Environmental Toxicology and Chemistry, Vol. 20, No. 11, pp. 2433–2442, 2001
q 2001 SETAC
Printed in the USA
0730-7268/01 $9.00 1 .00
2,3,7,8-TETRACHLORODIBENZO-p-DIOXIN EQUIVALENTS IN TISSUE SAMPLES
FROM THREE SPECIES IN THE DENVER, COLORADO, USA, METROPOLITAN AREA
KATHERINE K. COADY,* PAUL D. JONES, and JOHN P. GIESY
Department of Zoology, National Food Safety and Toxicology Center, and Institute for Environmental Toxicology, Michigan State University,
East Lansing, Michigan 48824, USA
( Received 13 December 2000; Accepted 16 March 2001)
Abstract—The Rocky Mountain Arsenal (RMA), a Superfund site near Denver, Colorado, USA, has a history of various industrial
processes that may have led to the release of polychlorinated dibenzo-p-dioxins (PCDDs), polychlorinated dibenzofurans (PCDFs),
and polychlorinated biphenyls (PCBs). The PCDDs, PCDFs, and non-ortho- and mono-ortho-substituted PCBs cause a common
set of toxic effects that are mediated through the aryl hydrocarbon receptor (AhR). The total AhR-mediated activity of complex
mixtures in biota samples from the RMA and surrounding reference areas was determined by both instrumental and bioanalytical
techniques. Mean concentrations of bioassay-derived 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD) equivalents (TCDD-EQ) in carp
eggs, owl livers, and kestrel eggs ranged from 0 to 8.5, 17 to 130, and 2.4 to 18 pg/g, respectively. For most samples analyzed,
concentrations of TCDD-EQ and instrumentally derived TCDD equivalents (TEQs) were not significantly different. In a few tissue
samples, concentrations of TEQs and TCDD-EQs were not equivalent. This can indicate the presence of AhR-active compounds
that were not identified or quantified by instrumental techniques or the existence of nonadditive interactions among congeners when
samples are analyzed by the bioassay. Overall, mean concentrations of TCDD-EQs in extracts of carp and kestrel eggs were not
significantly different between groups of samples collected on and off the RMA, whereas concentrations of TCDD-EQs in several
owl livers collected on the RMA were significantly greater than concentrations in owl livers from off the RMA.
Keywords—Dioxin equivalents
Aryl hydrocarbon receptor
Rocky Mountain Arsenal
H4IIE-luc bioassay
2,3,7,8-Tetrachlorodibenzo-p-dioxin
mixture multiplied by toxic equivalency factors (TEFs) derived
by the World Health Organization (WHO, Paris, France) (Eqn.
1) [10]. Since it is the archetypal and most potent of the PCDHs,
TCDD is used as the reference compound and is assigned the
TEF value of 1.0. Toxic equivalency factors for other dioxins,
furans, and PCBs are calculated and expressed relative to
TCDD. The values of TEFs vary, depending on the organism
and endpoint used. In this study, TCDD equivalents (TEQs)
derived using the WHO TEF values are designated as TEQWHO:
INTRODUCTION
Planar chlorinated diaromatic hydrocarbons (PCDHs), including polychlorinated dibenzo-p-dioxins (PCDDs), polychlorinated dibenzofurans (PCDFs), and some polychlorinated
biphenyls (PCBs), are trace contaminants that are widely distributed in the environment. The PCDHs are persistent, subject
to bioaccumulation, and can be toxic to biota at environmentally relevant concentrations [1–3]. A subset of these chemicals
is characterized by a common mechanism of action via binding
to the cytoplasmic aryl hydrocarbon receptor (AhR) [4,5].
Chemicals such as PCDDs, PCDFs, and some PCB congeners
can attain a planar configuration, bind to the AhR, and are
referred to as dioxin-like chemicals. On binding of the dioxin
or dioxin-like ligand, the AhR–ligand complex translocates to
the nucleus, where it binds to dioxin-responsive elements
(DREs) in the DNA, which results in gene expression [4,5].
Toxic effects of PCDHs include impaired reproduction of birds
and fish [3,6–8], severe weight loss, fatty deposition in the
liver, edema, fetotoxicity, and teratogenicity in laboratory animals and wildlife [1,2,4,9].
The total potency of dioxin-like chemicals in environmental
matrices or tissue samples can be estimated by calculating the
concentrations of 2,3,7,8-tetrachlorodibenzo-p-dioxin equivalents. The TCDD equivalents are measures of the total dioxinlike potency relative to the most potent AhR-binding congener,
TCDD. Two methods are commonly used to estimate concentrations of TCDD equivalents. In the first method, concentrations of TCDD equivalents are calculated as the sum of the
products of the concentration of each congener detected in a
TEQ WHO 5
O C · TEF
i
i
(1)
where Ci indicates the concentration of each congener and TEFi
is the TEF for that congener.
The TCDD equivalents can also be determined by use of
in vitro bioassay procedures based on the AhR-binding mechanism of action. In vitro bioassays have several advantages
over instrumental analyses. They are rapid, sensitive, and integrative measures of the total activity of complex PCDHcontaining mixtures [4,5,7]. In this study, the H4IIE-luc in
vitro cell bioassay was used to measure TCDD-equivalent concentrations. These TCDD-equivalent concentrations are designated as TCDD-EQs. The H4IIE-luc bioassay is a rat hepatoma cell line into which the luciferase gene, under control
of the AhR, has been stably transfected. The gene product,
luciferase, cleaves the added substrate, luciferin, and luminescence is produced in a dose-dependent manner [5].
Concentrations of TEQWHO are not directly comparable to
bioassay-derived TCDD-EQ concentrations because the TEFs
promulgated by WHO are designed for use in risk assessment
and are meant to be protective rather than predictive. The
expert panel assembled to evaluate the TEFs made consistently
* To whom correspondence may be addressed
(kemlerka@msu.edu).
2433
2434
Environ. Toxicol. Chem. 20, 2001
K.K. Coady et al.
conservative assumptions about relative toxicity and determined that values were always rounded up to the next higher
order of magnitude [10]. As a result, the TEFs derived only
estimate the relative biological potency of the congeners. Derivation of a mass balance relationship between concentrations
of TEQs and TCDD-EQs can be accomplished by applying a
relative effects potency (REP) to detected congener concentrations (Eqn. 2) [11]. The REPs are derived from the same
endpoint in the same bioassay system (H4IIE-luc) [5,11].
Equivalent concentrations (TEQREP) derived using REPs and
individual PCDH concentrations can be compared directly to
concentrations of TCDD-EQ measured in the same bioassay.
