‘
a

$51

’1

.1 e

{é
. I“?
we ..

iliof
. .. 3.. 1
‘ 2,. 1.
‘ . L. .13 v
3.. i

.1? m;

t

.. ;
3.":

.¥

, . kn.
.. ‘. , . ~
ha" hr ._.u_.ﬁ..£._
33:... . a.

“nr

. é ? a
i. 2.6.
V .Lrﬂxka.
(_ 4w.
”was.
6 A.
u ‘ A

2. t.-

5 .
.i ..
L! ..
.35..

.5.

 

 

 

‘ “nut-'45?

1003

ﬁq;;”f§@t?

LIBRARY
Michigan State 3

Ulnverﬂty This is to certify that the

thesis entitled

2 , 3 , 7 ,8 - TETRACIILORODIBENZO—rfP-DIOXIN EQUIVALENTS
IN TISSUE SAMPLES FRO! THREE SPECIES IN THE DENVER
METROPOLITAN AREA

presented by

Katherine Kemler

has been accepted towards fulﬁllment
of the requirements for

Master's _ Zoology
degree in

 

 

(ELF/ﬂaw,

Major pymr

Date Apr11 17, 2000

 

0-7639 MS U is an Afﬁrmative Action/Equal Opportunity Institution

 

PLACE IN RETURN BOX to remove this checkout from your record.
TO AVOID FINES return on or before date due.
MAY BE RECALLED with earlier due date if requested.

 

 

DATE DUE DATE DUE DATE DUE

 

 

 

 

 

 

 

 

 

 

 

 

 

 

6/01 cJCIRC/DateDuost-p. 15

 

2,3,7,8 — TETRACHLORODIBENZO-P—DIOXIN EQUIVALENTS IN TISSUE
SAMPLES FROM THREE SPECIES IN THE DENVER METROPOLITAN AREA
By

Katherine K. Kemler

A THESIS

Submitted to
Michigan State University
In partial fulﬁllment of the requirements
for the degree of

MASTERS OF SCIENCE

Department of Zoology

2000

 

 

ABSTRACT
2,3,7,8 — TETRACHLORODIBENZO-P-DIOXIN EQUIVALENTS IN TISSUE
SAMPLES FROM THREE SPECIES IN THE DENVER METROPOLITAN AREA
By

Katherine K. Kemler

Polychlorinated dibenzo-p-dioxins (PCDDS), polychlorinated dibenzofurans (PCDFS), and
polychlorinated biphenyls (PCBs) are possible contaminants of concern in areas with a history of
organochlorine pesticide production and various other industrial chlorination processes. The
Rocky Mountain Arsenal (RMA) is a Superfund site near Denver, CO. This site has a history of
various industrial processes that may have led to the release of these compounds and subsequent
exposure risk to surrounding biota. Non- and mono—orthO-substituted PCBS, PCDDs, and PCDFs
cause a common set of toxic effects that are mediated through the aryl-hydrocarbon receptor
(AhR). The total AhR-mediated activity of complex mixtures was determined by both
instrumental and bioanalytical techniques in a range of biota samples from the RMA.
Concentrations of 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD) equivalents (TCDD-EQ and
TEQ) were measured in carp eggs, great horned owl livers, and American kestrel eggs collected
from the RMA and from off-post reference sites. The tissue samples were collected from 1995
through 1998. Two carp egg samples were collected from reference areas and 16 samples were
collected from the RMA. Eleven great horned owl livers were collected from reference areas and
16 samples were collected from the RMA. Sixteen American kestrel eggs were collected from
reference areas and 30 samples were collected from the RMA. Detected concentrations of dioxin
equivalents were used to assess risk to CXposed wildlife through the use of Hazard Quotients
(HQS). Mean concentrations of TCDD-EQ and TEQ in extracts of carp and kestrel eggs were not
Signiﬁcantly different between groups of samples collected on and off the RMA. Concentrations
of TCDD-EQ in several owl livers collected on the RMA were signiﬁcantly greater than
concentrations in owl livers from off of the RMA. Signiﬁcant differences in TCDD-EQ

concentrations were also observed between different age classes of great horned owls.

 

 

ACKNOWLEDGEMENTS

Thanks to Dr. Paul Jones, Dr. John Giesy, and Dr. Alan Blankenship for their help and
encouragement while working on this project.
Thanks also to my committee members, Dr. Steve Bursian, Dr. Karen Chou, and Dr.
Patrick Muzzall, for their guidance.
I would also like to thank the Biological Advisory Subcommittee (BAS) at RMA for

funding the project.

 

TABLE OF CONTENTS

Introduction ............................................................................................ 1
Problem Formulation ......................................................................... 1
Endpoints ............................................................................................... 5
Assessment Endpoints ...................................................................... 5
Measurement Endpoints ..................................................................... 5
Analysis Plan ........................................................................................... 9
Sampling Design and Methods ............................................................... 9
Laboratory Methods ........................................................................ 13
Data Analysis — TEQs ...................................................................... 15
Data Analysis — TCDD-EQS ............................................................... 16
Statistical Methods to Detect for Differences in
TCDD-Equivalent Concentrations Among Locations ................................. 19
Exposure Assessment ............................................................................... 20
Quality Control/ Quality Assurance Results ............................................ 20
Residue Concentrations of TEQWHO in Carp Eggs ...................................... 21
Concentrations of TCDD-EQ in Carp Eggs .............................................. 22
Residue Concentrations of TEQwuo in Great Horned Owl Livers .................... 24
Concentrations of TCDD—EQ in Great Horned Owl Livers .............................. 25
Residue Concentrations of TEQWHO in American Kestrel Eggs .................... 29
Concentrations of TCDD-EQ in American Kestrel Eggs ............................. 30
Effects Characterization .......................................................................... 33
Derivation of the Carp Egg TRV ......................................................... 33
Derivation of the American Kestrel TRV ............................................. 34
Derivation of the Great Horned Owl Liver TRV ....................................... 35

Risk Characterization ............................................................................. 36

 

 

Risk Estimation Using Hazard Quotients ............................................... 36

 

Carp Eggs .................................................................................... 36
Owl Livers .................................................................................. 37
Kestrel Eggs ................................................................................. 38
Risk Description: Ecological Signiﬁcance of the Results .................................. 39
Uncertainties and Future Research Needs ................................................... 41
References ........................................................................................... 44
Appendices
Appendix 1: Standard Operating Procedure for the H4IIE-luc Bioassay
for the Detection of AhR Agonists (MSU) ...................................................... 50
Appendix 11: Sample Calculations-Detection Limit Method and Slope-Ratio
Method ................................................................................................ 72
Appendix III: Quality Control/ Quality Assurance Samples .................................. 86
Appendix IV: Complete Results for Carp Egg Samples .......................................... 89
Appendix V: Complete Results for Owl Liver Samples ....................................... 92
Appendix VI: Complete Results for Kestrel Egg Samples .................................... 95

Appendix VII: Calculation of Uncertainty Factors for TRV
Determination ....................................................................................... 98

LIST OF TABLES

Table 1: Mean TEano concentrations for carp egg samples ................................ 21
Table 2: Comparison of mean TCDD-EQ and TEQREp from carp egg samples ............ 23
Table 3: Signiﬁcance of differences between locations as determined by the

Student’s t-test (0. = 0.05) .......................................................................... 24
Table 4: Mean TEQWHO concentrations for great horned owl liver samples ................ 25

Table 5: Comparison of TCDD-EQ and TEQ concentrations in great horned
owl liver samples .................................................................................... 27

Table 6: Signiﬁcance of differences between age classes as determined

by the Mann—Whitney U test (on = 0.05) .......................................................... 28
Table 7: Signiﬁcance of differences between locations as determined by the

Mann-Whitney U test (Ot= 0.05) ................................................................... 28
Table 8: Mean TEQWHQ concentrations for American kestrel egg samples ................. 29

Table 9: Comparison of mean TCDD-EQ and TEQ concentrations from
American kestrel egg samples ..................................................................... 31

Table 10: Analysis of variance (ANOVA) results for comparison of
TEQWHOMAX concentration in kestrel egg samples among location ........................ 32

Table 11: Student’s t-test results for the comparison of on-site and reference

TEQMAX and TCDD-EQMAX concentrations in American kestrel egg samples .......... 32
Table 12: Hazard Quotients (HQs) for carp egg samples ...................................... 37
Table 13: Hazard Quotients (HQs) for great horned owl liver samples ..................... 38
Table 14: Hazard Quotients (HQs) for American kestrel egg samples ...................... 38
Table III-a: MSU Quality Control/ Quality Assurance Samples ............................... 87

Table IV-a: Complete listing of concentrations of TEQ and HQ values for
carp egg extracts ..................................................................................... 90

Table IV-b: Complete listing of concentrations of TCDD-EQ and TEQREP
values for carp egg extracts ........................................................................ 91

 

Table V-a: Complete listing of concentrations of TEQ and HQ values
for owl liver extracts ................................................................................ 93

Table V—b: Complete listing of concentrations of TCDD-EQ and TEQREP
values for owl liver extracts ........................................................................ 94

Table VI-a: Complete listing of concentrations of TEQ and HQ values
for kestrel egg extracts .............................................................................. 96

Table VI—b: Complete listing of concentrations of TCDD-EQ and
TEQREP values for kestrel egg extracts ........................................................... 97

 

Vii

LIST OF FIGURES

Figure 1: Rocky Mountain Arsenal ................................................................. 10
Figure 2: American kestrel nestbox locations on the RMA ................................... 11
Figure 3: Reference locations of great horned owl specimen collection ..................... 12
Figure 4: Comparison of carp egg TEQREP and TCDD-EQ values .............................. 23
Figure 5: Comparison of great horned owl TEQREp and TCDD-EQ values .................... 26
Figure 6: Comparison of American kestrel egg TEQREp and TCDD-EQ values ............... 30
Figure III-a: MSU method spike quality control for lots BBA, BBB, and BBC .............. 88

Figure II-b: MSU method and matrix spike quality control for lots KKA,
KB, and KC ....................................................................................... 88

viii

AhR

AN OVA
BAS
CERCLA
DRE

EPA
GC/MS

H4IIE-Iuc

HQ
LOAEL
MDL
MRI
MSU
NOAEC
N OAEL
PCB
PCDD
PCDF
QA/QC
REP

LIST OF SYMBOLS AND ABBREVIATIONS

Aryl hydrocarbon Receptor

Analysis of Variance

Biological Advisory Subcommittee, Rocky Mountain Arsenal
Comprehensive Environmental Response, Compensation and Liability Act
Dioxin Responsive Enhancer

Environmental Protection Agency

Gas Chromatography / Mass Spectrometry

Rat hepatoma cell line stably transfected with the luciferase gene under
control of AhR

Hazard Quotient

Lowest Observable Adverse Effect Level
Method Detection Limit

Midwest Research Institute

Michigan State University

No Observable Adverse Effect Concentration
No Observable Adverse Effect Level
Polychlorinated biphenyl
Polychlorinated dibenzo-p-dioxin
Polychlorinated dibenzoﬁirans

Quality Assurance/ Quality Control
Relative Effects Potency

Rocky Mountain Arsenal

 

RLU
TCDD
TCDD—EQ
TEF

TEQ
TEQWHO

TEQREP

TRV
TU
UF

WHO

 

Relative Luminescence Unit

Tetrachlorodibenzo—p-dioxin (usually reserved for 2,3,7,8-TCDD)
Tetrachlorodibenzo—p-dioxin equivalent (bioassay)

Toxic Equivalency Factor

2,3,7,8-tetrachlorodibenzo-p-dioxin equivalent (instrumental analysis)
2,3,7,8-tetrachlorodibenzo-p-dioxin equivalent based on WHO TEF values

2,3,7,8-tetrachlorodibenzo-p-dioxin equivalent based on relative potency
factors

Toxicity Reference Value
Toxic Unit, dimensionless unit used in Hazard Quotient calculations
Uncertainty Factor

World Health Organization

 

 

Introduction

This study was conducted to determine the concentrations of polychlorinated
diaromatic hydrocarbons (PCDHS), such as polychlorinated dibenzo-p-dioxins (PCDDS),
polychlorinated dibenzofurans (PCDFS), and non- and mono-ortho- substituted
polychlorinated biphenyls (PCBs) in tissues from three animal species from the Rocky
Mountain Arsenal (RMA) and adjacent areas. The overall objective of the investigation was
to determine whether concentrations of 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD)
equivalents in samples present on the RMA were signiﬁcantly greater than those ﬁ'om
reference sites in adjacent off-site locations. Concentrations of 2,3,7,8-tenachlorodibenzo-
p-dioxin equivalents were determined by instrumental and bioanalytical techniques (in vitro
bioassays) to measure TEQS and TCDD-EQs, respectively. Concentrations of TEQS were
compared to TCDD-EQS to determine if all of the AhR-mediated activity determined in the
H4IIE-luc assay could be accounted for by the identiﬁed PCDH congeners. Possible risks
that concentrations of TCDD-EQS or TEQS pose to wildlife on the RMA were also assessed.
The US. Environmental Protection Agency’s “Framework for ecological risk assessment”
was used as an outline for reporting the formulation and results of this study (US. EPA

1992)

Problem Formulation:
In 1996, the Colorado Department Of Public Health released results from a survey
of trace organic compounds on the grounds of the Rocky Mountain Arsenal. Anomalous

concentrations of PCBs, PCDDs, and PCDFs were detected in the Basin F wastepile, a

 

waste area adjacent to the North Plant facility in the core region of the RMA, and in some
biota samples. This suggested that dioxin-like compounds could be potential
contaminants of concern (COCs) on the RMA. Consequently, the Biological Advisory
Subcommittee (BAS) began an investigation of the prevalence and availability of dioxin-
like compounds to wildlife on the RMA (US. Army 1999).

PCDHS, including PCBS, PCDDS, and PCDFS, are widely distributed in the
environment. A subset of the PCB congeners, including non-ortho and some mono-
ortho-substituted congeners, are structurally similar to PCDDs and PCDFS and are
referred to as co-planar PCBs. PCDDS, PCDFS, and co-planar PCBs are persistent,
subject to bioaccumulation, and some classes are toxic to biota at small concentrations
(Giesy et al. 1994a, Giesy et al. 1994b, Peterson et al. 1993). A subset of these
chemicals is characterized by a common mechanism of action via binding to the
cyt0plasmic aryl-hydrocarbon receptor (AhR) (Giesy and Kannan 1998, Sanderson et al.
1996). PCDDS, PCDFS, and some PCB congeners that can attain a planar configuration
bind to the AhR. Upon binding of the ligand, the AhR-ligand complex translocates to the
nucleus where it dimerizes and binds to dioxin responsive enhancers (DREs) on the DNA
(Sanderson et al. 1996). Hence, specific changes in gene expression can then result from
receptor binding (Giesy and Kannan 1998). Toxic effects of PCDDS, PCDFS, and co-
planar PCBs include impaired reproduction of several avian and ﬁsh species (Jones et al.
1993, Peterson et al. 1993, Williams et a1. 1995, Xu et al. 1997), severe weight loss, fatty
deposition in the liver, edema, fetotoxicity, and teratogenicity in laboratory animals and
wildlife (Couture et al. 1990, Giesy et al. 1994a, Giesy et al. 1994b, Giesy and Kannan

1998).

 

 

The RMA is a 27-square-mile Superfund site located in Commerce City, CO. The
vegetation of the site is primarily composed of grassland ﬂora with some riparian
woodlands. The wildlife composition of RMA is diverse, including 26 species of mammals,
17 species of reptiles and amphibians, aquatic communities, and 127 species of birds have
been observed on-post (US. Army 1999). The avian community on the RMA consists of
raptors (eagles, hawks, great horned owls, burrowing owls), ground-nesting songbirds, as
well as bird species preferring open, grassy habitats. Due to the historical presence Of
multiple agencies and multiple chemical production facilities on this site, the RMA was
placed on the National Priorities List in accordance with the Comprehensive Environmental
Response, Compensation and Liability Act (CERCLA) in 1987. The RMA is an army
facility composed of two main chemical-producing plants, the South Plants and the North
Plants, and numerous support buildings and other infrastructures (Figure 1). These plants
were used to produce a diversity of chemicals throughout the Arsenal’s history (U .8. Army
1999). The plants are located in the central area of the RMA, which is designated as the
core area. The core area is 6 square miles and is surrounded by 21 square miles of periphery
area. It is expected that the periphery is relatively uncontaminated, since no major chemical
producing activities took place in the peripheral area. From 1942 to 1945, the RMA was
used to produce chemical warfare agents, such as Levinstein mustard and Lewisite by the
US. Army. During this time, incendiary bombs, organophosphate nerve agents, such as
Sarin, and Hydrazine rocket fuel were also manufactured at the RMA. From 1946 to 1982,
some of the RMA facilities were leased to various companies that used its plants to produce
pesticides, including dieldrin, aldrin, chlordane, carbamate insecticides, soil fumigants, and

various other chlorinated hydrocarbon pesticides (U .S. Army 1999).

 

 

The source of dioxin-like compounds on the RMA is unknown. However, PCDDS,
PCDFs, and PCBs have been detected in wastes from the North Plants. This waste
consisted of chemical byproducts like hydrogen chloride from nerve agent production
processes and was deposited into Basins A and F (US Army 1999). Dioxin-like chemicals
can be formed as byproducts from the production of pesticides, nerve agents, and other
processes involving chlorination (Kannan et al. 1998). PCDDS and PCDFs have never been
manufactured for any beneficial purposes, but rather were released as byproducts from many
human activities (Giesy and Kannan 1998). PCDDS and PCDFs were investigated on the
RMA to determine if they were released as byproducts in the production of other chlorinated
compounds. PCDDS and PCDFs are known to be released from chloralkali plants, the
manufacture of some types of pesticides, incineration of chlorine-containing wastes, and
fossil fuel combustion (Kannan et al. 1998). Routine application of various pesticides such
as pentachlorophenol (PCP) on the RMA may also have been a source for surface soil

contamination with these compounds (U .8. Army 1999).

 

 

533-925 '5' a..,.. -~—_._ _~ '~

Endpoints

Assessment Endpoints:

Assessment endpoints are the ultimate focus when characterizing risk in a given
environment. These endpoints are selected to link measured endpoints to an evaluation
of risk (US. EPA 1992). To assess the risk to wildlife on RMA, concentrations of
TCDD-equivalents in the tissues of sentinel species were selected as the assessment
endpoints and were used as an integrative measure Of the bioavailability and toxicity of
dioxin-like compounds in the RMA environment. A simpliﬁed risk assessment based on
hazard quotients (HQs) was utilized. HQs are ratios based on a reference dose of toxicant
determined to have some adverse effect on the organism of interest. Toxicity Reference
Values (TRVs) were used as the accepted reference doses in this study, and Hazard
Quotients were calculated based on the amount of toxicant detected in the sample extract
relative to the TRV derived for that organism. The HQ is measured in toxic units (TU),
and a HQ value greater than 1.0 is indicative of potential adverse effects on the organisms
of interest. In this study, HQS were calculated based on concentrations of both TCDD-
EQS, which were calculated from chemical concentrations determined by an in vitro

bioassay and TEQS, which were measured by instrumental analyses.

