Work
Assignment
3­
7
White
Paper
on
Species/
Strain/
Stock
in
Endocrine
Disruptor
Assays
Contract
No.
68­
W­
01­
023
JULY
25,
2003
RTI
Project
No.
08055.002.023
PREPARED
BY:

Sherry
P.
Parker,
Ph.
D.
Rochelle
W.
Tyl,
Ph.
D.,
DABT
REVIEWED
BY:

Jimmy
L.
Spearow,
Ph.
D.

FOR:

James
Kariya
WORK
ASSIGNMENT
MANAGER
U.
S.
ENVIRONMENTAL
PROTECTION
AGENCY
ENDOCRINE
DISRUPTOR
SCREENING
PROGRAM
WASHINGTON,
DC
BATTELLE
505
KING
AVENUE
COLUMBUS,
OH
43201
iii
TABLE
OF
CONTENTS
Page
1.0
Executive
Summary
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
1
2.0
Introduction
and
Background
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
1
2.1
Purpose
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
3
2.2
Literature
Search
Strategy
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
3
2.2.1
Databases
Searched
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
3
2.2.2
Database
Search
Strategies
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
4
2.2.3
Keywords
and
Phrases
Used
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
4
2.2.4
Summary
of
the
Review
Process
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
6
2.3
Definitions
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
6
2.3.1
Inbred
and
Outbred
Strains
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
7
2.3.2
Species
Selection
for
Endocrine
Disruption
Assays
and
Genetic
Variability
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
21
2.3.3
Confounders
Affecting
Comparisons
of
Reproductive
Toxicity
Data
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
23
2.4
Assays
Under
Consideration
of
the
EDSP
and
Associated
Endocrine
Endpoints
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
27
2.4.1
One­
Generation
Assay
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
30
2.4.2
Pubertal
Male
and
Female
Assays
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
30
2.4.3
In
Utero
Through
Lactation
Assay
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
33
2.4.4
Adult
Male
Assay
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
33
2.4.5
Two­
generation
Assay
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
33
2.5
Endocrine
Endpoints
Under
Consideration
for
EDSP
Assays
and
Intraspecies
Variability
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
34
2.5.1
Fertility
and
Gestational
Indices
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
35
2.5.2
Survival
and
Growth
Indices
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
36
2.5.3
Reproductive
Tract
Development
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
38
2.5.4
Anogenital
Distance
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
40
2.5.5
Urethral
Vaginal
Distance
(
UVD)
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
41
2.5.6
Retention
of
Nipples
in
Preweanling
Males
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
41
2.5.7
Puberty
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
42
2.5.7.1
Vaginal
Patency
in
Females
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
43
2.5.7.2
Age
of
First
Estrus
in
Females
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
44
2.5.7.3
Preputial
Separation
in
Males
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
44
2.5.8
Estrous
Cyclicity
and
Ovulation
Rate
in
Postpubertal
Females
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
44
2.5.9
Andrology
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
46
2.5.10
Organ
Weights
and
Histopathology
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
47
2.5.11
Behavioral
Assessments/
Clinical
Observations
.
.
.
.
.
.
.
.
.
.
.
.
51
2.5.12
Hormonal
Controls
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
52
2.5.13
Uterine
Weight
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
56
iv
TABLE
OF
CONTENTS
(
continued)
Page
3.0
Interspecies
Similarities
and
Differences
in
Endocrine
Endpoints
.
.
.
.
.
.
.
.
.
.
58
4.0
Summary
and
Conclusions
of
Intraspecies
and
Interspecies
Similarities
and
Differences
in
Endocrine
Endpoints
and
Conclusions
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
60
5.0
References
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
76
LIST
OF
TABLES
Page
Table
1.
Assays
Under
Consideration
by
the
EDSP
and
Associated
Endocrine
Endpoints
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
28
Table
2.
Intraspecies
Comparisons
of
Endocrine
Endpoints
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
62
Table
3.
Summary
of
Agent­
and
Endpoint­
Specific
Intraspecies
Differences
.
.
.
.
71
1
Intraspecies
and
Interspecies
Differences
in
Endocrine
Endpoints
in
In
Vivo
Assays
Under
Consideration
for
the
Endocrine
Disruptor
Screening
Program
1.0
Executive
Summary
This
white
paper
is
a
review
of
the
interspecies
and
intraspecies
similarities
and
differences
in
endocrine
endpoints
in
the
absence
and
presence
of
test
chemicals,
in
order
to
determine
whether
specific
species/
strains
should
be
preferred
or
avoided
when
screening
for
endocrine
activity.
There
is
much
evidence
that
different
species
and
strains
within
species
exhibit
differing
sensitivities
to
endocrine­
active
compounds,
specific
for
chemicals
and
endpoints
evaluated.
Thus
selection
of
appropriate
species
and
strain(
s),
or
at
least
understanding
the
differential
responsivity
of
them
is
crucial
to
detecting
effects
in
animal
models
which
are
extrapolable
to
human
risk.
This
white
paper
is
limited
to
the
species
being
considered
for
inclusion
in
the
Endocrine
Disruptor
Screening
Program
(
EDSP)
and
also
limited
in
scope
to
the
endocrine
endpoints
under
consideration.
Currently,
the
reproductive
and
developmental
toxicology
Environmental
Protection
Agency
(
EPA)
guideline
studies
recommend
using
the
rat
and
not
strains
with
low
fecundity.
The
most
commonly
used
rat
strain
for
these
Guideline
studies
is
the
CD
Sprague­
Dawley
(
SD)
rat
(
the
CD­
1
Swiss
mouse
is
also
frequently
used).
Though
the
majority
of
historical
data
exists
in
this
species/
strain,
there
is
evidence
that
endocrine­
active
chemicals
may
have
very
different
dose­
response
curves
for
certain
endocrine­
related
reproductive
endpoints,
which
may,
in
part,
be
due
to
a
differential
sensitivity
of
different
species/
strains
and
endpoints
in
these
species/
strains
to
these
chemicals.
Since
confounding
factors
make
interlaboratory
comparisons
of
data
problematic,
multi­
strain
studies
conducted
under
the
same
experimental
conditions
and
same
laboratory
were
primarily
used
in
the
species/
strain/
stock
comparisons.