The REP values used in this study were taken from a previous
study in which REP values were calculated with the H4IIEluc cell line [11]. Concentrations of TEQREP can be compared
to concentrations of TCDD-EQ in a mass balance to determine
if all the TCDD-EQ activity observed could be accounted for
by concentrations of the instrumentally measured PCDHs:
TEQ REP 5
O C · REP
i
i
(2)
where Ci indicates the concentration of the congener and REPi
is the REP for that congener.
The RMA is a 27-square-mile Superfund site located in
Commerce City, Colorado, USA, near Denver. The RMA was
an army facility composed of two main chemical-producing
plants, the South Plants and the North Plants, and numerous
support buildings and other infrastructure. These plants were
used to produce a diversity of chemicals throughout the arsenal’s history [12]. The plants were located in the central six
square miles of the RMA, designated as the core. The core
area is surrounded by 21 square miles, designated as the periphery. It was expected that if operations at the production
facilities had been a source of PCDHs, the periphery would
be less contaminated than the core. From 1942 to 1945, the
U.S. Army used the RMA to produce chemical warfare agents,
such as Levinstein mustard and Lewisite. During this time,
incendiary bombs, organophosphate nerve agents (e.g., Sarin),
and Hydrazine rocket fuel (U.S. Army, Commerce City, CO,
USA) were also manufactured at the site. From 1946 to 1982,
some of the RMA facilities were leased to various companies
that used the facilities to produce pesticides, including dieldrin,
aldrin, chlordane, carbamate insecticides, soil fumigants, and
various other organochlorine compounds [12]. Both PCDDs
and PCDFs are known to be released from the manufacture of
some chlorinated pesticides, incineration of chlorine-containing wastes, chloralkali plants, and fossil fuel combustion [13].
Routine application of various pesticides such as pentachlorophenol on the RMA may also have resulted in contamination
of surface soil by these compounds.
The overall objective of the investigation on which we report here was to determine whether concentrations of dioxin
equivalents in samples of biota from the RMA were significantly greater than those from adjacent reference sites. Eggs
from the American kestrel (Falco sparverius) and common
carp (Cyprinus carpio) and livers from adult and juvenile great
horned owls (Bubo virginianus) were collected to compare the
dioxin-like activity between the on- and off-site populations.
The working hypothesis for the use of these tissue samples
was that if PCDHs were released into the environment, they
could persist and possibly biomagnify to hazardous concentrations in some portions of the local food chain. The PCDHs
are lipophilic compounds that tend to accumulate in fats within
organisms [3] and can also be maternally transferred to off-
spring via the egg yolk in oviparous organisms [6,7]. For these
reasons, kestrel and carp eggs, as well as owl livers, were used
as indicators of the availability of PCDHs present in the RMA
environment. Kestrels and owls were selected as sentinel species because they are near the top of the food chain, and thus
concentrations of PCDHs in their tissues are useful integrative
measures of exposure to biologically available PCDHs in the
terrestrial environment. These birds prey on insects, small
mammals, and at times other birds. Carp eggs were used as
bioindicators of dioxin-like activity in the limited aquatic systems of the RMA.
Here concentrations of TEQ and TCDD-EQ were compared
by use of a mass balance between the two measures of concentrations in order to determine if all dioxin-like activity
present in the tissue samples was accounted for by PCDDs,
PCDFs, and PCBs. Congener-specific concentrations were also
examined by pattern analysis in order to identify the possible
source of dioxin-like contamination in tissue samples from the
RMA. Finally, concentrations of TCDD equivalents in tissue
samples from the RMA and reference populations were examined to determine if concentrations in samples from the
RMA were significantly different from those samples from
other areas of the Denver metropolitan area.
METHODS
Sampling techniques
Kestrel eggs were collected from nest boxes located on the
RMA and at selected reference areas off the RMA. Twentynine kestrel eggs were collected on-site, 10 eggs from the core
and 19 from the periphery. Sixteen kestrel eggs were collected
from nest boxes in the reference areas off the RMA. Reference
areas were selected to represent background PCDH concentrations in the Denver metropolitan area. The reference areas
from which kestrel eggs were collected included a fairground,
a cemetery, a wastewater facility, a lakeside area, a reservoir,
and a former airport.
Carp eggs were collected from lakes on and off the RMA.
Two samples of carp eggs were collected from reference sites,
while 16 samples of eggs were collected from carp collected
on the RMA. Carp were collected by electroshocking and gill
nets.
Livers were collected from juvenile and adult great horned
owls. Samples were collected when owls were found dead or
moribund as part of the U.S. Fish and Wildlife Service’s opportunistic sampling program. Eleven great horned owl livers
were collected from reference areas, and 16 livers were collected on-site. Reference areas from which owl specimens were
collected were within a 130-mile radius of the RMA.
Sample homogenization, extraction, and cleanup
Samples were shipped frozen to Michigan State University
(East Lansing, MI, USA) and stored at 2208C for subsequent
analysis. Homogenization of tissue samples was accomplished
by adding a sufficient amount of Na2SO4 to dry the sample
(usually in a 10:1 w/w ratio) and mixing until the tissues were
dry. The mixture was then blended at high speed (Omnimixt,
Sorvall, Norwalk, CT, USA) to produce a fine powder. Homogenized samples were Soxhlet extracted for 18 h with 300
ml of 1:1 acetone/hexane. After extraction, sample extracts
were concentrated to near dryness by rotary evaporation at
358C, and 5 ml of concentrated H2SO4 were added to each
sample to oxidize lipids. After separation of the aqueous and
Environ. Toxicol. Chem. 20, 2001
2,3,7,8-TCDD equivalents in wildlife tissue samples
organic phases, the hexane fraction was removed. The acid
phase was washed with an additional aliquot of hexane, which
was removed and combined with the first hexane fraction. The
hexane fraction was reduced to near dryness by rotary evaporation, and 5 ml of nanopure water were used to rinse the
extracts before the extracts were dried through Na2SO4 and
evaporated under a stream of nitrogen to 0.1 ml. Final extracts
were diluted to 1.0 ml with isooctane for use in the H4IIEluc bioassay [5]. Pigments and other compounds that interfere
with the H4IIE-luc bioassay were removed from extracts of
kestrel eggs by use of column chromatography with a silica,
acidic silica, and KOH-silica gel column [13].
Quality control and assurance for the extraction and cleanup
included method blanks and method spikes with each batch
of approximately 20 samples. Method blanks and spikes consisted of 50 g of Na2SO4. For the method spikes, a solution
containing a mixture of PCB congeners (Cambridge Isotopes,
Cambridge, MA, USA) was added to Na2SO4. Matrix spikes
and matrix blank samples were also analyzed with each batch
of kestrel egg samples. Matrix blanks and spikes consisted of
chicken egg homogenate, to which a known amount of PCB
congeners had or had not been added. Twelve quality control
quail egg samples spiked with PCB congener 126 were also
submitted blind to the laboratory with the RMA kestrel egg
samples.
