Full text of DOI:10.1002/etc.5620210812

Environmental Toxicology and Chemistry, Vol. 21, No. 8, pp. 1618–1627, 2002
2002 SETAC
Printed in the USA
0730-7268/02 $9.00 .00
᭧
ϩ
DISTRIBUTION OF CYTOCHROME P4501A1–INDUCING CHEMICALS IN SEDIMENTS
OF THE DELAWARE RIVER–BAY SYSTEM, USA
D
ANIEL L. M
C
C
OY,† JENNIFER M. JONES,‡ JACK W. ANDERSON,*† MICHELLE
AWED AMEEDI
†Columbia Analytical Services, 1185 Park Center Drive, Suite A, Vista, California 92083, USA
‡Department of Toxic Substance Control, 1011 North Grandview Avenue, Glendale, California 91201, USA
§National Oceanic and Atmospheric Administration, 1305 East–West Highway, Silver Spring, Maryland 20910, USA
HARMON,§ IAN HARTWELL,§ and
J
H
§
(
Received 25 July 2001; Accepted 4 February 2002)
Abstract—The Delaware River–Bay system, USA, was the subject of a study by the National Oceanic and Atmospheric Admin-
istration that involved chemical and biological analyses, including the use of the biomarker P450 human reporter gene system
(HRGS) to document the occurrence and distribution of cytochrome P450 (CYP) 1A1–inducing compounds. Sediment extracts
from 81 locations along the Delaware River, Delaware Bay and immediate coastline were tested by utilizing HRGS as an inexpensive
screening test, and were also analyzed for polycyclic aromatic hydrocarbons (PAHs) and polychlorinated biphenyls, with selected
stations analyzed for dioxins and furans. Benthic community degradation has been observed when benzo[
(B PEq) exceeded 60 g/g. The average levels of B PEq for the largely industrialized upper, middle, and lower regions of the
Delaware River were 107, 62, and 5 g/g, respectively, excluding outliers. Tributaries leading into river averaged 21 g/g B PEq,
PEq, respectively). The HRGS values were
0.81). Overall, contamination levels consistently
a]pyrene equivalents
a
␮
a
␮
␮
a
whereas the central Bay and open coast had relatively low values (2.0 and 0.5 ␮g/g Ba
highly correlated with total PAHs measured in the same sediment samples (r2
ϭ
decreased from the upper and middle river sites as collection locations progressed down through the lower river and bay to the
coast. Thus, despite the relatively high contaminant load in the river system, Delaware Bay and the immediate coastline seem to
have relatively low levels of contaminants, and, therefore, impacts on the benthic organisms in the bay and coast would not be
expected from these findings.
Keywords—Delaware River
Polycyclic aromatic hydrocarbons
Dioxins
Human reporter gene system
INTRODUCTION
Delaware Bay is one of the largest estuaries on the East
years, a human reporter gene system (HRGS) has been added
to determine the presence of contaminants that induce the cy-
tochrome P450 (CYP) 1A1 gene. In an investigation of San
Diego Bay, California, USA, Fairey et al. [9] found that sed-
Coast of the United States, with a vast drainage area that is
located in the states of Delaware, New Jersey, New York, and
Pennsylvania. Although impairment of coastal environmental
quality and degradation of coastal ecosystems due to popu-
lation increase and industrialization in the region were rec-
ognized in the 1930s, concerns about the widespread occur-
rence of toxicants in the bay are relatively more recent [1].
Several sediment toxicity studies have been carried out in the
bay during the past 10 years, focusing on specific sources of
contaminants or areas of concern [2–4].
As part of its environmental stewardship mission, the Na-
tional Oceanic and Atmospheric Administration (NOAA) con-
ducts studies to determine the extent and severity of environ-
mental contamination and its associated biological effects in
coastal waters and estuaries of the United States. These studies
primarily focus on sediment contamination and are based on
the sediment quality triad approach [5,6]. The approach con-
sists of measurements of sediment contaminant levels, sedi-
ment toxicity, and parameters of benthic community structure.
In 1997, NOAA initiated a geographically comprehensive
study to examine the spatial extent, patterns, and severity of
sediment toxicity in the Delaware River and Bay. Typically,
the amphipod mortality test, the sea urchin fertilization im-
iments exhibiting HRGS responses of 60
␮g/g benzo[a]pyrene
(B P) equivalents (B PEq) and greater contained infaunal
a
a
communities that were classified as degraded, accounting for
50% of the stations that were sampled for both parameters.
The induction of CYP 1A1 in the CYP gene family has
been widely used as a biomarker for exposure to toxic or
carcinogenic organic compounds in the aquatic environment.
Organic compounds such as dioxins, furans, coplanar poly-
chlorinated biphenyls (PCBs), and the higher molecular weight
polycyclic aromatic hydrocarbons (PAHs) are known to bind
to intracellular cytosolic protein referred to as the aryl hydro-
carbon receptor. Upon binding, this complex is translocated to
the nucleus of the cell, where it interacts with xenobiotic re-
sponse elements in the promoter region of the CYP 1A1 gene
and causes transcription of the P450 enzyme system.
The induction of the CYP gene family, specifically CYP
1A1, in response to PAHs has been extensively studied in fish
[10–12]. These studies have demonstrated that high levels of
P450 in fish are correlated with increased sediment contami-
nation, especially by PAHs, and histological and reproductive
effects on the fish. In recent NOAA studies, HRGS responses
have correlated well (
r
2 ϳ 0.7–0.8) with the concentrations of
0.85) with the
pairment test, and the Microtox௡ test (AZUR Environmental,
Carlsbad, CA, USA) have been used [7,8]. However, in recent
total PAHs in sediments, and even better (r2
ϳ
concentrations of PAHs containing between four and six ben-
zene rings (molecular weights ranging from 228 to 302) [13].
The HRGS assay has also been used to detect PAH contam-
* To whom correspondence may be addressed
(janderson@vista.caslab.com).
1618

CYP 1A1–inducing chemicals in the Delaware River and Bay, USA
ination in sediments from the coastal waters of Korea [14,15].
Environ. Toxicol. Chem. 21, 2002
1619
In addition, HRGS B
a
PEq obtained in analysis of extracts of
0.9) with the pres-
mussel tissue were highly correlated (r2
ϭ
ence of four- to six-ring PAHs reported from chemical ana-
lytical data [16]. Strong correlations have been demonstrated
between HRGS responses and subsequent high-resolution gas
chromatography–mass spectrometry analyses of chlorinated
compounds (dioxins and furans) in contaminated soil samples
[17,18].
The P450 HRGS utilizes a human liver cancer cell line,
which has been stably transfected with a plasmid containing
CYP 1A1 promoter regions fused to the firefly luciferase gene.
When the cells are exposed to compounds that induce CYP
1A1, luciferase is produced and can be detected by a simple
assay that measures relative light units with a luminometer.
