻ
0.05; litter size: Mann–Whitney test; birth weight: Dunnett’s
0.01).
t test).
p
Ͻ

2324
Environ. Toxicol. Chem. 20, 2001
B. Brunstro¨m et al.
Fig. 1. Numbers of two-week-old kits per mated female mink follow-
ing exposure to polychlorinated biphenyl (PCB) for 18 months. Boxes
extend from the 25th to the 75th percentile, with a horizontal line at
the median. The whiskers show the range of the data among mated
females in each group. The numbers of litters in each group are shown
below boxes and whiskers. Asterisks denote values that differ sig-
Fig. 3. Hepatic pentoxyresorufin-
O
-dealkylase (PROD) activities in
adult female mink exposed to various polychlorinated biphenyl (PCB)
preparations for 18 months. Values shown are means and standard
deviations for seven animals in each group. Asterisks denote values
that differ significantly from the control value (**
test).
p Ͻ 0.01, Dunnett’s
t
nificantly from the control value (*
Whitney test, one tailed).
p Ͻ 0.05, ***p Ͻ 0.001, Mann–
fresh weight, then the animals would be exposed to about 0.1
mg PCB/kg body weight/d (food consumption assumed to be
200 g/kg body wt/d). We found a marked effect on kit survival
and growth in the group exposed to this dose of PCB (A50
low), a dose that is slightly lower than the lowest-observed-
adverse-effect level (LOAEL) of 0.16 mg PCB/mink/d as re-
ported for mink fed carp from Saginaw Bay, Lake Huron,
Michigan, USA [8]. A comparison of the LOAEL values for
PCB obtained in our study and in that carried out by Heaton
and coworkers [8] is hampered by several discrepancies in
design of the two studies. We exposed the mink to a technical
PCB preparation for 18 months, whereas the mink in the study
by Heaton and coworkers were exposed for six months to the
weathered PCB present in the carp. The carp also contained
compounds other than PCBs, such as PCDDs/Fs, which prob-
ably contributed to the effects observed in that study (see the
following comparison of TEQ exposure). In our study, the
concentration in the feed was the same during the two repro-
duction seasons, resulting in a reduced frequency of whelping
females in the second season only. This finding suggests that
(lipid wt), differed in the various groups given different CB
mixtures. The BAFs were 2.1, 1.3, 2.3, 0.7, and 1.9 for the
groups A50 high, A50 low, 0–1-ortho, 2–4-ortho, and blubber
extract, respectively.
DISCUSSION
In this paper we show that long-term exposure to low doses
of PCB impairs reproduction in mink, manifested as fetal
deaths, abnormalities, decreased kit survival, and decreased
kit growth. The doses of A50 given (0.1 and 0.3 mg/mink/d)
are relevant in terms of wildlife exposure to PCB via the diet
in contaminated areas. Mean PCB concentrations in samples
of fish from Swedish coastal waters and lakes in 1991–1992
were as follows: in sea trout (Salmo trutta) from the Gulf of
Bothnia, 331
from the Baltic Sea, 430
alpinus) from Lake Va¨ttern, Sweden, 797
diet of fish-eating mammalian wildlife, such as otters, con-
sisted solely of fish with a PCB concentration of 500 g/kg
␮
g/kg fresh weight; in salmon (Salmo salar
g/kg; and in arctic char (Salvelinus
g/kg [31]. If the
)
␮
␮
␮
Fig. 2. Weights at birth and at two and five weeks of age in kits of
female mink exposed to polychlorinated biphenyl (PCB) for 18
months. Values shown are means and standard deviations of the litter
means in each group. The number of litters in each group is shown
below the bars. Note that there were no kits by two and five weeks
in A50 high. Asterisks denote values that differ significantly from the
Fig. 4. Hepatic ethoxyresorufin-O-dealkylase (EROD) activities in
adult female mink exposed to various polychlorinated biphenyl (PCB)
preparations for 18 months. Values shown are means and standard
deviations for seven animals in each group. Asterisks denote values
that differ significantly from the control value (*
0.01, Dunnett’s test).
p Ͻ 0.05, **p Ͻ
control value (*
p
Ͻ
0.05, **
p
Ͻ
0.01, Dunnett’s
t
test).