H4IIE-luc bioassay
The H4IIE-luc bioassay was conducted as previously described [5,14]. Analyses of individual extracts were conducted
in separate 96-well cell culture plates concurrently with a range
of 2,3,7,8-TCDD standards delivering doses of 30, 10, 3, 1,
0.3, and 0.1 pg TCDD/well. Extracts were applied in serial
dilution, which delivered 13, 0.33, 0.13, 0.033, 0.013, and
0.0033 diluted extracts to the wells. Cell viability was measured at the time of the luminescence assay to ensure that the
tissue extracts were not causing cell death. Viability was determined by both detailed visual inspection of cell growth and
a cell viability test kit (Live/Dead Viability/Cytotoxicity kit
for animal cells; Molecular Probes Eugene, OR, USA).
Instrumental analysis
Concentrations of individual congeners were measured by
gas chromatography/mass spectrometry (GC/MS) at Midwest
Research Institute (MRI, Kansas City, MO, USA) using standard methods. Concentrations of all 2,3,7,8-substituted PCDDs
and PCDFs and non-ortho and mono-ortho PCB congeners
(IUPAC 77, 81, 105, 114, 118, 123, 126, 156, 157, 167, 169,
and 189) were measured in kestrel eggs and owl livers. Monoortho-substituted PCB congeners were not measured in carp
egg samples. All data generated by MRI and Michigan State
University (MSU) were subject to U.S. Environmental Protection Agency quality assurance and quality control guidelines.
Data analysis: TEQs
Consensus TEFs developed by the WHO (TEFWHO) were
used in the calculation of TEQWHO values for carp egg, owl
liver, and kestrel egg extracts [10]. Avian TEF values were
applied to the owl and kestrel sample extracts, while fish TEF
values were applied to the carp sample extracts. Congener
concentrations were multiplied by TEFs for the PCDH analytes
measured [10]. This product was then summed to yield a total
TEQWHO for the sample (Eqn. 1). To determine the potential
2435
effects of samples for which concentrations of individual congeners were less than the minimum detection limits (MDL),
two TEQWHO values were calculated as follows: TEQWHO-MAX
and TEQREP-MAX were calculated by substituting the MDL for
those congeners whose concentrations were below the MDL,
and TEQWHO-MIN and TEQREP-MIN were calculated when congener concentrations below the MDL were set to 0.0. Therefore, the MAX values represent the greatest possible overestimate of the dioxin-like activity, while the MIN values represent the greatest possible underestimates of equivalent concentrations.
Data analysis: TCDD-EQs
Concentrations of TCDD-EQs were determined for each
sample by comparing the sample dose–response curve to the
TCDD standard dose–response curve analyzed on the same
plate using previously described methods [15]. Samples that
showed adequate dose–response curves were analyzed using
the slope-ratio method, which compares the probit-transformed
linear region of the dose–response curve. Samples that did not
show a full dose–response curve were evaluated using Dunnett’s test to identify extract concentrations showing a response
significantly (p 5 0.05) greater than the MDL [16]. The sample
response (at the least concentration significantly greater than
the MDL, which was usually the undiluted sample extract) was
then divided by the standard response (at the least concentration that differed significantly from the MDL, usually 0.1 pg
TCDD). This ratio was then multiplied by the TCDD concentration used in the standard response (usually 0.1 pg TCDD)
in order to determine the maximum possible TCDD-equivalent
amount present in the sample.
Statistical methods
The chi-square test was used to examine the mass balance
relationship between concentrations of TCDD-EQMAX and TEQREP-MAX in tissue samples. Concentrations of TEQWHO-MAX and
TCDD-EQMAX were compared between on-site and reference
areas using either the Mann–Whitney U test or the two-group
t test. The two-group t test was utilized if the data were normally distributed and had homogeneous variances. Separate
and pooled variances were examined with the t test. The Mann–
Whitney U test was used in cases where the assumptions of
a normal distribution were not met [17]. One-way analysis of
variance and Dunnett’s multiple comparison were used in cases
where more than one population of RMA samples (originating
from the core or periphery areas) was compared to reference
populations.
To determine whether patterns of relative concentrations of
PCDH congeners were different between RMA and reference
samples, principal components analysis (PCA) was used. Principal components analysis was performed using the full data
set containing all quantifiable data and half the MDL for concentrations below the MDL. For PCA, the data were first standardized to z scores; these values are calculated by setting the
mean and standard deviation of the data for each variable to
values of 0 and 1, respectively.
The level of statistical significance for all tests was set at
a 5 0.05. Power (1 2 b), the probability of avoiding a type
II error (b 5 0.20) [16], was set at 0.80. SYSTAT software
(SPSS 1998, Chicago, IL, USA) was used for statistical analyses, and Pass 60 computer software (NCSS, Kaysville, UT,
USA) was used to calculate the power of analyses.