The potencies of several PAHs and coplanar PCBs for this test
system have been described [19]. The HRGS (also known as
U.S. Environmental Protection Agency method 4425 [20]) has
gained acceptance as a rapid and inexpensive in vitro approach
to screening solvent extracts of environmental samples of soil,
sediment, tissue, and water. As an alternative to chemical anal-
ysis of all samples in an environmental assessment, the HRGS
detects the total amount of compounds that act via the aryl
hydrocarbon receptor to induce CYP 1A1. As such, the HRGS
provides a measure of the inducing chemicals in a sample with
the potential to cause harmful health effects. When only di-
oxins, furans, and coplanar PCBs are the concern, extracts are
first cleaned of PAHs with silica gel and carbon columns (ex-
traction method for U.S. Environmental Protection Agency
method 8290 [www.epa.gov/epaoswer/hazwaste/test/8290.pdf]).
The objectives of this paper are to document the occurrence
and distribution of CYP 1A1–inducing compounds in surficial
sediment of the Delaware River–Bay system; relate the HRGS
responses to the levels of PAHs, coplanar PCBs, and dioxins
and furans in the sediment; and infer likely adverse biological
effects associated with these chemicals.
Fig. 1. Map of Delaware River, Delaware Bay, USA, and immediate
coastline with labeled station numbers.
was evaporated to 2 ml. The dichloromethane extracts were
exchanged into dimethylsulfoxide to better facilitate the HRGS
P450 assay. Two 1-ml vials were prepared from each extract
for the purpose of testing one vial by HRGS and the second
vial by Microtox [21]. Additional sediment from each sample
location also was prepared and used to test sea urchin gamete
survival at three pore-water concentrations according to meth-
ods described by Carr [22]. A small subsample of each of the
81 sediments was used to measure percent solids. For purposes
of data interpretation and description, samples were placed
into six different categories based on their location within the
river, bay, or ocean (Table 1). The location of each sample can
be found on the map in Figure 1. Because stations 54 and 66
are located in the middle of the mouth of the central bay and
are subject to a tidal regime characteristic of stations along
the coastline, they were considered part of the open coast
region (Table 1).
MATERIALS AND METHODS
Sampling
The study area extended from Trenton, New Jersey, USA,
to the coastal area outside the mouth of the bay, encompassing
approximately 2,350 km2. Sampling sites were selected on a
random basis within well-defined sampling strata, whose
boundaries were established in consideration of regional ba-
thymetry, hydrography, and surficial sediment distribution, and
to include areas of concern as expressed by state and local
resource managers (Fig. 1). The total number of sampling sites
was 81.
Surface sediment samples were collected by NOAA in Sep-
tember 1997, by using a Kynar-coated modified van Veen grab.
Each sample was visually inspected to determine if the surface
of the sample was undisturbed and contained fine-grained ma-
terial. If acceptable, 2 to 3 cm of the surficial sediment was
removed with a sterile high-impact styrene scoop and placed
into a polyethylene bucket. When enough material had been
collected, the sample was homogenized with a polyethylene
paddle and distributed among the various sample containers.
Samples to be tested by HRGS P450 were sent to the Columbia
Analytical Services laboratory in Jacksonville (FL, USA) for
extraction.
Test methods
The HRGS cell culturing methodology used in this study
has been described previously [23–24]. Human liver cancer
cells (101L cells) were grown as monolayers in an atmosphere
of 5% CO2 and 100% humidity at 37
essential media (Mediatech, Herndon, VA, USA), supple-
mented with 10% fetal bovine serum, 2% -glutamine, 1%
ЊC in Eagle’s minimum
L
The extraction procedure was done according to the U.S.
Environmental Protection Agency method 3550 (www.epa.gov/
epaoswer/hazwaste/test/3550b.pdf) and the dichloromethane
sodium pyruvate, and 0.4 mg/ml of genticin (all from Sigma,
St. Louis, MO, USA). Cells were split upon reaching a max-
imum of 90% confluence, and discarded after 25 passages.

1620
Environ. Toxicol. Chem. 21, 2002
D.L. McCoy et al.
In preparation for testing, cells were subcultured into six-
well plates at a density of 2.5
grow for 72 h in the environment described above to reach a
density of approximately 1
106 cells/well. Triplicate test
wells were inoculated with 10 l of a given sample extract
before incubating the plates at 37 C for 16 h. Standards, con-
sisting of a 1 ng/ml solution of 2,3,7,8-tetrachlorodibenzo-
dioxin in dimethylsulfoxide, a 300 ng/ml B P spike in di-
ϫ
105 cells/well and allowed to
ϫ
␮
Њ
p
-
a
methylsulfoxide, and a dimethylsulfoxide solvent blank, were
also applied to wells in triplicate. At the end of the exposure,
the cells were rinsed with Hank’s balanced salt solution, and
lysed with 200 ␮l of luciferase 3ϫ cell lysis buffer (Promega,
Madison, WI, USA). Plates were scraped and the cell lysates
were centrifuged at 6,000 rpm for 10 s. Fifty-microliter ali-
quots of the supernatant were then applied to a 96-well plate,
followed by 100
actions were initiated by timed injection from the luminometer
of 100 l of substrate B (luciferin). Luminescence was mea-
␮l of substrate A (buffer and cofactors). Re-
␮
sured in relative light units with a ML2250 Luminometer (Dy-
natech Laboratories, Chantilly, VA, USA). Luciferase assay
buffers were purchased from Pharmingen (San Diego, CA,
USA). When extracts were tested at two time periods, two
replicates of each extract were applied to duplicate plates at
the same time, so that one plate was incubated for 6 h, and
the other for 16 h [23].
Means of the dimethylsulfoxide solvent blanks were used
as a fold induction baseline and set equal to 1, whereas the
fold induction of each sample extract and reference inducer
(2,3,7,8-tetrachlorodibenzo-p-dioxin and BaP) was normalized
by dividing the mean relative light units produced by the sam-
ple by the mean relative light units produced by the solvent
blank. As a quality assurance–quality control measure, a long-
term control chart for 2,3,7,8-tetrachlorodibenzo-p-dioxin has
been logged with each test run since the development of the
test. For the Delaware study, 13 test runs with 1.0 ng/mL of
2,3,7,8-tetrachlorodibenzo-p-dioxin produced a mean of
117.2-fold induction, a standard deviation of 15.0, and a co-
efficient of variation of 12.8%. These numbers illustrate the
low variability of cellular response to the reference inducer
and solvent blank. The test results for a sample were acceptable
if both the coefficient of variation was less than 20% and the
fold induction was less than 100. Any samples not meeting
these criteria were retested; if the fold induction exceeded 100,
the samples were first diluted and then retested. Fold induction
from the 2,3,7,8-tetrachlorodibenzo-p-dioxin standard over
100 is acceptable, because the concentration response curve
for dioxin does not flatten out above 100, as does the curve
for PAHs, which possibly is caused by toxicity. Only accept-
able tests and final dilutions were used to determine fold in-
duction and BaPEq.