t

Reproductive toxicity in chronically PCB-exposed mink
Environ. Toxicol. Chem. 20, 2001
2325
Madsen [34] reported that PCB concentrations exceeded 50
g/g in 19%, 30 g/g in 30%, and 10 g/g in 70% of tissue
␮
␮
␮
samples (71 liver, two muscle) from 73 otters found dead in
Denmark between 1980 and 1990. Median PCB concentrations
(
␮
g/g, lipid wt) in otters traumatically killed in 1990 to 1994
were 7.7 (range 0.6–75, 29) in northern Sweden and 29
(range 7–149, 10) in southern Sweden [35].
n
ϭ
n
ϭ
Our results show the decisive role played by the dioxin-
like PCB congeners in the PCB-induced impairment of repro-
duction in mink. Congener-specific PCB analysis of the various
oils made it possible to estimate their TEQ concentrations and
to assess TEQ exposures in the mink. A negative correlation
between TEQ exposure and overall reproductive performance
was found. The PCB congeners lacking affinity for the Ah
receptor had a very limited effect (if any) on mink reproduction
at the doses given. The finding that the reproductive toxicity
is due solely to the dioxin-like PCBs differs from the conclu-
sions by Kihlstro¨m et al. [13], who suggested that the con-
geners lacking Ah receptor affinity also contribute to the re-
productive toxicity of PCB. Adult mink are known to be highly
susceptible to TCDD, and Hochstein et al. [36] found a single
Fig. 5. Production of two-week-old kits per mated female mink and
hepatic 7-ethoxyresorufin-
females as functions of daily 2,3,7,8-tetrachlorodibenzo-
equivalent (TEQ) exposure for 18 months. □ ϭ number of kits,
EROD activity.
O
-dealkylase (EROD) activity in the adult
p
-dioxin
᭡
ϭ
the PCB concentration in the animals increased from the first
to the second reproduction season, showing the relevance of
long-term exposure for estimation of a LOAEL. In a recent
multigenerational study by Restum et al. [14], mink were ex-
posed to PCBs and other contaminants present in carp from
oral 28-d LD50 of 4.2 ␮g/kg body weight for males. In female
mink, LD50 values of 264 and 47 ng/kg/d were estimated for
28- and 125-d exposure periods, respectively [15]. Daily in-
traperitoneal injections of 0.1
␮g TCDD/kg body weight in
Saginaw Bay. At a feed concentration of 500 ␮g PCB/kg (0.1
newborn kits for 12 consecutive d significantly reduced body
weights and produced over 50% mortality within 10 weeks
[37]. In our experiment, the exposure of the mink in groups
A50 high and A50 low gave 8.4 and 2.9 ng TEQ/d, respec-
tively, corresponding to 65 and 22 pg TEQ/g feed (fresh wt)
and a total exposure over the 18 months of approximately 4.6
mg PCB/mink/d, assuming a feed consumption of 200 g/d),
litters had greater mortality and lower body weights than con-
trols.
We found a pronounced reproductive toxicity without any
obvious toxic effects in the dams, confirming that impaired
reproduction is a very sensitive marker for PCB-induced tox-
icity in the mink. The lack of obvious toxic effects in the dams
might be explained by the relatively low levels of PCB found
and 1.6
␮g TEQ/kg, respectively.
The LOAEL in our study was 2.9 ng TEQ/animal/d (about
2.4 ng/kg body wt/d) or 22 pg TEQ/g feed (fresh wt) because
we noted effects on kit production and kit growth in group
A50 low. According to Tillitt et al. [38], the mink in the group
receiving the lowest dose in the study by Heaton et al. [8]
(10% carp in the feed) were exposed to 21.2 pg TEQ/g feed.
The daily feed consumption in that group was reported to be
218 g/mink, resulting in an exposure of 4.6 ng TEQ/animal/d.
Kit body weights and survival at three and six weeks of age
were significantly reduced in the animals exposed to 10% carp
in the feed. Thus, a similarity exists in the effects on repro-
duction between our study and that carried out by Heaton and
coworkers, even though the animals were exposed to PCB from
different sources (technical and weathered) and the exposure
times differed. We found no significant adverse effect on re-
production in the mink exposed to the 2–4-ortho-CB fraction.