0.001
0.01
0.01
0.5
0.1
0.01
0.1
0.01
0.1
1.0
0.05
0.1
0.5
1.0
0.05
0.0001
0.0001
0.00005
0.005
0.0001
0.0005
0.000005
0.000005
0.000005
1234678-HpCDF
1234789-HpCDF
123478-HxCDD
123478-HxCDF
123678-HxCDD
123678-HxCDF
123789-HxCDD
123789-HxCDF
12378-PeCDD
12378-PeCDF
234678-HxCDF
23478-PeCDF
2378-TCDD
2378-TCDF
OCDD
OCDF
(PCB 169)
(PCB 126)
(PCB 77)
(PCB 81)
(PCB 105)
(PCB 114)
(PCB 118)
Fish
N
% Lipid
1234678-HpCDD
Parameter
0.00001
0.0001
0.0001
—
—
0.48
(0.17)
0.53
(0.28)
0.02
(0.01)
0.22
(0.01)
0.09
(0.02)
0.06
(0.03)
0.16
(0.10)
0.02
(0.01)
0.58
(0.09)
0.17
(0.10)
0.07
(0.01)
0.79
(0.09)
0.05
(0.01)
0.09
(0.05)
0.26
(0.16)
3.05
(0.50)
0.16
(0.03)
0.30
(0.11)
2.05
(1.72)
13.49
(10.8)
0.93
(0.32)
—
2
—
—
0.73
(0.22)
1.21
(0.72)
0.03
(0.02)
0.23
(0.09)
0.12
(0.05)
0.10
(0.06)
0.13
(0.03)
0.04
(0.03)
0.47
(0.16)
0.22
(0.09)
0.23
(0.11)
2.34
(1.07)
0.16
(0.07)
0.11
(0.07)
0.88
(0.42)
3.33
(1.28)
0.30
(0.18)
0.95
(0.50)
12.75
(5.96)
52.39
(19.5)
3.38
(1.43)
—
16
Carp eggs Carp eggs
(reference) (RMA)c
4
3.6
24.0
(19.2)
13.1
(8.36)
2.11
(2.56)
5.81
(3.69)
16.0
(20.2)
19.9
(14.8)
6.61
(8.22)
1.65
(1.29)
3.67
(1.60)
6.14
(3.95)
1.82
(1.65)
2.35
(2.20)
8.45
(5.78)
1.04
(1.09)
5.07
(1.82)
33.4
(24.0)
4.26
(2.23)
72.6
(70.0)
122
(142)
22.1
(6.62)
1.68
(1.00)
1,400
(1,510)
196
(174)
4,970
(5,310)
Owl livers
(reference
juvenile)
10
4.0
101
(163)
64.7
(82.2)
20.7
(29.0)
9.79
(11.2)
85.9
(89.6)
26.3
(28.6)
35.2
(337)
3.91
(5.00)
2.43
(1.97)
5.61
(5.50)
2.64
(2.97)
10.1
(9.37)
33.6
(50.0)
0.673
(0.696)
1.08
(0.912)
97.7
(152)
22.4
(25.6)
145
(239)
306
(328)
10.6
(4.99)
2.48
(2.80)
3,920
(8,970)
765
(2,097)
13,600
(33,600)
Owl livers
(RMA juvenile)
6
7.5
80.6
(165)
14.2
(19.4)
4.33
(6.19)
9.74
(14.2)
47.2
(100)
21.4
(29.2)
14.9
(29.0)
1.10
(1.40)
2.19
(1.95)
3.61
(1.79)
0.502
(0.528)
5.81
(8.35)
16.1
(25.9)
0.551
(0.297)
3.64
(3.68)
36.3
(45.4)
5.55
(7.11)
132
(101)
58.3
(42.4)
8.55
(8.24)
3.53
(6.10)
624
(865)
82.5
(70.5)
2,700
(4,190)
Owl livers
(reference adult)
3
2.3
1,320
(993)
257
(123)
137
(184)
312
(245)
1,860
(2,020)
1,020
(608)
838
(1,020)
90.9
(48.6)
34.6
(45.1)
186
(67.2)
20.3
(19.1)
162
(174)
586
(664)
11.3
(1.04)
3.13
(1.37)
594
(469)
60.6
(71.4)
336
(148)
423
(139)
3.17
(2.04)
0.17
(0.06)
2,700
(1,760)
590
(336)
11,900
(7,550)
Owl livers
(RMA adult)
3
4.05
58.2
(5.53)
29.6
(11.1)
4.48
(2.18)
11.7
(4.05)
15.6
(5.67)
28.7
(7.68)
8.93
(3.45)
3.75
(1.18)
0.639
(0.210)
9.12
(5.63)
0.262
(0.308)
6.91
(2.86)
7.27
(1.44)
1.03
(0.899)
0.410
(0.151)
60.0
(28.5)
14.1
(6.47)
92.9
(69.1)
74.9
(20.3)
8.53
(2.89)
0.930
(0.580)
938
(496)
202
(185)
3,970
(3,020)
Owl livers
(RMA unknown)
30
0.73
(0.97)
1.06
(0.98)
0.62
(0.45)
2.17
(2.90)
5.13
(4.61)
2.22
(1.87)
5.17
(9.94)
4.29
(8.43)
0.86
(0.35)
4.32
(4.25)
8.45
(9.28)
2.65
(3.59)
12.3
(30.9)
1.97
(3.30)
24.0
(34.5)
3.47
(4.90)
51.1
(81.3)
7.80
(5.26)
56.7
(66.6)
86.7
(122)
7.82
(7.25)
1,340
(2,060)
75.7
(103)
3,520
(5,110)
4.12
(1.11)
1.02
(1.43)
8.86
(31.3)
8.58
(22.0)
5.75
(9.10)
7.59
(22.1)
4.78
(9.04)
2.43
(3.10)
1.54
(1.33)
3.08
(4.62)
4.89
(5.96)
1.17
(1.57)
3.10
(3.82)
1.85
(2.06)
6.00
(5.41)
3.51
(6.47)
17.6
(11.1)
39.6
(110)
244
(734)
120
(153)
21.4
(34.6)
2,460
(5,980)
139
(299)
4,750
(7,440)
Kestrel eggs
(RMA)
16
Kestrel eggs
(reference)
Environ. Toxicol. Chem. 20, 2001
0.1
0.05
0.1
0.001
0.0001
0.0001
1.0
1.0
1.0
0.1
0.1
1.0
0.1
0.1
0.1
0.01
0.1
0.05
0.01
0.01
0.001
Avian
WHO TEF valuesb
Table 1. Sample information from various wildlife populations and mean concentrations of individual polychlorinated dibenzodioxins (PCDD), polychlorinated dibenzofurans (PCDF), and polychlorinated
biphenyl (PCB) congeners (pg/g, wet wt) and 2378 tetrachlorodibenzo-p-dioxin (TCDD) equivalents in wildlife tissues (standard deviation in parentheses)a
2436
K.K. Coady et al.