Analysis
The HRGS fold induction response at 16 h of exposure was
used to produce a quantitative estimate of the levels of CYP
1A1–inducing compounds in all sample extracts. All standard
curves and analyses of environmental samples have been pro-
duced from the standard 16-h time interval. During this study
a standard spike of 300 ng B
and these produced an average 14.4-fold induction with a stan-
dard deviation of 2.93. The HRGS determination of B PEq
aP/ml was used with the tests,
a
in all previous studies of coastal sediments for NOAA has
utilized a conversion factor of 60-fold equal to 1 part per

CYP 1A1–inducing chemicals in the Delaware River and Bay, USA
Environ. Toxicol. Chem. 21, 2002
1621
Chemical methods
Chemical analysis methods for quantifying PAHs and their
alkylated homologues by gas chromatography–mass spectrom-
etry in this study were performed by NOAA contractors ac-
cording to the appropriate methods under the National Status
and Trends Program [26]. Also following procedures described
for this program, PCB analyses were quantitatively determined
by NOAA contractors through capillary gas chromatography
with an electron-capture detector and dioxins and furans were
measured by high-resolution gas chromatography and mass
spectrometry. Toxic equivalency quotients (TEQs) were de-
termined from the analyses of dioxins and furans, by using
the World Health Organization (WHO) [27] TEQs. The TEQ
values for coplanar PCBs were not calculated, because only
5 (77, 126, 169, 105, and 118) of the 12 (nonortho- and mono-
ortho-) congeners evaluated by the WHO [27] were included
in the analyses of 24 samples and only 2 (105 and 118) were
measured in all samples. Of the five samples tested for dioxins
and furans, only one was analyzed for the content of three (of
the four) coplanar PCBs.
Fig. 2. Distribution of benzo[a]pyrene equivalents (BaPEq) taken
from 81 sediment samples along the Delaware River, Delaware Bay,
and Delaware, USA, coast. Categories were based on hydrogeography,
salinity, sediment type, and location. Samples exceeding 60
PEq/g represented by the line would be expected to exhibit de-
graded benthic communities [9].
␮g
B
a
million of B
curves [13]. The HRGS B
a
P, based on previous concentration–response
PEq are calculated as follows:
a
HRGS B PEq (fold induction/60)·[(Ve
e is the total extract volume, a is the volume of extract
applied to cells, and d is the dry weight of sample. The above
equation yields HRGS B PEq in g/g dry weight, and assumes
that all CYP 1A1–inducing compounds in the samples are
PAHs. Although the data were expressed as B PEq, the mea-
a
ϭ
/Va)/Wd]
RESULTS
where
V
V
Distribution of P450 HRGS responses
W
a
␮
The HRGS BaPEq for each portion of the Delaware River–
Bay system, as defined in Table 1, are plotted on a log scale
in Figure 2. Values ranged widely among the 81 samples,
a
sured responses at 16 h of exposure may have included in-
duction from coplanar PCBs, dioxins, or furans. However, ex-
pressing the HRGS results as BaPEq is appropriate because
chemical findings have shown that PAHs generally are the
dominant class of organic contaminants within estuarine and
marine sediments.
varying from 1,584
␮
g B
PEq/g in sample 51. The mean for all samples was 57.4
PEq/g, with an upper 99% confidence interval of 112
PEq/g and a standard deviation of 190.8 g B
a
PEq/g in sample 11 to 0.22
␮
␮
␮
g
g
g
B
B
B
a
a
a
␮
a
PEq/g. The
highest sediment value observed in this investigation was lo-
cated on the eastern bank of the Delaware River, across from
the Philadelphia (PA, USA) airport in close proximity to sev-
eral other high samples. The lowest value was found in the
central bay region. The vast majority of samples with high
To determine if samples with the highest levels of induction
were due to chlorinated hydrocarbons, two time interval tests
were performed on selected samples. Laboratory analysis and
testing have found that HRGS responses over time can predict
if induction is mainly a result of PAHs or chlorinated hydro-
carbons [25]. This is done by means of two-time-interval test-
ing at 6 h and 16 h. If 16-h responses drop by approximately
60 to 80% in relation to 6-h responses (a 6-h:16-h response
ratio of 5 or greater), the induction can be assumed to be a
result of PAHs (cells metabolize PAHs). Conversely, if the
fold induction increases or stays the same from 6 h to 16 h
(response ratio of 1 or less), a major portion of the response
can be attributed to chlorinated compounds such as dioxins,
furans, and coplanar PCBs. Response ratios between one and
five indicate approximately equal contribution from PAHs and
chlorinated organic compounds.
B
a
PEq values were found along the upper and middle portions
of the river. These sections of the river contained 10 of the 12
sites having higher than 60 g B PEq/g. The mean for the
upper river section, which contained no outliers was 106.7
PEq/g, whereas the middle river average changed from
162.6 to 62.3 g B PEq/g when outliers were excluded (Table
2). The lower portion of the river, comprised of 14 sampling
locations, had a mean of 27.7 g B PEq/g (5.3 g B PEq/g
excluding outliers) and contained only one site with a value
above 60 g B PEq/g. The central bay and open coastal areas
around the mouth of the bay showed primarily background
levels, ranging from 0.2 to 7.2 g B PEq/g (Fig. 2).
Table 2 lists the number of samples, the means, and standard
␮
a
␮
g
B
a
␮
a
␮
a
␮
a
␮
a
␮
a
Table 2. Summary of human reporter gene system responses to regional samplesa
Mean
B
a
g/g)
PEq
SD
g/g)
CV
(%)
Region
Stations
n
(
␮
(
␮
Upper river
Middle river
Lower river
Tributaries
Central bay
Open coast
1–7
8–25
26–39
84, 85, 87–92
40–61 (excluding 54)
54, 62–73
7
106.6
162.6
27.7
67.7
2
116.4
367.1
78.2
133.8
1.7
109.2
225.8
282.3
197.6
85.0
18
14
8
21
13
0.7
0.7
100.0
a B
(in percent).
a
PEq
ϭ
benzo[
a
]pyrene equivalents; SD ϭ standard deviation; CV ϭ coefficient of variation

1622
Environ. Toxicol. Chem. 21, 2002
D.L. McCoy et al.
h responses all decreased markedly as a result of the additional
dilution. These findings indicated that both PAHs and chlo-
rinated hydrocarbons (dioxins, furans, and coplanar PCBs)
were present in the samples.
Chemical analyses
Jones and Anderson [19] previously established HRGS tox-
ic equivalency factors (TEFs) for eight of the PAHs analyzed.
By using the concentrations of each of these PAHs and their
respective HRGS TEFs, the chemical BaPEq (expected HRGS
response) values were calculated for each sample. Table 3
summarizes information on the total concentrations of all mea-
sured PAHs, the P450 HRGS responses to extracts of these
Fig. 3. Average benzo[a]pyrene equivalents (BaPEq) values among
six regions of Delaware River basin (USA). Seven values determined
to be outliers within their locations were excluded from this figure.
sediment samples (in
dicted HRGS responses from chemical analyses. The HRGS
PEq are in general about an order of magnitude higher than
the chemical B PEq, which is likely due to induction from
compounds (alkylated PAHs) other than the eight inducing
PAHs previously studied [19].
