Dose-dependent enantiomeric enrichment of 2,2′,3,3′,6,6′- hexachlorobiphenyl in female mice

Article abstract of DOI:10.1897/07-359R.1

Nineteen of the 209 possible polychlorinated biphenyl (PCB) congeners are chiral and stable to racemization at ambient temperature. Chiral PCB congeners are important components of technical and environmental PCB mixtures, and some are highly toxic. Both environmental and laboratory studies have shown that these chiral PCB congeners undergo enantiomeric enrichment in many species; however, the processes and factors influencing the extent of this enantiomeric enrichment are poorly understood. We hypothesized that the exposure levels are an important factor affecting the extent of enantiomeric enrichment. To test this hypothesis, we investigated the levels and enantiomeric fractions of (±)-PCB 136 in selected tissues, feces, and urine of female C57B1/6 mice 3 d after oral administration of 2.5, 10, or 50 mg/kg body weight of (±)-PCB 136. The PCB 136 tissue levels typically increased with increasing dose. The extent of the enrichment of (+)-PCB 136 in tissues and feces, however, decreased with increasing dose, an observation that suggests a saturation of the disposition process responsible for the enantiomeric enrichment. Overall, the present study demonstrates that in addition to species, exposure source, exposure frequency, and other factors, levels of PCB exposure are an important determinant of the enantiomeric enrichment of PCBs in mice and, most likely, other species.

Full text of DOI:10.1897/07-359R.1

Environmental Toxicology and Chemistry, Vol. 27, No. 2, pp. 299–305, 2008
2008 SETAC
Printed in the USA
730-7268/08 $12.00 .00
᭧
0
ϩ
DOSE-DEPENDENT ENANTIOMERIC ENRICHMENT OF
,2 ,3,3 ,6,6 -HEXACHLOROBIPHENYL IN FEMALE MICE
2
Ј
Ј
Ј
I
ZABELA
Department of Occupational and Environmental Health, University of Iowa, College of Public Health, Iowa City, Iowa 52242, USA
Department of Environmental Chemistry and Technology, University of Silesia, Szkolna 9, 40-006 Katowice, Poland
Department of Civil and Environmental Engineering, University of Iowa, Iowa City, Iowa 52242, USA
KANIA-KORWEL,†‡ KERI C. HORNBUCKLE,§ LARRY W. ROBERTSON,† and HANS-JOACHIM LEHMLER*†
†
‡
§
(
Received 12 June 2007; Accepted 7 August 2007)
Abstract—Nineteen of the 209 possible polychlorinated biphenyl (PCB) congeners are chiral and stable to racemization at ambient
temperature. Chiral PCB congeners are important components of technical and environmental PCB mixtures, and some are highly
toxic. Both environmental and laboratory studies have shown that these chiral PCB congeners undergo enantiomeric enrichment
in many species; however, the processes and factors influencing the extent of this enantiomeric enrichment are poorly understood.
We hypothesized that the exposure levels are an important factor affecting the extent of enantiomeric enrichment. To test this
hypothesis, we investigated the levels and enantiomeric fractions of (
C57Bl/6 mice 3 d after oral administration of 2.5, 10, or 50 mg/kg body weight of (
typically increased with increasing dose. The extent of the enrichment of ( )-PCB 136 in tissues and feces, however, decreased
Ϯ
)-PCB 136 in selected tissues, feces, and urine of female
Ϯ
)-PCB 136. The PCB 136 tissue levels
ϩ
with increasing dose, an observation that suggests a saturation of the disposition process responsible for the enantiomeric enrichment.
Overall, the present study demonstrates that in addition to species, exposure source, exposure frequency, and other factors, levels
of PCB exposure are an important determinant of the enantiomeric enrichment of PCBs in mice and, most likely, other species.
Keywords—Polychlorinated biphenyls
ortho-Substituted polychlorinated biphenyls
Atropisomers
Enantiomeric fraction
Excretion
INTRODUCTION
one PCB atropisomer [6–10]. The results from these studies
suggest that this enantiomeric enrichment is caused by enan-
tioselective biotransformation processes.
Polychlorinated biphenyls (PCBs) are a group of 209 bi-
phenyl derivatives (or congeners) that differ in the number and
position of the chlorine substituents on the biphenyl nucleus
The above-mentioned studies also show significant differ-
ences in the extent of the enantiomeric enrichment between
studies. For example, in laboratory studies, the ratios of the
two PCB atropisomers typically range from 2:1 to 4:1, with a
more significant enantiomeric enrichment observed in the stud-
ies that employed a lower dose [6–12]. A comparable enan-
tiomeric enrichment, with atropisomer ratios of 2:1 to 4.5:1,
has been reported for a limited number of human samples [4,5].
Interestingly, environmental samples can show a much more
pronounced enantiomeric enrichment, with atropisomer ratios
as high as 6:1 to 9:1 in some species [13–20], whereas no
enantiomeric enrichment is noted in other species
[
1]. Polychlorinated biphenyl congeners with an asymmetrical
substitution pattern in both phenyl rings (relative to the bond
between the two phenyl rings) are, in principle, axially chiral;
however, rotational isomers (or atropisomers) that are stable
under ambient conditions are formed only in the presence of
three or four chlorine substituents in the ortho-position. Sev-
enteen of the 19 possible chiral PCB congeners can be found
in considerable amounts in commercial PCB mixtures [1]. For
example, Aroclor௡ (Monsanto, St. Louis, MO, USA) contains
from 6% (Aroclors 1242 and 1248) to 30% (Aroclor 1260) by
weight of chiral PCB congeners (calculated based on the work
of Frame [2]), which makes them important constituents of
technical and, ultimately, environmental PCB mixtures.