0.000005
0.000005
0.000005
0.000005
0.000005
(PCB 123)
(PCB 156)
(PCB 157)
(PCB 167)
(PCB 189)
0.00001
0.00001
0.0001
0.0001
0.00001
Avian
0.34
(0.14)
0.84
(0.11)
0.33
(0.11)
1.50
(0.14)
0.00
(0.00)
2.50
(0.71)
—
—
—
—
—
0.38
(0.12)
1.60
(0.32)
0.64
(0.17)
3.40
(1.1)
4.25
(11.4)
8.50
(10.4)
—
—
—
—
—
Carp eggs Carp eggs
(reference) (RMA)c
108
(90.9)
1,410
(1,550)
289
(321)
752
(906)
217
(264)
33.4
(20.2)
42.7
(26.6)
13.5
(9.59)
28.2
(16.1)
0.00
(0.00)
6.50
(8.42)
Owl livers
(reference
juvenile)
Owl livers
(RMA adult)
92.7
(66.5)
3,040
(1,150)
499
(141)
767
(581)
248
(90.0)
1,460
(1,130)
1,460
(1,140)
984
(976)
1,140
(1,020)
590
(426)
590
(426)
Owl livers
(reference adult)
72.8
(59.7)
724
(1,320)
132
(231)
341
(658)
117
(200)
36.0
(46.0)
41.5
(49.9)
21.4
(31.9)
31.8
(45.3)
31.2
(76.3)
35.8
(74.3)
Owl livers
(RMA juvenile)
113
(248)
3,820
(10,200)
589
(1,500)
667
(1,540)
272
(689)
87.3
(97.3)
89.2
(96.4)
66.0
(59.9)
69.9
(58.4)
30.7
(66.7)
34.9
(64.7)
33.2
(32.3)
2,004
(2,590)
306
(374)
451
(536)
331
(484)
29.7
(11.0)
31.9
(10.5)
24.9
(10.0)
30.0
(10.0)
0.00
(0.00)
5.33
(4.51)
Owl livers
(RMA unknown)
135
(267)
3,530
(10,100)
631
(1,450)
2,290
(5,990)
625
(2,170)
58.05
(108.14)
60.23
(107.78)
20.12
(34.2)
21.98
(34.1)
14.47
(35.4)
18.20
(34.0)
Kestrel eggs
(reference)
65.0
(84.0)
1,270
(2,130)
250
(254)
604
(6,560)
163
(331)
20.94
(17.6)
23.59
(17.7)
15.44
(20.4)
18.85
(21.1)
2.43
(11.4)
6.87
(11.0)
Kestrel eggs
(RMA)
a
HpCDD 5 heptachlorodibenzo-p-dioxin, HpCDF 5 heptachlorodibenzofuran, HxCDD 5 hexachlorodibenzo-p-dioxin, HxCDF 5 hexachlorodibenzofuran, PeCDD 5 pentachlorodibenzo-p-dioxin,
PeCDF 5 pentachlorodibenzofuran, TCDD 5 tetrachlorodibenzo-p-dioxin, TCDF 5 tetrachlorodibenzofuran, OCDD 5 octachlorodibenzo-p-dioxin, OCDF 5 octachlorodibenzofuran, PCB 5
polychlorinated biphenyl, TEQ 5 TCDD-equivalent value as determined by instrumental analyses, TEQREP 5 relative potency of instrumentally derived TCDD equivalents on the H4IIE-luc cell line,
TCDD-EQ 5 TCDD-equivalent value as determined by the H4IIE-luc cell line.
b WHO 5 World Health Organization; TEF 5 toxic equivalency factors.
c RMA 5 Rocky Mountain Arsenal.
TCDD-EQMAX
TCDD-EQMIN
TEQREP-MAX
TEQREP-MIN
TEQMAX
TEQMIN
Fish
Parameter
WHO TEF valuesb
Table 1. Continued
2,3,7,8-TCDD equivalents in wildlife tissue samples
Environ. Toxicol. Chem. 20, 2001
2437
2438
Environ. Toxicol. Chem. 20, 2001
K.K. Coady et al.
Table 2. Mean concentrations (pg/g wet wt) of TEQMAX and TCDDEQMAX in carp eggs from the RMA and reference locations (standard
deviation in parentheses). Significance of differences (p values)
between the two locations as determined by Student’s t testa
On-site
N
TEQMAXb
TCDD-EQMAXc
16
2
1.60 (0.32)
0.84 (0.11)
8.25 (10.4)
2.50 (0.70)
0.004
0.007
0.045
0.457
(Power 5 0.67)
p value
Separated
Poolede
RMA 5 Rocky Mountain Arsenal.
TEQ 5 TCDD-equivalent value as determined by instrumental analyses.
c TCDD-EQ 5 TCDD-equivalent value as determined by the H4IIEluc cell line.
d Separate 5 tests using separate variances.
e Pooled 5 tests using pooled variances.
a
b
Fig. 1. Concentrations of toxic equivalents (TEQ)REP as a function of
concentrations of 2,3,7,8-tetrachlorodibenzo-p -dioxin eqivalents
(TCDD-EQ) in wildlife tissues. All samples from on and off the Rocky
Mountain Arsenal in which TEQREP and TCDD-EQ were synoptically
measured. The closed line represents equivalence between TEQREP
and TCDD-EQ. (A) Carp eggs, (B) great horned owl livers, (C) American kestrel eggs.
RESULTS
Carp eggs
Mean concentrations of TEQWHO in carp eggs from the
RMA ranged from 0.38 to 1.6 pg TEQ/g wet weight (wet wt)
egg, while mean concentrations of TEQWHO in samples from
reference locations ranged from 0.34 to 0.84 pg TEQ/g wet
wt egg, depending on which proxy values were applied (Table
1). The mean contribution of PCBs to the total TEQWHO ranged
from 1.5 to 20% among samples and methods of calculation.
Mean concentrations of TEQREP in carp eggs from the RMA
ranged from 0.64 to 3.4 pg TEQ/g wet wt egg, while the mean
concentration of TEQREP in samples from reference locations
ranged from 0.33 to 1.5 pg/g wet wt egg (Table 1). Mean
concentrations of bioassay derived TCDD-EQs in carp eggs
from the RMA ranged from 4.3 to 8.5 pg TCDD-EQ/g egg,
while mean concentrations in carp eggs from reference areas
ranged from 0 to 2.5 pg TCDD-EQ/g wet wt egg, depending
on the choice of surrogate values chosen for nondetect values
(Table 1). Concentrations of TEQREP-MAX were significantly different from concentrations of TCDD-EQMAX when all data
points were considered in the chi-square analysis (x2 5 94.3,
df 5 17) (Fig. 1A). When three outlying points were removed
from the analysis, the remaining concentrations of TEQREP MAX
and TCDD-EQMAX were not significantly different from each
other (x2 5 21.2, df 5 14).
Since concentrations of TEQWHO in carp were normally distributed and the variances between the two groups were not
significantly different, the two-group Student’s t test was used
to compare the mean values of on-site and reference concentrations. Concentrations of TEQMAX in carp egg samples from
the RMA were significantly greater than those in samples from
reference locations using both separate and pooled variances
(Table 2). Concentrations of TCDD-EQMAX were also significantly different between locations when separate variances
were used, but no significant differences were observed when
the variances were pooled (Table 2). The small sample size of
carp eggs collected from reference areas greatly influenced the
difference in variance measurements. Therefore, the pooled
estimate is a more accurate test statistic.