Comparisons
revealed
a
lack
of
consistency
in
effects
produced
by
endocrine­
disrupting
chemicals
on
endocrine
endpoints
from
strain
to
strain.
Endocrine
effects
were
chemical
specific,
strain
specific,
endpoint
specific,
and,
in
some
cases,
laboratory
specific.
There
were
more
sensitive
and
less
sensitive
strains
to
endocrineactive
compounds
among
outbred
and
inbred
strains,
depending
on
the
chemical
used
and
the
endpoints
evaluated.
Clearly,
strain
(
genotype)
by
environmental
agent
by
endpoint
interactions
need
to
be
considered
in
selecting
the
appropriate
species/
strains
for
EDSP
assays.

2.0
Introduction
and
Background
In
1996,
the
Food
Quality
Protection
Act
was
enacted
by
Congress.
It
directs
the
United
States
Environmental
Protection
Agency
(
EPA)
to
screen
pesticides
for
endocrine
activity.
Thus,
the
EPA
is
implementing
an
Endocrine
Disruptor
Screening
Program
(
EDSP).
In
this
program,
comprehensive
toxicological
and
ecotoxicological
screens
and
assays
are
being
developed
to
identify
and
characterize
the
endocrine
effects
of
various
environmental
contaminants,
industrial
chemicals,
and
pesticides.
2
The
program's
aim
is
to
develop
a
two­
tiered
approach,
i.
e.,
a
combination
of
in
vitro
and
in
vivo
mammalian
and
ecotoxicological
screens
(
Tier
1)
to
identify
substances
with
the
potential
to
interact
with
the
endocrine
system,
and
a
set
of
definitive
apical
in
vivo
assays
(
Tier
2)
to
determine
whether
the
substances
identified
in
Tier
1
cause
adverse
effects,
identify
the
adverse
effects,
and
determine
the
quantitative
relationship
between
dose
and
adverse
effects.
The
EDSP
is
required
to
use
"
validated"
test
systems.
The
Endocrine
Disruptor
Methods
Validation
Subcommittee
(
EDMVS)
provides
technical
advice
on
the
validation
of
most
of
the
assays.

In
order
to
determine
necessary
modifications
to
standard
reproductive
and
developmental
toxicology
guidelines,
to
detect
and
characterize
the
effects
of
endocrine
disruptors,
the
National
Toxicology
Program/
National
Institute
of
Environmental
Health
Sciences
(
NTP/
NIEHS),
at
the
request
of
the
EPA,
organized
a
peer
review
panel
which
convened
in
October
2000.
The
panel
consisted
of
scientists
from
academia,
industry,
and
the
government.
One
of
the
subpanels
of
the
NTP's
Endocrine
Disruptor
Low
Dose
Peer
Review
Panel
investigated
the
concern
that
animal
models
used
in
assays
to
detect
endocrine
disruption
have
been
chosen
on
the
basis
of
convenience
and
familiarity,
and
species/
strains/
stocks
which
are
more
frequently
used
are
those
which
are
bred
specifically
for
robust
fecundity
and
likely
reduced
sensitivity
to
endocrine
perturbations
(
NTP's
Report
of
the
Endocrine
Disruptors
Low
Dose
Peer
Review,
2000).
The
subpanel
addressed
this
issue
with
respect
to
the
mammalian
twogeneration
assay
and
had
the
following
remarks:

On
animal
model
selection:
The
subpanel
recommended
that
the
selection
of
species
or
strain
for
future
studies
should
be
the
result
of
a
more
deliberate
thought
process,
rather
than
based
on
availability,
convenience,
or
familiarity.
Development
of
a
core
of
historical
data
across
mouse
and
rat
strains
(
inbred
and
outbred),
with
known
endocrine­
disrupting
chemicals
and
characterization
of
the
reproductive
endpoints
of
interest,
was
also
recommended.
These
data
could
be
used
to
modify
current
protocols
with
respect
to
the
number
and
doses/
concentrations
of
dose
groups,
group
size,
and
endpoint
selection.