Measurement Endpoints:
Measurement endpoints are directly measured components of an ecosystem that
can be related to assessment endpoints (US. EPA 1992). The speciﬁc measurement

endpoints used in this study were concentrations of individual PCDDS, PCDFs, and PCB

 

 

 

congeners (Van de Berg et al. 1998), patterns of the congeners, and measures of total
dioxin-like activity in tissue samples. Eggs from the American kestrel (Falco sparverius)
and common carp (Cyprinus carpio) and livers from adult and juvenille great horned owls
(Bubo virginianus) were selected to compare the dioxin-like activity between the on- and off
post populations. The working hypothesis for the use of these tissue samples was that if
PCDDs, PCDFS, and PCBs were released into the environment, they could persist and
possibly biomagnify to hazardous concentrations in some portions of the local foodchain.
PCDDS and PCDFs are lipOphilic compounds; they tend to accumulate in fats or fatty
stores within organisms (Peterson et al. 1993). Dioxin-like toxicants can also be
transferred maternally to offspring via the egg yolk (Williams et al. 1995, Jones et al.
1993). The liver often accumulates dioxin-like chemicals. For these reasons, kestrel and
carp eggs, as well as owl livers, were used as indicators of the availability of PCDDs,
PCDFs, and PCBs present in the RMA environment. Kestrels and owls were selected as
sentinel species because they are near the top of the foodchain, and thus are useful for
making integrative measures of the terrestrial environment. These birds prey upon
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. The chemical
composition of all tissue samples were considered to be representative Of the surrounding
environment. Great horned owl adults are the only organisms collected that may not be
indicative of the local environment in which they were sampled, since the adult bird has
relatively high dispersion ability.

In order to determine possible risks to RMA wildlife, two methods of estimating

TCDD-equivalents were applied. In the ﬁrst method, concentrations of T CDD-equivalents

 

 

were calculated as the sum product of Toxic Equivalency Factors (TEFS) derived by the
World Health Organization (WHO), which are multiplied by the concentration of each
congener detected in a mixture (Equation 1) (Van den Berg et al. 1998). This measure of
TCDD-equivalency is designated as TEQSWHQ In addition, TCDD-equivalents were
determined by use of the H4IIE-luc in vitro cell bioassay, and these equivalent

concentrations are designated as TCDD-EQS.

TEano = 23C. * TEF,- (l)
where C, indicates the concentration Of the congener and TEF,- is the TEF for that
congener

To assess the potential risk of TCDD-equivalents to wildlife, consensus TEFS,
derived from a variety of species and endpoints, were applied (Van den Berg et al. 1998).
TEFS are relative values assigned to dioxin-like compounds based on the most potent
AhR binding dioxin known, which is 2,3,7,8-TCDD. As it is the most potent PCDH,
TCDD is assigned the TEF value of 1.0, and T EF 5 for other dioxins, furans, and PCBs
are calculated relative to TCDD. Determination of TEF values vary depending on the
organism and endpoint used. TEF values relevant to birds and ﬁsh were applied for
samples of avian tissues and ﬁsh tissues respectively. This was done because it is known
that ﬁsh, birds, and mammals respond differently to dioxin-like compounds (Van den
Berg et al. 1998).

Concentrations of TEQWHO are not directly comparable to bioassay derived TCDD-
EQ concentrations. Derivation of a mass balance relationship between concentrations of

TEQS and TCDD-EQs was accomplished by applying a relative potency factor (REP) to

 

"'""" 1‘12.- J¢.£.Hezw;’-‘- ~_ “hens... “L .. amt—5'“ -' ~~1 »

detected congener concentrations (Equation 2). The REPS used are derived from the same

endpoint in the same bioassay system (H4IIE-luc) (Sanderson et al. 1996, Giesy et al.

1997). These REPS allowed calculation of concentrations Of TEQREP that could be

compared directly to concentrations of TCDD-EQ measured in the same bioassay.
TEQREP = 2C1 * REPI, (2)

where C, indicates the concentration of the congener and REP,- is the REP for that
congener

 

Total AhR-mediated activity in complex mixtures (TCDD-EQ) was determined by
use of an integrating bioassay, the H4IIE-luc (CALUX) assay (Sanderson et al. 1996).
TCDD-EQs calculated from the H4IIE-luc bioassay were used in conjunction with TEQS
calculated from concentrations of individual congeners (Midwest Research Institute;
MRI) on the same set of tissues. This in vitro bioassay is an integrative measure of all the
dioxin-like activity of PCB, PCDD, and PCDF mixtures in a given sample. Concentrations
of TCDD-EQ are a single aggregate measure of toxicity to examine the AhR-mediated
activities of complex mixtures (Giesy and Kannan 1998, Sanderson et al. 1996, Williams et
al. 1995). In the H4IIE-luc cell line, the luciferase gene has been stably transfected into rat
hepatoma cells so that AhR binding causes promoter activation and upregulation of the
luciferase gene. The luciferase gene increases production of the enzyme product, luciferase.
Luciferase cleaves luciferin (a substance dosed onto the cells on the day of analysis) and
luminescence is produced in a dose-dependent manner. Therefore, mixtures of dioxin-like
chemicals are measured by the upregulation of the luciferase gene and consequent increase
of luminescence intensity. In vitro bioassays are advantageous due to their rapidity and

decreased cost as compared to instrumental analyses. Bioassays also integrate various

 

interactions among the toxicants of interest and are therefore more biologically relevant
(Giesy and Kannan 1998).

Analysis Plan

Sampling Design and Methods:

Kestrel eggs were collected from nestboxes located on the RMA and at selected
reference areas off the RMA. At the RMA, kestrel nest boxes were located at the comers
of the sqaure mile demarcations (28 nestboxes in all) (Figure 2). Sixteen kestrel eggs
were collected from nestboxes in the reference areas off of the RMA. The reference
areas for kestrel egg collection included a fairground, a cemetery, a wastewater facility, a
lakeside area, a reservoir, and a former airport. Twenty-nine kestrel eggs were collected
on-site; ten eggs were collected from the core region of the RMA and 19 samples
originated from the periphery (Figure 2).

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 carp eggs were collected from
the RMA. Eggs were collected from Banner Lake and Lower Derby Lake on the RMA.
Carp were collected by use of electroshocking procedures and gill nets. Eggs were
collected from the adult carp.

Livers were collected from juvenile and adult great horned owls. Samples were
collected when owls were found dead or moribund as part of the US Fish and Wildlife
Services’ opportunistic sampling program. Eleven great horned owl livers were collected
from reference areas (Figure 3) and 16 livers were collected on-site. Reference areas

from which owl specimens were collected were within a 130-mile radius of the RMA.

 

 

 

 

 

 

 

 

 

’- - —?r.m—:_s,—— ' -— ' "" ' ‘7 ————— ‘1.
* l
19 20 I
I
I
. l
l
I
30 I 29 I
l
I
/ l
{ I
31 ' 32
I ,
I I
l _____
l
I I
l 6 5
I I I
I - -- $— ----- |
l l ' I
I I r I l
I 9 _I ! .11 _ 12 ; 7 a l
I e. ' _ ,~"

, I
L_.__-_I'"L.-_-_'_-_-__.__._l

 

Figure 1: Rocky Mountain Arsenal (RMA)

10

 

  
  
 

I .'
r . ' ‘ _ _ _ _
, . .

I Inninlc a

("1 "mum

"%

— — a —— - _
’E 4 -' - .c'nil 14:311.“: 2.. [galllllm M “an {€14}: g
L“, "- .' I.q'. ‘1‘“. 6’. '

/ I II

23 ‘ 24 . 19 ‘ 20

l
“$5935. 4'5
, - — . ‘3‘: '1 .. . L I. h. 2‘
" .f i. g I
‘ V‘s-.4 K45;
ﬁrmer . 7 - '

Banal" :
, ,I

f
North i
I
I
L

 

   

“1"“ 30 29

,J1 M"

is”

 

 

‘r', .. I“
l; 1'

‘24“ Kestrel Box

Active

i; ;’ Kestrel Box
(No Activity)

Figure 2: American kestrel nestbox locations on the RMA

ll

 

Rocky Mountain Arsenal _
3,.“ w National Wildlife Refuge b
v b 1

a I. ,,
_. Q " I. ' ' 4'
v 5
, -.~ .O'
' II l."v.\."
' I

h
I _ I
LJ-I-h—uI r. .—
r—ntn.¢p'lL J
: .r—If'u
l "
I
" 'H
@ (Xx C I H i a.
Di~lJl‘.. 1‘ ‘MW‘ RP. a'.‘ .Khilfllzel" Al'srl‘Cll " I‘- L. . "
I -~ l L. .uI'. ,p -1, '\ /
\asrnrkv-renw Rt...“ rJ ’ i L;
' a
9 1' 'I
- (“‘l“ "i. Lli'l’l '3.

Not to Sail: (“a 1’) : 9 ’: l
v'

Figure 3: Reference locations of great horned owl specimen collection

12

 

 

 

Laboratory Methods:

Carp and kestrel egg samples, as well as owl livers, were delivered to Michigan State
University (MSU) for chemical extraction, clean-up, and determination of TCDD-EQ
concentrations using the H4IIE-luc in vitro bioassay. Coded tissue samples were
supplied “blind” to MSU so that analyses were performed without knowledge of sample
origin. Homogenization of tissue samples was accomplished by adding a sufﬁcient
amount of Na2SO4 to dry the sample (usually in a 10:1 ratio to the tissue weight). The
tissue—Na2804 mixture was mixed until the tissues were dry. The mixture was then
blended at high speed in a homogenizer to produce a ﬁne powder.

With each batch Of approximately twenty samples, method blank and method spike
samples were extracted to determine the background and method recovery for dioxin-like
chemicals. The method blanks and spikes consisted of 50 grams of Na2804,
approximately equivalent to that used for each sample. For the method spikes, twenty-
ﬁve LIL of method spike solution, containing a mix of PCB congeners, was added to
NaZSO4. When extracting kestrel egg tissue, matrix spikes and matrix blank quality
control samples were analyzed with each block of 20 samples. These samples consisted
of approximately ﬁve grams of chicken egg homogenate to which a known amount of
PCB congeners had been added. Twelve quality control quail egg samples were also
included in the kestrel egg sample group. Eight of these samples were spiked with
known amounts of PCB 126, while four un-spiked quail eggs served as blanks. With
each batch of approximately twenty samples, method blank and method spike samples

were extracted to determine the background and method recovery for dioxin-like

chemicals.

13

 

 

After homogenization, samples were extracted in a soxhlet apparatus with 300 ml of
1:1 acetone/hexane for 18 hr. After extraction, sample extracts were concentrated to near
dryness by rotary evaporation at 35° C. Five ml of concentrated H2804 was added to
each sample to remove lipids by oxidation. Each sample was then left for approximately
3 hr to allow separation of acidic (lipid) and hexane layers. When the layers had
separated, the hexane portion, containing the analytes of interest, was removed from the
acidic lipid layer and placed in a round bottom ﬂask. The acid phase was washed with an
additional aliquot of hexane, which was removed and combined with the ﬁrst hexane
fraction. The volume of the hexane extract was again reduced to near dryness by rotary
evaporation, and ﬁve ml of nanopure water was used to rinse the extracts. After rinsing,
hexane extracts were evaporated under a stream of nitrogen to 0.1 ml and diluted to 1.0
ml with isooctane for use in the H4IlE-luc bioassay (Sanderson et al. 1996).

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. This column removes almost all compounds
except for dioxins, furans, and coaplaner PCBs (dioxin-like chemicals) (Kannan et al.
1998)

Concentrations of TCDD-EQS were determined by the response of rat hepatoma
cells (H4IIE) transfected with the luciferase gene under the control of the dioxin-
responsive enhancer (DRE) (Sanderson et al. 1996). The H4IIE-luc bioassay was
conducted over a period of 5 (1. Cells were plated on day 1, dosed on day 2 with 1.25 uL
of the sample extract, and dioxin-like activity and cell viability were measured on day 5.

Analyses of individual extracts from tissue samples were conducted in separate 96-well

l4

 

cell culture plates concurrently with a range Of concentrations of 2,3,7,8—TCDD as the
standard. The TCDD standard was made from a 10 ppm stock solution. T CDD standard
dilutions were made to deliver doses of 30, 10, 3, 1, 0.3, and 0.1 pg TCDD/well. Extracts
were applied in serial dilution, which delivered 1x, 0.3x, 0.1x, 0.03x, 0.01x, and 0.003x
diluted extracts to the wells. A series of wells that contained cells but were not used in
the analyses were interspersed between the active wells to minimize cross-talk between
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 (Molecular Probes Inc.). A
standard Operating procedure for the H4IIE—luc bioassay is included (Appendix 1).
Concentrations of individual congeners were measured by gas chromatography and
mass spectrometry (GC/MS) using standard analytical techniques. Midwest Research
Institute (MRI) chemically analyzed all sample types. These chemical analyses were
conducted under a separate contract between the BAS and MRI. All data generated by
MRI was subject to US-EPA quality assurance and quality control assessment and

validation before use in this report.

Data Analysis - T EQs:

Consensus TEFS were used in the calculation of TEQWHO values for carp egg, owl
liver, and kestrel egg extracts to assess risks to exposed species (Van den Berg et al.
1998). Avian TEF values were applied to the owl and kestrel sample extracts, while ﬁsh
TEF values were applied to the carp sample extracts. TEFS were multiplied by the

congener concentrations of the dioxin-like analytes that were chosen for measurement

15

 

 

v _ Ia$_¢-Vp—lf 7" ‘ ._ '

(Van den Berg et al., 1998). This product was then summed to yield a total TEano for
the sample. Several concentrations of TEme were calculated based on varying
manipulations Of the congener data below the method detection limit (MDL). TEQWHO-
MAX was calculated by substituting the MDL for those congeners that were below the
MDL. TEQWHOMIN was calculated when congeners below the MDL were set to 0.0.
Therefore, the TEQWHOMAX values are the most conservative and greatest possible
overestimates of the dioxin-like activity present in the samples, while the TEQWHoMm

values are the least conservative and lowest possible underestimates reported.

Data Analysis — T CDD—EQs:

Bioassay derived TCDD-EQS were determined for each sample by comparing mean
luminescence readings of sample extracts to luminescence readings from the TCDD
standard curve analyzed on the same plate. Data were plotted and evaluated using
Dunnett’s test to identify samples where the extract response was greater than
background and greater than the method detection limit (MDL) (Kuehl 1994). Samples
for which the response of the undiluted extract was less than the least standard (0.1 pg
TCDD/well) were determined to exhibit no detectable response (“non-detect”) because
extrapolation to concentrations less than those included in the standard curve was not
acceptable (Villeneuve et al. in press). “Non-detect” samples were evaluated using the
detection limit method to determine the sample-Speciﬁc TCDD-EQ based on the MDL.
Concentrations of TCDD-EQ were determined for samples for which a signiﬁcant

response was achieved by use of the SIOpe-ratio method.

16

 

 

The detection limit method involves the calculation of sample-speciﬁc MDLs.
MDLs for the analyses were based on the response of the sample in undiluted extract (1x)
and the response of the minimum standard (0.1 pg TCDD/well). These responses varied
among assays and among plates because of differences in samples sizes and
responsiveness of the H4IIE-luc cells. To determine the MDL, the relative luminescence
units (RLUs) measured in the solvent blank wells were subtracted from the sample

response, and from the response of the lowest standard, usually 0.1 pg TCDD. The

 

sample response was then divided by the standard response to determine the maximum
possible TCDD-equivalent amount present in the sample well. This TCDD amount was
then converted to a tissue concentration by determining the gram equivalents of sample in
the well. The measured dioxin-like activity calculated using this method for “non-detect”
samples was reported as the TCDD—EQMAX value. The TCDD-EQMIN reported “non-
detect” samples with a value of zero. Sample calculations for this method are provided
(Appendix II).

In the SIOpe-ratio method, the un—transformed TCDD standard and sample dose
response curves were evaluated to determine the linear portion of the curve (Villenueve
in press). The sample dilution series was ﬁrst converted to gram equivalents/well using
initial tissue weight, ﬁnal extract volume and dilution factor. It was sometimes necessary
to remove ‘saturated’ response values and the blank values from the standard curve. This
was due both to the non-linear effect of plateau values and an initial non-linear threshold
effect in the dose response curve due to the undue inﬂuence of the six blank values (3
solvent blanks and 3 no-treatment blanks) on the curve containing at most 12 other data

points. Statistical rigor in the analysis was achieved by determining the correlation

 

_-_.‘_~w.. - .

coefﬁcient for the line, after each data elimination. When the maximum r2 value was
achieved, the SIOpe of the standard curve was calculated and used in subsequent
calculations. The same assessment was performed for the sample to determine the slope
of the linear portion of the dose—response relationship for each sample. The ratio of the
slopes (sample/standard) was then used to determine the relative potency of the sample.
Sample calculations for the SIOpe-ratio method are provided (Appendix 11). These
calculations determined the mass of sample required to cause the same response as a
speciﬁc mass of TCDD in a well. The result of the analysis was an equivalent
concentration of 2,3,7,8-TCDD (TCDD-EQ) expressed as pg/g wet weight of tissue.
Since most undiluted extracts of carp eggs contained concentrations of TCDD-EQ
less than the minimum TCDD standard concentration, the greater proportion of the carp
egg extracts were analyzed by use of the detection limit method. Three extracts of carp
eggs were analyzed using the slope-ratio method, since their luciferase induction was
within the range of the TCDD standard curve. Concentrations of TCDD-EQs in owl
livers and kestrel eggs were calculated in the same manner as the carp egg samples,
where concentrations of TCDD-EQs were calculated by both the detection limit and
slope-ratio methods, depending on the level of AhR-induced luciferase activity of the

samples in comparison to the TCDD standard curve.

18

 

 

Statistical Methods to Detect for Differences in TCDD-Equivalent Concentrations Among
Locations:

In order to establish whether or not tissue samples collected from the RMA had
greater concentrations of TCDD-equivalents than tissue samples from reference areas,
concentrations Of TEQWHOMAX 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 set was normally distributed and had variances that
were not statistically different. The Mann-Whitney U test was used in cases where the
assumptions of a normal distribution were not met (Rand 1995). The one-way ANOVA
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. The level of statistical signiﬁcance for all tests was set at or:

0.05. Power, the probability of avoiding a type II error (Kuehl 1994), was set at 0.80 ([3 =
0.20). For any analyses for which sufﬁcient power (0.80) was not achieved, the results
were deemed inconclusive. SYSTAT software (SPSS Inc. 1998) was used for statistical
analyses, and Pass 60 computer soﬁware was used to calculate the power of each

analysis.

19

 

 

Exposure Assessment

Quality Control/ Quality Assurance Results:

MRI quality control samples were analyzed in accordance with US-EPA
regulations and will not be reported here. Quality control/quality assurance samples
performed at MSU are examined in this document. The BAS supervised and authorized
all quality control and quality assurance measures taken during the course of this study.
All quality control and quality assurance criteria established by the MSU aquatic
toxicology lab were met. Method spikes for carp eggs and great horned owl livers were
evaluated together to determine their acceptability. Likewise, method and matrix spikes
for American kestrel samples were evaluated as a group. Mean method and matrix spikes
for each group were determined, and spike responses were expected to be within 20% of
the mean value. All method and matrix spikes were within the expected range. Quail
eggs, included in the kestrel egg sample blocks, were spiked with varying concentrations
of PCB 126 and did not show signiﬁcant dioxin-like activity when measured by the
H4IIE-luc bioassay. These results reﬂect the low dioxin-like induction potential of PCB-
126 in the H4IIE-luc cell line (Giesy et al. 1997). Method blank samples were less than
the least TCDD standard concentration as expected, and did not display any dose-
response relationships. A summarization of quality control/quality assurance samples is

listed in Appendix III.