␮g BaPEq/g), and calculations of pre-
deviations of the 6 selected sections of the Delaware River–
Bay system. The means for the different sections of the wa-
tershed have been plotted in Figure 3 to illustrate the rela-
tionships between regions of the system and the dramatic de-
crease in contamination from the upper and middle river to-
ward the open coast. To reduce the effect of extremely high
or low values skewing the regional averages in Figure 3, out-
liers were removed within each set of data. Outlier values were
determined by multiplying the interquartile range by 1.5 and
then subtracting or adding the value of the first or third quartile,
respectively, within each region. By this method, samples 11,
13, 89, 29, 26, 40, and 66 were found to be outliers and were
excluded from Figure 3.
B
a
a
Table 4 illustrates results of the PCB congener analyses
performed by the NOAA contractor on all samples. Although
the analysis of noncoplanar PCB congeners was performed on
all 81 samples, coplanar PCB congeners were analyzed in only
24 randomly chosen samples. Coplanar PCB analyses included
PCB congeners 77, 126, and 169, but not 81, which has been
shown to induce HRGS at a lowest concentration of any PCB
[19]. Samples 1, 3, 4, 7, 13, and 91 were the only samples
analyzed in which the levels of coplanar PCBs (at about 0.3
ng/g and greater) would be expected to induce the P450 HRGS.
Dioxin and furan chemical analyses were performed on five
samples that were suspected of containing significant amounts
of chlorinated hydrocarbons, based on two-time-period HRGS
testing. The chemical TEQs (in parts per trillion; pg/g) for
samples 10, 11, 20, 29, and 89 are based on the potency of
the 17 dioxin and furan compounds that induce the CYP 1A1
promoter gene (Table 5). The TEQ values were derived from
the concentration of each chemical and the WHO [27] TEFs.
The HRGS TEFs are not necessarily equivalent to the WHO
TEFs, because the latter were derived from data on many
species and a wide variety of biological responses. However,
two years of daily testing with HRGS (D. McCoy, unpublished
data) have shown that fold induction is equivalent to the WHO
TEQ (in pg/ml) of the test solution. Among the five samples,
sample 29 had the highest TEQ value (48.3 pg/g), followed
by samples 20 and 89 (26.9 and 20.5 pg/g TEQ, respectively).
Samples 10 and 11 contained only small quantities of dioxins
and furans (0.3 and 3.8 pg/g TEQ, respectively), which would
likely be responsible for only a small percentage of the ob-
served induction.
6- and 16-h testing
To determine if the 11 samples producing the highest HRGS
responses contained significant levels of chlorinated hydro-
carbons, two-time-period tests were performed (Fig. 4). These
samples required high dilution to bring their fold induction
values within the test range. At 16 h, fold induction should
be less than 100, and at 6 h it should be below about 50.
Samples 4 and 11 initially required dilutions of 1:300. Fold
induction values decreased by 75 to 85% in every sample from
initial 6-h values to final 16-h values with the exception of
sample 11. Sample 11 decreased by just 39% and yielded a
response ratio of 1.65, which could indicate the presence of
chlorinated hydrocarbons. However, because the 6-h response
exceeded 50-fold, a spillover of PAHs or oversaturation effect
could have occurred. In this case, the response ratio is skewed
toward a low value because of insufficient time for the cells
to metabolize the high amount of PAHs present. For this rea-
son, samples 11, 13, and 29 (the samples with the highest 16-
h values) were diluted once more and retested (Fig. 4). In each
case, the 6-h responses decreased somewhat, whereas the 16-
Correlations with analytical chemistry data
The correlation of P450 HRGS data with the subsequent
chemical analyses (Table 3) of the total measured PAHs in the
sediment samples is shown in Figure 5. A strong correlation
value (r2
ϭ 0.81) suggests that total measured PAHs are the
primary contributors to CYP 1A1 induction, despite the fact
that other compounds such as coplanar PCBs, dioxins, and
furans also were present (Tables 4 and 5) and may have con-
tributed to the overall HRGS response. In Figure 6, the HRGS
responses (except samples 11 and 17) are plotted against the
predicted responses (chemical BaPEq), based on the conver-
sion of data on the specific PAHs known to induce the CYP
1A1 gene in this assay, and the magnitude of the expected
Fig. 4. Fold induction responses of the 11 highest samples at 6-h and
16-h test periods, with dilution factors in parentheses.

CYP 1A1–inducing chemicals in the Delaware River and Bay, USA
Environ. Toxicol. Chem. 21, 2002
Table 3. Continued
Chemical
1623
Table 3. Measured and calculated human reporter gene system (HRGS)
responses and total polycyclic aromatic hydrocarbons (TPAHs)
HRGS
PEqa
HRGS
PEqa
Chemical
B
a
B
a
PEq
TPAH
(␮g/g)
B
a
B
a
PEq
TPAH
(␮g/g)
Sample
(
␮
g/g)
(
␮
g/g)
Sample
(
␮
g/g)
(
␮
g/g)
71
72
73
84
85
87
88
89
90
91
92
0.3
0.4
0.7
14.3
48.1
3.8
0.02
0.00
0.01
0.82
0.92
0.06
0.96
12.33
1.74
1.40
0.98
0.0
0.0
0.0
1.3
1.5
0.3
1.2
10.9
1.7
2.1
1
2
3
4
5
6
7
8
9
38.4
83.3
37.9
313.0
0.7
47.6
225.5
226.6
8.4
9.83
3.80
8.5
6.1
9.1
11.62
20.30
0.36
4.12
9.40
9.03
2.80
4.70
5.48
0.74
8.61
7.88
0.33
4.91
1.7b
17.2
0.4
4.5
14.0
12.9
3.5
6.5
10.0
0.9
18.1
5.4
0.4
11.1
1.6b
2.8
6.1
14.6
4.0
1.3
3.7
2.8
2.4
2.2
2.0
1.2
12.7
2.2
0.2
0.4
0.4
1.0
0.5
1.1
1.0
0.5
0.1
0.6
0.2
0.3
0.5
0.2
0.0
0.1
0.0
0.1
0.0
0.2
0.0
0.0
0.0
0.1
0.1
0.4
0.7
0.1
0.1
0.6
0.0
0.0
0.0
0.0
0.0
0.1
0.1
0.0
0.0
0.0
6.7
396.9
28.3
27.8
15.5
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
115.1
1,584.0
2.4
2.0
a B
a
PEq
ϭ
benzo[a]pyrene equivalents.
b Chemistry provided was a factor of 10 higher than listed and was
suspect.