It is well established that chiral molecules can enantiose-
lectively interact with biomolecules, such as proteins. Because
of these enantioselective interactions, pure enantiomers not
only have different biological effects but also are subject to
enantioselective absorption, distribution, biotransformation,
and elimination processes [3]. Not surprisingly, both environ-
mental and laboratory studies have shown that chiral PCBs,
like other chiral xenobiotics, are subject to such enantioselec-
tive disposition processes. For example, an enantiomeric en-
richment of several chiral PCB congeners has been reported
in wildlife [1] and in humans [4,5]. Similarly, laboratory stud-
ies have investigated the enantioselective disposition of in-
dividual PCB congeners and found a significant enrichment of
[
13,14,16,21,22]. Currently, it is difficult to explain these dif-
ferences in the enantiomeric enrichment of PCBs in different
species. This results not only from the limited data that are
available but also from our incomplete understanding about
the complex processes resulting in an enantiomeric enrichment
in various species.
All active biological processes are dose dependent and can,
for example, be saturated at high concentrations of a substrate.
Because laboratory studies that use lower PCB doses show a
more pronounced enantiomeric enrichment [6–12], we hypoth-
esize that in addition to the above-mentioned factors, the PCB
dose (in laboratory studies) or the biota PCB concentration (in
environmental studies) is an important determinant of the degree
of enantiomeric enrichment. In the present study, we investigated
the disposition of PCB 136 atropisomers in female mice using a
high (139
␮mol/kg body wt), a medium (28
␮mol/kg body wt),
and a low (6.9
␮
mol/kg body wt) dose of PCB 136 at 3 d after
*
To whom correspondence may be addressed
PCB administration. The high dose was used in a previous study
from our laboratory [9] and represents an environmentally rele-
(
hans-joachim-lehmler@uiowa.edu).
Published on the Web 10/08/2007.
2
99

3
00
Environ. Toxicol. Chem. 27, 2008
I. Kania-Korwel et al.
Fig. 1. Dose-dependent changes in polychlorinated biphenyl (PCB) congener 136 (2,2
Ј
,3,3
Ј
,6,6
Ј
-hexachlorobiphenyl) levels in mouse tissues. A
single asterisk denotes a significant (analysis of variance, Bonfferoni multiple-comparison test, ␣ ϭ 0.05,
dose group; two asterisks denotes a significant (analysis of variance, Bonfferoni multiple-comparison test, ␣ ϭ 0.05,
the medium-dose group. ⅷ ϭ average of six tissue samples, with the standard deviation represented by the error bar; ⅜ ϭ pooled samples or
samples with PCB levels below the detection limit; – – – samples with PCB 136 levels below the detection limit.
p
ϭ
0.05) difference from the low-
p
ϭ
0.05) difference from
ϭ
vant dose [23,24]. The low dose was targeted to a level that
would allow detection of PCB 136 at least in storage tissues,
such as the liver and adipose tissue.
entire cookie sample was eaten. Typically, the entire cookie was
consumed within 30 min.
The animals were than transferred to metabolic cages (treat-
ment groups,
n ϭ 3 animals/cage; control group, n ϭ 4 ani-
MATERIAL AND METHODS
mals/cage) and given ad libitum access to water and chow.
Feces and urine samples were collected daily from each cage
Chemicals
and stored in glass vials at
Ϫ20ЊC before PCB extraction and
2,2Ј,3,3Ј,6,6Ј-Hexachlorobiphenyl (PCB 136) was synthe-
sized from 2,3,6-trichloro-1-iodobenzene using the Ullman
coupling reaction [25] and purified by column chromatography
analysis. Animals were killed 3 d after PCB administration by
carbon dioxide asphyxiation followed by cervical dislocation.
Tissues (abdominal adipose tissue, brain, heart, intestine, kid-
ney, liver, lung, tight muscle, skin, spleen, and uterus) were
excised en bloc, and wet weight was determined. All tissue
on silica gel (eluent
ϭ n-hexanes) followed by charcoal treat-
ment to remove dibenzofuran impurities [9]. The purity was
confirmed by gas chromatography as being greater than 99%
based on relative peak area.
samples were stored in aluminum foil at
extraction and analysis.
Ϫ20ЊC before PCB
Animal treatment
PCB 136 and lipid extraction from tissues and feces
All animal procedures were approved by the Institutional An-
imal Care and Use Committee of the University of Iowa. Twenty-
two female C57Bl/6 mice (age, eight weeks; weight, 17.7
g) were obtained from Harlan (Indianapolis, IN, USA) and ini-
tially housed individually. The animals were allowed to accli-
matize for a week, followed by random assignment to one control
and three treatment groups (Fig. 1). The treatment groups received
A pressurized solvent extraction system (ASE 200; Dionex,
Sunnyvale, CA, USA) was used for a combined PCB extraction
and cleanup step, as well as for lipid extraction, as reported
by Kania-Korwel et al. [9]. The lipid content was determined
gravimetrically in adipose tissue, feces, kidney, and liver. The
Ϯ 1.0
nonvolatile residue in solvent blanks was 0.42
7) and was significantly lower than the amount of lipid in
tissues (67 32 mg in adipose tissue, 34 10 mg in feces,
1 mg in kidney, and 23 6 mg in liver).
Ϯ 0.31 mg (n
ϭ
a low (2.5 mg/kg body wt; 6.9
mg/kg body wt; 28 mol/kg body wt), or high (50 mg/kg body
wt; 139 mol/kg body wt) oral dose of PCB 136 on a Vanilla
Wafer cookie (7.5 g/kg body wt; Kebler, Battle Creek, MI, USA).