20

 

Residue Concentrations of TEQWHO in Carp Eggs:

The average total dioxin-like activities (reported as TEQwuo) for the carp egg
samples were not greater than 1.6 pg/g. The inclusion of MDL values for congeners with
values less than the MDL inﬂuenced the resulting TEQWHO. The average concentration
of TEQWHO in carp eggs from the RMA ranged from 0.38 to 1.6 pg TEQ/g egg, while the
average concentration of TEQWHO in samples from reference locations ranged from 0.34
to 0.84 pg TEQ/g egg depending on what surrogate proxy values were applied. The
average contribution of PCBs to the total dioxin-like activity ranged from 1.5 to 20%
among samples and methods of calculation (Table 1). Only three non-ortho substituted
PCB congeners were measured in carp egg samples. The contribution of mono-ortho-
substituted PCBs to the TEQwuo was not determined as these congeners were not
measured in the carp egg samples. Detailed results of concentrations of TEQ in

individual samples of carp eggs are summarized in Appendix IV.

Table 1: Mean TEQ concentrations for carp egg samples
(i 95% conﬁdence intervals)

 

 

 

Sample Origin N Total %PCBs Total TEQero- %PCBS
TEQWHOMIN (min) MAX (max)
Pg/g Pg/g
RMA 16 0.38i0.07 20i6.3 1.6-£0.17 4.510.10
Reference 2 0.34i1.3 4.6-: 43 0.84i 0.97 15:12

 

Total TEQwuo= Total concentration of TEQ (PCBs, PCDDs, and PCDFs wet weight)
MIN= 0 used for “non-detect” congeners
MAX = MDL used for “non-detect” congeners

21

 

 

Concentrations of T CDD-EQ in Carp Eggs:

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 egg depending on the choice Of
surrogate values chosen for “non-detect” values (Table 2). Concentrations of TEQREP
were calculated by multiplying concentrations of individual congeners by their H4IIE-luc
REP values. Concentrations of T EQREP were compared with concentrations of TCDD-
EQs. Differences between TCDD-EQ and TEQREp were a function of surrogate values
applied. When “non-detect” congeners were assigned MDL values (denoted MAX), there
was better agreement between concentrations of TCDD-EQ and TEQREP. This indicates
that congeners that occurred at concentrations less than the MDL had the potential to
contribute Signiﬁcantly to the total dioxin-like activity detected in the H4IIE-luc
bioassay. The concentrations of TEQREPMAX were found to be signiﬁcantly different
(Ot=0.05) from concentrations of TCDD-EQMAX when all data points were considered in
the chi-square analysis (X2 = 94.3, df = 17) (Figure 4). 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 = 21.2, df = 14). The three
outlying carp egg samples were collected on the RMA and had greater TCDD-EQ
concentrations than predicted by the TEQREp concentrations (Appendix IV). This is to
be expected since the bioassay measures the total activity of complex mixtures and
accounts for interactions among congeners (Sanderson et al. 1996). Instrumental
analyses measured only target analytes. The H4IIE-luc bioassay can also detect

additional dioxin-like activity from compounds not measured in analytical chemistry

22

 

 

procedures. Complete information on concentrations of TCDD-EQ and T EQREP in

samples of carp egg are summarized (Appendix IV).

 

 

 

 

40 r r T ,r
30 i—- ’IIII —1
TEQREP MAX 20 L— ”x” _ Expected 1:1 ratio
10 *- T
O «’3‘... . l . L .4 .
0 10 20 30 40
TCDDEQMAX

Figure 4: Comparison of carp egg TEQREPMAX
and TCDD-EQMAx values

Table 2: Comparison of mean TCDD-EQ and TEQREP concentrations in carp egg
samples
(i 95% conﬁdence intervals)

 

 

 

sample N TCDD- TEQREP-MIN TC DD-EQMAX TEQREP-MAX
Origin EQMIN (98/ g) (98/ g) (pg/g) (Pg/ g)
RMA 16 4.3 16.1 0.64i0.10 8.5i5.5 3.4:059
Reference 2 0:0.0 0.33i0.95 25:64 15:13

 

TEQ“, = sum of concentrations of individual congeners multiplied by their respective H4IIE
relative potencies
MIN = 0 used for “non-detect” congeners
MAX = MDL used for “non-detect” congeners

23

 

 

Since the concentrations of TEQWHO in carp were normally distributed and the
variances between the two groups were not signiﬁcantly different, the 2-group Student’s
t-test was used to compare the mean values of on—site and reference concentrations. At a
= 0.05, TEQMAX concentrations from carp egg samples on the RMA were signiﬁcantly
lower from reference samples using both separate and pooled variances. TCDD-EQMAX
concentrations were also signiﬁcantly different between locations when separate
variances were used, but there were no signiﬁcant differences when the variances were
pooled (Table 3). The small sample size of carp eggs collected from reference areas
greatly inﬂuenced the difference in variance measurements. Therefore, the pooled
estimate may therefore be a more accurate test statistic. Power (1-l3) was designated as

0.80 in this study. The power of this analysis is low due to small sample sizes.

Table 3: Significance of differences (p values) between locations as determined by
the Student’s t-test (a. = 0.05)

 

 

 

N TEQMAx TCDD-EQMAX
On-site Reference Sep Pool l-B Sep Pool 1-3
16 2 0.004 0.007 1.0 0.045 0.457 0.67
Sep = tests using separate variances Pool = tests using pooled variances

Signiﬁcant differences are in bold.

Residue Concentrations of T EQWHO in Owl Livers:
Average concentrations of TEQWHO in samples of owl livers from the RMA
ranged from 3.3x102 to 3.4x102 pg TEQWHO /g liver, while the average concentration of

TEme in owl livers from reference areas ranged from 35 to 42 pg TEQWHo/g liver

24

 

depending on the proxy values used for congeners below the MDL (Table 4). The
average contribution Of PCBs to total TEano ranged from 24 to 39%, again depending
on the proxy value chosen (Table 4). The relative contribution of PCB congeners to the
total TEQwuo was greater in owl livers than in carp egg samples. Overall, there was
greater variation in TEQwuo concentrations among individual owl liver samples collected
on and off the RMA than was evident in carp egg TEQWHO concentrations. Complete
information on concentrations of TCDD-EQS and TEQS in samples of owl livers are

summarized (Appendix V).

Table 4: TEme concentrations in owl liver samples
(: 95% conﬁdence intervals)

 

 

 

Sample Origin N Total % Total % PCBMAX
TEano-MIN PCBMIN TEQWHO-MAX
(pg/g) (Pg/g)
RMA 16 3.3x102: 39:9.0 3.4 x lo2 : 34:6.0
3.7x102 3.7x102
Reference 1 1 35:1 8 29:13 42:29 24:90

 

Total TEano= Total concentration of TEQwuo (PCBs, PCDDS, and PCDFs)
MIN = 0 used for “non-detect” congeners
MAx= MDL used for “non-detect” congeners

Concentrations of T CDD-EQ in Owl Livers:

Average concentrations Of TCDD-EQ in owl livers collected in reference areas
ranged from 17 to 23 pg TCDD-EQ/g liver and the average concentration in samples
from the RMA was 1.31:102 pg TCDD-EQ/g liver (Table 5). Again, concentrations of

TCDD-EQS were variable among samples of individual owl livers collected on and off of

25

 

 

the RMA (Appendix V). Concentrations of TEQREP were compared with TCDD-EQ. In
this case, concentrations of TEQREP were, on average, greater than concentrations of
TCDD-EQ (Figure 5). 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 (Sanderson et al. 1996). Using the Chi-square analysis, it was
found that concentrations of TCDD-EQMAx and TEQREaMAx differ signiﬁcantly at a=
0.05 (X2 = 86, df = 26) when all data points are considered. By removing a single
outlying point from the analysis (Appendix V), TCDD-EQMAx and TEQREPMAX
concentrations were found to not be statistically different (X2 = 14.2, df = 25). Removal
of the outlying point illustrates that most of the great horned owl samples examined had

similar results when analyzed by instrumental and bioanalytical techniques.

 

 

 

 

3000 l l
2000_ gff' —
TEQREPMAX
1000 r _ Expected l :1 ratio
Ok .3“... I I
0 1 000 2000 3000

TCDD-EQMAx

Figure 5: Comparison of great horned owl TEQREEMAX
and TCDD-EQMAx values

26

 

 

Table 5: Comparison of TCDD-EQ and TEQREP concentrations in great horned owl
liver samples

(: 95 % conﬁdence intervals)

 

 

 

 

 

Sample TCDD- TEQREP-M]N TCDD-EQMAX TEQRELMAX
Orig'n N EQMIN tpgtgl (MEI (931g) M
RMA l6 1.3x 102 2.3x 102 1.3x 102 2.6x 102
:l.5x102 :2.8x102 :1.5x102 :3.1x102
Reference 1 l 17:38 18:18 23:37 30:25

 

TEQREP = sum of concentrations of individual congeners multiplied by their respective H4IIE
relative potencies
MIN = 0 used for “non-detect” congeners
MAX = MDL used for “non-detect” congeners

Since the great horned owl liver TCDD-equivalent concentrations were not
normally distributed, the Mann-Whitney U test was used to test for signiﬁcant differences
between adult and juvenile owl populations (Rand 1995). In order to investigate the
possible outcomes of the analysis between age classes, owls of unknown age were treated
as adults for the ﬁrst analysis, treated as juveniles for the following analysis, and
eliminated in the ﬁnal analysis (Table 6). There were no signiﬁcant differences among
age classes in concentrations of TEQWHOMAX. Therefore, comparisons between
concentrations of TEQWHOMAX from on-site and reference areas were performed on
pooled age classes. Signiﬁcant differences in concentrations of TCDD-EQMAx were
detected among age classes. Therefore, comparisons of concentrations of TCDD-EQMAX

between on-site and reference locations were performed separately. Concentrations of

27

 

 

 

TEQWHOMAX between locations were not signiﬁcantly different. However,

concentrations of TCDD-EQMAX in livers of adult owls were signiﬁcantly different
between locations (Table 7). The power of this analysis was found to be less than 0.8,

therefore the results are inconclusive, and it could not be determined if there were any

differences among locations.

Table 6: Signiﬁcance of differences between age classes as determined by the Mann-
Whitney U test (or = 0.05).

 

 

Manipulation of owl N TEQMAx TCDD-

specimens of unknown Adults Juveniles EQMAx
age
Unknowns treated as 13 (14) 13 0.144 0.046
adults
Unknowns treated as 9 17 (18) 0.146 0.020
juveniles
Unknown eliminated 9 13 0.117 0.021

 

Values in parentheses indicate sample sizes for the TCDD—EQMAX analysis.
Signiﬁcant differences are in bold.

Table 7: Signiﬁcance of differences between locations as determined by the Mann-
Whitney U test (a= 0.05).

 

 

N TEQMAX TCDD’EQMAX
On-site Reference
Analysis across age classes 16 10 0.077
Adults 4 5 . 0.014
Juveniles 8 5 0.604
Power (1-13) 0.50 0.44

 

Signiﬁcant differences are in bold.

28

 

Residue Concentrations of T EQWHO in American Kestrel Eggs:

Mean concentrations of TEQWHO in American kestrel eggs collected from the

RMA were in the range of 21 to 24 pg TEQ/g egg. Concentrations of TEQWHO in

American kestrel eggs collected from reference areas were in the range of 55 to 57 pg

TEQ/g egg (Table 8). Concentrations of TEQWHO in American kestrel eggs collected

from reference areas were variable (Appendix VI), possibly reﬂecting the diversity of L

reference locations utilized in this study. The average contribution of PCBs to the total

dioxin-like activity ranged from 21 to 57% among samples and methods of calculation, Pi

 

which is similar to the results reported for the great horned owl livers. Complete

information on concentrations of TCDD-EQs and TEQS in samples of kestrel eggs is

summarized (Appendix VI).

Table 8: Mean TEano concentrations in American kestrel egg samples
(: 95 °/o conﬁdence intervals)

 

 

 

 

 

Sample Origin N Total TEQwuo. % Total %
MIN PCBMIN TEQWHO-MAX PCBMAX
(22181 mglg)
RMA 30 21:66 55:86 24:6.6 46:6.8
Reference 16 55:58 60:14 57:57 54:13

 

Total TEQwuo= Total concentration of TEQ (PCBS, PCDDs, and PCDFs wet weight)
MIN = 0 used for “non-detect” congeners
MA, = MDL used for “non-detect” congeners

29

Concentrations of T CDD-EQ in American kestrel eggs:

Average concentrations of TCDD-EQ in samples of American kestrel eggs from
the RMA contained 2.4 to 6.7 pg TCDD-EQ/g egg. Average concentrations of TCDD-
EQ in kestrel eggs collected in reference areas ranged from 14 to 18 pg TCDD-EQ/g egg
(Table 9). The variability in concentrations of TCDD-EQ were among individual kestrel
egg samples collected from reference areas as is apparent in the large 95% conﬁdence
intervals for these samples (Table 9). This phenomenon may reﬂect the diversity of
reference sites. Concentrations of TEQREP were compared with TCDD-EQ. As with the
owl livers, concentrations of TEQREP in the kestrel eggs were, on average, greater than
concentrations of TCDD-EQ (Figure 6). Concentrations of TEQREPMAX values were
statistically different from TCDD—EQMAx when all data points were analyzed by the Chi-
square test (X2 = 668, df = 56, ct = 0.05). By removing a single outlier from the analysis,

TCDD-EQMAX and TEQREPMAX concentrations were not signiﬁcantly different (X2 = 40,

df = 55, (F 0.05). Complete information on TCDD-EQ and TEQREp concentrations is

 

 

 

provided (Appendix VI).
H l ,
” Expected 1:1 ratio
0 l l
1 00 1 50 200
TCDD-ECJNIAX

Figure 6: Comparison of concentrations of TEQRmMAx with concentrations
of TCDD-EQMAX in American kestrel eggs

30

 

j I ' 7'1.“ Irrati- r" .H

Table 9: Comparison of TCDD-EQ and TEQREP concentrations in American
kestrel egg samples

(: 95 °/o conﬁdence intervals)

 

Sample N TCDD-

 

 

 

TEQREP-MIN TCDD- TEQREP-MAX
Origin EQMIN (pg/g) Eme (Pg/g)
tests) leafs)
RMA 31 2.4:4.3 15:7.6 6.9:4.l 19:7.9
Reference 16 14:19 19:18 17:18 21:18

 

TEQ“, = sum of concentrations of individual congeners multiplied by their respective H4IIE
relative potencies
mm = 0 used for “non-detect” congeners
MAX = MDL used for “non-detect” congeners

A one-way ANOVA followed by a Dunnett’s multiple comparison were used to
test kestrel eggs collected from the core and periphery areas of the RMA as well as the
reference areas. No signiﬁcant differences in concentrations of TEQWHQ-MAX or TCDD-
EQMAX were found among locations (OF-0.05) (Table 10). Concentrations of both TEQ
and TCDD-EQ in eggs from core and peripheral area were grouped together in order to
compare all on-site and reference areas by use of a Student’s two group t-test. NO
signiﬁcant differences in concentrations of either TEQ or TCDD-EQ were observed (or =

0.05). However, the power level was not sufﬁcient in the t-test to be conclusive (Table
11).

Table 10: Analysis of variance (ANOVA) results for comparison of TEQMAx
concentrations in American kestrel egg samples among location

 

Source Sum of Squares df Mean Square F-ratio p
Site 11470 2 5735 1.345 0.271

 

Error 183282 43 4262

 

Table 11: Student’s t-test results for the comparison of on-site and reference
TEQMAX and TCDD-EQMAX concentrations in American kestrel egg samples

 

N TEQMAx TCDD-EQMAX

 

On-site Reference Sep Pool [-3 Sep Pool 1 -B

30 16 0.241 0.105 0.219 0.246 0.126 0.214

 

Sep = tests using separate variances Pool = tests using pooled variances

32

Effects Characterization

Possible adverse effects of the TEQ and TCDD-EQ concentrations measured in
carp eggs, owl livers, and kestrel eggs are determined by calculation of hazard quotients
(HQs). Observed concentrations were compared to toxicant-speciﬁc Toxicity Reference
Values (TRVS). TRVS are protective values that are calculated based on either a NO
Observable Adverse Effect Level (NOAEL) or a Lowest Observable Adverse Effect
Level (LOAEL) of the toxicant on the species of interest (Equation 3). A NOAEL is
deﬁned as the greatest dose of toxicant tested at which no signiﬁcant deleterious effect
has been observed, and the LOAEL is deﬁned as the lowest dose of toxicant tested at
which a deleterious effect has been observed (Rand et al. 1995). The TRVS used in this
study were derived from well-crafted preexisting studies providing NOAEL and LOAEL
data. Uncertainty factors (UFS) were applied in TRV calculations if the NOAELS and

LOAELS used were in some way non-representative of the species or toxicant of interest.

TRV = NOAEL or LOAEL/ UFs (3)

Derivation of the Carp Egg TR V:

Fish embryos are often the most sensitive life stage to dioxins and dioxin-like
compounds (EPA 1993). The TRV used for carp egg samples was based on a
combination of studies determining sensitivities of various freshwater ﬁsh species to
dioxin-like compounds (Elonen et al. 1998, Henry et al. 1997, Walker and Peterson
1994). The No Observable Adverse Effect Concentration (NOAEC) from these studies

was found to be between 170 and 500 pg T CDD/g egg wet weight. The UF for this TRV

33

 

was based on the extrapolation of laboratory studies to ﬁeld situations, but only amounted
to 1.0. Thus, a TRV for the carp eggs was calculated and found to be in the range of 170
to 500 pg/g egg T CDD-equivalent (Equation 4). The calculation of UFs is explained in

Appendix VII.

TRV = [NOAEL / UF]

TRV = [170-500pg/g 11.01 = 170-500 pg/g (4)

Derivation of the American Kestrel TR V:

The TRV for American kestrel egg samples was derived from a study in which
PCB 126 was injected into kestrel eggs (Hoffman et al. 1998). Eggs were examined for
embryotoxic effects at the close of the study. The route of exposure, egg injection, is
assumed to closely mimic natural, maternal contribution to the developing embryo.
Kestrel embryos were used as the test subject, since it has been observed that they are
more sensitive to dioxin-like compounds than are hatchlings (Hoffman et al. 1996). Of
all the studies examined for TRV derivation, the Hoffman et al. 1998 study presented the
fewest uncertainties. The test organism was the species Of interest, the American kestrel,
and even though PCB 126 was used as the test compound instead of dioxin, a TEF value
of 0.1 was used in order to make the conversion to dioxin equivalents. The uncertainty
factor was determined to be 3.5 based on uncertainties in the extrapolation of a lab study
to ﬁeld conditions and the uncertain NOAEL endpoint. The NOAEL for effects of PCB

126 on kestrel embryos was found to be 2300 pg PCB 126/g egg wet weight. This

34

 

NOAEL value was used to calculate a TRV value of 70 pg/g TCDD-equivalent
(Equation 5).
TRV = [NOAEL * TEF/ UF]

TRV =.- [2300 pg/g * 0.1/ 3.5] = 70 pg TCDD/g egg (5)

Derivation of the Great Horned Owl Liver TR V:

No studies of the effects of TCDD on owls were found in the literature. For this

 

reason, the derivation of the great horned owl TRV was based on the American kestrel
TRV (Hoffman et al. 1998). More uncertainties were associated when determining the
TRV for the great horned owl. Since the NOAEL value had to be extrapolated from
kestrels to owls, an uncertainty factor of 5.0 was applied. Other uncertainties included
extrapolation from the lab to the ﬁeld (UF = 2), an unclear NOAEL endpoint (UF= 1.5),
and an uncertain organ ratio (UF = 1). The organ ratio for converting from egg tissue to
liver tissue was determined to be 1.5 (on a lipid basis) in herring gulls (Braune and
Norstrom 1989). In total the uncertainty factor amounted to 22.5 ((2+1+1.5)*5). Based
on these parameters, the TRV for great horned owls was found to be 15 pg/g wet weight

TCDD-equivalent (Equation 6).