344.7
18.0
3.4
130.6
62.3
47.4
124.2
183.1
7.9
16.3
18.8
12.5
20.4
25.2
7.9
13.0
298.4
9.3
1.9
1.3
1.7
5.3
1.1
6.9
5.2
8.9
1.1
7.2
1.8
4.2
1.5
3.4
1.0
1.2
0.7
0.4
0.3
2.7
0.2
0.6
0.6
0.6
2.77
5.24
6.31
3.93
1.01
2.61
2.38
1.70
1.08
1.06
0.83
4.98
1.32
0.03
0.01
0.01
0.65
0.27
0.64
0.57
0.27
0.07
0.57
0.10
0.27
0.49
0.22
0.03
0.05
0.02
0.05
0.04
0.11
0.00
0.02
0.01
0.07
0.04
0.30
0.52
0.12
0.05
0.59
0.01
0.01
0.01
0.00
0.00
0.05
0.02
0.01
0.01
0.00
response (HRGS TEFs). The correlation coefficient in Figure
6 (r2
ϭ 0.63) is not as high as that of Figure 5, because a few
high HRGS values are not explained by the concentrations of
inducing PAHs. Samples with significant amounts of dioxins,
furans, or coplanar PCBs would be expected to fall above the
trend line in Figure 6. Of the three samples located substan-
tially above the trend line (samples 13, 29, and 89), only
sample 13 was analyzed for coplanar PCBs and was found to
contain more than four times the amount of any other analyzed
sample (2 ng/g coplanar congeners
ϩ 660 ng/g other conge-
ners; Table 4). As noted above (Table 5), samples 29 and 89
contained significant amounts of dioxins and furans (TEQs of
48 and 20 pg/g, respectively), which may explain the high
HRGS response. Sample 17, with suspect analytical chemical
data, was not included in Figure 6. The highest HRGS value
(sample 11 at 1,584
6 because it would have expanded the scale significantly. This
sample was found to contain about 4 pg TEQ/g of dioxins or
␮g BaPEq/g) was not plotted in Figure
furans and 10 ␮g/g of PAHs, but was not analyzed for coplanar
PCBs, which may have also contributed to the high induction
observed. Some samples were well below the trend lines of
Figures 5 and 6, and the best explanation is likely differences
in both the known and unknown inducing PAH compounds in
these extracts. Some suppression or toxicity is possible, con-
sidering the interactions between chemicals in these complex
mixtures.
Correlations with biological analyses
A recent report from NOAA [28] describes the full amount
of biological and chemical data gathered on samples taken
from the Delaware River–Estuary system. In its analysis, the
report compares P450 HRGS results with both chemical data
and the findings of other bioassays and benthic community
descriptions. In the estuarine strata, the HRGS responses were
negatively correlated (p Ͻ 0.01) with the total number of taxa
as well as the diversity index of benthic organisms, whereas
in the freshwater strata this high correlation was only asso-
ciated with the number of taxa. The B
tively correlated with sea urchin fertilization success (Spear-
man rank correlation of 0.663 to sea urchin sperm exposed
to 50% pore water) and to the 50% effects concentration of
Microtox [28]. Not surprisingly, the results of 10-d amphipod
2.8
2.9
3.7
1.7
2.3
1.8
1.3
0.2
0.5
0.2
0.4
3.0
1.0
0.4
0.4
0.4
aPEq also were nega-
Ϫ
mortality tests did not correlate with BaPEq, because HRGS

1624
Environ. Toxicol. Chem. 21, 2002
D.L. McCoy et al.
Table 4. Coplanar and total polychlorinated biphenyl (PCB) congeners.
Coplanar PCBs only were measured in those samples listed
Table 4. Continued
Coplanar PCBs
Noncoplanar
PCBs (ng/g)
Total PCBs
(ng/g)
Coplanar PCBs
(ng/g)
Noncoplanar
PCBs (ng/g)
Total PCBs
(ng/g)
Sample
(ng/g)
Sample
72
73
84
85
87
88
89
90
91
92
0.25
0.37
0.25
0.37
1
2
3
4
5
6
7
8
9
0.320
142.99
20.35
237.18
188.51
11.63
52.37
229.52
252.49
53.65
92.20
112.57
13.32
660.61
30.24
17.29
466.23
112.58
46.68
138.78
319.42
100.80
40.20
110.77
15.49
70.50
54.77
3.73
34.38
32.41
60.50
2.94
17.28
2.24
26.26
5.28
33.13
28.16
17.46
3.68
9.61
3.64
4.56
3.54
4.54
1.50
1.74
2.10
3.72
1.46
2.76
0.27
0.91
0.88
1.90
1.88
143.31
20.35
237.63
188.91
11.63
52.37
230.11
252.49
53.65
92.21
112.57
13.33
662.67
30.61
17.31
466.23
112.58
46.89
138.78
319.42
101.15
40.31
110.77
15.49
70.50
54.77
3.73
34.38
32.41
60.50
2.94
17.28
2.24
26.26
5.28
33.13
28.34
17.46
3.70
9.61
3.64
4.56
3.56
4.54
1.50
1.74
2.10
3.72
1.46
2.76
0.27
0.91
0.88
1.92
1.88
38.05
37.67
12.04
30.66
279.07
71.29
59.95
28.51
38.05
37.67
12.04
30.66
279.07
71.29
60.44
28.76
0.446
0.401
0.589
0.005
0.487
0.254
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
0.013
2.064
0.371
0.021
measures the levels of hydrophobic toxic organic compounds,
which, in situ, may be released slowly over time.
As already described above, a strong correlation was found
between the HRGS and PAHs, but because metals co-occurred
with PAHs, this correlation also was high. Because the HRGS
assay is conducted on organic solvent extracts of sediments,
only the organometals such as tributyl tin would be expected
to be present in the extract applied to the cells.
0.214
0.348
0.111
DISCUSSION
Samples that generated an HRGS response equal to or
above 60
␮g BaPEq/g are an indication of areas in which
environmental degradation is expected [9]. This is based on
the levels of CYP 1A1–inducing compounds observed in a
San Diego Bay (CA, USA) study that were highly correlated
with degradation of the benthic community [9,13]. This deg-
radation was based on low species diversity as well as the
presence of significant numbers of opportunistic species in
tandem with the absence of nonopportunistic species. Both the
upper and middle Delaware River regions had a significant
number of sites that approached or exceeded the 60
g level. Of the total 81 samples tested in this project, 13 (16%)
were above 60 g B PEq/g. Also, of the 47 samples taken
from the Delaware River and its tributaries, 28% were above
60 g B PEq/g, including 42% of the samples from the upper
and middle river areas. These regions are located in close
proximity to major metropolitan and urban areas, with a large
number of refineries along the river’s banks. Therefore, not
surprisingly, nearly all of the samples in the Delaware River
watershed with high HRGS values were located in the middle
and upper reaches of the river, adjacent to the urban areas of
Philadelphia (PA, USA), Wilmington (DE, USA), and Trenton
(NJ, USA).