The TEQ exposure in that group was 0.41 ng TEQ/animal/d
(about 0.35 ng/kg body wt/d) or 3.2 pg/g feed (fresh wt, TEQ
concentration in the basal feed not included), which represents
no-observed-adverse-effect levels in terms of reproductive ef-
fects in our study. However, both EROD and PROD activities
were induced in that group, and hence a biological effect ex-
isted also in those animals.
in the animals, 10 to 50 ␮g PCB/g in muscle (lipid wt). Using
experimental data on PCB levels and reproduction outcome in
female mink, Leonards et al. [32] estimated the median effect
level (EC50) to be 1.2
a lipid concentration of 2 to 3% in muscle, this value corre-
sponds to a concentration of 40 to 60 g PCB/g (lipid wt).
␮g total PCB/g wet weight. Assuming
␮
The data from groups A50 high and A50 low in our study
suggest a lower EC50 with regard to reproductive impairment
when animals are chronically exposed. The PCBs in European
otter (Lutra lutra) were studied by Smit et al. [33], who re-
ported that concentrations above 50
␮g/g lipid weight are
found in many European countries. For instance, Mason and
The BAFs differed for the CB mixtures given to the various
groups. The mink in group A50 high were exposed to a three-
fold higher PCB dose (0.3 mg/d) than the animals in group
A50 low (0.1 mg/d). The greater-than-threefold difference in
total PCB levels between females from these groups (54 vs
12 ␮g/g, BAFs: 2.1 and 1.3, respectively) most probably shows
that the animals in A50 high did not rid any PCB via the milk,
whereas the dams in A50 low had one to five kits that suckled
Fig. 6. Total polychlorinated biphenyl (PCB) concentrations (␮g/g,
lipid wt) in muscle from adult female mink exposed to various PCB
preparations for 18 months. Values shown are means and standard
deviations for seven animals in each group.

2326
Environ. Toxicol. Chem. 20, 2001
B. Brunstro¨m et al.
expert animal care and Agneta Bostro¨m, Britt-Marie Ba¨cklin, Erika
Gustafsson, Katarina Hjelm, Margareta Matsson, and Elisabeth Pers-
son for valuable assistance during the experiment. Ake Bergman is
gratefully acknowledged for advice and support concerning fraction-
ation and chemical analyses. Mats Olsson is thanked for providing
seal blubber and Lars Fo¨rlin for Clophen A50. This study was sup-
ported by the Swedish Environmental Protection Agency.
for five weeks. The animals treated with the 2–4-ortho-CB-
fraction were exposed to more total PCB than those given the
0–1-ortho-CBs or the low dose of A50. By the end of the
study, the residual levels of PCB were similar in these groups,
and the estimated BAFs were 2.3 (0–1-ortho-CBs), 1.3 (A50
low), and 0.7 (2–4-ortho-CBs). These differences in accu-
mulation probably reflect partly a greater persistence of the
0–1-ortho-CBs than of the 2–4-ortho-CBs in mink and were
partly a consequence of only a few kits suckling from the
females exposed to the 0–1-ortho-CBs. It has been shown
previously that many 2–4-ortho-CBs, notably those with un-
substituted 3,4 positions, are readily metabolized by mink [22].
The Baltic gray seal blubber contained low relative concen-
trations of both 0- and 1-ortho-CBs (compare Table 1), indi-
cating a high capacity of the gray seal to metabolize these
CBs. There might consequently be considerable differences
between mink and seals with respect to PCB metabolism. The
BAF for total CBs in the blubber extract was fairly high (1.9),
reflecting the high persistence of the 2–4-ortho-CBs that had
accumulated in the blubber.
˚
REFERENCES
1. Olsson M. 1986. PCBs in the Baltic environment. In Waid JS,
ed, PCBs and the Environment, Vol 3. CRC, Boca Raton, FL,
USA, pp 181–208.
2. Sandegren F, Olsson M, Reuterga˚rdh L. 1980. Der Ru¨ckgang der
Fischotterpopulation in Schweden. In Reuther C, Festetics A, eds,
Der Fischotter in Europa—Verbreitung, Bedrohung, Erhaltung.