␮mol/kg body wt), medium (10
Ϯ
Ϯ
␮
9
Ϯ
Ϯ
␮
௡
Gas chromatography
The control animals received the vehicle (i.e., cookie) alone. A
detailed description of the vehicle preparation was reported pre-
viously by Kania-Korwel et al. [9]. Animals were monitored
closely after receiving the PCB-treated cookie to ensure that the
Tissue levels and enantiomeric fractions (EFs) of PCB 136
were determined using a 6890N gas chromatograph (Agilent
Technologies, Santa Clara, CA, USA) equipped with a Ni ␮-
6
3

Dose-dependent enantiomeric enrichment of PCB 136
Table 1. Tissue levels of 2,2 ,3,3 ,6,6
Environ. Toxicol. Chem. 27, 2008
301
Ј
Ј
Ј
-hexachlorobiphenyl (PCB 136) expressed as a percentage of total dosea
Tissue
Low (2.5 mg/kg)
Medium (10.0 mg/kg)
High (50.0 mg/kg)
Adipose tissue
Brain
Heart
Intestine
Kidney
Liver
Lung
Muscles
Skin
Spleen
Uterus
3.1
0.005
0.005
0.11
0.02
0.02
0.03
5.6
Ϯ
Ϯ
Ϯ
Ϯ
Ϯ
Ϯ
Ϯ
Ϯ
Ϯ
Ϯ
Ϯ
Ϯ
0.6
6.5
0.007
0.02
0.19
0.03
0.04
0.10
2.8
Ϯ
Ϯ
Ϯ
Ϯ
Ϯ
Ϯ
Ϯ
Ϯ
Ϯ
Ϯ
Ϯ
Ϯ
1.0b
0.001
0.01
11.5
0.008
0.01
0.29
0.05
0.07
0.09
3.3
Ϯ
Ϯ
Ϯ
Ϯ
Ϯ
Ϯ
Ϯ
Ϯ
Ϯ
Ϯ
Ϯ
Ϯ
0.7bc
bc
0.001b
0.001
0.04
0.01
0.01
0.01
0.1
b
0.001
b
c
0.01
0.13
0.01
0.01
0.01
0.24
0.02
b
bc
0.02
0.07
b
bc
b
0.1
0.8
5.4
2.8
0.1
0.001
0.02
5.9
1.1
0.01
0.01
2
8.7
b
0.005
0.04
12
0.01
0.02
15
0.01
0.03
23
0.01
0.03
bc
Sum % total dose
1
4
d
%
total dose in feces
1.2 (1.4, 0.9)
2.0 (3.5, 0.5)
0.4 (0.7, 0.2)
a
The percentage of body weight was assumed to be 5.85% for blood (L. Hansen, Department of Veterinary Biosciences, University of Illinois,
Urbana-Champaign, IL, USA, personal communication), 5.9% for adipose tissue, 38.4% for muscles, and 16.5% for skin [9]. The weights of
all other organs used to adjust the PCB levels were wet weights determined in the present study.
b
c
d
Significantly different from the 2.5 mg/kg treatment group (analysis of variance, Bonferroni multiple-comparison test, ␣ ϭ 0.05,
Significantly different from the 10 mg/kg treatment group (analysis of variance, Bonferroni multiple-comparison test, ␣ ϭ 0.05,
Numbers in parentheses represent the percentage of the total dose in feces for the individual metabolism cages.
p
p
ϭ
0.05).
ϭ 0.05).
electron-capture detector (Agilent Technologies) and a Chir-
asil-Dex column (length, 25 m; internal diameter, 0.25 mm;
film thickness, 0.25 ␮m; ␤-cyclodextrin chemically bonded to
dimethylpolysiloxane; Varian, Palo Alto, CA, USA) [9]. The
EF was calculated using the following formula [26]:
Differences in body and organ weights, PCB 136 levels
(expressed as wet wt, total dose, and lipid adjusted), and EFs
between treatment groups and control were analyzed with anal-
ysis of variance using the Bonfferoni multiple-comparison test.
Analysis of variance with the Tukey multiple-comparison test
was employed to determine significant differences between
EF
ϭ
Area(ϩ)-PCB136/(Area(ϩ)-PCB136
ϩ
Area(Ϫ)-PCB136)
(1)
tissues within the same treatment group. One-sample
used to test the significant differences between tissue EFs and
racemic standard (EF 0.5). For all statistical comparisons,
0.05 was considered to be significant.
t test was
The elution order of the (
ϩ
)- and ( )-atropisomer was deter-
Ϫ
mined based on the published elution order of PCB 136 atro-
pisomers on the same column [27]. The racemic PCB 136 had
ϭ
p
Յ
an EF of 0.50
resolution (R) of PCB 136 enantiomers aws calculated as
Ϯ 0.01 (average of 58 measurements). The
RESULTS
Effects of PCB 136 on body and tissue weights
R
ϭ
(RT1 RT )/0.5(BW1 BW2)
Ϫ
ϩ
(2)
2
Transferring the animals into metabolic cages immediately
after PCB 136 administration resulted in weight loss over the
where RT and RT are the retention times of the first- and second-
1
2
eluting atropisomer peaks, respectively, and BW and BW are
1
2
3
1
-d study period (see Tables S1–S4 [http://dx.doi.org/10.
897/07-359.S1]). The PCB 136 treatment itself, however, did
the respective base widths. The resolution was 0.64
Ϯ 0.1.
not appear to have an effect on body and tissue weights com-
pared to those of control animals (Table 1). One notable ex-
ception was liver weight, which was significantly higher in the
high-dose group.