0.004
0.182
0.018
␮g BaPEq/
␮
a
␮
a
0.017
Based on the analyses of previous HRGS data from several
coastal areas [13], detrimental biological effects may occur
among benthic, epibenthic, or bottom-feeding fish communi-
0.018
0.053
6.61
21.93
3.43
1.12
14.50
1.11
0.34
0.33
0.18
0.43
2.28
2.95
0.54
0.43
0.19
0.28
6.61
21.93
3.43
1.12
14.55
1.11
0.34
0.33
0.18
0.43
2.28
2.95
0.54
0.43
0.19
0.28
ties when BaPEq levels are above 37 ␮g BaPEq/g. This es-
timate is based on the mean upper 99% confidence interval
derived from more than 912 samples from 12 different study
locales (Table 6). Five of the samples from this investigation
(6%) were greater than 37 but less than 60 ␮g BaPEq/g.
Samples taken from the marshes and inner portions of small
tributaries along the middle portion of the river also had el-
evated contaminant levels, including sample 89, which had a
0.002
0.003
BaPEq level of 396.8 ␮g BaPEq/g. Lower flow rates, de-
creased water volume, industrial discharge, and runoff from
both Dover Air Force Base, and the city of Dover (DE, USA)
might be contributing to pockets of high contamination in these
areas. These data would be consistent with those of other
0.003
0.003

CYP 1A1–inducing chemicals in the Delaware River and Bay, USA
Environ. Toxicol. Chem. 21, 2002
1625
Table 5. Chemical toxic equivalency quotients (TEQs) of dioxins and furans contained in Delaware Bay, USA, samples (pg/g; parts per trillion)a
Samples
TEQ
WHO
TEFs
Analyte
10
TEQ
11
TEQ
20
29
TEQ
89
TEQ
2,3,7,8-Tetrachlorodibenzo-
1,2,3,7,8-Pentachlorodibenzo-
1,2,3,4,7,8-Hexachlorodibenzo-
1,2,3,6,7,8-Hexachlorodibenzo-
1,2,3,7,8,9-Hexachlorodibenzo-
p
-dioxin
-dioxin
1.0
1.0
0.1
0.1
ND 0.00
ND 0.00
ND 0.00
0.31 0.03
ND 0.00
14.09 0.14
ND 0.00
ND 0.00
ND 0.00
3.56 0.36
3.09 0.31
62.13 0.62
ND 0.00
4.12 4.12
7.36 0.74
ND 0.00
ND 0.00
ND 0.00
ND 0.00
ND 0.00
ND 0.00
p
p
p
p
-dioxin
-dioxin
-dioxin
12.4
1.24
5.53 0.55
8.72 0.87
137.98 1.38
26.90 2.69
23.93 2.39
599.94 6.00
0.1
0.01
0.0001 118
0.1
0.05
0.5
0.1
0.1
0.1
8.67 0.87
222.04 2.22
1,2,3,4,6,7,8-Heptachlorodibenzo-
Octachlorodibenzo- -dioxin
2,3,7,8-Tetrachlorodibenzofuran
p
-dioxin
p
0.01 1,103.98 0.11 4,902.63 0.49 4,530.75 0.45 13,272.54 1.33
0.46 0.05
ND 0.00
1.23 0.06
2.7 1.35
3.27 0.33
2.55 0.26
1.93 0.19
ND 0.00
25.95 0.26
ND 0.00
39.44 0.00
ND 0.00
8.09 0.40
12.83 6.42
29.97 3.00
17.63 1.76
14.39 1.44
ND 0.00
407.29 4.07
14.66 0.15
512.65 0.05 4,488.92 0.45
9.86 0.99
25.27 1.26
24.07 12.04
124.67 12.47
58.33 5.83
30.18 3.02
ND 0.00
15.88 1.59
ND 0.00
8.14 4.07
7.98 0.80
9.06 0.91
ND 0.00
ND 0.00
72.27 0.72
ND 0.00
128.73 0.01
1,2,3,7,8-Pentachlorodibenzofuran
2,3,4,7,8-Pentachlorodibenzofuran
0.50 0.03
ND 0.00
ND 0.00
ND 0.00
ND 0.00
ND 0.00
1.95 0.02
ND 0.00
ND 0.00
1,2,3,4,7,8-Hexachlorodibenzofuran
1,2,3,6,7,8-Hexachlorodibenzofuran
2,3,4,6,7,8-Hexachlorodibenzofuran
1,2,3,7,8,9-Hexachlorodibenzofuran
1,2,3,4,6,7,8-Heptachlorodibenzofuran
1,2,3,4,7,8,9-Heptachlorodibenzofuran
Octachlorodibenzofuran
0.1
0.01
0.01
0.0001
787.78 7.88
114.76 1.15
Total chemical TEQ (pg/g)
0.27
3.85
26.96
48.33
20.51
a WHO
ϭ
World Health Organization; TEF
ϭ
toxic equivalency factor; ND
ϭ
not detected at 0.5 pg/g.
NOAA studies done in south and central Puget Sound (WA,
USA) and Florida, USA, in which areas of high contamination
were frequently found in creeks, canals, inlets, and back bays
that were often subjected to the aforementioned phenomena
[13].
method, the Delaware River and Bay study produced a mean
similar to that of southern Puget Sound, and higher than several
other regions (Table 6) [13]. The Delaware system had a great-
er percentage of samples above 60
␮g BaPEq/g than most
regions previously studied. Aside from samples 20, 29, and
89, which had elevated amounts of dioxins and furans, and
sample 13 with high PCBs, nearly all of the HRGS induction
appears to be the result of PAH contamination. However, be-
cause only 24 of the 81 samples were analyzed for PCBs, and
only 5 samples were analyzed for dioxins and furans, the pos-
sibility exists that contributions from chlorinated hydrocarbons
may have been responsible for other high induction values not
explained by PAHs.
The lower river region also contained elevated contaminant
levels (mean
ϭ 27.8 ␮g BaPEq/g), although not to the extent
of the upper or middle river or tributary regions. The main
reason for this is likely that it is situated further away from
the urban sprawl and refineries upstream. Only one sample
(sample 29) exhibited greater than 60
value is considered an outlier, the overall mean for the lower
river drops to 6.8 g B PEq/g.
␮g BaPEq/g and if this
␮
a
In contrast to the upper and middle river and tributary re-
gions, and to a lesser extent the lower river, the central bay
and open coast sites contained concentrations of chemicals
producing low induction levels. Of the 34 samples taken from
Overall, contamination levels consistently decreased from
the upper and middle river sites on down through the lower
river and central bay to the open coast. Of the 2,347 km2
covered in this investigation, examination of the HRGS data
indicates that 63% of the system (1,825 km2) did not contain
levels of sediment chemicals found to be linked to moderate
the central bay or open coast regions, none were above 10
PEq/g, and 50% were below 1 g B PEq/g. Analyses of
previous data from many regions [13] indicate that biological
effects would not be expected at B PEq levels less than 11
g B PEq/g. At these low levels, relatively little contamina-
␮g
B
a
␮
a
a
␮
a
tion with respect to dioxins, furans, coplanar PCBs, and PAHs
would be present.