On
species/
strain
selection:
The
subpanel
asserted
that
although
there
is
an
abundance
of
historical
control
data
in
CD­
1
mice
and
SD
rats
collected
in
reproductive
and
developmental
toxicology
studies,
inbred
F1
strains
such
as
the
B6C3F1
mouse
may
provide
less
variable
responses
to
endpoints
assessed.
In
addition,
the
advantage
of
historical
data
may
be
compromised
by
genetic
drift
and/
or
selective
breeding.

In
the
December
2001
meeting
of
the
EDMVS,
committee
members
discussed
strains
and
stocks
and
concluded
that
the
EPA
should
prepare
a
white
paper
summarizing
what
is
known
about
intraspecies
strain/
stock
similarities
and
differences
in
the
neuroendocrine
control
of
reproduction/
development
and
in
responses
to
endocrine­
active
chemicals,
and
provide
the
rationale
for
strain/
stock
selection.
This
review
is
the
"
white
paper"
requested
by
the
EDMVS,
designed
to
assess
the
interspecies
and
intraspecies
variability
of
endocrine
endpoints
in
in
vivo
assays
under
consideration
by
the
EDSP.
Please
note
that
the
uterotrophic
assay
and
the
3
Hershberger
assay
are
under
consideration
by
the
EDSP,
for
inclusion
in
testing
guidelines
but
are
being
standardized
and
validated
in
a
cooperative
effort
between
US
EPA
and
OECD
and
participating
laboratories
on
three
continents;
therefore,
these
assays
will
not
be
standardized
or
validated
in
this
project,
and
will
not
be
discussed
in
this
white
paper.

2.1
Purpose
The
purpose
of
this
review
is
to
summarize
the
interspecies
and
intraspecies
similarities
and
differences
in
response
to
endocrine
endpoints,
in
order
to
determine
whether
specific
species/
strains
should
be
preferred
or
avoided
when
screening
for
endocrine
activity.
Currently,
the
recommended
species
for
reproductive
and
developmental
toxicology
EPA
guideline
studies
is
the
rat.
Though
the
majority
of
historical
data
exists
in
the
SD
rat
and
CD­
1
mouse
strains,
there
is
evidence
that
endocrine­
active
chemicals
may
have
very
different
dose­
response
curves
for
responses
to
certain
endocrine­
related
reproductive
endpoints,
and
this
may
in
part
be
due
to
a
differential
sensitivity
of
different
species/
strains
to
these
chemicals.
Whether
or
not
non­
monotonic
dose­
response
curves
are
eventually
shown
to
be
a
general
phenomenon
of
endocrine
disruptors
(
or
a
class
of
them),
it
is
appropriate
to
ask
whether
screening
for
endocrine
activity
is
being
carried
out
in
appropriately
sensitive
test
systems.

A
literature
review
was
performed
to
identify
key
references
on
the
following
two
general
topics:

1)
Influence
of
rat
strain/
stock
on
endocrine
endpoints
measured
in
the
mammalian
in
vivo
assays
being
considered
for
the
EDSP
2)
Interspecies
similarities
and
differences
in
neuroendocrine
control
of
reproduction/
development
and
in
responses
to
endocrine­
active
chemicals
(
reported
since
1986)

2.2
Literature
Search
Strategy
Literature
databases,
accessible
through
the
RTI
Information
Technology
Services,
were
searched
for
published
peer­
reviewed
and
nonpeer­
reviewed
articles
using
on­
line
electronic
literature
databases,
followed
by
a
focused
literature
screening
process.
Full
citations,
including
abstracts
(
if
available)
were
retrieved
for
review.

2.2.1
Databases
Searched
MedLine,
PubMed,
Biological
Abstracts,
Chemical
Abstracts,
Toxline
including
DART
(
Developmental
and
Reproductive
Toxicology)
4
2.2.2
Database
Search
Strategies
°
English
language
articles.
The
literature
search
was
performed
to
include
all
applicable
English
language
articles.
°
Foreign
language
articles
with
English
abstracts.
The
literature
search
was
performed
to
exclude
foreign
language
articles
with
foreign
language
abstracts.
However,
the
literature
search
was
performed
in
a
manner
that
accepted
foreign
language
articles
that
also
have
English
language
abstracts.
This
strategy
was
used
to
allow
authors
to
review
some
literature
published
in
any
number
of
foreign
languages.