Great horned owl livers
When data from all age classes were considered together,
mean concentrations of TEQWHO in owl livers from the RMA
ranged from 330 to 340 pg TEQWHO/g wet wt liver, while the
mean concentration of TEQWHO in owl livers from reference
areas ranged from 35 to 42 pg TEQWHO/g wet wt liver, depending on the proxy values used for congeners occurring at
concentrations less than the MDL (Table 3). In samples from
Environ. Toxicol. Chem. 20, 2001
2,3,7,8-TCDD equivalents in wildlife tissue samples
Table 3. A range of mean concentrations (pg/g, wet wt) of TCDD
equivalents in great horned owl livers and the significance of
differences among locations as determined by the Mann–Whitney U
testa
Concentration ranges of TCDD
equivalents in combined age classes
On-site
Reference
p value
Combined
Adults
Juveniles
a
TEQMIN–
TEQMAX
TEQREP-MIN–
TEQREP-MAX
TCDD-EQMIN–
TCDD-EQMAX
330–340
35–42
230–260
18–30
130–130
17–32
0.077
—
—
—
—
—
—
0.014
0.604
TEQ 5 TCDD-equivalent value as determined by instrumental analyses, TEQREP 5 relative potency of instrumentally derived TCDD
equivalents on the H4IIE-luc cell line, TCDD-EQ 5 TCDD-equivalent value as determined by the H4IIE-luc cell line.
the RMA and reference areas, mean proportions of total concentrations of TEQWHO contributed by PCBs ranged from 24
to 39%, depending on the proxy values chosen. The relative
contribution of PCB congeners to the total TEQWHO was greater
in owl livers than in carp egg samples. Overall, there was
greater variation in TEQWHO concentrations among individual
owl liver samples collected on and off the RMA than was
evident in TEQWHO concentrations of carp egg samples (Table
1). Specific congener concentrations and concentrations of
TCDD equivalents were calculated for separate age classes
and location of collection (Table 1). When owl liver lipid
contents were stratified by age and location, the lipid content
of adult owls on site was less than the lipid content of all other
groups, including juveniles (Table 1).
When concentrations of TEQREP in livers of owls of all ages
were considered together, the mean concentrations of TEQREP
in great horned owl livers from the RMA ranged from 230 to
260 pg TEQ/g wet wt liver, while the mean concentration of
TEQREP in samples from reference locations ranged from 18
to 30 pg/g wet wt liver (Table 3). Mean concentrations of
TCDD-EQ in owl livers collected from reference locations
ranged from 17 to 23 pg TCDD-EQ/g wet wt liver, and the
mean concentration in samples from the RMA was 130 pg
TCDD-EQ/g wet wt liver (Table 3).
Concentrations of TEQREP-MAX were, on average, greater
than concentrations of TCDD-EQMAX (Fig. 1B). Concentrations of TCDD-EQMAX and TEQREP-MAX were significantly different (x2 5 86, df 5 26) when all data points were considered.
When a single, outlying point was removed from the analysis,
TCDD-EQMAX and TEQREP-MAX concentrations were not statistically different (x2 5 14.2, df 5 25).
Since concentrations of TEQ in livers of individual great
horned owls were not normally distributed, the Mann–Whitney
U test was used to test for significant differences between adult
and juvenile owls [16]. Because the ages of some owls could
not be determined, statistical tests for differences in concentrations of TCDD equivalents between owls of different ages
were conducted several ways. When owls of unknown age
were classified as adults, classified as juveniles, or eliminated
altogether, no significant differences were observed in concentrations of TEQWHO-MAX among age classes. Therefore, comparisons between concentrations of TEQWHO-MAX from on-site
and reference areas were performed on pooled age classes
(Table 3). Concentrations of TCDD-EQMAX were significantly
2439
different among age classes. Therefore, comparisons of concentrations of TCDD-EQMAX between the RMA and reference
locations were performed separately for adults and juveniles.
Concentrations of TEQWHO-MAX between locations were not significantly different. However, concentrations of TCDD-EQMAX
in livers of adult owls from the RMA and reference areas were
significantly different (Table 3).
American kestrel eggs
Mean concentrations of TEQWHO in American kestrel eggs
collected from the RMA ranged from 21 to 24 pg TEQ/g wet
wt egg, while those collected from reference locations were
between 58 and 60 pg TEQ/g wet wt egg (Table 1). The mean
relative proportions of TEQWHO contributed by PCBs ranged
from 21 to 57% among samples and methods of calculation,
which is similar to the proportion observed in great horned
owl livers.
Mean concentrations of TEQREP in American kestrel eggs
from the RMA ranged from 15 to 19 pg TEQ/g wet wt egg,
while the mean concentration of TEQREP in samples from reference locations ranged from 20 to 22 pg/g wet wt egg (Table
1). Mean concentrations of TCDD-EQ in samples of American
kestrel eggs from the RMA contained 2.4 to 6.9 pg TCDDEQ/g wet wt egg, while those from reference locations ranged
from 14 to 18 pg TCDD-EQ/g wet wt egg (Table 1).
Concentrations of TEQREP-MAX were compared with TCDDEQMAX (Fig. 1C). As with the owl livers, concentrations of
TEQREP-MAX in the kestrel eggs were, on average, greater than
concentrations of TCDD-EQMAX. Concentrations of TEQREPMAX were statistically different from TCDD-EQMAX when all
data points were analyzed (x2 5 668, df 5 56, a 5 0.05). By
removing a single outlier from the analysis, TCDD-EQMAX and
TEQREP-MAX concentrations in American kestrel eggs from onsite and reference areas were not significantly different (x2 5
40, df 5 55, a 5 0.05).
Concentrations of TCDD equivalents in kestrel eggs collected from the core and periphery areas of the RMA as well
as the reference locations were not statistically different among
locations (df 5 2,43, F ratio 5 1.345; p 5 0.271). Therefore,
concentrations of both TEQ and TCDD-EQ in eggs from core
and peripheral area were grouped, and concentrations in all
kestrel eggs from the RMA were compared to those in the
eggs pooled from all the reference locations. Concentrations
of TEQWHO-MAX or TCDD-EQMAX between locations were not
significantly different. However, the power of the test was
insufficient to allow definitive conclusions to be drawn (Table
4).
Pattern analysis of PCDD and PCDF congeners
Pattern recognition of the relative concentrations of PCDD
and PCDF congeners was used to determine if the PCDDs and
PCDFs were from a local or more regional source. The results
of the PCA is presented by plotting the first three principal
components (Figs. 2 and 3). Because of the small number of
reference samples and the small concentrations of PCDDs and
PCDFs measured in the carp eggs, pattern recognition techniques could not be usefully applied to the carp data set.
No distinct cluster was apparent among adult owls from
the RMA (Fig. 2). In fact, the samples ordinate in different
directions away from the central cluster. This observation supports the conclusion that the PCDD and PCDF congeners detected in tissue samples from the RMA were more likely to
be due to a regional rather than a local source.
2440
Environ. Toxicol. Chem. 20, 2001
K.K. Coady et al.