Quality assurance
The limit of detection of PCB 136 was calculated based on
the method blanks [9] and was equal to 6.3 ng. The limit of
quantification, calculated as 10-fold the limit of detection, was
equal to 63 ng. 2,2
04) was used as internal standard for the quantification of
PCB 136. The recovery rate of the surrogate standard
,3,4,4 ,5,6-hexachlorobiphenyl (PCB 166) was 92% 14%
399). Levels of PCB 136 were corrected if the recovery
Ј
,3,4,4
Ј
,5,6,6
Ј
-Octachlorobiphenyl (PCB
Levels of PCB 136 in tissues
2
Independent of how the PCB 136 tissue levels were ex-
pressed (i.e., based on wet wt, percentage of total dose, or for
some tissue, lipid content), PCB levels decreased as the dose
decreased (Tables 1 and 2 and Fig. 1). For example, the per-
2
(n ϭ
Ј
Ϯ
rate was less than 100%.
Statistical analysis
Table 2. Dose-dependent lipid adjusted levels (␮g/g lipid) of
2
,2
Ј
,3,3
Ј
,6,6
Ј
-hexachlorobiphenyl (PCB 136) in adipose tissue,
kidney, and liver
All statistical analyses were performed using SAS௡ soft-
ware (Ver 9.1; SAS Institute, Cary, NC, USA). Data are pre-
sented as the mean standard deviation with exception of
Ϯ
Low
(2.5 mg/kg)
Medium
(10 mg/kg)
High
(50 mg/kg)
Tissue
several samples that were pooled for analysis because of low
PCB tissue levels (i.e., brain, heart, lungs, muscles, and spleen
in the medium-dose group and lungs, muscles, and skin in the
low-dose group). In a few instances, when the PCB 136 levels
in the pooled sample were less than the detection limit (i.e.,
brain, heart, and spleen in the low-dose group), the method
detection limit value was used for statistical analysis. Data
points recognized by the SAS software as being outliers were
removed before the statistical analysis.
3.0a
0.9
0.3
134
21
13
Ϯ
Ϯ
Ϯ
9
6
4
ab
Adipose tissue
Kidney
Liver
2.0
0.4
0.3
Ϯ
Ϯ
Ϯ
0.4
0.1
0.1
15.0
2.5
1.3
Ϯ
Ϯ
Ϯ
ab
ab
a
Significantly different from the 2.5 mg/kg treatment group (analysis
of variance, Bonfferoni multiple-comparison test, ␣ ϭ 0.05,
.05).
p
ϭ
0
b
Different from the 10 mg/kg treatment group (analysis of variance,
Bonferroni multiple-comparison test, ␣ ϭ 0.05, 0.05).
p
ϭ

3
02
Environ. Toxicol. Chem. 27, 2008
I. Kania-Korwel et al.
Fig. 2. Dose-dependent changes in the enantiomeric fractions in mouse tissues. The brain, heart, and spleen in the 2.5 mg/kg group were below
the detection limit. The dotted lines represent the enantiomeric fraction of the racemic polychlorinated biphenyl (PCB) congener 136 (2,2 ,3,3 ,6,6
hexachlorobiphenyl) that was administered to the animals. A single asterisk denotes a significant (analysis of variance, Bonfferoni multiple-
comparison test, ␣ ϭ 0.05, 0.05) difference from the low-dose group; two asterisks denotes a significant (analysis of variance, Bonfferoni
multiple-comparison test, ␣ ϭ 0.05, 0.05) difference from the medium-dose group. ⅷ ϭ average of six tissue samples; ⅜ ϭ pooled samples.
Ј
Ј
Ј-
p
ϭ
p
ϭ
centage of the total dose remaining in tissues decreased in the
order of 23% 15% 12% for the high-, medium-, and low-
tively) were significantly different from those in both the me-
dium-dose group (0.07 0.02 and 0.14 0.02, respectively)
0.01 and 0.16 0.03, re-
Ͼ
Ͼ
Ϯ
Ϯ
dose groups, respectively, although the latter two groups were
not significantly different from each other (Tables 1 and 2 and
Fig. 1). Levels of PCB 136 in the medium-dose group usually
were higher in all tissues compared to those in the low-dose
group. The only exception was the muscle, which contained
a significantly higher percentage of the total dose in the low-
dose group compared to that in the medium-dose group (5.6%
vs 2.8% of the total dose) (Table 1).
and the high-dose groups (0.05
spectively).
Ϯ
Ϯ
EF of PCB 136
The enrichment of (
from all treatment groups (Fig. 2). In the high-dose group, the
highest enrichment was observed in the liver (0.71 0.01)
and the brain (0.68 0.02). The liver and brain EFs were
ϩ)-PCB 136 was observed in all tissues
Ϯ
Ϯ
The concentrations of PCB 136 in the tissues were dissim-
ilar across dosage groups and across tissues even when nor-
malized to the lipid content in the tissue. For example, in the
high-dose group, PCB 136 levels were significantly higher in
the adipose tissue and the skin compared to all other tissues.
In the lower-dose groups, the PCB 136 levels in the adipose
tissue and the skin tended to be higher than in other tissues,
but this difference did not reach statistical significance for most
tissues. The lowest PCB 136 levels were found in brain and
heart, followed by kidney, liver, lung, and uterus (Table 1 and
Fig. 1).
significantly higher than those in most tissues from this treat-
ment group. Similarly, the most pronounced enantiomeric en-
richment was observed in the liver of the low- and the medium-
dose groups (0.80
Ϯ 0.02 and 0.77 Ϯ 0.01, respectively). These
EFs were only significantly different for a few tissues, how-
ever, such as the intestine and kidney.