2.2.3
Keywords
and
Phrases
Used
Articles
were
identified
through
the
use
of
keywords
in
the
literature
search.
Individual
sets
of
keywords
were
selected
for
each
of
the
topics
listed
in
"
Objectives"
above.
Combinations
of
keywords
in
Column
A
were
combined
with
keywords
in
Column
B
for
Task
2
in
order
to
identify
key
articles
addressing
the
influence
of
rat
strain
and
stock
on
endocrine
endpoints
from
pubertal
male
and
female,
in
utero
lactational,
adult
male,
and
two­
generation
reproductive
toxicity
studies.
Initially,
"
rat
strain"
and
"
rat
stock"
were
combined
with
keywords
in
Column
B.
When
more
than
100
hits
were
found
per
combination
of
keywords,
additional
terms
from
Column
A
were
added
to
limit
the
search.
Since
the
rabbit
is
not
a
species
under
consideration
for
use
in
the
EDSP,
it
has
been
deliberately
omitted
from
the
white
paper.

A.
Rat
strain
Rat
stock
(
supplier)
Rat
genetic
variation
Sprague­
Dawley
(
CD,
SD)
Long­
Evans
(
hooded,
LE)
Alderly
Park
(
ALK
or
ALP)
Dark
agouti
Norway
Fisher
(
F344)
Wistar
Lewis
Noble
Holtzman
(
HTZ)
Outbred
rats
Inbred
rats
B.
1
Reproductive
toxicity
Anogenital
distance
Nipple
retention
Ovarian
corpora
lutea
Precoital
interval
Sperm,
sperm
production
Estrous
cyclicity
Areolar
retention
Vaginal
patency
Preputial
separation
Uterotrophic
Embryo
loss
Testes
(
Leydig,
and
Sertoli
cells)
Lactational
exposure
Thyroid
development
Thyroid
development
and
pregnancy
Lactation
Blood
testis
barrier
Spermatogenesis
Mammary
glands
17$­
Estradiol
(
E2)
Estrus
Litter
size
Gestation,
gestational
loss
Pregnancy
5
INSL3
Reproductive
tract
development
Müllerian
ducts,
Müllerian
Inhibitory
Substance
Wolffian
ducts
Reproductive
and
accessory
organs
Thyroid
hormones
Hypothalamic,
pituitary,
and
gonadal
hormones
Puberty
Implantation
Fetal
survival,
mortality
1keywords
were
obtained
from
approved
EDSP
protocols
and
preliminary
search
results.

In
addition
to
searches
performed
with
combinations
of
both
columns
of
keywords,
papers
by
the
following
authors
were
searched
for
relevant
articles:

vom
Saal,
FS;
Ashby,
J;
Odum,
J;
Gray,
LE;
Ostby,
J;
Cooper,
RL;
Spearow,
JL;
Michna,
H;
Diel,
P;
Festing,
M.

The
focus
of
the
search
was
on
"
rat
strain."
When
there
was
a
paucity
of
references
pertaining
to
a
general
endocrine
endpoint,
"
mouse
strain"
was
added
to
the
search.

For
interspecies
comparisons
of
reproductive
and
developmental
endpoints,
combinations
of
keywords
in
Column
C
were
combined
with
keywords
in
Column
D,
and
the
search
was
limited
to
articles
published
since
1986,
in
order
to
identify
key
articles
addressing
interspecies
differences
in
neuroendocrine
control
of
reproduction/
development,
with
emphasis
on
parameters
addressed
by
the
NTP's
Endocrine
Disruptor
Low
Dose
Peer
Review
Subpanel
on
Biological
Factors
and
Study
Design.

C.
Reproductive
toxicity,
endocrine
Developmental
toxicity,
endocrine
Mouse/
mice
Rat
Interspecies
D.
2Intrauterine
position
(
IUP)
Estradiol
LH
LHRH
(
GnRH)
FSH
ACTH
Hypothalamic­
pituitary­
gonadal
(
HPG)
axis
Puberty
Prolactin
Testosterone
(
T)
Thyroid
hormone
(
T3,
T4)
TSH
Oxytocin
Growth
hormone
Diet
Caging
Bedding
Genetic
variability
Gene
expression
Strain
differences
Genotype
6
Oogenesis
Meiosis
Mitosis
Endocrine­
mediated
nondisjunction
Hypospadias
Steroidogenesis
Reproductive
tract
malformation
Cryptorchidism
Spermatogenesis
Nipple
retention
Anogenital
distance
Uterus
Progesterone
Dihydrotestosterone
(
DHT)
Gestation
Pregnancy
Gestational
loss
Embryo
loss
Implantation
Fetal
survival,
mortality
Androsterone
Androstenedione
Prostate
Hypothalamic­
pituitary­
thyroid
(
HPT)
axis
Areolar
retention
Vaginal
patency
Preputial
separation
(
balanopreputial
Epispadia
Ovary,
ovaries
Testis,
testes
Gonad
Aromatase
2Keywords
were
obtained
from
the
NTP's
Report
of
the
Endocrine
Disruptors
Low
Dose
Peer
Review
(
Subpanel
on
Biological
Factors
and
Study
Design),
October
2000,
and
preliminary
search
results.