Selbstverlag, Oderhaus and Go¨ttingen, Germany, pp 107–113.
3. Olsson M, Sandegren F. 1991. Is PCB partly responsible for the
decline of the otter in Europe? In Reuther C, Ro¨chert R, eds,
Proceedings, Fifth International Otter Colloquium, Hankensbu¨t-
tel, Germany, November 24–27, 1989. Habitat 6:223–227.
4. Olsson M, Sandegren F. 1991. Otter survival and toxic chemi-
cals—Implications for otter conservation programmes. In Reuther
C, Ro¨chert R, eds, Proceedings, Fifth International Otter Col-
loquium, Hankensbu¨ttel, Germany, November 24–27, 1989. Hab-
itat 6:191–200.
Induction of EROD proved to be a sensitive marker for
dioxin-like biological effects in the mink, and the non- and
mono-ortho-chlorinated CBs were potent inducers. The frac-
tion that consisted mainly of 2–4-ortho-CBs and the blubber
extract also caused significant induction, showing that these
fractions contained Ah receptor agonists in concentrations high
enough to cause a dioxin-like biological effect. The inverse
relationship between EROD activity and production of kits
shown in Figure 5 supports EROD as a biomarker predicting
toxic effects in the mink, and it also shows that reproductive
impairment and EROD are induced by fairly similar doses of
Ah receptor agonists. In mink exposed to PCB-contaminated
carp, a dose-dependent EROD induction was noted, and EROD
was suggested as a potential biomarker for PCB exposure in
mink [39]. Another marker for dioxin-like effects is a reduced
thymus weight. We noted such a reduction in newborn kits of
mothers exposed to the two highest TEQ concentrations in the
feed (groups A50 high and 0–1-ortho) for six months, but the
lack of effect in the five-week-old kits in the second season
suggests that thymus weight is not a useful biomarker in ju-
venile mink. The PROD activity was strongly induced in the
group treated with the 2–4-ortho-CBs but not in the group
exposed to the 0–1-ortho-CBs. This is in accordance with
PROD as being induced by noncoplanar PCB congeners, and
induction of this enzyme may consequently be used as an
indicator of PCB exposure.
5. Erlinge S. 1969. Food habits of the otter and mink in a trout
water in southern Sweden. Oikos 20:1–7.
6. Aulerich RJ, Ringer RK. 1977. Current status of PCB toxicity to
mink, and effect on their reproduction. Arch Environ Contam
Toxicol 6:279–292.
7. Hornshaw TC, Aulerich RJ, Johnson HE. 1983. Feeding Great
Lakes fish to mink: Effects on mink and accumulation and elim-
ination of PCBs by mink. J Toxicol Environ Health 11:933–946.
8. Heaton SN, et al. 1995. Dietary exposure of mink to carp from
Saginaw Bay, Michigan. 1. Effects on reproduction and survival,
and the potential risks to wild mink populations. Arch Environ
Contam Toxicol 28:334–343.
9. Giesy JP, et al. 1994. Contaminants in fishes from Great Lakes-
influenced sections and above dams of three Michigan rivers. II:
Implications for health of mink. Arch Environ Contam Toxicol
27:213–223.
10. Platonow NS, Karstad LH. 1973. Dietary effects of polychlori-
nated biphenyls on mink. Can J Comp Med 37:391–400.
¨
11. Jensen S, Kihlstro¨m JE, Olsson M, Lundberg C, Orberg J. 1977.
Effects of PCB and DDT on mink (Mustela vison) during the
reproductive season. Ambio 6:239.
12. Bleavins MR, Aulerich RJ, Ringer RK. 1980. Polychlorinated
biphenyls (Aroclors 1016 and 1242): Effects on survival and
reproduction in mink and ferrets. Arch Environ Contam Toxicol
9:627–635.
˚
13. Kihlstro¨m JE, Olsson M, Jensen S, Johansson A, Ahlbom J, Berg-
˚
man A. 1992. Effects of PCB and different fractions of PCB on
the reproduction of the mink (Mustela vison). Ambio 21:563–
569.