Comparison between groups showed an increase in the EFs
with decreasing dose in the adipose tissue, intestine, liver, lung,
spleen, and uterus (Fig. 2). The extent of the enantiomeric
enrichment tended to increase from the high- to the medium-
dose group in the kidney, muscle, and skin but remained sim-
ilar or even decreased from the medium- to the low-dose group.
Furthermore, our data suggest that the EFs decrease or remain
similar from the high- to the medium-dose group in the brain
and the heart.
The concentrations of PCB 136 in the liver tissues were
similar across the dosage groups when normalized to the ad-
ipose tissue concentration. Specifically, the relative liver to
adipose tissue constants were 0.007
Ϯ 0.002, 0.010 Ϯ 0.003,
and 0.011 0.003 for the high-, medium-, and low-dose
Ϯ
Levels of PCB 136 in feces
groups, respectively, and were similar among treatment groups
regardless of the dose administered. This relationship has been
observed in other laboratory studies [9,28,29]. In contrast, the
relative muscle to adipose and liver to muscle tissue ratios in
Overall, fecal excretion was a minor route of excretion,
with less than 2% of the total dose of PCB 136 being excreted
independent of the dose (Table 1). The excretion profile in
feces in the high- and medium-dose groups were comparable
the lowest-dose group (0.20
Ϯ 0.05 and 0.04 Ϯ 0.01, respec-

Dose-dependent enantiomeric enrichment of PCB 136
Environ. Toxicol. Chem. 27, 2008
303
high- and medium-dose groups and was comparable between
both groups independent of the dose. The EF of PCB 136 in
feces in the low-dose group was higher than that in the other
two groups on day 1. The EFs for the other days 2 and 3 could
not be determined, because as mentioned earlier, PCB 136
levels fell below the detection limit on those days.
DISCUSSION
Comparison to earlier studies
Overall, the tissue PCB 136 levels and the EFs in the high-
dose group of the present study are comparable to the results
of an earlier study by Kania-Korwel et al. [9]. Specifically,
the adipose tissue had the highest PCB 136 levels (based both
on tissue wet wt and lipid content) and contained a compar-
atively high percentage of the total dose in both studies (Table
1
). In both studies, the lowest PCB 136 levels were observed
in the brain, spleen, and uterus, which is in agreement with
toxicokinetic studies of this congener in rats [30]. As indicated
by the relative liver to adipose tissue constant (0.011 in the
present study and 0.005 in the study by Kania-Korwel et al.
[
9]), PCB 136 was enriched in the adipose tissue compared to
the liver, the major site of PCB biotransformation.
Furthermore, both studies showed a significant enrichment
of (ϩ)-PCB 136 compared to racemic PCB 136 in all tissues.
The highest enantiomeric enrichment was observed in the liver,
with EFs of 0.71 (present study) and 0.75 [9]. The brain, which
is a target organ for neurotoxic, chiral PCB congeners [31,32],
also displayed a more pronounced enrichment of (ϩ)-PCB 136
compared to other tissues. Similarly, the brain of female mice
also showed a pronounced enrichment of one PCB 84 atro-
pisomer, (ϩ)-PCB 84, in a similar study by Lehmler et al. [10].
Two other disposition studies investigated the enantiomeric
enrichment of PCB 132 [7] or PCB 139 [6] in the liver but
not the brain of rats; however, it cannot be excluded that one
PCB atropisomer also was enriched in the brain in these stud-
ies.
In contrast, the EFs of chiral PCBs from human brain sam-
ples were racemic for PCB 95 and near racemic for PCBs 132
and 149 [4]. Rats treated with complex PCB mixtures also did
not display a significant enantiomeric enrichment of any chiral
PCB congener in the brain, although, for example, PCBs 95
and 147 were enriched in tissues such as the liver and the
adipose tissue [8]. These differences in the extent of the en-
antiomeric enrichment between these studies are startling but,
considering the neurotoxicity of these PCB congeners, im-
portant observations.
Fig. 3. Dose- and time-dependent fecal levels (A) and enantiomeric
fractions (EFs) (B) of polychlorinated biphenyl (PCB) congener 136
(
2,2
samples from two metabolism cages with three animals each). The
method detection limit for PCB 136 ( ) is shown for fecal samples
Ј,3,3Ј,6,6Ј-hexachlorobiphenyl) in female mice (average of fecal
⅜
below the detection limit. The dotted line (– – –) in B represents the
EF of the racemic PCB 136 administered to the animals. ⅷ ϭ 2.5
mg/kg; ᭢ ϭ 10 mg/kg; Ⅵ ϭ 50 mg/kg.
Levels of PCB 136 in feces
It is well established that PCBs have an excellent oral bio-
availability [33]. Indeed, the present study shows that ap-
proximately 0.4 to 2.0% of the total PCB 136 dose was ex-
creted with the feces within 3 d independent of the dose ad-
ministered. The comparatively high fecal PCB concentrations
and the near-racemic EFs at the 24-h time point suggest that
the majority of the PCB 136 excreted by this route was caused
by nonabsorbed and, therefore, racemic PCB 136. Although
the percentage of the total PCB 136 dose excreted with the
feces was comparable between groups, the PCB 136 levels in
feces showed clear dose dependence. The fecal PCB 136 levels
in the medium- and high-dose groups showed an identical time
course, but fecal PCB 136 levels in the lower-dose group were
lower compared to those in the other two groups, especially
within the first 24 h after exposure.