The
focus
of
the
search
was
reproductive
and
developmental
toxicity
and
endocrine
effects.
When
a
search
of
combined
terms
from
Columns
C
and
D
yielded
more
than
100
hits,
specific
animal
species
names
(
in
Column
C)
were
added
to
limit
the
search.

2.2.4
Summary
of
the
Review
Process
After
key
articles
were
identified,
individual
references
were
retrieved
and
further
searched
to
identify
additional
key
articles
(
i.
e.,
tree
search).
References
of
interest
were
electronically
downloaded
to
Reference
Manager,
retrieved,
evaluated,
organized
by
topic,
and
retained
for
use
in
preparing
the
white
paper.

2.3
Definitions
The
Institute
for
Laboratory
Animal
Resources
(
ILAR)
defines
a
strain
as
"
inbred"
when
it
has
been
mated
brother
x
sister
for
20
or
more
consecutive
generations
(
ILAR
Journal,
1992).
"
To
ensure
isogenicity,
as
well
as
homozygosity,
a
single
brother
x
sister
pair
must
be
selected
in
the
twentieth
or
a
subsequent
generation
to
perpetuate
the
strain.
Parent
x
offspring
matings
may
be
substituted
for
brother
x
sister
matings,
provided
that
in
the
case
of
consecutive
parent
x
offspring
matings,
each
mating
is
to
the
younger
of
the
two
parents;
this
will
prevent
repeated
backcrossing
to
a
single
individual"
(
ILAR
Journal,
1992)
.
A
strain
is
defined
as
"
outbred"
when
it
has
been
maintained
as
a
closed
colony
for
at
least
four
generations.
"
To
minimize
changes
caused
by
inbreeding
and
genetic
drift,
the
population
should
be
maintained
in
such
7
numbers
as
to
give
less
than
1
percent
inbreeding
per
generation.
Under
these
conditions,
a
heterozygous
breeding
population
is
expected
to
reach
equilibrium
and
to
produce
a
stock
of
stable
genetic
composition.
Formerly
inbred
strains
may
be
included
after
four
generations
of
closed
outbreeding,
provided
that
continued
outbreeding
is
intended.
Outbred
stocks
are
not
necessarily
highly
variable
genetically.
The
degree
of
genetic
variability
of
any
individual
stock
can
only
be
determined
by
studying
the
appropriate
genetic
markers"
(
ILAR
Journal,
1992).
"
Wild
type"
refers
to
the
genotype
or
phenotype
that
is
found
most
commonly
in
nature
or
in
the
standard
laboratory
stock
for
a
given
organism,
before
mutations
are
introduced.

2.3.1
Inbred
and
Outbred
Strains
The
term
"
strain"
refers
to
a
closed
population
of
organisms
of
the
same
species,
with
distinctive
hereditary
characteristics
that
distinguish
them
from
other
groups
within
the
species.
"
Strains"
are
artificially
maintained
to
promote
certain
characteristics
by
manipulation
of
population
size,
mating
system,
as
well
as
the
intensity
and
direction
of
artificial
selection
(
Lynch
and
Walsh
1998).
The
terms
"
strain,"
"
stock,"
and
"
line"
are
used
somewhat
interchangeably
on
inbred
and
outbred
strains.
The
most
commonly
used
outbred
strains
of
rats
include
Wistar,
Long
Evans,
SD,
and
CD
(
caesarean­
derived
CR
®
SD).
Outbred
mice
include
ICR,
SENCAR,
NMRI,
CFW,
CF1,
CD­
1,
and
"
Swiss"
mice.
The
most
commonly
used
inbred
rat
strains
include
Fischer
(
F­
344),
Brown
Norway
(
BN),
ACI,
Lewis
(
LEW),
Noble,
DA,
Copenhagen,
Dahl
Salt
Sensitive
(
SS),
spontaneously
hypertensive
rat
(
SHR),
and
WKY.
Commonly
used
inbred
mouse
strains
include
A,
AKR,
BALB/
c,
CBA/
Ca,
CBA,
C3H/
He,
C57BL/
6,
C57BL/
10,
DBA/
2,
FVB,
NOD,
NZB,
SJL,
and
SWR.
Commonly
used
isogenic
rat
F1
crosses
include
F­
344
x
BNF1
(
FBNF1)
and
LEW
x
BN
F1
(
LBNF1).
Commonly
used
isogenic
mouse
F1
crosses
include
B6D2F1,
B6C3F1,
B6AF1,
CAF1,
CB6F1,
and
NZBWF1.