In comparison to other recent studies that utilized the HRGS
Fig. 6. Correlation between observed human reporter gene system
(HRGS) benzo[
a]pyrene equivalents (BaPEq) and the expected
B
a
PEq, based on the presence of specific polycyclic aromatic hydro-
carbons (PAHs) and their potency in the HRGS assay [19]. Samples
11 and 17 were excluded. Samples above the trendline would be
expected to contain compounds other than PAHs. Labeled stations
were found to contain significant concentrations of polychlorinated
biphenyls and dioxins and furans. Stations below the trend-line may
contain fewer unidentified compounds, which induce cytochrome
P450 1A1.
Fig. 5. Correlation between human reporter gene system (HRGS)
benzo[a]pyrene equivalents (BaPEq) and total polycyclic aromatic
hydrocarbons (PAHs). Sample 11 was excluded because of the very
high value (1,584) and sample 17 was excluded because of suspect
chemical data.

1626
Environ. Toxicol. Chem. 21, 2002
D.L. McCoy et al.
Table 6. Distribution of data from P450 human reporter gene system (HRGS) analyses of sediment samples from regional surveys. Values listed
represent HRGS responses ( g benzo[
]pyrene equivalents/g) and percentages of samples exceeding estimated thresholdsa
␮
a
No.
stations Stations Total area
60 60
(km2)
%
Locations in
United States
No.
stations
Mean
SD
6.2
6.0
8.5
27.3
48.3
207.7
21.1
14.3
190.8
31.0
11.9
14.9
49.0
99% CI Lower CI Upper CI
Ͼ
Ͼ
Southern California bays
Galveston Bay, TX
29
75
5.2
5.2
8.2
2.5
1.8
2.7
3.4
7.7
7.0
0
0
0
0.0
0.0
0.0
5.0
1,351.1
271.4
806.2
737.5
857.7
245.9
7.3
Biscayne Bay, FL (1996)
Northern Puget Sound
Central Puget Sound
Southern Puget Sound
Sabine Lake, TX
121
100
100
100
65
9
81
63
69
2.0
7.0
6.2
4.0
25.1
0.0
7.8
10.2
18.1
50.0
106.3
21.2
30.2
112.0
36.0
12.0
18.7
35.8
11.1
37.5
52.8
14.5
16.3
57.0
25.9
8.3
4
4.0
12.5
53.5
6.7
14.0
23.8
10.1
37
19
14
5
19.0
14.0
7.7
Winyah Bay, SC
2.3
0
0.0
Delaware River and Bay
Chesapeake Bay (1998)
Chesapeake Bay (1999)
San Francisco Bay (2000)
Mean values
58.7
15.8
4.6
11
11.8
13
6
1
16.0
9.5
1.4
2,346.8
2,266.0
NA
100
912
15.7
21.5
3.8
11.8
2
5.3
2.0
6.1
1,200.0
917.7
Total
ϭ
Charleston Harborb
San Diego Bayb
20
30
23
52
268
30.1
31.4
0.9
63.5
18.5
26.8
35.5
1.9
171.3
58.8
17.3
16.2
1.02
61.2
8.8
12.7
15.2
0
2.3
9.7
47.4
47.6
1.9
124.6
27.3
3
15
0
11
15
15.0
50.0
0.0
21.2
5.6
6.6
5.5
NA
Columbia Riverc
Willamette Riverc
NA
Southern California Bight (1998)c
3,808.0
Total samples
1,305
a SD
ϭ
standard deviation; CI
ϭ
confidence interval; NA ϭ estimate of area not currently available.
b Samples from the NOAA Projects on Charleston Harbor and San Diego Bay, USA, were selected from a larger group of samples, and therefore
are not representative of the random sampling approach used in other NOAA surveys.
c Projects funded by sources other than NOAA.
or strong biological effects. Much of the low contaminant
levels in the central bay likely can be attributed to a high rate
of tidal flushing in concert with an up-estuary movement of
sediment transport within areas of strong tidal flow. The deep
channels in the bay run northwest from the mouth of the bay.
At the mouth of the bay, Cape May shoals restrict the entrance
to the bay on the north side and tidal flow velocities are stron-
gest in the channels. As a consequence, net sediment flux is
actually in the upstream direction in the main bay area because
ebb tidal velocities are relatively weaker due to the shoreline
configuration and Coriolis effects [29]. Further upstream, the
estuary narrows significantly and the influence of lateral water
movement on depositional patterns is reduced. Sediment-
bound contaminants associated with fine-grained, high organic
content materials possibly are prevented from migrating to the
lower estuary by the circulation system. Thus, despite the rel-
atively high contaminant loads in the river system leading into
it, the Delaware Bay and immediate coastline seem to have
relatively low levels of contaminants and likely exhibit healthy
populations of marine organisms.
It is important to keep in mind that because the solvent
extracts used in HRGS testing remove all hydrophobic con-
taminants from the samples and does not take bioavailability
into consideration, the values derived by HRGS may represent
a worst-case scenario. The HRGS values may not always cor-
relate well with acute toxicity studies, but other studies have
shown that impacts on epibenthic and infaunal species exposed
to sediment-bound compounds can be significant when con-
sidered over the course of months or years [30]. We believe
the best use of the HRGS assay is an initial screening of all
stations in a sediment survey and a subsequent comparison to
measured PAHs. Those stations with HRGS responses well
above the HRGS–total measured PAHs correlation curve
should then be investigated for contamination from chlorinated
CYP 1A inducers (dioxins, furans, and coplanar PCBs), thus
concentrating the expensive analytical costs on the key sta-
tions.
REFERENCES
1. Sutton CC, O’Herron JC II, Zappalorti RT. 1996. The scientific
characterization of the Delaware Estuary. Delaware River Basin
Commission Project 321; HA File 93.21. Delaware Estuary Pro-
gram, Philadelphia Pike, Wilmington, DE, USA.
2. Delaware River Basin Commission. 1993. Sediment contaminants
of the Delaware River estuary. Estuary Toxics Management Pro-
gram, West Trenton, NJ, USA.
3. Little AD. 1994. Distributions of chemical contaminants and acute
toxicity in Delaware Estuary sediments. Final Report 94-08. Del-
aware Estuary Program, Philadelphia Pike, Wilmington, DE,
USA.
4. Paul JF, Gentile JH, Scott KJ, Schimmel SC, Campbell DE, Lat-
imer RW. 1999. EMAP–Virginian Province four-year assessment
report (1990–93). EPA/620/R-99/004. U.S. Environmental Pro-
tection Agency, Washington, DC.
5. Chapman PM. 1996. Presentation and interpretation of sediment
quality triad data. Ecotoxicology 5:327–339.