(
Fig. 3A), with a similar percentage of total remaining dose
excreted over the entire study period. The highest levels of
PCB 136 were excreted within the first 24 h after administra-
tion. Fecal PCB 136 levels dropped 22- and 25-fold in the
medium- and high-dose groups, respectively, on day 2 and
remained almost constant on day 3. Compared to the other
two groups, the concentration of PCB 136 in the low-dose
group was sixfold lower on day 1 and was less than the de-
tection limit on days 2 and 3. Despite these differences in the
fecal excretion over time, fecal excretion of PCB 136 still
accounts for approximately 1.2% of the total dose, which is
similar to the other two treatment groups.
An enrichment of (
ϩ)-PCB 136 was observed in all fecal
samples (Fig. 3B). The fecal EF increased with time in the

3
04
Environ. Toxicol. Chem. 27, 2008
I. Kania-Korwel et al.
Dose-dependence of PCB 136 tissue levels
routes of administration (i.e., oral vs intraperitoneal) ultimately
result in different EFs because of different PCB 136 tissue
levels and, thus, different levels at the binding site responsible
for the enantioselective disposition (e.g., biotransformation) of
PCB 136 [9].
Levels of PCB 136 in tissue and feces tended to decrease
with decreasing dose. In the low- and medium-dose groups,
several tissues even contained PCB levels below the detection
limit and needed to be pooled for the PCB analysis. Levels of
PCB 136 were less than the detection limit even in some pooled
samples. The muscle was the only tissue that did not entirely
follow this trend. Although the PCB 136 levels (adjusted by
tissue wet wt) decreased with decreasing dose (Fig. 1), the
estimated percentage of the total remaining dose in the muscle
was significantly higher in the low-dose group compared to
the medium- and high-dose groups (Table 1). A difference in
the distribution of PCB 136 also was apparent from the relative
muscle to tissue constants (e.g., the relative muscle to adipose
and liver to muscle tissue constants), with the low-dose group
displaying a significantly higher levels of PCB 136 in the
muscle compared to other tissues, such as adipose tissue and
the liver.
Considering that the muscle compartment plays an impor-
tant role during the initial disposition of PCBs [30,34,35], it
was not surprising that the muscle tissue did not entirely follow
the trends observed for the other tissues at the 3-d time point
selected for the present study. It is well established that im-
mediately after administration, PCBs are distributed by the
blood into highly perfused tissues as well as the muscles, which
make up a large part of the body [9]. Subsequently, PCBs are
redistributed by the blood into peripheral organs with a high
fat content, such as adipose tissue. Because muscle tissue has
no high affinity for PCBs, the comparatively higher PCB 136
content of the muscle in the low-dose groups most likely re-
flects the large volume of the muscle compartment and the
relatively low perfusion rate [35].
Interestingly, such an inverse relationship between PCB
tissue levels (or PCB dose) and EFs has not been observed in
environmental samples. For example, no correlation between
hepatic PCB levels and the EF was observed in cetaceans from
the Mediterranean Sea [18]. On the contrary, in the study by
Jimenez et al. [18], one sample from a bottlenose dolphin not
only had the highest EFs but also the highest PCB level. Sim-
ilarly, a study by Wong et al. [15] found no relationship be-
tween EFs and PCB levels in aquatic and riparian biota. The
differences between our laboratory study and these environ-
mental studies may not only reflect species differences but
also result from the complicated character of environmental
exposures of cetaceans and other species. Other factors, such
as species differences, the complexity of environmental PCB
mixtures, exposure to enantiomerically enriched PCBs, or re-
peated PCB exposures of varying magnitude, also may con-
tribute to these apparently contradictory results that, ultimate-
ly, reflect our limited understanding of the processes involved
in their enantioselective disposition.
CONCLUSION
The present study investigated dose dependence as an im-
portant factor determining the enantiomeric enrichment of
PCBs in mice. Overall, the EFs in various tissues increased
with decreasing doses, which indicates that the process re-
sponsible for the enantiomeric enrichment is saturated at higher
doses. Similarly, (environmental) exposures to PCBs are ex-
pected to be dose dependent in other species, including hu-
mans, as well. Because environmental exposures typically are
lower compared to laboratory studies, it therefore is not sur-
prising that some of the highest enantiomeric enrichments have
been found in environmental samples. Some studies, however,
also report that the enantiomeric enrichment, in contrast to the
present study, was not correlated with the level of exposure
in some species. These differences result from factors that have
not yet been studied under rigorous laboratory controls (i.e.,
species, exposure source, exposure frequency, and others).
Systematic studies of these factors are needed to further en-
hance our understanding of relevant chiral processes and to
allow environmental researchers the use of chiral analyses for
the investigation of environmental processes.
Effect of PCB 136 dose on enantiomeric enrichment
The present study shows, to our knowledge for the first
time, that the enrichment of a PCB atropisomer is dose de-
pendent and, in most tissues and in feces, decreases with in-
creasing dose and increasing tissue levels. The present study
shows a significant enrichment of one atropisomer (i.e., (ϩ)-
PCB 136) in all tissues and in feces, which is in agreement
with the results of previous laboratory studies in mice [9,10]
and rats [6–8]. This enantiomeric enrichment is thought to
result from enantioselective biotransformation processes,
which may or may not differ among species. Currently, how-
ever, other enantioselective disposition processes, such as en-
antioselective binding to (transport) proteins or active transport
of one PCB 136 atropisomer into (or out of) tissues, cannot
SUPPORTING INFORMATION
be excluded as the cause of the enrichment of (ϩ)-PCB 136
in mice. Overall, our results suggest that the enantioselective
disposition processes of PCB 136 can be saturated by high
doses of PCB 136, an observation that helps us to understand
the findings from earlier disposition studies in rodents.