Strain
differences,
in
response
to
xenobiotics
and
hormonally
active
compounds,
are
an
extremely
common
finding
in
the
few
studies
that
have
compared
several
strains
of
mice
or
rats
(
Festing
1979;
Festing
1987;
Festing
1995;
Steinmetz
et
al.
1998;
Spearow
et
al.
1999;
Long
et
al.
2000).
Since
most
toxicologists
and
physiologists
only
use
a
single
strain
of
mice
or
rats,
the
amount
of
genetic
variation
between
strains
is
not
usually
apparent.
Many
assume
that
finding
a
fairly
uniform
response
to
a
given
hormonally
active
toxicant
in
a
commercially
available
outbred
strain
indicates
a
lack
of
genetic
variation
in
susceptibility
in
the
species.
As
will
be
discussed
in
detail,
such
assumptions
are
often
invalidated
by
the
extremely
narrow
genetic
base,
history
of
long­
term
selection
for
large
litter
size,
and
correlated
changes
in
reproductive
development
and
function
traits
characteristic
of
many
commercially
available
outbred
strains
of
mice
and
rats.
This
section
will
consider
the
features
and
selection
history
that
have
defined
available
inbred
and
outbred
mouse
and
rat
strains,
in
particular,
factors
relating
to
genetic
variation
in
susceptibility
to
endocrine
disruption.

One
criticism
against
utilizing
the
most
commonly
used
outbred
strains
is
based
on
the
fact
that
these
animals
have
been
bred
specifically
for
robust
fecundity
and
8
relative
insensitivity
to
endocrine
perturbations.
Selective
breeding
can
alter
reproductive
traits,
including
natural
and
hormone­
induced
ovulation
in
the
rat,
litter
size,
testis
weight,
and
sperm
production
(
Bradford
1969;
Johnson
et
al.,
1994;
Okwun
et
al.,
1996a;
Okwun
et
al.,
1996b;
Spearow
and
Barkley
2001).
Selection
for
large
litter
size
can
affect
the
hypothalamic­
pituitary­
gonadal
axis,
resulting
in
differential
sensitivity
of
the
ovaries
to
gonadotropins
and
changes
in
follicular
steroidogenesis
in
female
mice,
increased
testis
weight
in
males,
and
altered
sensitivity
of
testis
weight
to
estrogen
in
males
(
Spearow
1985;
Spearow
et
al.,
1987;
Spearow
et
al.,
2001).

Inbred
strains
have
several
features
that
make
them
valuable
for
biomedical
research
(
Festing,
1979).
They
are
produced
by
at
least
20
generations
of
brother
x
sister
matings,
with
all
individuals
of
a
strain
in
the
20th
or
subsequent
generation
tracing
back
to
a
single
common
ancestral
breeding
pair
(
Festing
1987).
While
a
small
number
of
genes
may
continue
to
segregate
as
residual
heterozygosity,
especially
in
the
20th
to
30th
generation
of
inbreeding,
practically
speaking,
all
members
of
an
inbred
strain
are
isogenic,
i.
e.,
essentially
genetically
identical
individuals
(
Festing
1987).
Thus,
each
isogenic
inbred
strain
represents
an
infinitely
repeatable,
genetically
defined
set
of
identical
twins
which
are
homozygous
at
essentially
all
loci.
A
complete
mouse
genomic
map
and
mouse
genomic
DNA
sequence
are
available
for
the
C57BL/
6J
inbred
mouse
strain
(
www.
informatics.
jax.
org;
www.
ensembl.
org/
Mus_
musculus/).
In
addition,
a
rat
genomic
map,
and
a
rat
genomic
DNA
sequence
are
in
development
for
the
Brown
Norway
BN/
SsNHsd/
MCW
(
BN)
rat
strain
(
Rattus
norvegicus)
(
www.
ensembl.
org/
Rattus_
norvegicus/;
also
see
http://
bacpac.
chori.
org/
rat230.
htm).

Genetically
defined,
isogenic
inbred
strains
are
an
important
resource
for
determining
the
toxicological
effects
of
chemicals
on
biodiversity
within
such
species.
Because
of
their
very
high
level
of
homozygosity,
inbred
strains
stay
genetically
uniform
and
constant
over
many
generations,
with
only
a
slight
amount
of
genetic
drift
due
to
new
mutations
(
Festing
1987).
Their
utility
in
toxicological
research
is
due
to
the
highly
consistent
and
reproducible
genotype,
ideal
for
testing
many
different
chemicals
for
toxicity
over
time
(
Festing
1979;
Festing
1987;
Festing
1995).
Inbred
strains
are
genetically
monitored
to
confirm
genetic
integrity
of
each
strain
stain,
and
to
ensure
they
have
not
been
accidentally
outcrossed
by
using
coat
color,
biochemical,
immunogenetic,
as
well
as
microsatellite
and
Single
Nucleotide
Polymorphism
(
SNP)
molecular
genetic
markers.

F1
crosses
between
inbred
strains
are
also
commonly
used
in
toxicological
research,
including
B6C3F1
mice
by
the
NTP.
Isogenic
F1
crosses
are
uniformly
heterozygous
at
all
loci,
differing
between
parental
inbred
strains.
As
parental
inbred
strain
genotypes
and
genetic
sequence
become
increasingly
defined,
the
genotype
of
F1
crosses
of
genetically
defined
inbred
strains
can
be
accurately
predicted
in
silico
(
Manly
et
al.
2001;
Marshall
et
al.,
2002).
This
enables
the
production
of
an
almost
infinite
number
of
genetically
defined
isogenic
F1
cross
animals
with
identically
defined
homozygous
and/
or
heterozygous
genotypes
at
all
known
loci.
While
the
F1
crosses
of
inbred
strains
are
isogenic,
their
F1
+
F1
crosses
to
generate
F2
animals
and/
or
9
backcrosses
produce
offspring
which
are
segregating
at
many
loci.
The
increased
genetic
and
phenotypic
variability
in
such
segregating
crosses
confounds
the
description
of
treatment­
related
effects
and
thereby
decreases
the
sensitivity
of
detecting
treatment
effects,
and
therefore
limits
their
use
in
toxicological
studies
with
offspring
of
F1
parents.