14. Restum JC, Bursian SJ, Giesy JP, Render JA, Helferich WG,
Shipp EB, Verbrugge DA, Aulerich RJ. 1998. Multigenerational
study of the effects of consumption of PCB-contaminated carp
from Saginaw Bay, Lake Huron, on mink. 1. Effects on mink
reproduction, kit growth and survival, and selected biological
parameters. J Toxicol Environ Health A 54:343–375.
15. Hochstein JR, Bursian SJ, Aulerich RJ. 1998. Effects of dietary
Several conclusions may be drawn from this study. Long-
term exposure of mink to PCB impairs reproduction when PCB
residue levels in the mothers are considerably below 50 ␮g/g,
a value often referred to as a threshold for reproductive toxicity
in mink. The PCB congeners that are Ah receptor agonists
were responsible for the reproductive dysfunction of the ani-
mals in this study. Furthermore, kit survival and kit growth
were sensitive endpoints for toxicity caused by the Ah receptor
agonists. Induction of EROD in adult female mink appears to
be a good marker for reproductive effects, as reproductive
performance decreased concomitant with an increase in en-
zyme activity. In the otter, coplanar PCB congeners bioaccu-
mulate to a high degree [40], and exposure to these compounds
is probably one important factor behind declines of otter pop-
ulations.
exposure to 2,3,7,8-tetrachlorodibenzo-p-dioxin in adult female
mink (Mustela vison). Arch Environ Contam Toxicol 35:348–
353.
16. Aulerich RJ, Bursian SJ, Breslin WJ, Olson BA, Ringer RK. 1985.
Toxicological manifestations of 2,4,5,2
Ј,4Ј,5Ј-, 2,3,6,2Ј,3Ј,6Ј-,
and 3,4,5,3 ,4 ,5 -hexachlorobiphenyl and Aroclor 1254 in mink.
Ј
Ј Ј
J Toxicol Environ Health 15:63–79.
17. Ballschmiter K, Mennel A, Buyten J. 1993. Long chain alkyl-
polysiloxanes as non-polar stationary phases in capillary gas chro-
matography. Fresenius Z Anal Chem 346:396–402.
18. Sundstro¨m G. 1973. Polychlorinated biphenyls II. Synthesis of
some tetra- and pentachlorobiphenyls. Acta Chem Scand 27:600–
604.
19. Smith BV, Waldron NM. 1980. Vogels Elementary Practical Or-
ganic Chemistry 1: Preparations, 3rd ed. Longman, New York,
NY, USA.
Acknowledgement—We thank mink rancher Gottmar Mattsson for

Reproductive toxicity in chronically PCB-exposed mink
Environ. Toxicol. Chem. 20, 2001
2327
20. Athanasiadou M, Jensen S, Klasson Wehler E. 1991. Preparative
fractionation of a commercial PCB product. Chemosphere 23:
957–970.
lund-Glynn A. 1996. CB 153 as indicator for congener specific
determination of PCBs in diverse fish species from Swedish wa-
ters. Chemosphere 33:1459–1464.
˚
21. Jensen S, Athanasiadou M, Bergman A. 1992. A technique for
32. Leonards PEG, de Vries TH, Minnaard W, Stuijfzand S, de Voogt
P, Cofino WP, van Straalen NM, van Hattum B. 1995. Assessment
of experimental data on PCB-induced reproduction inhibition in
mink, based on an isomer- and congener-specific approach using
separation of xenobiotics from large amounts of lipids. Orga-
nohalogen Compounds 8:79–80.
˚
22. Bergman A, Athanasiadou M, Bergek S, Haraguchi K, Jensen S,
Klasson Wehler E. 1992. PCB and PCB methyl sulfones in mink
treated with PCB and various PCB fractions. Ambio 21:570–576.
23. Brunstro¨m B, Ha˚kansson H, Lundberg K. 1991. Effects of a tech-
nical PCB preparation and fractions thereof on ethoxyresorufin
2,3,7,8-tetrachlorodibenzo-
Toxicol Chem 14:639–652.
p-dioxin toxic equivalency. Environ
33. Smit MD, Leonards PEG, van Hattum B, de Jongh AWJJ. 1994.
PCBs in European otter (Lutra lutra) populations. Report R-94/
7. Institute for Environmental Studies, Amsterdam, The Nether-
lands.