This explains, at least in part, the above-mentioned differ-
ences in the enantiomeric enrichment among several laboratory
studies, in which the ratios of the two atropisomers range from
Table S1. Changes in body mass and organ weights in mice
treated with polychlorinated biphenyl (PCB) congener 136.
Table S2. Levels of polychlorinated biphenyl (PCB) con-
gener 136 and enantiomeric fractions in the feces of C57Bl/6
mice after oral administration of PCB 136. Each number rep-
resents the respective value for one metabolism cage with four
animals.
2
:1 to 4:1 [6–10]. Although a comparison of the atropisomer
Table S3. Significant differences in levels of polychlori-
nated biphenyl (PCB) congener 136 between tissues in mice.
The ‘‘v’’ indicates pairs of tissues that have significantly dif-
ratios from these studies is difficult because of the different
experimental designs (e.g., differences in the PCB congener
and the species studied), a review of the literature shows the
most pronounced enantiomeric enrichment in studies employ-
ing a comparatively low PCB dose [7]. In addition, the inverse
relationship between PCB 136 tissue levels and EFs observed
in the present study supports the hypothesis that different
ferent concentrations (
p Ͻ 0.05).
Table S4. Significant differences in enantiomeric fractions
of polychlorinated biphenyl (PCB) congener 136 between tis-
sues in mice. The ‘‘v’’ indicates pairs of tissues that have
significant different enantiomeric fractions (
p Ͻ 0.05).

Dose-dependent enantiomeric enrichment of PCB 136
All found at DOI: 10.1897/07-359.S1 (90 KB PDF).
Environ. Toxicol. Chem. 27, 2008
305
pesticides and polychlorinated biphenyl in standard and certified
reference materials. Chemosphere 49:1139–1347.
1
7. Wong CS, Mabury SA, Whittle DM, Backus SM, Teixeira C,
DeVault DS, Bronte CR, Muir DCG. 2004. Organochlorine com-
pounds in Lake Superior: Chiral polychlorinated biphenyls and
biotransformation in the aquatic food web. Environ Sci Technol
38:84–92.
Acknowledgement—We thank Regine Garcia Boy and Wei Xie for
their assistance with the animal procedures, Collin Just for his help
with the gas chromatographic analysis, Holly Moriarty and Allison
Smith for help with the analytical work, and Botond Banfi for use of
his metabolism cages. The research was supported by grants ES05605,
ES013661, and ES012475 from the National Institute of Environ-
mental Health Sciences and Major Research Instrumentation grant
BES-0420378 from the National Science Foundation.
18. Jimenez O, Jimenez B, Gonzales MJ. 2000. Isomer-specific poly-
chlorinated biphenyl determination in cetaceans from the Medi-
terranean Sea: Enantioselective occurrence of chiral polychlori-
nated biphenyl congeners. Environ Toxicol Chem 19:2653–2660.
19. Harju A, Bergman A, Olsson M, Roos A, Haglund P. 2003. De-
termination of atropisomeric and planar polychlorinated biphe-
nyls, their enantiomeric fractions and tissue distribution in grey
seals using comprehensive 2D gas chromatography. J Chroma-
togr 1019:127–142.
REFERENCES
1
2
. Lehmler H-J, Robertson LW. 2001. Atropisomers of PCBs. In
Robertson LW, Hansen LG, eds, Recent Advances in the Envi-
ronmental Toxicology and Health Effects of PCBs. University
Press of Kentucky, Lexington, KY, USA, pp 61–65.
. Frame GM. 1997. A collaborative study of 209 PCB congeners
and 6 Aroclors on 20 different HRGC columns. 2. Semiquanti-
tative Aroclor congener distributions. Fresenius J Anal Chem
2
0. Serrano R, Fernandez M, Rabanal R, Hernandez LM, Goznales
MJ. 2000. Congener-specific determination of polychlorinated bi-
phenyls in shark and grouper livers from Northwest African At-
lantic Ocean. Arch Environ Contam Toxicol 38:217–224.
1. Hoekstra PF, Wong CS, O’Hara TM, Solomon KR, Mabury SA,
Muir DCG. 2002. Enantiomer-specific accumulation of PCB atro-
pisomers in the bowhead whale (Balaena mysticetus). Environ
Sci Technol 36:1419–1425.
2
3
57:714–722.
3
4
5
6
. Brocks DR. 2006. Drug disposition in three dimensions: An up-
date on stereoselectivity in pharmacokinetics. Biopharm Drug
Dispos 27:387–406.
. Chu S, Covaci A, Schepens P. 2003. Levels and chiral signatures
of persistent organochlorine pollutants in human tissues from Bel-
gium. Environ Res 93:167–176.
. Harrad S, Ren J, Hazrati S, Robson M. 2006. Chiral signatures
of PCBs 95 and 149 in indoor air, grass, duplicate diets, and
human feces. Chemosphere 63:1368–1376.
. P u¨ ttmann M, Mannschreck A, Oesch F, Robertson L. 1989. Chiral
effects in the induction of drug-metabolizing enzymes using syn-
thetic atropisomers of polychlorinated biphenyls (PCBs). Bio-
chem Pharmacol 38:1345–1352.
22. Reich S, Jimenez B, Marsili L, Hernandez LM, Schurig V, Gon-
zales MJ. 1999. Congener specific determination and enantio-
meric ratios of chiral polychlorinated biphenyl in striped dolphins
(
Stanella coeruleoalba) from Mediterranean Sea. Environ Sci
Technol 33:1787–1793.
23. Bruckner JV, Wen-Der J, Brown JM, Putcha L, Chu CK, Stella
VJ. 1977. The influence of ingestion of environmentally encoun-
tered levels of a commercial polychlorinated biphenyl mixture
(Aroclor 1254) on drug metabolism in the rat. J Pharmacol Exp
Ther 202:22–31.