A
wide
variety
of
specialized,
genetically
defined,
inbred
strain
genetic
resources
are
also
available,
and
are
especially
useful
for
characterizing,
mapping,
and
identifying
genes
controlling
strain
differences
in
a
wide
variety
of
traits
(
Silver
1995).
Two
strains
are
referred
to
as
consomic
when
they
differ
by
one
complete
chromosome
pair,
and
a
congenic
strain
when
it
carries
a
small
genetic
region
(
ideally
a
single
gene)
from
another
strain,
but
which
is
otherwise
identical
to
the
original
inbred
strain.
Several
congenic
inbred
strains
of
mice
and
rats
are
available,
each
with
a
single
chromosomal
region
from
one
strain
backcrossed
onto
the
genetic
background
of
another
strain.
Several
consomic
(
chromosome
substitution)
inbred
strains
are
also
available,
each
with
a
single
chromosome
from
one
strain
introgressed
by
backcrossing
onto
the
genetic
background
of
another
strain.
Available
rat
consomic
strains
include
several
SS.
BN
chromosome­
specific
consomic
strains
with
individual
BN
chromosomes
on
the
Dahl
Salt­
Sensitive
(
SS)
genetic
background.
Available
mouse
consomic
strains
include
a
full
set
of
C57BL/
6J­
A
chromosome
substitution
strains,
each
with
a
single
A/
J
chromosome
substituted
on
the
C57BL/
6J
(
B6)
genetic
background
(
Nadeau
et
al.,
2000).
Several
mouse
and
rat
recombinant
inbred
(
RI)
strain
sets
are
also
available.
These
RI
strains
are
formed
by
crossing
two
inbred
parental
strains,
intermating
the
F1
to
produce
F2s,
and
then
producing
a
set
of
inbred
lines
by
inbreeding
ad
infinitum
from
each
of
several
F2
mating
pairs.
Available
mouse
RI
strain
sets
include
BXD
RI,
AXB,
BXA
RI,
AKXD
RI,
and
AKXL
RI
(
http://
jaxmice.
jax.
org/
jaxmicedb/
html/
rcbinbred.
shtml).
Available
rat
RI
strain
sets
include
the
SHR
x
BN
(
HXB/
BXH)
RI
and
the
LEW
x
BN
(
LXB)
RI.

Outbred
strains
are
also
commonly
used
in
biomedical
research
and
are
intended
to
be
genetically
more
diverse
by
maintenance
of
large,
heterogenous,
random
mating
populations
and
avoiding
inbreeding.
Yet,
the
diversity
of
many
outbred
mouse
strains
is
limited,
in
part,
by
the
narrow
genetic
base
of
the
founders
and,
in
some
cases,
early
inbreeding
programs
to
select
against
deleterious
recessive
genes.
For
example,
laboratory
stocks
of
Norway
rats
(
Rattus
norvegicus)
were
developed
from
albino
mutants
that
had
been
bred
by
fanciers
(
Gray
1977).
Around
1900,
H.
H.
Donaldson
obtained
rats
from
fanciers
for
a
breeding
colony
at
the
University
of
Chicago.
This
stock
was
transferred
to
the
Wistar
Institute
in
1906
and
maintained
as
a
closed
colony.
This
Wistar
stock
contributed
to,
and
is
therefore
related
to,
several
other
outbred
strains.
In
1915,
a
few
albino
Wistar
Institute
females
were
crossed
with
a
single
wild
gray
male
and
then
used
to
develop
the
Long­
Evans
rat
stock
(
http:/
www.
criver.
com/
03CAT/
rm/
rats/
longevansRats.
html).