6. Long ER, Chapman PM. 1985. A sediment quality triad: Measures
of sediment contamination, toxicity and infaunal community com-
position in Puget Sound. Mar Pollut Bull 16:405–415.
7. Long ER, Robertson A, Wolfe DA, Hameedi J, Sloane GM. 1996.
Estimates of the spatial extent of sediment toxicity in major USA
estuaries. Environ Sci Technol 30:3585–3592.
8. Long ER, et al. 1998. Magnitude and extent of sediment toxicity
in selected estuaries of South Carolina and Georgia. Technical
Memorandum NOS ORCA. National Oceanic and Atmospheric
Administration, Silver Spring, MD, USA.
9. Fairey R, et al. 1996. Chemistry, toxicity, and benthic community
conditions in sediments of the San Diego Bay region. Final Re-
port. California State Water Resources Control Board, Sacra-
mento, CA, USA.
10. Stegeman JJ, Hahn ME. 1994. Biochemistry and molecular bi-
ology of monooxygenases: Current perspectives on forms, func-
tions, and regulation of cytochrome P450 in aquatic species. In
Malins DC, Ostrander GK, eds, Aquatic Toxicology: Molecular,
Biochemical, and Cellular Perspectives. Lewis, Boca Raton, FL,
USA, pp 87–206.
11. Collier TK, Anulacion BF, Stein JE, Goksoyr A, Varanasi U. 1995.

CYP 1A1–inducing chemicals in the Delaware River and Bay, USA
Environ. Toxicol. Chem. 21, 2002
1627
A field evaluation of cytochrome P4501A as a biomarker of con-
taminant exposure in three species of flatfish. Environ Toxicol
Chem 14:143–152.
on a human cell line. SW 846 Methods, Update IVB. Office of
Solid Waste, Washington, DC.
21. Johnson BT, Long ER. 1998. Rapid toxicity assessment of sed-
iment from estuarine ecosystems: A new tandem in vitro testing
approach. Environ Toxicol Chem 17:1099–1106.
22. Carr RS. 1998. Marine and estuarine porewater toxicity testing.
In Wells PG, Lee K, Blaise C, eds, Microscale Testing in Aquatic
Toxicology. CRC, Boca Raton, FL, USA, pp 523–538.
23. American Public Heath Association, American Water Works As-
sociation, Water Pollution Control Federation. 1998. Standard
Methods for the Examination of Water and Wastewater, 20th ed.
American Public Health Association, Washington, DC.
24. American Society for Testing and Materials. 1999. Standard for
measuring the presence of planar organic compounds which in-
duce CYP1A, reporter gene test systems. E 1853M-98. In Annual
Book of ASTM Standards, Vol 11.05. Philadelphia, PA, pp 1383–
1389.
12. Collier TK, Krones CA, Krahn MM, Stein JE, Chan S, Varanasi
U. 1996. Petroleum exposure and associated biochemical effects
in subtidal fish after the Exxon Valdez oil spill. Am Fish Soc Sym
18:671–683.
13. Anderson JW, Jones JM, Hameedi J, Long E, Tukey R. 1999.
Comparative analysis of sediment extracts from NOAA’s bioef-
fects studies by the biomarker, P450 RGS. Mar Environ Res 48:
407–425.
14. Kim GB, Anderson JW, Bothner K, Lee J-H, Koh C-H, Tanabe
S. 1997. Application of P450RGS (reporter gene system) as a
bioindicator of sediment PAH contamination in the vicinity of
Incheon Harbor, Korea. Biomarkers 2:181–188.
15. Koh C-H, Kim GB, Maruya KA, Anderson JW, Jones JM, Kang
S-G. 2001. Induction of the P450 reporter gene system biomarker
by polycyclic aromatic hydrocarbons in Ulsan Bay (South Korea)
sediments. Environ Pollut 111:437–445.
25. Jones JM, Anderson JW, Tukey RH. 2000. Using the metabolism
of PAHs in a human cell line to characterize environmental sam-
ples. Environ Toxicol Pharmacol 8:119–126.
26. Lauenstein GG, Cantillo AY, eds. 1998. Sampling and analytical
methods of the National Status and Trends Program Mussel Watch
Project: 1993–1996 update. NOAA Technical Memorandum,
NOS ORCA 130. National Oceanic and Atmospheric Adminis-
tration and National Status and Trends Program, Silver Spring,
MD, USA.
27. Van den Berg M, Birnbaum L, Bosveld ATC, Brunstro¨m B, Cook
P, Feeley M, Giesy JP, Hanberg A, Hasegawa R, Kennedy SW,
Kubiak T, Larsen JC, van Leeuwen FXR, Djien AK, Nolt C,
Peterson RE, Poellinger L, Tillitt D, Tysklind M, Younes M,
Wærn F, Zacharewski T. 1998. Toxic equivalency factors (TEFs)
for PCBs, PCDDs, PCDFs for humans and wildlife. Environ
Health Perspect 106:775–792.
28. Hartwell SI, Hameedi J, Harmon M. 2000. Magnitude and extent
of contaminated sediment and toxicity in Delaware Bay. NOS/
NCOS/CCMA 148. National Oceanic and Atmospheric Admin-
istration, BioEffects Branch, Silver Spring, MD, USA.
29. Kenebel HJ. 1989. Modern sedimentary environments in a large
tidal estuary, Delaware Bay. Mar Geol 86:119–136.
30. Long ER, Mac Donald DD, Smith SL, Calder FD. 1995. Incidence
of adverse biological effects within ranges of chemical concen-
trations in marine and estuarine sediments. Environ Manage 19:
81–97.
16. Anderson JW, Jones JM, Steinert S, Sanders B, Means J, Mc
Millin D, Vu T, Tukey R. 1999. Correlation of CYP1A1 induction,
as measured by the P450 RGS biomarker assay, with high mo-
lecular weight PAHs in mussels deployed at various sites in San
Diego Bay in 1993 and 1995. Mar Environ Res 48:389–405.
17. Jones JM, Anderson JW. 1998. Cost-effective site characterization
using P450 reporter gene system (RGS). In Wickramanayake GB,
Hinchee RE, eds, Risk, Resources, and Regulatory Issues: Re-
mediation of Chlorinated and Recalcitrant Compounds. Battelle,
Columbus, OH, USA, pp 1–6.
18. Anderson JW, Jones JM, McCoy DL. 2000. Cost-effective dioxin
site characterization using a P450 human reporter gene system
(HRGS; EPA 4425). In Denison MS, ed, Organohalogen Com-
pounds, Vol 45. University of California, Davis, CA, USA, pp
208–211.
19. Jones JM, Anderson JW. 1999. Relative potencies of PAHs and
PCBs based on the response of human cells. Environ Toxicol
Pharmacol 7:19–26.
20. U.S. Environmental Protection Agency. 2000. Method 4425:
Screening extracts of environmental samples for planar organic
compounds (PAHS, PCBS, PCDDS/PCDFS) by a reporter gene