TRV = [(NOAEL * (organ ratio) *TEF)/ UF]

TRV = ((2300 pg/g * (1.5) *0.1)/ 22.5] = 15 pg TCDD/g liver (6)

35

Risk Characterization

Risk Estimation Using Hazard Quotients:

The potential for adverse effects of TEQS and TCDD-EQS concentrations on
wildlife was assessed by calculating a hazard Quotient (HQ). The HQ was determined as
the ratio of the concentration of TCDD-equivalent in the tissue of interest to the reference
dose of toxicant (TRV) determined to have some meaning within a given organism. The

values of HQs are reported as dimensionless Toxic Units (TU) (Equation 7).

HQ (TU) = concentration detected in the tissue/ TRV (7)

Carp Eggs:

Values of HQs for samples of carp egg from neither on nor off of the RMA
exceeded 1.0. The upper 95% conﬁdence intervals for carp HQs were an order of
magnitude less than 1.0 indicating quite conservatively that there are no signiﬁcant
adverse effects to carp as a result of exposure to dioxin-like chemicals (Table 12 and

Appendix IV).

36

 

Table 12: Mean Hazard Quotients (HQs) for carp egg samples

(: 95 "/6 conﬁdence intervals)

 

 

 

 

Sample HQMIN HQMAX HQMIN HQMAx
Origin N (TEQ based) (TEQ based) (TCDD-EQ (TCDD-EQ
based) based)
RMA 16 0002:0001 0.01:0.001 0.02:0.035 0.05:0.032
Reference 2 0002:0008 0005:0006 0000:0000 0015:0064
Owl Livers:

 

Mean HQ values for owl livers collected from reference areas were less than
mean HQ values for owl livers collected on-post for all methods of calculation employed
(Table 13). HQ values for several liver samples collected from on or off of the RMA
exceeded 1.0 TU. The HQMAX (TEQ based) exceeded 1.0 TU for 8 samples from
reference areas and 14 samples from the RMA. The HQMAX (TCDD-EQ based) exceeded
1.0 TU for 3 owl liver samples from reference Sites and 5 samples from on the RMA
(Appendix V). HQ values differed considerably among individual owl liver samples
collected on and off the RMA (Appendix V). This variation could be a result of the

diversity in age, location, and cause of death of the owl specimens.

37

Table 13: Mean Hazard Quotients (HQs) for great horned owl liver samples
(1: 95 % conﬁdence intervals)

 

 

 

 

 

Sample HQMIN HQMAX HQMIN HQMAX
Origin N (TEQ based) (TEQ based) (TCDD-EQ (TCDD-EQ
based) based)
RMA 16 22:25 22:25 8.7:10 8.9:10
Reference 11 2.3:1 .7 28:1 .9 l.l:2.5 16:25
American Kestrel Eggs:

Mean HQ values for samples of kestrel egg from neither on nor off of the RMA
exceeded 1.0. The results of the risk assessment indicate that there is no Signiﬁcant
adverse effects toward American kestrels as a result of exposure to dioxin-like chemicals

(Table 14 and Appendix VI).

Table 14: Mean Hazard Quotients (HQs) for American kestrel egg samples
(: 95 % conﬁdence intervals)

 

 

 

Sample HQMIN HQMAx HQMIN HQMAX
Origin N (TEQ (TEQ based) (TCDD-EQ (TCDD-EQ
based) based) based)
RMA 30 (31) 0.30:0.10 0.34:0.09 0.04:0.06 0.10:0.06
Reference 16 0.78:0.82 0.81:0.82 0.19:0.27 0.24:0.26

 

The value in parenthesis indicates sample size for TCDD-EQ analyses

38

Risk Description: Ecological Signiﬁcance of the Results

If the HQ is greater than 1.0 TU, then the detected concentrations in tissue
samples are greater than the desired threshold concentration. Due to the inclusion of
uncertainty factors, TRVS are conservative values calculated to be protective of wildlife.
Therefore, adverse population effects may not occur at levels of contamination resulting
in HQ values slightly exceeding 1 TU, but TU values of 10-20 can be indicatative of
population-level effects, depending on the slope Of the dose-response curve (Giesy and
Kannan 1998).

The results of the risk assessment, based on HQS, for the carp eggs indicate that
aquatic systems on the RMA are not at risk as a result of exposure to dioxin-like
compounds. Carp eggs collected from reference areas contained statistically less
concentrations of TEQWHOMAX and TCDD-EQMAX than did carp eggs collected on the
RMA. However, the small sample sizes utilized in this analysis (n =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. A larger sample size in the reference population
would allow for more representative results from this location.

In this study, livers of great horned owls contained the greatest concentrations of
TEQS and TCDD-EQS, and exhibited the greatest HQ values of the three species studied.
The terrestrial RMA environment had varying results in the risk assessment analysis.
Some individual great horned owls on the RMA contained sufﬁcient concentrations of
TCDD-equivalents to result in HQ values greater than 1.0. No owl livers collected from

reference areas had HQS greater than 20 T Us. However, depending on the use of proxy

39

 

values, HQ values of 2 to 4 samples of owl livers from the RMA were greater than 20
TUs (Appendix V). This result indicates that great horned owls on the RMA may be at
risk due to exposure to dioxin-like chemicals. Adult great horned owl livers collected
from the RMA had statistically greater concentrations of TCDD-EQMAX than the adult
owl livers collected from surrounding reference areas. This may indicate that there is a
localized source of dioxin-like compounds in the terrestrial environment at the RMA.
However, the results are inconclusive due to the low power associated with the
comparison of concentrations of TEQ and concentrations of TCDD-EQ among locations.
An increased sample size would be required to increase the power to 0.80.

HQ values for samples of American kestrel eggs were not greater than 1.0 for any
locations. Concentrations of TEQ and TCDD-EQ in kestrel eggs also did not Show
signiﬁcant differences between on-site and 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 kestrel egg and owl liver
samples may indicate that there are separate terrestrial pathways for exposure to dioxin-

like chemicals.

40

 

Study Uncertainties

Quantiﬁcation of concentrations of TCDD-EQS by use of the H4IIE-luc in vitro
bioassay is based on luminescence readings and is thus an indirect measure of the relative
dioxin potency of the sample extract. Activation of the luminescence gene only indicates
that translocation and binding of the AhR has occurred in the nucleus. The bioassay was
not used as a direct measure of the potential toxic effect of dioxin-like activities, but
rather used as an integrative measure of the total concentration of TCDD-EQ. The
responses to the effects Of TCDD-EQS vary among species and endpoints. Thus, there is
some uncertainty in interpreting the toxicological relevance of the TCDD-EQS measured
by the induction of luciferase under control of a dioxin-responsive enhancer (DRE).
Another uncertainty associated with the application of the H4IIE-luc bioassay is that the
responses detected can not be solely attributed to PCDD, PCDFs, and PCBs since this
bioassay is an integrative measure of all of the individual AhR-active compounds and all
of the potential interactions among both AhR-active and inactive congeners (Giesy and
Kannan 1998, Williams et al 1995). Uncertainties also exist when results are
extrapolated from an in vitro situation to an in vivo situation. Complex interactions, like
metabolism by endogenous enzymes, may occur in vivo to cause activation or
inactivation of dioxin-like chemicals, and this phenomenon may not be measured in an in
vitro test. In the risk assessment, concentrations of TCDD-EQs estimated in the bioassay
were compared to TRV values.

TEQ values calculated by the application of TEFS or REPS are uncertain because

there is an inability to account for chemical interactions that occur in complex mixtures.

41

 

TEFS and REPS assume that interactions between chemicals are additive, and these two
factors can not account for synergistic or antagonistic interactions that may increase or
decrease toxic activity.

There are also inherent uncertainties in the derivation of TRV values for the three
species utilized in this experiment. These uncertainties, including lab to ﬁeld
extrapolation, species extrapolation, inadequate study designs, and uncertain endpoints
are taken into account when calculating a ﬁnal TRV, but do not lead to accurate values.
TRVS are formulated in order to be protective of organisms, therefore they are
conservative estimates. In cases such as the great horned owl, there is not sufﬁcient
experimental evidence on the test species for the estimation of an accurate TRV.

The three species selected in this study were chosen to act as sentinels for the
entire RMA wildlife population. These sentinel species yielded general information on
the distribution of TCDD-EQ and TEQ concentrations in tissue samples from the RMA
environment, but could not accurately represent every species present on the RMA.
Therefore, some uncertainty in the results is associated with the inability to measure
concentrations of TCDD-EQS and TEQS in every species present on and off the RMA. In
addition, great horned owl specimens were collected opportunistically, which meant that
some owls were found dead in response to an environmental stressor independent from
PDH exposure. The type of death of some owl specimens may have affected the
measurement of TCDD-equivalents in these samples. For example, loss of lipid content
occurs in the liver tissue of organisms that expire due to dieldrin poisoning (Stickel et al.

1969). This loss of lipid content in dieldrin-poisoned owls may have led to an increased

42

 

partitioning of TCDD-equivalents in liver samples, resulting in higher observed TCDD-
equivalent concentrations.

These uncertainties should be taken into account when interpreting the
results of this study. However, the dioxin-equivalent concentrations determined by
instrumental analyses and the H4IIE—luc bioassay were similar for most of the collected
tissue samples, giving strength the accuracy of the results. Overall, some signiﬁcant
differences in dioxin-equivalent concentrations were found between RMA and reference
locations depending on which dioxin-equivalent methodology was used. TEQ
concentrations in carp eggs and TCDD-EQ concentrations in adult great horned owl
livers differed signiﬁcantly among RMA and Reference locations. In contrast, dioxin-
equivalent concentrations in juvenile owl livers and in kestrel eggs did not vary
signiﬁcantly among locations. These conﬂicting results, the small sample sizes, and the
low power associated with most of the analyses, render the study inconclusive as to
whether or not dioxin-like chemicals are contaminants of concern on the RMA. It is
apparent from the Hazard Quotient results that carp and American kestrel populations are
not likely at risk due to dioxin-like exposure. However, some great horned owls from
both RMA and Reference locations do appear to be at risk if the calculated TRV is
reliable. It appears that adult great horned owls from the RMA may be at risk due to a

combination of both dieldrin poisoning and dioxin-like contamination.

43

 

References

l. Braune, B. G. and R. J. Norstrom. Dynamics of organochlorine compounds in
herring gulls III. Tissue distribution and bioaccumulation in Lake Ontario,
Canada! USA. Environmental Toxicology and Chemistry 1989. 8:957-

968.

2. Couture, L. A., B. D. Abbott, and L. S. Bimbaum. A critical review of the

 

developmental toxicity and teratogenicity of 2,3,7,8-tetrachlorodibenzo-p-
dioxin: recent advances toward understanding the

mechanism. T eratologv 1990. 42: 619-627.

3. Elonen, G. E., R. L. Spehar, G. W. Holcombe, R. D. Johnson, J. D. Fernandez, J.E.
T ietge, and P. M. Cook. Comparative toxicity of 2,3,7,8-tetrachlorodibenzo-
p-dioxin to seven freshwater ﬁsh species during early life-stage development.

Environmental Toxicology and Chemistry. 1998 17:472-483.

4. Giesy, J. P., D. A. Verbrugge, R. A. Othout, W. W. Bowerman, M. A. Mora, P. D.
Jones, J. L. Newsted, C. Vandervoort, S. N. Heaton, R. J. Aulerich, S. J.
Bursian, J. P. Ludwig, G. A. Dawson, T. J. Kubiak, D. A. Best, and D. E.
Tillitt. Contaminants in ﬁshes from Great Lakes inﬂuenced sections and
above dams of three Michigan rivers. 1: Concentrations of organochlorine

insecticides, polychlorinated biphenyls, dioxin equivalents, and mercury.

Arch.Environ. Contam. Toxicol. 1994. 27: 202-212.

44

5. Giesy, J. P., D. A. Verbrugge, R. A. Othout, W. W. Bowerman, M. A. Mora, P. D.
Jones, J. L. Newsted, C. Vandervoort, S. N. Heaton, R. J. Aulerich, S. J.
Bursian, J. P. Ludwig, G. A. Dawson, T. J. Kubiak, D. A. Best, and D. E.
Tillitt. Contaminants in ﬁshes from Great Lakes — inﬂuenced sections and
above dams of three Micigan rivers. II: Implications for health of mink.

Arcives of Environmental Contamination and Toxicology. 1994. 27: 213-223. L

T- IJ.‘.WJ . I .L‘V"
._ _... _ .
l . -

6. Giesy, J. P., D. J. Jude, D. Tillitt, R. W. Gale, J. C. Meadows, J. L. Zajieck, P. H.

 

Peterrnan, D. A. Verbrugge, J. T. Sanderson, T. R. Schwartz, and M. L.
Tuchman. Polychlorinated dibenzo-p—dioxins, dibenzofurans, biphenyls and
2,3,7,8-tetrachlorodibenzo-p—dioxin equivalents in ﬁshes from Saginaw Bay,

Michigan. Environmental Toxicology and Chemistry 1997. 16:713-724.

7. Giesy, J. P., and K. Kannan. Dioxin-like and non-dioxin-like toxic effects of
polychlorinated biphenyls (PCBs): implications for risk assessment. Critical

Reviews in Toxicology 1998. 28:511-569.

8. Henry, T. R., J. M. Spitsbergen, M. W. Homung, C. C. Abnet, and R. E. Peterson.
Early life stage toxicity of 2,3,7,8-tetrachlorodibenzo-p-dioxin in zebraﬁsh

(Danio rerio). Toxicology and Applied Pharmacology. 1997 142:56-68.

9. Hoffman, D.J., MJ. Melacon, P.N. Klein, C.P. Rice, J .D. Eisenmann, R.K. Hines,

J.W. Spann, and G. W. Pendleton. Developmental toxicity of PCB-126 (3, 3’,

45

 

4, 4’, 5, 5’, pentachlorobiphenyl) in nestling American kestrels (Falco

sparverius). Fundamental and Applied Research 1996. 34: 188—200.

10. Hoffman, D.J., M.J. Melacon, P.N. Klein, J.D. Eisenmann, and J .W. Spann.
Comparative developmental toxicity of planar PCB congeners in chickens,
American kestrels, and common terns. Environmental Toxicology and

Chemistry 1998 17: 747-757.

11. Kannan, K., I. Watanabe, and J. P. Giesy. Congener proﬁle of
polychlorinated/brominated dibenzo-p-dioxins and dibenzofurans in soil and
sediments collected at a former chlor-alkali plant. Toxicological and

Environmental Chemistry 1998 67: 135-146.

12. Kuehl, R. 0. Statistical Principles of Research Design and Analysis. 1994

Wadsworth Publishing Co., Belmont

13. Jones, P. D., J. P. Giesy, J. L. Newsted, D. A. Verbrugge, D. L. Beaver, G. T.
Ankley, D. E. Tillit, K. B. Lodge, and G. J. Niemi. 2,3,7,8
tetrachlorodibenzo—p-dioxin equivalents in tissues of birds at
Green Bay, Wisconsin, USA. Arch.Environ. Contam. Toxicol. 1993. 24: 345-

354.

46

 

 

l4. Peterson, R. E., H. M. Theobold, and G. L. Kimmel. Development and
reproductivetoxicity of dioxins and related compounds: cross-species

comparisons. Critical Reviews in Toxicology. 1993 23: 283-335.

15. Rand, G. M. Fundamentals of Aquatic Toxicology. 1995. Taylor and Francis,

Washington, DC.

16. Sanderson, J. T., J. M. M. J. G. Aarts, A. Brouwer, K. L. Froese, M. S. Dennison,
and J. P. Giesy. 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

hydocarbons. Toxicologyand Applied Pharmacology 1996. 137:316-325.

17. Stickel, W. H., L. F. Stickel, and J. W. Spann. Chemical Fallout; Current Research

on Persistent Pesticides Proceedings of the First Rochester Conference on

Toxicity. 1969. Springﬁeld. Chapter 8.

18. US. Army. Rocky Mountain Arsenal: Turning Vision into action. [Online]

Available December 16, 1999.

19. United States Environmental Protection Agency (US. EPA). Interim report on data

and methods for assessment of 2,3,7,8 —-tetrachlorodibenzo-p-dioxin risks

47

 

 

to aquatic life and associated wildlife. Ofﬁce of Research and Development,

Washington, DC. EPA/600/R-93/055. 1993.

20. United States Environmental Protection Agency (US. EPA). Framework for

ecological risk assessment. Ofﬁce of Research and Development, Washington,

DC. EPA/832/R/92/008a. 1992.

21. Van den Berg, M., L. Bimbaum, A. T.C. Bosveld, B. Brunstrom, P. Cook, M.
F eeley, J. P. Giesy, A. Hanberg, R. Hasegawa, S. W. Kennedy, T. Kubiak, J. C.
Larsen, F .X. Rolaf van Leeuwen, A.K. D. Liem, C. Nolt, R. E. Peterson, L.
Poellinger, S. Safe, D. Schrenk, D. Tillitt, M. Tysklind, M. Younes, F. Waem, and
T. Zacharewski. Toxic Equivalency Factors (TEFS) for PCBs, PCDDs, PCDFs

for humans and wildlife. Environmental Health Perspectives 1998. 106: 775-792.

22. Villeneuve, D. L., A. L. Blankenship, and J. P.Giesy. Derivation and application of

relative potency estimates based on in vitro bioassay results. Environmental

Toxicology and Chemistry In press.

23. Walker, M. K. and R. E. Peterson. Aquatic toxicity of dioxins and related

chemicals. In: Dioxins and Health. 1994 Plenum Press, New York. Chapter

11.

48

24. Williams, L. L., J. P. Giesy, D. A. Verbrugge, S. Jurzysta, and K. Stromborg.
Polychlorinated biphenyls and 2,3,7,8-tetrachlorodibenzo-p-dioxin
equivalents in eggs of double crested cormorants from a colony near Green
Bay, Wisconsin, USA. Archives of Environmental Contamination and

Toxicology. 1995 29:327-333.

25. Xu, Y., W.Z. Wu, W. Li, K.-W. Schramm, and A. Kettrup. 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. 1997.

33: 371-376.

49

Appendix 1: Standard Operating Procedure for the H4IIE-luc Bioassay for

the Detection of AhR Agonists (MSU)

50

. Michigan State University
National Food Safety and Toxicology Center
Aquatic Toxicology Laboratory

W
H4IIE-luc Bioassay For The Detection Of Ah Receptor Agonists
Version 1 September 14, 1998

Alan Blankenship, Dan Villeneuve, J. Thomas Sanderson,
Sarah Cholger, Katherine Kemler, and John P. Giesy

Supported through:

 

National Food Safety and Toxicology Center,
Institute for Environmental Toxicology,
and the Department of Zoology

Correspondence to:

Aquatic Toxicology Laboratory
224 National Food Safety and Toxicology Center
Michigan State University
East Lansing, MI 48824-1222 USA
T: (517) 432-6333
F: (517) 432-2310

 

SOP No.