24. Narbonne JF. 1980. Time course of induction of microsomal en-
zymes following dietary administration of a polychlorinated bi-
phenyl (Phenoclor DP6). Toxicol Appl Pharmacol 56:1–7.
7
8
9
. Norstr o¨ m K, Eriksson J, Haglund J, Silvari V, Bergman A. 2006.
Enantioselective formation of methyl sulfone metabolites of
2
5. Shaikh N, Parkin S, Lehmler H-J. 2006. The Ullman coupling
reaction: A new approach to tetraarylstannanes. Organometallics
2
7
,2
Ј,3,3Ј,4,6Ј-hexachlorobiphenyl in rat. Environ Sci Technol 40:
2
5:4207–4214.
649–7655.
2
6. Harner T, Wiberg K, Norstrom R. 2000. Enantiomer fractions are
preferred to enantiomer ratios for describing chiral signatures in
environmental analysis. Environ Sci Technol 34:218–220.
. Kania-Korwel I, Garrison AW, Avants JK, Hornbuckle KC, Rob-
ertson LW, Sulkowski WW, Lehmler H-J. 2006. Distribution of
chiral PCBs in selected tissues in the laboratory rat. Environ Sci
Technol 40:3704–3710.
. Kania-Korwel I, Shaikh N, Hornbuckle KC, Robertson LW,
Lehmler H-J. 2007. Enantioselective disposition of PCB 136
2
2
7. Haglund P. 1996. Isolation and characterization of polychlorinated
biphenyl (PCB) atropisomers. Chemosphere 32:2133–2140.
8. Birnbaum LS. 1985. The role of structure in the disposition of
halogenated aromatic xenobiotics. Environ Health Perspect 61:
(
2,2
and intraperitoneal administration. Chirality 19:56–66.
0. Lehmler H-J, Price DJ, Garrison AW, Birge WJ, Robertson LW.
003. Distribution of PCB 84 enantiomers in C57Bl/6 mice. Fre-
Ј,3,3Ј,6,6Ј-hexachlorobiphenyl) in C57Bl/6 mice after oral
1
1–20.
2
9. Saghir SA, Hansen LG, Holmes KR, Kodavanti PR. 2000. Dif-
ferential and non-uniform tissue and brain distribution of two
1
1
2
1
4
distinct C-hexachlorobiphenyls in weanling rats. Toxicol Sci 54:
0–70.
0. Birnbaum LS. 1983. Distribution and excretion of 2,3,6,2
and 2,4,5,2 ,4 ,5 -hexachlorobiphenyl in senescent rats. Toxicol
Appl Pharmacol 3:262–272.
senius Environ Bull 12:254–260.
1. Buckman AH, Wong CS, Chow EA, Brown SB, Solomon KR,
Fisk AT. 2006. Biotransformation of polychlorinated biphenyls
6
3
3
Ј,3Ј,6Ј-
Ј
Ј Ј
(
PCBs) and bioformation of hydroxylated PCBs in fish. Aquat
Toxicol 78:176–185.
1. Lehmler H-J, Robertson LW, Garrison AW, Kodavanti PRS. 2005.
1
1
1
1
1
2. Wong CS, Lau F, Clark M, Mabury SA, Muir DCG. 2002. Rain-
bow trout (Oncorhynchus mykiss) can eliminate chiral organo-
chlorine compound enantioselectively. Environ Sci Technol 36:
3
Effects of PCB 84 enantiomers on [ H]phorbol ester binding in
rat cerebellar granule cells and Ca -uptake in rat cerebellum.
Toxicol Lett 156:391–400.
45
2ϩ
1
257–1262.
3
2. Pessah IN, Hansen LG, Albertson TE, Garner CE, Ta TA, Do Z,
Kim KH, Wong PW. 2006. Structure–activity relationship for non-
coplanar polychlorinated biphenyl congeners toward the ryano-
3. Warner NA, Norstrom RJ, Wong CS, Fisk AT. 2005. Enantiomeric
fractions of chiral polychlorinated biphenyls provide insights on
biotransformation capacity of Arctic biota. Environ Toxicol Chem
2
ϩ
dine receptor–Ca channel complex type 1 (RyR1). Chem Res
Toxicol 19:92–101.
2
4:2763–2767.
4. Gomara B, Gonzalez MJ. 2006. Enantiomeric fractions and con-
gener specific determination of polychlorinated biphenyls in eggs
of predatory birds from Donana National Park (Spain). Chemo-
sphere 63:662–669.
5. Wong CS, Garrison AW, Smith PD, Foreman WT. 2001. Enan-
tiomeric composition of chiral polychlorinated biphenyl atropi-
somers in aquatic and riparian biota. Environ Sci Technol 35:
3
3. Gobas FAPC, Wilcockson JB, Russel RW, Haffner GD. 1999.
Mechanism of biomagnification in fish under laboratory and field
conditions. Environ Sci Technol 33:133–141.
34. Brandt I, Mohammed A, Slanina P. 1981. Persistence of 2,3,6-
substituted pentachlorobiphenyls in the lung parenchyma: A new
structure-dependent tissue localization of polychlorinated biphe-
nyls in mice. Toxicology 21:317–322.
35. Matthews HB, Tuey DB. 1980. The effect of chlorine position
on the distribution and excretion of four hexachlorobiphenyl iso-
mers. Toxicol Appl Pharmacol 53:377–388.
2
448–2454.
6. Wong CS, Hoekstra PF, Karlsson H, Backus SM, Mabury SA,
Muir DCG. 2002. Enantiomer fractions of chiral organochlorine