Table 4. Mean concentrations (pg/g, wet wt) of TEQMAX and TCDDEQMAX in American kestrel eggs collected from the RMA or reference
locations (standard deviation in parentheses). Student’s t test results
for the comparison of concentrations in eggs from the RMA and those
from reference locationsa
On-site
Reference
p value
Separated
Poolede
N
TEQMAX
TCDD-EQMAX
30
16
24 (17.7)
60 (107.8)
6.9 (11.0)
18 (34.0)
0.241
0.105
0.246
0.214
RMA 5 Rocky Mountain Arsenal.
TEQ 5 TCDD-equivalent value as determined by instrumental analyses.
c TCDD-EQ 5 TCDD-equivalent value as determined by the H4IIEluc cell line.
d Separate 5 tests using separate variances.
e Pooled 5 tests using pooled variances.
a
b
The results of the PCA for American kestrel eggs is presented (Fig. 3). This graphical representation is dominated by
a close grouping of both reference and RMA samples. This
suggests that the patterns of relative concentrations of PCDDs
and PCDFs in the eggs result from a common regional or global
source, not a local source. This grouping may, to some extent,
be the product of a significant number of nondetect values that
can cause clustering. Ordinated apart from the central cluster
were several samples from both the RMA and the reference
locations. The samples that are separate from the central cluster
do not form a separate grouping but are distributed around the
graph.
DISCUSSION
Two different methodologies to estimate TCDD equivalents
were applied in this study to ensure that no unaccounted for
AhR-mediated activity was present in the tissue samples. Both
bioanalytical and instrumental techniques have advantages and
disadvantages. In comparison to instrumental analyses, the
H4IIE-luc bioassay can be more cost effective and rapid, and
Fig. 2. Principal components analysis (PCA) of great horned owl liver
polychlorinated dibenzofurans (PCDD/F) profiles. ou 5 on-post, unknown age; oa 5 on-post adult; oj 5 on-post juvenile; ra 5 reference
adult; and rj 5 reference juvenile.
Fig. 3. Principal components analysis (PCA) for American kestrel
eggs (C 5 core, P 5 periphery, R 5 reference).
it integrates all the dioxin-like activity in complex mixtures.
However, instrumental analyses can identify individual congeners. Depending on the study objectives, one method for
determining TCDD-equivalent concentrations may be preferred over the other. Discovering the point source and assigning causality to contamination would require a pattern
analysis of specific congener concentrations among samples,
which is accomplished by instrumental analyses. If a rapid and
biologically relevant assay is desired to screen for TCDDequivalent concentrations in samples, then the H4IIE-luc bioassay would be favored.
Concentrations of TCDD-EQ and TEQREP values were similar in most of the tissues analyzed in this study. Significant
differences in concentrations of TCDD equivalents were observed, depending on which method was used to estimate
TCDD equivalents. Three outlying carp egg samples collected
on the RMA contained greater concentrations of TCDD-EQ
than predicted by concentrations of TEQREP. This can be explained by the fact that the bioassay measures total activity of
complex mixtures and accounts for interactions among congeners, while instrumental analyses measured only target analytes [5]. The H4IIE-luc bioassay can also detect additional
dioxin-like activity from compounds not specifically measured
in analytical chemistry procedures. The greater concentration
of TCDD-EQ in carp eggs may have been due to unidentified
AhR-active compounds. In a few samples of kestrel eggs and
owl livers, concentrations of TEQREP were greater than concentrations of TCDD-EQ. This outcome was possibly observed
due to the fact that congeners, when tested as a mixture in the
bioassay, can antagonize one another, causing interference and
reduced measures of dioxin-like activity that are not predictable in the additive TEQREP model [5]. Thus, in the absence
of unquantified AhR-active compounds, the concentration of
TCDD-EQ would be expected to be slightly less than that of
TEQREP. Overall, the good correlation between TCDD-EQ and
TEQREP values indicated that little AhR-mediated activity occurred that was due to compounds other than PCDDs, PCDFs,
and PCBs.
Carp eggs collected from reference locations contained significantly lesser concentrations of TEQWHO-MAX and TCDDEQMAX than did carp eggs collected on the RMA. However,
2,3,7,8-TCDD equivalents in wildlife tissue samples
the small sample sizes available for this analysis (n 5 2 from
reference areas) may not be representative of the locations
from which they were collected. Therefore, it is inconclusive
whether concentrations of TCDD equivalents are greater in
samples of carp eggs from the RMA. While the average concentrations were significantly different, they were small. The
contribution of TCDD-EQ that was contributed by proxy values was considerable. Thus, the differences observed may be
due more to differences in sample-specific detection and quantification limits than to actual concentrations of PCDHs. In
addition, the concentrations measured are similar to or lower
than background concentrations measured in national and
global surveys [18,19].
In this study, livers of great horned owls contained the
greatest concentrations of TEQs and TCDD-EQs. Concentrations of TEQs and TCDD-EQs in livers from juvenile great
horned owl populations did not significantly differ among the
RMA and reference locations. However, concentrations of
TCDD-EQMAX in livers of adult great horned owls collected
from the RMA were significantly greater than those of adult
owl livers collected from surrounding reference areas. This
may indicate that a localized source of dioxin-like compounds
exists in the terrestrial environment at the RMA. However,
some uncertainty persists in this result since great horned owl
specimens were collected opportunistically. Some owls were
found dead in response to an environmental stressor other than
exposure to PCDHs. Based on the concentrations of dieldrin
measured in several adult owls, it was concluded that these
specimens had expired because of dieldrin poisoning. Loss of
lipid content occurs in the liver tissue of organisms that die
because of dieldrin poisoning [20]. This loss of lipid content
in dieldrin-poisoned owls may have led to an increased partitioning of TCDD equivalents in liver samples. In addition,
several studies with rainbow trout have shown that pretreatment with dieldrin can cause alterations in the tissue distributions of subsequent doses of dieldrin, namely increased dieldrin concentrations in the liver [21,22]. Therefore, dieldrinpoisoned owls may have had increased concentrations of
TCDD equivalents partitioned to their livers.
Concentrations of TEQ and TCDD-EQ in American kestrel
eggs collected from the RMA were not significantly different
from those collected from reference locations, but the power
in these analyses was low so that the results were inconclusive.
The differing results in overall TCDD-EQ concentrations and
TEQ concentrations between American kestrel eggs and great
horned owl livers may indicate that separate terrestrial pathways exist for exposure to dioxin-like chemicals among species.