Document Control No.

Dale Issued:

lfthis is printed in RED. this is a controlled document
and is NOT to be reproduced

 

 

 

 

 

 

51

APPROVAL PAGE

Revisions to an existing SOP, addition of an SOP change form, or preparation of a new
SOP must be reviewed, approved, and signed by the following:

 

 

 

 

 

 

 

 

Authored By: Date:
Supervisor Review By: Date:
Reviewed By: Date:
(QA Coordinator)

Reviewed By: Date:
(CALS Program Manager)

52

AhR

ATL
DCCFBS
ECSO
FBS
H4IIE-luc
luciferase

PBS
RLU
TCDD

DEFINITIONS AND ACRONYMS

Aryl hydrocarbon receptor

Aquatic Toxicology Laboratory (Michigan Sate University)

Fetal bovine serum that has been charcoal-stripped

Conentration of test agent that causes 50% of maximal response
Fetal bovine serum

Rat hepatoma cells stably transfected with an AhR-controlled

reporter gene construct
Phosphate-buffered saline
Relative luminescent units
2,3,7,8-tetrachlorodibenzo-p-dioxin

53

TABLE OF CONTENTS

1.0. PURPOSE ................................................................................................................ 5
2.0. SCOPE AND APPLICATION ............................................................................... 5
3.0. SAFETY CONSIDERATIONS ............................................................................. 7
4.0. EQUIPMENT, MATERIALS, AND REAGENTS .............................................. 7
4.1. Maintenance, preparation and use of H4IIE-luc cells ................................... 7
4.2. Instruments ...................................................................................................... 8
4.3. Supplies and Biochemicals ............................................................................. 9
4.4. Media Preparation ........................................................................................... 10
5.0. METHOD, PROCEDURES, AND REQUIREMENTS ....................................... 11
5.1. Sample Preparation ......................................................................................... 11
5.2. Standards Preparation ..................................................................................... 11
5.3. Plating Cells (Day 1) ....................................................................................... 12
5.4. Dosing Cells (Day 2) ....................................................................................... 13
5.5. Conducting the Bioassay (Day 5) ................................................................... 14
5.6. Protein Determination ..................................................................................... 16
5.7. Data Analysis .................................................................................................. 17
6.0. RECORDS, DOCUMENTATION, AND QC REQUIREMENTS ................... 17
6.1 Records and Documentation ......................................................................... 17
6.2 QC Requirements and Data Quality Objectives ........................................... 18
7.0 RESPONSIBILITIES ........................................................................................... 18
8.0 REFERENCES ..................................................................................................... 18
ATTACHMENTS

Attachment 201-1: Figure 1 - T CDD Dose-response Curve with H4IIE-luc Cells
Attachment 201-2: Figure 3 - Viability Assay Sample Data Set (Raw Data)
Attachment 201-3: Figure 4 - Viability Assay Sample Data Set (Calculated Data)
Attachment 201-4: Figure 5 - Luciferase Assay Sample Data Set (Raw Data)
Attachment 201-5: Figure 6 - Luciferase Assay Sample Data Set (Calculated Data)

54

1.0. PURPOSE

2.0

The H4IIE-luciferase induction assay is an in vitro technique for the identiﬁcation of
aryl hydrocarbon (Ah) receptor-active compounds. The technique uses rat
hepatoma cells (H4IIE-luc) stably transfected with an AhR-controlled
luciferase reporter gene construct were developed at Michigan State
University by Dr. lac Aarts (Univ. of Wageningen, The Netherlands;
Sanderson et al., 1996). The assay is also referred to as the Chemical
activated luciferase gene expression (CALUX) system (Murk et al., 1996).
These cells express ﬁreﬂy luciferase in response to Ah receptor agonists. Luciferase
activity is measured conveniently and with high sensitivity as light emission using a
plate-scanning luminometer. Luciferase induction potential is assessed by comparison
of the response to that of 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD), the most potent
agonist for the mammalian Ah receptor.

SCOPE AND APPLICATION

Ah receptor agonists include polyhalogenated aromatic hydrocarbons (PHAHs),
such as polychlorinated biphenyls (PCBs), dibenzo-p-dioxins (PCDDS) and
dibenzofurans (PCDFs) which are persistent environmental contaminants found in
all parts of the world. A number of these compounds cause a variety of adverse
effects in laboratory studies on rats and mice (reviewed by Poland and Knutson
1982). These include hepatotoxicity, certain types of cancer, thymic atrophy and
other immunotoxicities, a wasting syndrome, reproductive toxicities, terata and the
induction of enzymes and porphyrins. A number of these toxicities have also been
observed in wildlife in areas with elevated levels of PHAHS, particularly in ﬁsh-
eating birds in the Great Lakes (reviewed by Gilbertson et al. 1991; Giesy et al.
1994a;1994b)

Interest exists in assessing the risk posed by these PHAHS to ﬁsh and wildlife,
which may also reﬂect the risk to humans. One aspect of risk assessment is the use
of bioanalytical assays to detect and determine the toxicity of complex mixtures of
these chemicals in extracts of environmental compartments such as soil, water and
biota. Quantitative instrumental analysis of complex mixtures of these compounds
is a difﬁcult and expensive task. Furthermore, demonstrating the presence of one or
many of these compounds in samples provides only limited information on their
biological potency, particularly when present in a complex mixture with many
potential interactions.

In order to develop a suitable bioassay, an understanding of the mechanism of
action for the compounds is required. In the case of a number of PHAHS,
considerable knowledge of the mechanism by which they cause their toxicities has
been acquired. As previously mentioned, PHAHS are persistent agonists for the Ah
receptor (Poland and Knutson 1982). Binding of agonist to receptor results in an
activated receptor-ligand complex that translocates to the nucleus. Here it interacts
with speciﬁc sequences on the DNA, termed dioxin-responsive elements (DREs;

55

also called xREs or AhREs), in order to alter gene transcription (reviewed by
Whitlock 1990; Okey et al. 1994). A rapid and sensitive response that is under
direct regulation of the Ah receptor is the induction of cytochrome P-450 1A
isoenzymes (CYP1A1 and 1A2) and their associated ethoxyresoruﬁn O-deethylase
(EROD) activity (Nebert and Gonzalez 1987). Good correlations exist between the
Ah receptor-binding afﬁnity of persistent PHAHS and their EROD-inducing
potency in vitro, and, dependent on the endpoint, their toxic potential in vivo (Safe
1986; 1990). 2,3,7,8-Tetrachlorodibenzo-p-dioxin (TCDD), in particular, binds
tightly to the Ah receptor, and is a potent inducer of EROD activity and highly
toxic. For the above reasons, the capacity of single compounds or complex
environmental mixtures to induce EROD activity is considered to be a reasonable
measure of their toxic potential. This mechanistic knowledge has been applied in
vivo in biological monitoring of ﬁsh and birds, and in vitro in the use of bioassays
for the screening of environmental extracts for Ah-active components. The H4IIE
rat hepatoma cell bioassay (Tillitt et al. 1991) is widely used for this purpose. In
this assay, EROD-inducing potencies (ED50 values) of single compounds and
environmental samples are determined from complete dose-response curves and
compared to that of TCDD in order to express the biological potency of the tested
samples in TCDD-equivalents (TCDD-EQS). The bioassay integrates potential
non-additive interactions among Ah receptor agonist and between Ah receptor
agonists and other compounds by measuring a ﬁnal receptor-mediated response
(Giesy et al. 1994a).

For the same purpose as the currently used wild-type H4IIE bioassay
(H4IIE-wt), a recombinant H4IIE cell line (H4IIE-Inc) has been developed that
exhibits Ah receptor-mediated luciferase expression (Aarts et al. 1995). This cell
line has been stably transfected with a luciferase reporter gene under
transcriptional control of several DRES from mouse (Aarts et al. 1993; Denison et
al. 1993). These DRE sequences are highly conserved among species, unlike the
Ah receptor which can exhibit greatly different ligand-binding properties among
species. A preliminary report using luciferase-transfected mouse Hepalclc7
hepatoma cells indicated that these cells are more sensitive to Ah receptor
agonists than the wild-type cells (Aarts et al. 1993; Sanderson et al., 1996). It
has been suggested that luciferase-transfected cell lines would have more
favorable properties than their respective wild-types, such as greater selectivity,
sensitivity and dynamic range. This has been postulated because the Ab receptor-
mediated expression of luciferase, being foreign to the cell, is probably not
affected by post-transcriptional and -translational events which inﬂuence
CYP1A1 expression. Furthermore, the recombinant cells would not be dependent
on a ﬁrnctional CYP1A1 gene or protein for responsiveness, although the Ab
receptor-mediated pathway would still need to be present. Another theoretical
consideration is that luciferase is assayed on the basis of light production for
which extremely sensitive detectors exist; also, the turnover number or molecular
activity of luciferase is so high that it allows the detection of very few molecules
of the enzyme, relative to CYP1A1. Further, recombinant cells are readily

56

amenable to further improvements in responsiveness by genetic engineering of the
reporter gene construct.

The threshold dose (i.e., detection limit) and EDso (i.e., effective dose to elicit
50% of the maximal response) for luciferase induction in H4IIE-Inc cells were
approximately 0.1 and 1.2 pg/well, respectively, as determined from 41 separate
standard TCDD curves analyzed in 1997 (Figure 1). Coefﬁcients of variation
(standard deviation/mean x 100) for the assay were under 10% at all
concentrations tested. For a sample size of 20 g tissue and a ﬁnal extract volume
of 0.25 ml, the H4IIE-Inc assay will detect 1 part per trillion (ppt; pg/g, wet
weight) TCDD-equivalents. With a sample size of 5 g tissue, 4 pg/g (wet
weight) TCDD-equivalents will be detected.

3.0. SAFETY CONSIDERATIONS

T CDD and many related compounds have been found to be carcinogenic. In addition,
the ethidium homodimer used in the cell viability assay is a powerful mutagen. Care
should be taken to minimize exposure. According to institutional guidelines (refer to
the Safety Manual for the Aquatic Toxicology Laboratory at Michigan State
University), medium should be collected in a liquid trap for disposal as hazardous
waste.

4.0. EQUIPMENT, MATERIALS, AND REAGENTS

4.1. Mamiemnmparahenandmfﬂﬂllﬂucsclls

The assay uses a stably transfected cell line developed by Dr. Jac Aarts at
Michigan State University (Sanderson et al., 1996). Brieﬂy, rat hepatoma
cells [American Type Culture Collection (ATCC) catalog #CRL 1548)] were
stably-transfected with an inducible reporter plasmid, pGudLucl.l. This
plasmid contains the ﬁreﬂy luciferase gene under PHAH-inducible control of
four DRES. Exposure of these cells to Ah receptor-active chemicals results in
induction of luciferase activity in a time-, dose-, and AhR-dependent manner.

4.1.1. Maintain adherant cells in continuous culture in 100 mm tissue

culture plates (Corning #25020-100, Cambridge, MA; 1-800-492-

1110), 75 cm2 ﬂasks (Corning #25113-75) or any appropriate vessel

at a maximum density of 80-90% conﬂuence in 10% Full Medium

(See Media Preparation). Incubator conditions are 37 C, 5% C02
humidiﬁed atmosphere.

4.1.2. Subculture cells 1:6 every week (depending upon density)
maintaining a minimum cell density of 15-25%.

57

4.2. Instmmems

4.2.1. Dynatech ML3000 Luminometer
(Dynatech Technical Support: (800) 336-4543; Chantilly, VA)

4.2.2. Cytoﬂuor 2300/2350 Fluorescence Measurement System
(Millipore Technical Support: (800) 645-5476; Bedford, MA)

NOTE: All users of the Dynatech ML3000 Luminometer and the Cytoﬂuor 2300 must read
and be familiar with the Operators Manual before using the instrument. A working
knowledge of Microsoft Windows is also necessary. For both the luminometer and
ﬂuorescence measurement system, instrument use is recorded in a log book (located next to
the instrument). For each use, the person, date, number of samples, type of samples, and
results of proﬁciency standards are recorded. In addition, any abnormal operation of the
instrument is recorded. Calibration is performed by analyzing a positive control or sample
with known activity each day that the assay is run. If the positive control exceeds : 20% of
the on-going average (as determined by a comparison to a proﬁciency curve maintained in
the instrument log book), the positive control will be rerun. If exceedance is conﬁrmed with
the second analysis, the manufacturer will be contacted as a corrective action so that the

instrument can be recalibrated.

4.2.3. Pipets

Eppendorf Repeat Pipetter (Brinkrnann Instruments #22 26 000-
6; Westbury, NY) with sterile, Biopur 12.5 ml combitips
(Brinkrnann Instruments #22 49 520-8) for dispensing cells into
the 96-well plate and for changing media. Calibration checked
weeldy by weight of water check and the results are entered in
calibration log book. Adjustments made by MSU Biochemistry
Instrument Shop when accuracy exceeds manufacturer’s

speciﬁcations of: 0.3%.

a.

b. Rainin Pipetrnan Pipets (0.5 - 10 pl capacity, Rainin #P-10; 20 -
200 pl, Rainin #P-200 ; 100 - 1000 pl, Rainin #P-1000;
Wobum, MA) for making sample dilutions, dosing cells, etc.
Calibration checked weekly by weight of water check and the
results are entered in calibration log book. Adjustments made by
MSU Biochemistry Instrument Shop when accuracy exceeds
manufacturer’s speciﬁcations of i 2.5% for P-lO, 1.0 % for P-

200, and 3.0% for P-1000.

Brinkmann Eppendorf Multichannel Pipetter 30-300 ul
(Brinkmann Instruments #22 45 120-1) for washing cells and
adding reagents for viability and Luclite reagents. Calibration
checked weekly by weight of water check and the results are
entered in calibration log book. Adjustments made by MSU

58

Biochemistry Instrument Shop when accuracy exceeds

manufacturer’s speciﬁcations of i 1.5%.

d. Drummond Pipet-Aid Filler/Dispenser (Drummond #4-000~111-
TC, available from Fisher #13-681-150) for dispensing volumes
greater than 1 ml, such as for changing media on cells.

NOTE: For all pipettors, except the Drummond Pipet-A id, calibration test results are
recorded weekly, along with the name of the person and any abnormal operation of the

instrument.

4.3. Supplmmﬂmhemicals

4.3.1.

4.3.2.

4.3.3.

96-Well ViewPlatesTM (Packard Instruments #6005181; Meriden,

CT); white 96-well plates, sterile, tissue culture treated, with lids and
self-adhesive sticker for bottom of plates

Cost: $297/50 plates;
Ordering: (800) 856-0734
Technical support: (800) 323-1891

Cell viability assay reagents; sold either as a kit from Molecular
Probes (#L-3224; Eugene, OR) or as individual components:

Calcein AM (Molecular Probes #C-3100);
MW = 994.87; made up as 4000x (2 mM) stock (50 pg/ 12.56 pl

DMSO)

Ethidium homodimer I (Molecular Probes #E l 169)
MW = 857; made up as 2000x (1 mM) stock in DMSO

Ordering: (800) 438-2209
Technical support: (541) 465-8353

LucLitem Kit (Packard Instruments # 6016911 - 1000 assay kit;
Meriden, CT). Make fresh on same day as assay. Dissolve one
bottle of lyophilized reagent with 10ml buffer (supplied) for every

133 assays (individual wells) to be analyzed.

Cost: $420/l333 assays using 75 pl/assay (or 1000 assays if using
the manufacturer’s suggested volume of 100 pllassay. Preliminary

studies showed equivalent responses at both 75 pl and 100 pl).

Ordering: (800) 856-0734
Technical support: (800) 323-1891

59

4.3.4.

4.3.5.

10x Trypsin-EDTA solution (Sigma #T-4174, St. Louis, M0) for
dissociation of cells from plates. From a 10x concentrated solution,
a 1x working solution is prepared using sterile PBS as the diluent.

Dulbecco’s phosphate-buffered saline (PBS) with Ca2+ and Mg2+
First, make up 10 L of Cay-free and Mg2+- free Dulbecco’s PBS:

2.0 g KCl

2.0 g KH2P04
80.0 g NaCl
11.5 g NazHPO4
10.0 L H20a

“distilled/deionized or Nanopure biological grade

To each 1 L of PBS, add 0.1 g anhydrous CaClz and 0.1 g MgClz 0
61120 Keep at room temperature (good for at least 2 - 4 weeks).

4.4. mm

4.4.1 .

4.4.2.

10% Full Medium

Dulbecco’s Modiﬁed Eagle’s Medium (Sigma #D-2902) without
nhenglnd (known to be estrogenic and with unknown inﬂuence on
H4IIE cells), and sodium bicarbonate. Prepare as instructed by the
manufacturer, adjust the pH to ~7.3, and then add :

10% fetal bovine serum (Hyclone deﬁned FBS # 8113007003;
Logan, UT; 1-800-492-5663)

10% DCCFBS Medium
Prepared as for the full medium, except the fetal bovine serum is

replaced with dextran/charcoal-stripped fetal bovine serum (available
from Hyclone #SH30068.03).

6O

5.0 METHOD, PROCEDURES, AND REQUIREMENTS

5.1.

5.2.

Sample Preparation

Two types of samples may be assayed: pure compounds and mixtures derived
from environmental or tissue samples. For pure compounds, sample preparation
consists of dissolving the material in an appropriate solvent. The solvent of
choice is isooctane (because of its low toxicity and for direct comparison to
TCDD, which is dissolved in isooctane). However, if the material is insoluble in
isooctane, the following solvents can be tried:, ethanol, acetone, pdioxane,
acetonitrile, and methanol. A stock solution should be prepared at 5 mM for
compounds of unknown activity and stored in amber glass vials at -20°C.
However, it may be necessary to test concentrations near the limit of solubility
of the compound. Ideally, a volume of 1-2 ml at the highest possible
concentration should be obtained for standards and samples. This volume is
sufﬁcient for splitting samples for instrumental analysis and for preparation of
serial dilutions.

Samples derived from environmental matrices such as tissue, water, or sediment
should be extracted and concentrated according to appropriate protocols (in
accordance with SOP #211 - Extraction and Analysis of PCBs and Non-ortho
Coplanar PCBs in Biological Matrices). The volume of sample should be
recorded before re-dissolution in the assay solvent so that a dilution or
concentration factor can be calculated. To prepare dilutions, 0.5 ml of isooctane
(assuming that this is the solvent of choice) is added to ﬁve appropriately labeled
2, ml amber GC glass vials with teﬂon lined caps. Then 0.5 ml of the 1x extract
(original extract that is undiluted) is added to the ﬁrst vial (labeled 0.5x), the lid
enclosure tightened, and then the sample is vortexed well. Then 0.5 ml of the
0.5x stock is added to the next vial (labeled 0.25x), the lid enclosure tightened,
and then the sample is vortexed well. These steps are repeated until all dilutions
are prepared. The sample dilutions will be 100, 50, 25, 12.5, 6, and 3% of the
extract. If less than 1 ml of original 1x extract is available, the above mentioned
volumes should be proportionately reduced. After dissolution in the assay
solvent, the samples should be stored at -20°C.