34. Mason CF, Madsen AB. 1993. Organochlorine pesticide residues
and PCBs in Danish otters (Lutra lutra). Sci Total Environ 133:
73–81.
O
-deethylase activity, vitamin A levels and thymic development
in the mink (Mustela vison). Pharmacol Toxicol 69:421–426.
24. Pohl RJ, Fouts JR. 1980. A rapid method for assaying the me-
tabolism of 7-ethoxyresorufin by microsomal subcellular frac-
tions. Anal Biochem 107:150–155.
25. Brunstro¨m B, Andersson L. 1988. Toxicity and 7-ethoxyresorufin
35. Roos A, Greyerz E, Olsson M, Sandegren F. 2001. The otter
O
-deethylase-inducing potency of coplanar polychlorinated bi-
(Lutra lutra) in Sweden—Population trends in relation to ⌺DDT
and total PCB concentrations during 1968–99. Environ Pollut
111:457–469.
phenyls (PCBs) in chick embryos. Arch Toxicol 62:263–266.
26. Lubet RA, Mayer RT, Cameron JW, Nims RW, Burke MD, Wolff
T, Guengerich FP. 1985. Dealkylation of pentoxyresorufin: A rap-
id and sensitive assay for measuring induction of cytochrome(s)
P-450 by phenobarbital and other xenobiotics in the rat. Arch
Biochem Biophys 238:43–48.
27. Lowry OH, Rosebrough NJ, Farr AL, Randall RJ. 1951. Protein
measurement with the Folin phenol reagent. J Biol Chem 193:
265–275.
36. Hochstein JR, Aulerich RJ, Bursian SJ. 1988. Acute toxicity of
2,3,7,8-tetrachlorodibenzo-p-dioxin to mink. Arch Environ Con-
tam Toxicol 17:33–37.
37. Aulerich RJ, Bursian SJ, Napolitano AC. 1988. Biological effects
of epidermal growth factor and 2,3,7,8-tetrachlorodibenzo- -di-
p
oxin on developmental parameters of neonatal mink. Arch En-
viron Contam Toxicol 17:27–31.
28. Van den Berg M, et al. 1998. Toxic equivalency factors (TEFs)
for PCBs, PCDDs, PCDFs for humans and wildlife. Environ
Health Perspect 106:775–792.
29. Jensen S, Reuterga˚rdh L, Jansson B. 1983. Analytical methods
for measuring organochlorines and methyl mercury by gas chro-
matography. FAO Fisheries Technical Paper 212. Food and Ag-
riculture Organization of the United Nations, Rome, Italy, pp 21–
33.
38. Tillitt DE, et al. 1996. Dietary exposure of mink to carp from
Saginaw Bay. 3. Characterization of dietary exposure to planar
halogenated hydrocarbons, dioxin equivalents, and biomagnifi-
cation. Environ Sci Technol 30:283–291.
39. Shipp EB, Restum JC, Giesy JP, Bursian SJ, Aulerich RJ, Hel-
ferich WG. 1998. Multigenerational study of the effects of con-
sumption of PCB-contaminated carp from Saginaw Bay, Lake
Huron, on mink. 2. Liver PCB concentration and induction of
hepatic cytochrome P-450 activity as a potential biomarker for
PCB exposure. J Toxicol Environ Health A 54:377–401.
40. Leonards PEG, Zierikzee Y, Brinkman UAT, Cofino WP, van
Straalen NM, van Hattum B. 1997. The selective dietary accu-
mulation of planar polychlorinated biphenyls in the otter (Lutra
lutra). Environ Toxicol Chem 16:1807–1815.
30. Schultz DE, Petrick G, Duinker JC. 1989. Complete character-
ization of polychlorinated biphenyl congeners in commercial Ar-
oclor and Clophen mixtures by multidimensional gas chromatog-
raphy-electron capture detection. Environ Sci Technol 23:852–
859.
¨
31. Atuma SS, Linder C-E, Andersson O, Bergh A, Hansson L, Wick-