Pattern analysis of specific congener occurrences in wildlife
tissue samples and in soil samples from the RMA (Biological
Advisory Subcommittee, Commerce City, CO, USA, unpublished results) revealed that no consistent pattern of congeners
was associated with the core or periphery of RMA that would
distinguish these samples from samples collected in reference
areas. This indicates that a single distinct pattern of PCDD/F
congeners common to these samples does not exist, as would
be expected if a point source of contamination existed on the
RMA.
In addition to measuring the concentrations of TCDD
equivalents in tissue samples from the RMA and surrounding
reference areas, this study was designed to compare the results
of chemical and bioanalytical methods for measuring PCDHs.
While in general agreement, the two procedures also dem-
Environ. Toxicol. Chem. 20, 2001
2441
onstrated differences. Notably, when PCDH concentrations
were small, as in the carp egg samples, estimated TCDD-EQ
concentrations were greater than TEQ. However, when PCDH
concentrations were large, as in the great horned owl, TEQs
appeared to overestimate the total PCDH activity. These differences could be attributed to deficiencies in the bioassay
method due to false positives at low PCDH concentrations,
nonadditivity, or saturation at high PCDH concentrations. Differences between concentrations of TEQs and TCDD-EQs
could also be attributed to deficiencies in the TEQ method,
including nonmeasured dioxin-like chemicals at low PCDH
concentrations and the inability to account for interactive effects or saturation at high PCDH concentrations. These differences in interpretation again highlight the fact that the two
methods provide complementary, not necessarily comparative,
data.
Acknowledgement—Funding for this project was provided by the
Rocky Mountain Arsenal, U.S. Army. Major technical oversight of
this project was provided by the Biological Advisory Subcommittee
that was established to evaluate ecological risks at the Rocky Mountain Arsenal.
REFERENCES
1. Giesy JP, et al. 1994. Contaminants in fishes from Great Lakes
influenced sections and above dams of three Michigan rivers. I:
Concentrations of organochlorine insecticides, polychlorinated
biphenyls, dioxin equivalents, and mercury. Arch Environ Contam Toxicol 27:202–212.
2. 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.
3. Peterson RE, Theobold HM, Kimmel GL. 1993. Development
and reproductive toxicity of dioxins and related compounds:
Cross-species comparisons. Crit Rev Toxicol 23:283–335.
4. Giesy JP, Kannan K. 1998. Dioxin-like and non-dioxin-like toxic
effects of polychlorinated biphenyls (PCBs): Implications for risk
assessment. Crit Rev Toxicol 28:511–569.
5. Sanderson JT, Aarts JMMJG, Brouwer A, Froese KL, Dennison
MS, Giesy JP. 1996. Comparison of Ah receptor mediated luciferase and ethoxyresorufin-O-deethylase induction in H4IIE cells:
Implications for their use as bioanalytical tools for the detection
of polyhalogenated aromatic hydrocarbons. Toxicol Appl Pharmacol 137:316–325.
6. Jones PD, et al. 1993. 2,3,7,8 Tetrachlorodibenzo-p-dioxin equivalents in tissues of birds at Green Bay, Wisconsin, USA. Arch
Environ Contam Toxicol 24:345–354.
7. Williams LL, Giesy JP, Verbrugge DA, Jurzysta S, Stromborg K.
1995. Polychlorinated biphenyls and 2,3,7,8-tetrachlorodibenzop-dioxin equivalents in eggs of double crested cormorants from
a colony near Green Bay, Wisconsin, U.S.A. Arch Environ Contam Toxicol 29:327–333.
8. Xu Y, Wu WZ, Li W, Schramm KW, Kettrup A. 1997. Toxic
effects of 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD) during a
different life stage test with Chinese rare minnow (Gobiocypris
rarus). Organohalogen Compounds 33:371–376.
9. Couture LA, Abbott BD, Birnbaum LS. 1990. A critical review
of the developmental toxicity and teratogenicity of 2,3,7,8-tetrachlorodibenzo-p-dioxin: Recent advances toward understanding
the mechanism. Teratology 42:619–627.
10. 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.
11. Giesy JP, et al. 1997. Polychlorinated dibenzo-p-dioxins, dibenzofurans, biphenyls and 2,3,7,8-tetrachlorodibenzo-p -dioxin
equivalents in fishes from Saginaw Bay, Michigan. Environ Toxicol Chem 16:713–724.
12. U.S. Fish and Wildlife Service. 1996. Rocky Mountain Arsenal
National Wildlife Refuge comprehensive management plan. Commerce City, CO.
13. Kannan K, Watanabe I, Giesy JP. 1998. Congener profile of polychlorinated brominated dibenzo-p-dioxins and dibenzofurans in
2442
14.
15.
16.
17.
18.
Environ. Toxicol. Chem. 20, 2001
soil and sediments collected at a former chlor-alkali plant. Toxicol
Environ Chem 67:135–146.
Kemler K. 2000. 2,3,7,8-Tetrachlorodibenzo-p-dioxin equivalents
in tissue samples in three species from the Denver metropolitan
area. MS thesis. Michigan State University, East Lansing, MI,
USA.
Villeneuve DL, Blankenship AL, Giesy JP. 2000. Derivation and
application of relative potency estimates based on in vitro bioassay results. Environ Toxicol Chem 19:2835–2843.
Kuehl RO. 1994. Statistical Principles of Research Design and
Analysis. Wadsworth, Belmont, CA, USA.
Rand GM. 1995. Fundamentals of Aquatic Toxicology. Taylor &
Francis, Washington, DC, USA.
Buckland SJ, Jones PD, Ellis HK, Salter RT. 1998. Organochlorines in New Zealand: Ambient concentrations of selected or-
K.K. Coady et al.
19.
20.
21.
22.
ganochlorines in rivers. Ministry for the Environment, Wellington, New Zealand.
U.S. Environmental Protection Agency. 1992. National study of
chemical residues in fish. EPA 823-R-92-008a. U.S. Environmental Protection Agency, Washington DC.
Stickel WH, Stickel LF, Spann JW. 1969. Chemical fallout; current
research on persistent pesticides. Proceedings, 1st Rochester Conference on Toxicity. Springfield, IL, USA, pp 174–204.
Gilroy DJ, Miranda CL, Siddens LK, Zhang Q, Buhler DR, Curtis
LR. 1995. Dieldrin pretreatment alters [14C]dieldrin and [3H]7,12dimethylbenz[a]anthracene uptake in rainbow trout liver slices.
Fundam Appl Toxicol 30:187–193.
Gilroy DJ, Carpenter HM, Siddens LK, Curtis LR. 1993. Chronic
dieldrin exposure increases hepatic disposition and biliary excretion of [14C]dieldrin in rainbow trout. Fundam Appl Toxicol 20:
295–301.