Standards Preparation

A large range of standards should be prepared to deliver a ﬁnal concentration of
TCDD between 0.03 - 100 pg/well in a volume of 1.25 pl (i.e., make 200x stock
solutions). Generally, 6 concentrations will achieve a full dose response curve
(ﬁnal = 0.1, 0.3, l, 3, 10, 30 pg/well) for TCDD. Ideally, standards should be
dissolved in the same solvent as the samples, but this is not always possible. In
this case, be sure to conduct assays with both solvent controls and compare them
to blanks (see Dosing Cells). A stock of TCDD in isooctane is maintained in
Room 181, URCF, in a 1 liter volumetric ﬂask. The concentration is written on
the ﬂask and is tested for purity and accuracy by GC/MS and GC/ECD by

61

comparison to commercially available certified standards. To make dilutions,
prepare the following stock concentrations (assuming a stock concentration of

 

 

10000 ng/ml):
ﬁnal mass. stock
in wells concentration
(pg/125 19 jug/HID
12500 10000
10000 8000
1000 800
100 80

The following six concentrations are
used for the bioassay:

30 24
10 8
3 2.4
1 0.8
0.3 0.24
0.1 0.08

 

NOTE: Adherence to these procedures will insure consistent response of cells. Timing and
cell density are critical. An assay may be completed in 5 days.

5 .3. Plating Cells (Day 1)

Prior to conﬂuence, write down the passage number of the cells, aspirate media
from cell culture dishes, wash the cells with sterile PBS without Ca + and Mg2+,
and trypsinize the cells with 3 ml of 1x sterile trysin-EDTA solution for 5
minutes at 37°C. Add trysinized cell suspension to 27 ml of “10% Full
Medium” and determine the number of cells/m1 with a hemacytometer (for more
information on cell counting, refer to “Cell Culture: A Manual of Basic
Techniques”, by Dr. Ian Freshney, 1996) . Dilute the cell solution to a
concentration of 60,000 cells/m1 with media. Add 0.25 ml of cell suspension to
each well (15,000 cells) of a 96-well ViewPlateTM using an EppendorfrM repeat
pipettor. Care must be taken that the cell suspension is uniform each time that
the pipettor is reﬁlled. This is done by gently inverting the tube or bottle of cell
suspension end-over-end several times immediately prior to reﬁlling the pipettor.
If the outer 36 wells are not being used for the experiment (recommended), ﬁll
them with either sterile media or PBS to maintain humidity consistently across
the plate. Use of the outer wells is not recommended because of an edge effect
caused by inconsistent growth of cells in these wells.

NOTE: cell number per well is one of the largest contributors to variation in the data - so

take plenty of time to do this step properly. Cell passage number should be noted to monitor
long-term changes in the responsiveness and growth characteristics of the cells.

62

5.4. Dosing Cells (Day 2)

Dose cells 24 hours after plating and continue exposures for 3 days. First,
examine cells to ensure consistent plating. Aspirate media (attach p-lO tips to
the suction line to minimize cell scraping), and replace with 0.25 ml of
DCCFBS media that is prewarmed to 37°C.

NOTE: the use of cold media should be avoided because it may “shock” or stress the cells
and it may cause precipitation of test agents that have poor water solubility).

A typical plate design is shown in Figure 2. Use at least three replicates per
treatment. Note that with TCDD, only two replicates are used for a typical plate
design. Each control and TCDD concentration are averaged for all plates within
a given experiment. Use a negative control with no treatment (blank) and a
solvent control treated with pure solvent only. Use at least ﬁve concentrations of
each compound or extracts tested. Prepare 200x chemical stocks in the
appropriate solvent (see Sample Preparation). In general, it is best to dose cells
directly in the well; however, for water miscible solvents, the test agent can be
added to a sterile glass vial containing 2 ml of media, mixed, and then added to
each well. Each well shall receive 1.25 pl of sample or standard or solvent
control as noted in the plate design below.

Earlier versions of this assay determined that cross-talk can occur when high
activity samples are directly adjacent to low activity samples. Therefore, as a
corrective action, the plate design shown in Figure 2 was developed. Note that
two blank columns border samples, so that there is no cross-talking between
samples and TCDD and between samples and controls.

63

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Row/ 2 3 6 7 8 10 ll 12
Col
A
B 0.1 pg 0.1 pg C1 C1 C1 solvent
TCDD TCDD Conc. Conc. Conc. control
1 1 1
C 0.3 pg 0.3 pg C1 C1 C1 solvent
TCDD TCDD Conc. Cone. Conc. control
2 2 2
D 1 pg 1 pg C1 C1 C1 solvent
TCDD TCDD Conc. Conc. Cone. control
3 3 3
E 3pg 3pg C1 C1 C1 blank
TCDD TCDD Cone. Conc. Cone.
4 4 4
F 10 pg 10pg C1 C1 C1 blank
TCDD TCDD Conc.5 Conc.5 Conc.5
G 30 pg 30 pg C1 C1 C1 blank
TCDD TCDD Conc. Conc. . Conc.
6 6 6
H

 

 

Figure 2. A typical plate design for determination of Ah receptor agonist activity in H4IIE-
luc cells. Note that blanks, solvent controls, and a standard curve for TCDD can be
analyzed on the same plate with one test sample (labeled Cl, at six different concentrations).
Two blank columns are recommended between samples and either TCDD or controls to
prevent any possible cross-contamination.

64

5.5. Conducting the Bioassay (Day 5)

5.5.1.

Preparation steps (prior to assay)

Inspect plates visually with and without microscope — check
degree of conﬂuence, homogeneity from well-to-well, and any
signs of cytotoxicity or altered morphology

Set up cytoﬂuor for viability assay:

set 1: excitation = B emission = B sensitivity = 3
set 2: excitation = C emission = E sensitivity = 4

NOTE: sensitivity can be adjusted if values are too high (9999) or low (not diﬂrerent

 

 

 

from blanks)
c. Preparation of viability assay reagent (refer to Supplies and
Biochemicals Section):
Each plate will need 1.8 ml or 3 ml, depending on whether 36
or 60 wells are used per plate, respectively plus a little extra (2
ml). Dilute the appropriate amounts of calcein and ethidium
with the appropriate volume of media without F BS as shown
below:
Number of Volume of viability assay reagent needed
plates [total volume (ml); calcein stock (pl); Ethidium stock
(111)]
using 36 wells Using 60 wells
1 3.8 ml; 0.95 pl; 1.9 pl 5 ml; 1.25 pl; 2.5 pl
2 5.6 ml; 1.4 pl; 2.8 pl 8 ml; 2 pl; 4 pl
3 7.4 ml; 1.85 pl; 3.7 pl 11 ml; 2.75 pl; 5.5 pl
4 9.2 ml; 2.3 pl; 4.6 pl 14 ml; 3.5 pl; 7 pl
5 11 ml; 2.75 pl; 5.5 pl 17 ml; 4.25 pl; 8.5 pl
6 12.8 ml; 3.2 pl; 6.4 pl 20 ml; 5 pl; 10 pl
7 14.6 ml; 3.65 pl; 7.3 pl 23 ml; 5.75 pl; 11.5 pl
8 16.4 ml; 4.1 pl; 8.2 pl 26 ml; 6.5 pl; 13 pl

 

CAUTION: ethidium homodimer is a powerful mutagen - handle with care and throw
contaminated tips, etc., into biohazard bags

65

 

Set up luminometer:

Mode = Cycle; Pause = 2 sec;

Data = All; Mix = Off;

Gain = High; Temp. = 30°C;

Cycle = 1-3; AID reads = 20 (number of times
that each well is analyzed per cycle)

Set up vacuum aspirator (attach p-10 tips to each suction line to
minimize scraping surface area and only use the plate washer
to aspirate and not to dispense PBS)

Prepare LucLite substrate solution and luciferase positive
control (must be used on the same day as prepared)

1) Reconstitute lyophilized substrate solution by adding 10
ml of Assay buffer solution A (provided with kit) to one
vial of lyophilized substrate (each vial is enough for 133
assays if 75 pl is used). Agitate gently until a
homogeneous solution is formed (a slight turbidity is
acceptable). Equilibrate to room temperature before use.

NOTE: if more than one vial is reconstituted, combine and mix to prevent any variation
from the substrate between plates.

2) Reconstitute the lyophilized luciferase positive control
with 200 pl of distilled (or nanopure water). Each vial
contains sufﬁcient luciferase for 20 controls.

5.5.2. Cell Viability Assay Procedure (process one plate at a time)

remove plate from incubator and aspirate media, then rinse 1
time with PBS

Add 50 pl of PBS with Ca2+ and Mg2+ to all wells using a 8-
channel pipet

Add 50 pl of viability assay reagent to all wells using a 8-
channel pipet

Incubate at room temperature for 10 minutes and then scan
plate in the Cytoﬂuor instrument

66

 

 

e. Export/print data (check that values are appropriate, otherwise
adjust sensitivity and rescan). Password protect data ﬁles with
a project-speciﬁc password. Data analysis are discussed in
Section E along with an example raw data ﬁle and a Microsoft
Excel spreadsheet version of a sample data analysis.

f. Aspirate viability reagent/PBS, and rinse 1 time with PBS
using vacuum
aspirator

g. Seal the bottom of the ViewPlates with self-adhesive TopSeal
(or tape very well). This increases the signal detection in the
luminometer.

h. Add 75 pl of PBS with Ca2+ and Mg2+ (or media without serum
or phenol red) to all wells using an 8-channel pipet

i. Working under SUBDUED light conditions, add 75 pl/well of
reconstituted LucLite substrate solution, cover with foil, and
agitate gently.

j. Wait twenty minutes before counting the plate to allow full
signal generation.

k. Measure luminescence on the Dynatech ML3000 luminometer

1. Save, print, and copy data to disk. Password protect data ﬁles
with a project-speciﬁc password. Data analysis are discussed
in Section E along with an example raw data ﬁle and a
Microsoﬁ Excel spreadsheet version of a sample data analysis.

5.6. Bmteinlletennination

Protein determination is readily accomplished in the same plates with same cell
lysates after measuring luminescence. A ﬂuorescamine-based protein assay
(Sanderson et al., 1996) or Micro-BCA Assay (available from Pierce (800) 874-
3723) can be used. Degree of confluency of wells was verified
microscopically, and in most cases this made normalization to protein
unnecessary; therefore, luciferase activity is reported as either relative
light units (RLU) or percent of solvent control.

67

5.7. mamas

5.7.1.

5.7.2.

1,. 1'1' 13

Average the three measurements for calcein AM and ethidium. Divide
the average calcein AM ﬂuorescence for each sample by its ethidium
homodimer ﬂuorescence to obtain a live to dead ratio. Graph the
average calcein AM ﬂuorescence and standard deviation for the
negative control, solvent control, and each concentration tested.
Examine the calcein AM data visually. If the blank has greater
viability than the other treatments, the solvent may be toxic to the cells.
If viability decreases with increasing concentration of the test
substance, the test substance may be toxic to the cells. In either of
these cases, the luciferase data must be regarded with great suspicion.
If the solvent is toxic, try a different solvent or a lower concentration of
solvent. If the test substance is toxic, try extracting the toxic
component (e.g., removing sulfur compounds from sediment extracts),
or conclude that cytotoxicity is likely to preclude any dioxin-like
effects.

LueiferaseData

Calculate averages and standard deviations for each treatment.
Graph the change in response and its standard deviation against
concentration. Relative potencies should be calculated through
conversion of the data to probit values or any other appropriate
transformation that linearizes the sigmoidal dose-response curve.
For a complete description and decision tree for data analysis
methods, refer to the Standard Operating Procedures entitled,
“Estimation of Relative Potencies Based on In Vitro Bioassay
Results”. To convert data to probit values, the concentration
producing maximal induction is set at 100% and the responses of all
other concentrations are converted to a percentage of this maximal
level. A lookup table function can be used in a spreadsheet program
such as Excel or Lotus 1,2,3 to convert percentages to probit values
(contact the Aquatic Toxicology Laboratory for a copy of this table
on disk).

6.0. RECORDS, DOCUMENTATION, AND QC REQUIREMENTS

6.1

Records and Documentation

The primary analyst shall document any anomalies and/or deviation from the
speciﬁed method in a bound, serially numbered, laboratory notebook with tear-
out carbon copies.

Aquatic Toxicology Laboratory at Michigan State University and a duplicate

All electronic ﬁles and hardcopies will be kept at the

68

copy will be kept in the Archive Room of Dr. John Giesy (Dept. of Zoology,
Michigan State University). This information will also be recorded in the data
package and listed in the Case Narrative Form (in accordance with SOP 802).
The primary analyst will sign and date any forms as the analyst.

The technical reviewer will record any problems noted during the technical
review. The technical reviewer will return the items to the analyst for
corrections prior to inclusion into the data package. The technical reviewer will
sign and date all forms as the reviewer.

6.2 QC Requirements and Data Quality Objectives

The threshold dose and EC50 for luciferase induction in H4IIE-Inc cells were
approximately 0.1 and 1.2 pg/well, respectively, as determined from 41
separate standard TCDD curves analyzed in 1997 (Figure l). Coefﬁcients of
variation (standard deviation/mean x 100) for the assay were under 10% at all
concentrations tested for any single day of experiments. A proﬁciency curve
is maintained for the EC50 and threshold doses for TCDD standard curves
(every plate has a standard curve). If for a particular plate, the ECso or
threshold for the TCDD standard exceeds :1; 20 % of the proﬁciency curve,
then the sample on the plate in question will be reanalyzed. For a sample size
of 20 g tissue and a ﬁnal extract volume of 0.25 ml, the H4IIE-luc assay will
detect 1 part per trillion (ppt; pg/ g, wet weight) TCDD-equivalents. With a

sample size of 5 g tissue, 4 pg/g (wet weight) TCDD-equivalents will be
detected.

7.0. RESPONSIBILITIES

The primary analyst will complete the analysis as speciﬁed in this SOP and
provide documentation of raw data and any anomalies and provide data to the
data analyst who will perform data calculations in accordance with SOP 202.

The technical reviewer will determine if data quality objectives were met, notify
the analyst if any problems were found.

69

8.0. REFERENCES

Aarts, J.M.M.J.G., Denison, M.S., De Haan, L.H.J., Schalk, J.A.C., Cox, MA. and
Brouwer, A. 1993. Ah receptor-mediated luciferase expression: a tool for
monitoring dioxin-like toxicity. Short paper, Dioxin ’93, 361-364.

Aarts, J.M.M.J.G., Denison, M.S., Cox, M.A., Schalk, M.A.C., Garrison, P.A., Tullis, K.,
De Haan, L.H.J., and Brouwer, A. 1995. Species-speciﬁc antagonism of Ah
receptor action by 2,2’,5,5’-tetrachloro— and 2,2’,3,3’,4,4’-hexachlorobiphenyl. Eur.
J. Pharmacol. in press.

Aquatic Toxicology Laboratory/Michigan State University, Laboratory Quality Control
Plan (LQCP), September 1998.

Aquatic Toxicology Laboratory/Michigan State University, Safety Manual, September
1998.

Denison, M.S., El-Fouly, M.H., Aarts, J.M.M.J.G., Brouwer, A., Richter, C. and Giesy,
JP. 1993. Production of a novel recombinant cell line bioassay system for the
detection of 2,3,7,8-tetrachlorodibenzo-p-dioxin-like chemicals. Short paper,
Dioxin ’93, 365-368.

Giesy, J.P., Ludwig, JP. and Tillitt, D.E. 1994a. Dioxins, dibenzofurans, PCBs and
Colonial ﬁsh-eating water birds. In: Dioxins and Health, Chapter A. Schechter, ed.,
Plenum Press, N.Y..

Giesy, J.P., Ludwig, J .P. and Tillitt, D.E. 1994b. Deformities in birds of the Great Lakes
region: assigning causality. Environ. Sci. Toxicol. 28, 128A-135A.

Gilbertson, M., Kubiak, T. and Ludwig, J. 1991. Great Lakes embryo mortality, edema,
and deformities syndrome (GLEMEDS) in colonial ﬁsh-eating birds: similarity to
chick-edema disease. J. T oxicol. Environ. Health, 33, 455-520.

Murk, A.J., Jonas, A., Brouwer, A., Leonards, P.E.G. and Denison, MS. 1996.
Application of the CALUx (chemical activated luciferase gene expression) assay
for measuring TCDD-equivalents in sediment, pore water, and blood plasma
samples. Organohalogen Compounds 27, 291-296.

Okey, A.B., Riddick, BS. and Harper, RA. 1994. The Ah receptor: mediator of the
toxicitiy of 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD) and related compounds.
T oxicol. Lett. 1Q, 1-22.

Nebert, DW. and Gonzalez, F .J. 1987. P450 genes: structure, evolution, and regulation.
Ann. Rev. Biochem, 56, 945-993.

Poland, A. and Knutson, J.C. 1982. 2,3,7,8-Tetrachlorodibenzo-p-dioxin and related
halogenated aromatic hydrocarbons: examination of the mechanism of toxicity.
Annu. Rev. Pharmacol. T oxicol., 22, 517-544.

Program Manager Rocky Mountain Arsenal, Chemical Quality Assurance Plan (CQAP),
April 1996.

Safe, SH. 1986. Comparative toxicology and mechanism of action of polychlorinated
dibenzo-p-dioxins and dibenzofurans. Ann. Rev. Pharmacol. T oxicol., 26, 371-399.

Safe, SH. 1990. Polychlorinated biphenyls (PCBs), dibenzo-p-dioxins (PCDDS) and
dibenzofurans (PCDFs) and related compounds: environmental & mechanistic
considerations which support the development of toxic equivalency factors (TEFS).
CRC Crit. Rev. T oxicol., 2_1, 51-71.

70

 

Sanderson, J.T., J. M. M. J. G. Aarts, A. Brouwer, K. L. Froese, M. S. Denison and J. P.
Giesy. 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. W
Rhannagol. 137(2):316-325.

Scatchard, G. 1949. The attraction of proteins for small molecules and ions. Ann. NY.

Acad. Sci., 51, 660-672.

Tillitt, D.E., Giesy, JP. and Ankley, GT. 1991. Characterization of the H4IIE rat
hepatoma cell bioassay as a tool for assessing toxic potency of planar halogenated
hydrocarbons in environmental samples. Environ. Sci. T echnol. 25, 87-92.

Whitlock, J.P.Jr. 1990. Genetic and molecular aspects of 2,3,7,8-tetrachlorodibenzo-p-
dioxin action. Ann. Rev. Pharmacol. T oxicol., 39, 251-277.

ATTACHMENTS

Attachment 201-1: Figure 1 - TCDD Dose-response Curve with H4IIE-luc Cells
Attachment 201-2: Figure 3 - Viability Assay Sample Data Set (Raw Data)
Attachment 201-3: Figure 4 - Viability Assay Sample Data Set (Calculated Data)
Attachment 201-4: Figure 5 - Luciferase Assay Sample Data Set (Raw Data)
Attachment 201-5: Figure 6 - Luciferase Assay Sample Data Set (Calculated Data)

71

 

 

Appendix 11: Sample Calculation of the Detection Limit Method and the

Slope-Ratio Method

72

Appendix II - Sample Calculation: Detection Limit Method

Note: Luminescence was measured in relative units. The term ‘units’ was used to
describe this in the calculations below.

To determine sample-speciﬁc detection limits the following calculations re performed:

Lot: BBC

Sample: 96RFGH04

Tissue: Owl liver

Tissue weight: 0.912 g

Extract volume: 1000 pL

pl extract added per well: 1.25

Mean blank value: 0.005128

Mean standard response at 0.1 pg/well: 0.018183 (least standard concentration)

Mean sample response undiluted: 0.005711 (greatest sample concentration)

Gram equivalents/pL extract M = Q._1_22g = 0.000912 g/pL
Final volume lOOOpL

Therefore g-eq added to well in 1.25 pL = 0.000912*1.25 = 0.00114 (A)

Relative response sample/standard

(mean std at 0.1 pg/well — mean blank)

(0.018183-0.005128)

= 0.04466

Standard was 0.1 pg therefore the well contains 0.04466*0.1 pg TCDD

= 0.004466 pg TCDD
Tissue equivalents in the well = 0.00114 g (A)
Equivalents concentration = 9.095.506.1211
0.00114 g

= 3.918 pg/g TCDD-EQ

TCDD-EQMAx Reported detection Limit = 4 pg/g T CDD-EQ w/w

TCDD-EQMIN Reported detection Limit = 0 pg/ g TCDD-EQ w/w

73

 

Appendix II- Sample Calculation: Slope Ratio Method.

This method requires the determination of the slope of the linear portion of the dose
response curve for the standard and the sample run on the same plate. This evaluation is
best performed in SYSTAT where interactive data point selection, graphing and
regression analysis are available. The determination of the linear portion of the curve is a
judgement and experience based assessment not amenable to strict standardization. To
attain statistical rigor in the analysis, two criteria are applied to each sample analyzed by
this method.

1. Minimize data removal to obtain slope
2. Maximize r2 for the regression line.

Once slopes for the two lines have been determined the simple ratio of these slopes is the
TCDD-EQ concentration of the sample, since the dependent for the standard is pg/well
and the dependent for the sample is g-eq/well.

Lot: 7 BBB
Sample: 96FGH007
Tissue: owl liver
Tissue Weight: 3.9553 g
Extract Volume: 1000 pL
pL extract added per well: 1.25
Slope Standard dose response: 0.073
Slope sample dose response: 78.268
Gram equivalents/pL extract = M = 3.22512
Final volume 1000 pL
= 0.0039553 geq/pL
Sample concentration = 53111121151912:
Standard Slope
= 18.263
0.073

= 1072.16 pg/g TCDD-EQ w/w

Reported value = 1072 pg/g TCDD-EQ WM

74

 

O
.‘ 1111111111'111', _ .1 1.11.1'1’1 '11111

SYSTAT Rectangular ﬁle R:\lot BBB\96fgh007 positive.SYD,
created Tue Oct 05, 1999 at 15:47: 18, contains variables:

SAMPS TYPES SAM_DIL SAM_RESP STD_CON STD_RESP
POS TISSUE GEQ
'1' 1111' 1111'111'11311 I_' 1’111'11'11.’

 

 

 

 

0.20 I l l r l l
0.15"” r T
g, 6
LU
m10.10 '7; 7'
O
'—
(D
0.05; 7
0
ti 1 1 1 1 1 1
.00
005101520253035

STD_CON

75

 

 

 

 

 

 

015 I I I I I I
(L010 —
(I)
LU
ml
0
17;

005 -

000 l l l 1 l l

00 (15 1!) L5 2t).25 3t) 35
STD_CON

t .n’ 1 1'4 H1 ’111’11I1_'!11.1I’1’ 1'1,

 

 

1 1
04 05
STD_CON

l

08

 

76

1
10

 

12

t1 ’ 1 1!. 1!’ 1 1’ II 11. 1 .’ 1 ’I. 1' 1”11 I 1’ 3’ 1111 .11.” -

Dep var: STD_RESP N: 12 Multiple R: 0.980 Squared multiple R:
0.961

Adjusted squared multiple R: 0.958 Standard error of estimate: 0.007

Effect Coefficient Std Error Std Coef Tolerance t
P(2 Tail)

CONSTANT 0.010 0.002 0.000 . 4.039
0.002

STD_CON 0.087 0.006 0.980 1.000 15.776
0.000

Analysis of Variance

Source Sum—of-Squares df Mean-Square F-ratio P
Regression 0.012 1 0.012 248.879
0.000

Residual 0.000 10 0.000

Durbin-Watson D Statistic 0.233

First Order Autocorrelation 0.76

77

 

Plot of Residuals against Predicted Values

 

0.010 I c-I I I I I I I I

l
1

0.005

0.000 “““““““““““““ —

RESIDUAL

-0.005 - 9 r

 

 

l l l l l l 1

 

l 1
700100 N '2. '3 a 1 ‘b a o
(,9 99 09 0.0 09‘ (,9 0.09 9.0 9.0 0.0 (,3

ESTIMATE

O O 0
r’,’ 1’11 1 11’1111. ’11! '1. ’1 1 ’1}!!! .1’1

 

0.10 I I I I I I I I
0.09
0.08
0.07

ESP

m. 0.06
0

ST

0.05
0.04

 

0.03:

 

 

l l l l l l l l

. 2
000.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0
STD_CON

 

Dep Var: STD RESP N: 6 Multiple R: 0.998 Squared multiple R:
0.996

78

Adjusted squared multiple R: 0.995 Standard error of estimate: 0.002

Effect Coefficient Std Error Std Coef Tolerance t
P(2 Tail)
CONSTANT 0.021 0.001 0.000 . 14.971
0.000
STD_CON 0.073 0.002 0.998 1.000 31.278
0.000

Analysis of Variance
Source Sum-of—Squares df Mean—Square F-ratio P
Regression 0.005 1 0.005 978.342
0.000
Residual 0 000 4 0 000
*** WARNING ***
Case 31 is an outlier (Studentized Residual = -
3 .306)
Durbin-Watson D Statistic 2.099

First Order

Autocorrelation ..... .0.256

Plot of Residuals against Predicted Values

 

0.003

0.002 ”

0.001

RESIDUAL

-0.001 “

-0.002 “

 

(1000 ------------------------

l

O

 

1 l l l l l

 

-0.003

0.02 0.03 0.04 0.05 0.06 0.07 0.08 0.09 0.10

ESTIMATE

O I I
111’11’1111’1’111 111.! 1

79

 

 

005

53
C)
.h

F’
c:
or

STD RESP
F’
c:
k)

 

 

 

l l l

 

Dep Var: STD_RESP
0.945

0.1 0.2 0.3 0.4
STD_CON

N: 10 Multiple R: 0.972 Squared multiple R:

Adjusted squared multiple R: 0.938 Standard error of estimate: 0.004

Effect Coefficient Std Error Std Coef Tolerance t
P(2 Tail)

CONSTANT 0.007 0.002 0.000 . 4.067
0.004

STD_CON 0.134 0.011 0.972 1.000 11.741
0.000

Analysis of Variance

Source Sum-of—Squares df Mean-Square F-ratio P
Regression 0.002 1 0.002 137.845
0.000

Residual 0 000 8 0 000

*** WARNING ***

Case 36 is int outlier (Studentized Residual =
2.680)

Durbin-Watson D Statistic 0.595

First Order Autocorrelation 0.469

80

 

Plot of Residuals against Predicted Values

 

 

 

 

0.010 I I I I
0.005— _
.J
<
D
Q
(0
DJ
0:
0.000 “'3 -------------------- __
7; 7)
-0 005 1 I I I
' 0.00 0.01 0.02 0.03 0.04 0.05
ESTIMATE

. O I
1111111'1. 1'1 ’1111111..I '111 1

: l 11 '22ZZ'12=2222

81

I
1’11!

W

l l

 

0.09
0.08 b
0.07 ‘
0.06 "
0.05 b
0.04 ‘
0.03
0.02
0.01

SAM_RESP

I 00 I

WIU)

 

I l l

l

 

 

0.00
0.000

0.001 0.002 0.003 0.004

GEQ

O
1’14’1111’111.’ 11’

1

0.005

e
" I’ll]

 

0.09
0.08
0.07
0.06
0.05
l
5 0.04
0.03
0.02 ”

0.01 %

0.00
0.0000

l l l

l

RESP
I

S

1

Of»

 

l 1

0.0005 0.0010
GEQ

82

 

 

0.0015

I “I
1
A hi
0
[I

0 O
1 .1 181’ I. 1 1 1 1’ 1111 1 1’! 111 1 ’t. 1' 1' .1'1111 -

 

(106 I I I I

005

c:
E

1003

SAM RESP

002

001

 

A .

 

 

l l l l

000
0.0000 0.0001 0.0002 0.0003 0.0004 0.0005
GECI

 

Dep var: SAM_RESP N: 18 Multiple R: 0.985 Squared multiple R:
0.970

Adjusted squared multiple R: 0.968 Standard error of estimate: 0.003

Effect Coefficient Std Error Std Coef Tolerance t
P(2 Tail)

CONSTANT 0.007 0.001 0.000 . 8.101
0.000

GEQ 87.224 3.860 0.985 1.000 22.597
0.000

Analysis of Variance

Source Sum-of—Squares df Mean-Square F—ratio P
Regression 0.004 1 0.004 510.628
0.000

Residual 0.000 16 0.000

*** WARNING ***

Case 9 is an outlier (Studentized Residual = -
2.986)
Durbin-Watson D Statistic 1.572

First Order Autocorrelation 0.193

83

 

Plot of Residuals against Predicted Values

 

 

0.010

0.005

0.000

RESIDUAL

-0.005

 

l

l l

_--------n—-—----———---—-

 

 

-0010

0.00

'11411'

0.01

1111131’

0.02
ESTIMATE

0.03

0.04

I '11I1- I.

 

0.06

.0
o
or

I

SAM_RESP

F3
o
N

I

 

9
O
A

l

p
O
(as)

l

l l

 

0.05

111! ’ 1"

 

l

 

. 1
00,0000 0.0001 0.0002 0.0003 0.0004 0.0005
GEQ

84

Dep Var: SAM_RESP N: 9 Multiple R: 0.984 Squared multiple R:
0.968

Adjusted squared multiple R: 0.963 Standard error of estimate: 0.003

Effect Coefficient Std Error Std Coef Tolerance t
P(2 Tail)

CONSTANT 0.010 0.002 0.000 . 6.286
0.000

GEO 78.268 5.377 0.984 1.000 14.556
0.000

Analysis of Variance

Source Sum—of-Squares df Mean—Square F—ratio P
Regression 0.002 1 0.002 211.877
0.000

Residual 0.000 7 0.000

*** WARNING ***

Case 9 is an outlier (Studentized Residual = -
2.647)
Durbin-Watson D Statistic 2.401

First Order Autocorrelation -0.238

Plot of Residuals against Predicted Values

 

0004 I_ I l ‘1
0.003 _ 7
0.002 r '7
0.001 r 7 7
0.000 """""""""""""" .5,
-0001 — C’
-0002 “'
. -0.003 "
-0004
0.008

RESIDUAL

l

 

 

1 1 l
01 0.02 0.03 0.04 0.05
ESTIMATE

 

85

 

Appendix 111: Quality Control/ Quality Assurance Samples

86

Table III-a: MSU Quality Control / Quality Assurance Samples

 

 

Sample number Tissue TCDD-EQMAx TCDD-EQMIN
Method Blank (lot BBA) carp eggs 1 0
Method Blank (lot BBB) Carp eggs/ owl livers l 0
Method Blank (lot BBC) owl livers l 0

Method Blank (KKA) kestrel eggs 1 0
Method Blank (lot KKB) kestrel eggs 1 0
Method Blank (lot KKC) kestrel eggs 2 0

Matrix Blank (KKA) kestrel eggs 5 0

Matrix Blank (KKB) kestrel eggs 1 0
Matrix Blank (KKC) kestrel eggs 2 0
Mean 1.67 0
Method Spike (lot BBA) carp eggs 1818 1818
Method Spike (lot BBB) Carp eggs/ owl livers 2169 2169
Method Spike (lot BBC) owl livers 1856 1856
Mean 1943 1943
Min 1555 1555
Max 2332 2332

Method Spike (KKA) kestrel eggs 908 908
Method Spike (lot KKB) kestrel eggs 866 866
Method Spike (lot KKC) kestrel eggs 986 986

Matrix Spike (KKA) kestrel eggs 1120 1120
Matrix Spike (KKB) kestrel eggs 1065 1065
Matrix Spike (KKC) kestrel eggs 888 888
Mean 972 972
Min 778 778
Max 1167 1167

 

 

87

 

 

 

 

 

 

 

 

Irv/7
W
***-—

F Method Spike Analysis

 

 

 

 

 

 

 

 

  
 

 

 

 

 

 

 

 

 

2500 'F
I
2000
I I
1500 ——___._ ---------------- ;.____——~ I yam

 

 

 

 

 

_. _ .. - min (mean-20%) .

1000 m - — — - max (mean+20°/o)

 

 

 

 

 

 

 

 

 

 

 

 

500

 

 

 

 

 

 

 

 

 

 

0 , __I

LotBBA LotBBB LotBBC

Figure III-a: MSU method spike quality control for lots BBA, BBB, and BBC

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

i7 Method and Matrix Spike Analysis i
1400 i
1200 _____________________ ~_ _

‘ A
1000 —-—— .
I I ‘ I I method
800 ' """""""""""""" -—-- min (mean-20%) '
600 - - - - max (mean+20%) .
A matrix
400
200 —
0 r I .
Lot KKA Lot KKB Lot KKC
wH‘~I_I__I_J

 

 

Figure III-b: MSU method and matrix spike quality control for lots KKA, KKB,
and KC

88

Appendix IV — Complete Results for Carp Egg Samples

89

2:52:83 085 you >E 05 3 tom: 3.: n—GUH. “R.— c: ._.
9.2—owes 98331.2... a8 63.: an: n 2: "Eu—Swag steam—TEE: .8.— voo... c u 5:

aha—Um 6:: Jan—UA— JmUm he ecu—32230 ﬂ FOP

 

 

 

 

 

 

90

3:3 3:3... 3:3. 3:3. 3:3 3:...” 3:3 3:3” 85566 25.:
3:3 3:3 3:3 3:3. 3:3 3:3 3:3 3:4.N :03:
62:3 3:33 3:5. 3:3 3:? 33.3 3:3 3:3. ozomam

3:283—

323 L325». 3:: 3:3. 3:3. 3:3 3:3 k3:3 98.2.9 25:2
3:3 3:3 3:5; 3:3 3:2 33.. 3:3 3:3 2595:
3:3 3:3 ”.235; 3:3 3:3 3:3 3:: 3:3 288%
3:3,. 3:3 "3:3 3:56 3:3 3:2 3:: 3:2“ 3208a:
N3:3 3:3 3:53 3:3 3:3 3:5; as :3 3:. .4 2586:
3:3 3:3 32: 3:3 3:; 3:2 3:3 3:3 :76er
335.3” 3:3 3:3 3:3 3:2 3:5 3:3 3:66 2.68%
3:3 3:3 3:3 3:2. 3:3 o2:3 3:3 3:3. 288%
3:3 3:3 3:3 3:3 3:3” 33.. 3:3. 3:3 2.683:
3:3 3:2: 3:5.” 3:3 33a 3:3 3:3 3:2. 2.6an
3:3,,“ 3:3 3:3 3:3 3:3 3:3 3:: 3:3 2.58%
3:2 3:3 3:53 3:3 02 a... 3:3 3:3 3:3. 2,6821
3:: 3:35 3:3 3:3 3:5 3:3 3:3 3:3. 8895:
3:3 3:3 3:o._ 3:3 33% 3:3 3:46 3:3.” Saoooa:
3:2 3:3 3:55 3:3 33.». 3:3 3:: 3:: 8388:
3:3. 3:3. 3:3 3:3 3:3 3:3 _o:m.m 3:3.m 888%
3:3 3:3 3:3 3:3 3:3 323a; 3:3 3:3 Sauces:

A??? mEmv dzﬂwhmm «menu—.0
€23 ORA—not €32 Oman—ob 833 9...: 682 OP: is 5:05. a: 520.5

02:0: a 2.20: + .220: + 220: + $5.. .x. hoe £0.— .\. .53 532%

as? 3a.: 322:8 ”we 9:8 .5. 83a.» 0: can UHF .3 2.3355259 mo wig: 829.30 31>— 03.2.

$33.2... Baa—5.30 a... a. 83...... ”we 9:3 warn—2... 3.3%... «
.aaowccu 9.3.2....2... .8.— 685 an: n .32 3.2.09.3 3082......5. .8 to... c u 2.2
8.2.32. «>329. Hﬂvl .3 v3.3.3... 3.58.. 9.3.32.9 1833...... u .85an.
338.. 3.3.2.8.. n OMAR—09

 

 

 

 

 

 

 

 

 

 

 

 

 

3:2 3:3 3:3 3:... .8538... 23....
3:3,. 3:... 3:3 3:3... 288..
3:... 3:... 3:3... 3:3... 3.8mm...

3:95.3—

33...” 3:3. 33.... 3:3. $2.3... 7.5.2
3:... 3:... ._ 3:8 3:... 2,583:
3:... 3:... 3:38 3:3... 3:08.:
3:2 3:... 3:... 3:3... 888....
3:... 3:3... 3:3,... 3:3... 2.68.:
3:... 3:... .. 3:; 3:... 2.08....
3:... 3:36 3:... 3:3... :88...
3:... 3:... 3:... 3:3... 3:68....
3:3 3:3... 33$ 3:..... 2,683..
3:... 3:9. 3:8 3:... 3.68....
3:... 3:3,. 3:... 3:3... 2,683..
3:... 3:3,. .33.. 3:3... . 768$.
3:... 3:... .. 3:3 3:3... 8.88%.
3:3... 3:3... 3:3 3:3... 8.683:
3:: 3:... 3:2. 3:3... 8.08....
3:... 3:8 3:2. 3:3... 838.:
3:3... 3:8 3:8 3:3... 5.68....

3% 3% .33 m3... 3%... .8..-.5

.2: .96.? 5.20...-303 z... 50...... z... 0.2.8.. 5.2%

82.....8 an» 9.3 .8 83.3 .5599 2... OH

A509 .3 gaunt—.359 .3 was... 0.0—@500 5..»— 03:9

91

Appendix V: Complete Results for Owl Liver Samples

 

 

92

Eon—nus”. :3uo?:o=3 .5 33.: ACE u 5.: ”9.2.9939 $930.75.... .5. won: a .I. 2.:

aha—U.— .Ea inn—U.— .mM—Um he acuséhacu u POP
9»: 5.55.5. ND 3:52;. n _. 31:? u <

 

 

 

 

 

 

93

3:2 3:2 3:3 3:3 3:3 3:3 3:3 3:3 .oocllalleao 72::
3:2 3:2 3:2 3:2 3:2 3:3 3:2 3:2 < $338
3:3 3:3 3:3 3:5. 3:: 3:2 3:2 3:3 H :35?
3:2 33.3 3:3 3:3 3:3 3:2. 3:3 3:3 H 2:928
3:m._ 3:3 3:2 3:3 3:2. 3:: 3:3 3:3 H 8:038
3:3 3:3 . . . . . a 8:928
3:2” 3:3 3:2 3:3 3:3 3:2 3:3 3:2 < 8:038
3:3 3:3 3:2 3:2 3:: 3:3 3:2“ 3:3 < 8:923
3:3 3:3 3:3 3:3 3:2 3:3 3:2 3:3 H 3:923
3:: 333 3:3 33.: 3:2 3:2 3:2 3:2 < 8:033
3:2 3:3 3:2 3:; 3:2 3:2 3:3 3:2 H 8:928
3:3 3:3 3:2 3:; 3:3 3:3 3:3 3:2 < 8:928

van—0.5.3.“
3:3 3:3 3:3 3:2 3:3 3:3 3:3 3:3 c8342 72::
3:3 333 3:2 3:2 3:3 3:3 3:3 3:3 H 332:0
3:2 3:3 3:3 3:2 3:3 3:3 3:3 3:3 H 832:0
3:3 3:3 3:2 3:2 3:3 3:3 3:3 3:3 H 22:03
3:3 3:3 3:3 3:3 3:3 3:3 3:3 3:3 H 22:88
3:3 3:3 3:3 3:3 3:3 3:2 3:3 3:2 3 32:93
3:3 3:3 3:3 3:3 3:3 3:2. 3:3 3:2. 3 2:93
3:3 33.3 3:2 3:m._ 3:3 3:3 3:3 3:3 3 220:8
3:3 3:3 3:3 3:3 3:3 3:2 3:3 3:: H 3:93
3:3 3:3 3:3 3:2 3:3 3:3 3:3 3:3 3 3:83
3:2 3:2 3:2 3:2 3:3 3:3 3:3 3:3 < 11:93
3:3 3:3 3:2 3:2 3:3 3:2 3:3 3:2 H 3:03.:
3:2 3:3 3:3 3:3 3:3 3:3 3:3 3:3 H 230:8
3:3 3:3 3:m.m 3:3 3:3 3:3 3:2. 3:3 H 38:93
3:2 3:2 3:3 3:3 3:3 3:; 3:3 3:3 < 23:93
3:3 3:3 3:2 3:2 3:2 3:3 3:2 3:3 < 58:88
3:3 3:3 3:3 3:3 3:3 3:3 3:3 3:3 < 8393

H22 3.: dz: 3.: 33.5

38.2 92:63 €82 3.395 88.3 95 €9.23 0:: 520:3 2.203

5:0: 5.5: .35: 5:0: 3.8.. .3 hos 2.28m .3 30,—. E? 2:23

33.53 .55 .3: 3.. 83.; O: E:— Omh .3 333.5538 .3 «:3: 829.80 3.5 23,—.

 

3932.: onus—5-20 05 E 0353. Mama—=5 2: 826:2: « ”2239.8 $8820.55: .8.— ..8: an: H .22 "9.2.0939 398%...2... :8 tow: a u 2.:
33:32— ?529. m::- an 63:51:: 31.8.. 35252.0 .3333...“ 320mb ”3.30.. 33233 n UNA—QUE "own 5505...: u D "05:33. u _. ”:33. u <

 

 

 

 

 

 

 

 

94

3:3. 3:2 3:3 3:2 H8553: 72:2
3:3 3:2 3:2 3:2 < 2:928
3:3 3:2 332 3:3 H ::928
3:3 3:2 3:3 3:3 H 2:928
3:3 3:2 3:2 3:3 H 8:028

. . 3:2 3:3 a 8:928
3:2 3:2 338 3:3 < 3:028
3:2 3:3 3:3 3:3 < 8:028
3:3 3:3 3:3 3:3 H 3:928
3:2 3:3 3:2 3:3 < 8:928
3:2 3:3 3:3 3:3 H 8:928
3:2 3:3. 3:2 3:3 < 8:928

3:383—
3:3 3:3 3:2 3:2 ©8250 72::
3:3 3:3 3:3 3:3 H 82820
3:3 3:2 3:3 3:3 H 8mm_2:0
3:2 3:3 3:3 333 H 22:98
3:3 3:3 3:3 3:3 H 22:98
3:2 3:2 3:3 3:3 0 38:88
3:_ .3 3:3 3:2 3:3 3 229:8
3:3 3:2 3:2 3:3 0 3:98
8:2 3:3 3:2 3:3 H 8598
3:3 3:3 3:3 3:3 H. 3:028
3:2 3:2 3:2 3:2 < «2:98
3:2 3:2 3:2 3:3 H 3:028
3:3 3:3 3:2 3:3 H 2398
3:3 3:3 3:2 3:2 H 83:98
3:3 3:3 3:3 3:3 < 23:98
3:3... 3:2 3:2 3:2 < 38:98
3:3 3:3 3:3 3:3 < 83:98
9ng HE; my: 9?me.: 3: «:5 .835
3 23 02. 7:2 25: 0:3 022 UNA—:03 23 3-953 :0< 2.3%

83.58 .52. .30 .3.— 823 mug—Om: was OMAR—Uh. .3 233.5533 .3 ”:3: 329—80 54’ «Ban.

Appendix VI: Complete Results for Kestrel Egg Samples

 

95

Table VI-a: Complete listing of concentrations of TEQ and HQ values for kestrel
egg samples

 

 

 

 

 

TOT % TOT % HQMIN H H
SAMPLE SITE TEQMIN PCBMIN TEQMAX PCBW (TEQ (31318 (TCSEIEQ 0233:ng
(M) (pg_/_g) based) based) based) based)
AKEGOO3 P 6.0xlOO 6.4x10‘ 8.3 x100 44 ° 4 4
AKEGOO8 P and mud 14x10' 641:8" i153:3" 5338" (3.2%? PXIOJ‘:
AKEGOI4 P 1.2x10' 6.4x10‘ 1.3x10l 7'5x100 1'7x10'l 1.910'l 'x 0 mm;
AKEGOIS P 3.9x10' 9.3x10‘ 4.3x10' 3'6x10' 5'6xlo" 6.2x10“ 0.0x100 WHO-2
AKEG018 P 9.1x10o 6.0x10‘ 1.0x10' 5:4x100 1'3x10‘l 'x -' O'OXIOO ”Km-2
AKEGOZO P 1.4x10' 4.7x10‘ 1.5x10' 6.8xlO° 2.0xlO" {3:34 8'3ﬂ30 :3):ng
AKEGOZZ P 1.5x10' 3.7x10' 1.5x10l 5.5x10° 2.1x10" 22x10" 0'0 10° 'x '2
AKEGOZ4 P 3.7x101 3.2x10' 3.7x10‘ 12x10' 5210‘l 5'3 " ' x 0 1.4x10_2
AKEGOZS P 5.9x10° 6.4x10' 7.6x100 3:8x100 8.5:10'2 {1’13" 3'3XIO" IMO-2
AKEGOZ9 P 6.5x10" 6.1x10‘ 7.8xlO° 39x10° 9°2x10'2 P1X " ' “0° “Km-l
“(56030 P 3.5x10' 13 1 . . o . -. .x10_ 0.0x10 1.4x10
.xlO 3.6x10 4.8xlO 5.lxlO 5.1x10' com" 5710'2
AKEGO36 P 8.6x100 9.5x10' 1.2x10' 8.2x10" 12x10" 18x10" 0'0 10° 'x "
AKEGO37 P 1.2x10‘ 3.5x10‘ 1.2x10' 4.2x10° 15x10" 1.8x10" o'oxloo ”Km-2
AKEG039 P 3.1x10‘ 2.9x10' 3.1x10' 9.0x10° 4'4xlo" 4'5x10'I o'ox10° “Km"
AKEGO45 P 4.8x10° 7.9x10l 7.0x10° 3.8xlO° 6.8x10'2 1.0x10" o'oxlo0 ”Km-2
AKEGO48 P 4.1xlo' 6.9x10' 4.3x10' 2.8x10' 5'9x10'I 6.1x10" o'oxlo0 “HO-2
AKEGOSZ P 4.3x10' 6.6x10' 4.6x10' 2.8x10' 6.2x10" 6'6xlO" o°ox10° 3"“0-2
AKEGOS4 P 3.0x10l 7.7x10' 4.0x10' 2.4x10' 4:4xlO" 527x10" 0:0:10" £334
AKEGOSS P 3.1x10; 2.3xlO: 3.3x10' 7.1xlO° 4.4xlO" 4.8xlO" 0.0x100 1:6xlO"
AKEGOOS C 6.5x100 8.1x10 1.1x10' 5.3x10" 9.3x10'2 l.6x10" 0.0x10° 14x102
AKEGOO7 C 4.4x10 9.6x10‘ 9.8x10° 7.7x10° 6.3x10'2 1.4xlO" 89x10" 8.910"
AKEGOll C 6.6x10° 6.0x10‘ 7.1x100 3.9x10° 9.4x10'2 10x10" 0'0x10° 2'9x10'2
AKEGOl7 C 1.4x10' 4.7x10' 1.5x10' 6.4xlO° 2.0x10" 2'2x10'l 0'0x100 2'9x10'2
AKEGOZ7 C 3.8x10' 2.3x10‘ 3.8x10' 8.8x10" 5.4xlO" 524x10" 0:0x10" {4:10-l
AKEGOBS C 1.4x10' 4.5x10' 1.5x10' 6.2x10° 2.0x10" 2.1x10" 00x10° 1'0x10'I
AKEGO40 C 1.1x10: 5.1x10: 12le 5.5x100 1.6xlO" 1.7xlO" OZOxlOO 7:1x10'2
agggzz g 39:3 33:13: 33:31 22*:33 33*:321
AKEG051 C 1.2 10' 5. 1 . 1 . x o .x -1 4'6XIO4 0.0x100 5.7x10-
. x .6x10 2.3x10 6.8x10 1.7xlO 3.2x10 0.0x10 5.7x102
AKEGOS7 C 8.6x10' 7.6x10' 8.8x10' 6.5xlO' 1.2x10° 1.3x10° 0.0x10° 1.0m"
MeaLOnsite 2.1x10l 5.5x10‘ 2.4x10' 1.1x10‘ 3.0x10" 3.4x10" 3.5x10'2 sumor
AKEGOOI R 2.8x10I 7.8xlO' 3.0x10' 2.2x10' 4.1x10“ 4.2x10'I 0.0x10° 2.9x10'2
AKEGOOZ R 1.1x10l 4.7x10' 1.3x10' 7.0x10° 1.6x10" 1.9xlO" 0.0x10° 1.9x10'1
AKEGOO4 R 7.0x100 8.9x10‘ 1.4x10‘ 1.0x10' 1.0xlO" 2.0xlO" 0.0x100 4.3x10'2
AKEGOIO R 6.2x100 6.8x10' 7.9x10° 4.7x10° 8.9x10'2 l.lxlO" 0.0x10° 4.3x10'2
AKEGOIZ R 2.3x102 1.3x10l 2.3x102 3.1x10' 3.3x10" 3.3x100 1.2x10° 1.2x100
AKEGOI6 R 2.0x10' 8.7x10' 2.2x10' 1.7x10' 2.8x10" 3.2xlO" 0.0m0 7.1x10'2
AKEGOZ] R 1.7x10' 7.1x10l 2.0x10' 1.2x10' 2.4x10'l 2.9x10'l 0.0xlO° 1.4x10'2
AKEGOZS R 3.2x10' 7.5x10' 3.3x10l 2.4x10' 4.5x10" 4.7xlO" 0.0xlO° 1.0m"
AKEGOBI R 4.1x10' 2.4x10' 4.2x10' 9.9x10" 5.8x10" 6.0xlO" 0.0x10° 1.1x10-l
AKEGOBZ R 9.2x10° 5.4x10' 1.1x10' 4.9x10° 1.3xlO" 1.5x10" 0.0x10° 4.3x10'2
AKEGO33 R 1.3x10l 8.3x10' 1.7x10' 1.1x10' 1.8xlO" 2.5m" 1.9x10" 1.9x10"
AKEGOBS R 3.2x10° 5.1x10° 3.7x10° 1.6m" 4.5x10'2 5.3x10'2 0.0xlO° 7.1x10'2
AKEGO41 R 4.0x102 8.7x10' 4.1x102 3.5x102 5.8x100 5.8x10" l.7xlO° 1.7x100
AKEGO43 R 8.4x10° 5.5x10' 8.6x100 4.6xlO° 1.2x10'l l.2x10" 0.0xlO° 8.6x10'2
AKEGO44 R 1.8x10' 8.0x10' 2.2x10' 1.4x10' 2.5xlO" 3.1x10" 0.0xlO° 4.3x10'2
AKEGO47 R 2.4x10' 4.1x10' 2.5x10' 9.9x100 3.5xlO" 3.5x10" 0.0x10° 2.9x10'2
Mean Ref 5.8x10' 6.3x10' 6.0x10' 3.6x10‘ 7.8x10" 8.1m" 1.9x10" 2.4x10"

 

R= reference sites; P = periphery of the RMA; C = core of the RMA
TOT = Contribution of PCBs, PCDDs, and PCDFs
mm = 0 used for “non-detect” congeners
m; = MDL used for “non-detect” congeners

96

Table VI—b: Complete listing of concentrations of TCDD-EQ and TEQREP values for
kestrel egg samples

 

 

 

 

SAMPLE SITE TCDD-E -
(D QM“ TCD? EQMAX TEQREP MIN TEQREP MAX
AKEGO 0 12 M2) ‘
03 P 0 0x10 ° °
. 2.0x10 1.6x10 32
AKEGOO8 P 1 1 101 I ' XIO
. x 1.1x10 5.7 xlO° 78 10°
AKEGOI4 P 0 0 10° 0 ' x
. x 1.0 x10 5.4 x10° 62 °
AKEGOIS P 00 10° 0 ' x10
. x 1.0x10 6.3 x10° 80 °
AKEGOIB P 0 0 10° 0 ' XIO
. x 1.0x10 3.971100 44 °
4.1030020 P 00 1 ° ° ' XIO
. x 0 2.0x10 5.2 x10° 63 10°
AKEGOZZ P 0 0 10° 0 - X
. x 1.0x10 1.4x10' 14 10I
AKEGOZ4 P 0 0 ° ° ' x
(3025 . x100 5.0x10 3.111101 3.3 11101
AKE P 0.0 x10 5.0x10° 6.8 x10° 78 10°
AKEG029 P 0 0 ° 1 ' x
603 . 11100 1.0 x10 6.0 x10° 7.0x10°
AKE 0 P 0.0 11100 4.0 x10° 5.9 x10' 5.9 x10'
AKEGO36 P 0.0 x100 1.2 x10' 5.6 x10° 7.7 x10°
AKEGO37 P 0.0 x100 5.0 x10° 1.5 x10' 1.6 x101
AKEGO39 P 0.01110o 1.2 x101 2.4x10' 3.15110l
AKEGO45 P 0.011100 3.0 x10° 1.2 x10° 3.3 x10°
AKEG048 P 0.0 x100 5.0 x100 9.9 x10° 1.2 x10'
AKEGOSZ P 0.05110O 5.0xlO° 1.451101 2.151101
AKEGOS4 P 0.0 51100 4.0 x10° 6.5 x10° 1.4 x10'
AKEGOS8 P 00,1100 1.11110l 20x10I 2.4:110l
AKEGOOS C 0.0x10 1.0x10° 3.2 x10° 5.4 x10°
AKEGOO7 C 6.2 x10' 6.2 x101 1.1 x10° 2.4 x10°
AKEGOll C 0.0x10: 2.0x10° 4.0 x10° 4.5 x10°
AKEGOl7 C 0.0x10 2.0 x10° 6.7 x10° 7.9 x10°
AKEGOZ7 C 0.0 xlo0 1.0 5110l 1.0 11102 1.0 x102
AKEGO38 c 0.0 x100 7.0 x10° 2.1 x10‘ 2.2 x10l
AKEGO40 C 0.0 x10° 5.0 x10° 2.0 x10l 2.1 x101
AKEGO49 C 0.0 x10° 2.0 x10° 1.5 x101 2.4 x10'
AKEGOSO C 0.0 x100 4.0 x100 2.3 5110I 5.2 51101
AKEGOSI C 0.0x10° 4.0x10° 7.2 x10° 1.6x10‘
AKEGOS7 C 0.0 x10° 7.0 x10° 1.8 x10' 2.3 x10l
Mean (On-Post) 2.4 11100 6.9 x10° 1.5 x10‘ 1.9 1110T
AKEGOOI R 0.0 x10° 2.0 x10° 6.6 x10° 8.3 x10°
AKEGOOZ R 0.0 x10° 1.3 x10‘ 3.6 x10° 4.8 x10°
AKEGOO4 R 0.0 51100 3.0 x10° 1.3 xlO° 3.5 x10°
AKEGOIO R 0.0x10° 3.0x10° 1.5 x10° 3.2 x10°
1414130012 R 8.2 51101 8.2 x10' 1.1x102 1.151102
AKEGOIG R 0.0x10° 5.0x10° 4.0x10° 5.9x10°
AKEGOZI R 0.0 x10° 1.0 x10° 1.0 x10' 1.3 x101
AKEGOZS R 0.0x10° 7.051100 1.4x10‘ 1.55110l
AKEGO31 R 0.0 x10° 8.0 x10° 2.6 x10l 2.7 x101
AKEGOBZ R 0.0 x10° 3.0 x10° 3.2 x10° 6.0 x10°
AKBGO33 R 1.3 x10' 1.3 5110l 5.2 x10° 7.8 x10°
AKEGO35 R 0.0 x10° 5.0 x10° 4.0 x10° 4.5 x10°
AKEGO41 R 1.251102 1.251102 9.7 x10' 1.051102
AKEGO43 R 0.0 x10° 6.0 x10° 2.5 x10° 3.5 x10°
AKEGO44 R 0.0 x10° 3.0 x10° 4.4 x10° 7.4 x10°
AKEGO47 R 0.0x10° 2.0x10° 9.3 x10° 1.05110l
Mean (Ram 1.4 1110' 1.8 x10‘ 2.0 x10' 2.2 noI

 

R= reference sites; P = periphery of the RMA; C= core area of RMA
TCDD-EQ = bioassay results
TEQREP= analytical chemistry results multiplied by H4IIE relative potencies
mm = 0 used for “non-detect” congeners
MAX = MDL used for “non-detect” congeners

97

Appendix VII- Calculation of Uncertainty Factors for TRV Determination

98

Overall Uncertainty Factors (UFS) are calculated and applied to existing NOAEL
or LOAEL data in order to determine conservative Toxicity Reference Values (TRVS) for
a given species and toxicant of interest. UF S are based on several parameters, which are
assigned a certain uncertainty value within a speciﬁed range. UFs are based on the
following parameters:

1. Intertaxon extrapolation

2. Study duration

3. Study endpoint
Uncertainty factors are also based on the following modifying factors:

1. Threatened species

2. Relevance of endpoint

3. Extrapolating lab to ﬁeld

4. Co-contaminants

5. Unclear endpoint

6. Study species sensitivity

7. Organ ratios

8. lntraspecies variability
Values are assigned to the ﬁrst three parameters listed depending on the amount of
uncertainty associated with that parameter. These values are multiplied together and the
product is multiplied by the sum of all values assigned to the modifying parameters.
Modifying parameters are assigned values pertinent to TRV calculation (EPA 1993). The

following is an example of an Overall Uncertainty Factor calculation:

99

Calculation of the great horned owl TRV from the Hoffmann et al. 1998 study using

American kestrel egg and PCB126

 

Uncertainty Factors

NOAEL =2,3OO pg/g

 

 

 

 

 

 

 

 

 

 

 

 

Intertaxon extrapolation 5
Study Duration 1
Study endpoint l

Modifying Factors
Threatened species 0
Relevance of endpoint 0
Extrapolating lab to ﬁeld 2
Co-contaminants O
Endpoint Unclear 115
Study Species sensitivity 0
Organ ratios 1
Intraspecies variability O

 

 

 

Overall UF for TRV (1*1*5)*(2+1.5+1)=
22.5
Proposed TRV TRV= NOAEL * organ ratio*TEF/U F

 

=(2300pg/g*(l.5)*0.l)/22.5

=15 pg TCDD/ g liver

 

100

 

IIIIIIIIIIIIIIIIIIIIIIIIIIIIIII

'11111111111111111111111111111 '

   

 

 

 

 

. iii