In
1925,
Robert
Dawley
crossed
a
single
hybrid
hooded
male
of
unknown
origin
with
an
albino
female
of
the
Douredoure
strain
(
probably
from
Wistar).
This
single
hooded
foundation
male
was
subsequently
backcrossed
to
his
albino
female
offspring
10
for
seven
successive
generations.
Multiple
daughter
lines,
developed
by
inbreeding,
were
then
crossed
to
form
the
stable
heterogeneous
SD
stock,
which
was
then
maintained
as
a
closed
outbred
population
(
http://
www.
harlan.
com/).
Thus,
while
the
SD
rat
strain
can
be
traced
back
to
a
single
(
most
likely
a
Wild
x
Wistar)
hooded
hybrid
male
and
a
Wistar
female,
due
to
the
repeated
backcrossing
of
successive
generations
of
daughters
to
the
hooded
hybrid
foundation
male,
the
vast
majority
of
the
genes
in
this
bred
strain
originated
from
the
single
hooded
hybrid
male.
Furthermore,
the
process
of
inbreeding
in
each
of
multiple
lines
enables
enhanced
selection
against
deleterious
recessive
genes
present
in
the
initial
population,
and
enables
increased
litter
size
means
and
improved
selection
responses
in
litter
size
following
crossing
(
Falconer
1971;
Eklund
and
Bradford
1977;
Falconer
1989).
In
other
words,
by
inbreeding
with
selection
in
several
lines
followed
by
crossing
these
lines,
Robert
Dawley
was
very
likely
to
have
eliminated
deleterious
recessive
genes
commonly
found
in
outbred
populations.
By
then
selecting
for
increased
fecundity
and
docility,
he
was
able
to
develop
the
highly
productive
SD
strain.
Charles
River
Laboratories
obtained
SD
breeders
in
1950,
Cesarean
derived
them
in
1955,
and
selected
long­
term
for
large
litter
size
and
vigor
to
develop
the
CD
®
(
SD)
rat
strain
(
http:/
www.
criver.
com/
03CAT/
rm/
rats/
longevansRats.
html).
Given
this
breeding
and
selection
history,
the
SD
strain,
and
the
SD­
derived
CD
strain
have
less
genetic
diversity
and
less
deleterious
recessive
genes
than
what
is
found
in
wild
type
populations.
As
discussed
below,
long­
term
selection
mainly
for
increased
prolificacy
may
have
resulted
in
even
greater
genetic
divergence
for
reproductive
and
correlated
traits.

A
stock
of
Swiss
mice
was
also
developed
from
two
male
and
seven
female
non­
inbred
albino
mice
by
Dr.
de
Coulon,
Lausanne,
Switzerland.
This
stock
was
imported
into
the
U.
S.
by
Dr.
Carla
Lynch
at
Rockefeller
Institute
in
1926,
and
transferred
to
the
Institute
for
Cancer
Research
in
Philadelphia
in
1948,
where
it
was
selected
for
high
production
and
growth
rate
(
http://
www.
taconic.
com/
addinfo/
icrorigin.
htm;
http://
www.
criver.
com/
03CAT/
rm/
mice/
cd1Mice.
html).
This
closed
ICR
Swiss
stock
was
used
to
establish
several
ICR
strains.
ICR
was
Caesarean
derived
in
1959
by
Charles
River
Laboratories
(
CRL)
and
used
to
bred
the
CD­
1
strain
which
was
also
selected
long­
term
for
large
litter
size
and
vigor.

Other
outbred
strains
of
mice
are
also
from
a
very
narrow
genetic
base.
For
example,
the
CRL
CF­
1
®
strain,
which
likely
originated
from
non­
Swiss
mice
at
Rockefeller
Institute,
was
intensively
inbred
for
over
20
generations
by
Carworth
and
then
outbred
from
a
single
mating
pair
(
http://
www.
criver.
com/
03CAT/
rm/
mice/
cf1Mice.
html).
This
outbred
stock,
like
all
the
other
outbred
strains
at
CRL,
was
long­
term,
selected
primarily
for
high
prolificacy.

Population
Genetics:

Mutations,
inbreeding,
genetic
drift,
and
differential
selection
may
account
for
genetic
differences
in
the
same
outbred
strains
provided
by
different
suppliers
or
even
different
colonies
of
the
same
supplier.
In
outbred
strains,
spontaneous
mutations
and
11
recombinations
between
alleles
supplement
each
other
in
generating
and
maintaining
a
multiple
allelic
system,
which
provides
even
more
genetic
variation
on
which
selection
can
act.
The
amount
of
genetic
variability
in
a
randomly
selected
and
randomly
mated
strain
or
population
depends,
in
part,
on
the
initial
heterozygosity,
the
rate
of
inbreeding,
and
the
number
of
generations
of
inbreeding
(
Pirchner
1969;
Falconer
1989).
Inbreeding
is
the
change
in
genotype
frequencies
resulting
from
the
mating
of
related
individuals.
The
rate
of
inbreeding
is
dependent
on
the
effective
population
size
(
Pirchner
1969;
Falconer
1989).
For
pair­
mated
species,
the
rate
of
inbreeding
per
generation
=
1/(
2N),
where
N=
total
number
of
unrelated
individuals
in
a
population
(
Pirchner
1969).
For
populations
in
which
males
are
mated
to
several
females,
the
rate
of
inbreeding
per
generation
=
(
1/(
8Nm))
+
(
1/(
8Mf)),
where
Nm=
number
of
males
and
Nf=
number
of
females
(
Pirchner
1969).

Genetic
drift
is
the
change
of
allele
frequencies
as
a
result
of
sampling
populations
of
limited
size.
A
colony
of
limited
population
size
also
undergoes
genetic
drift
and
therefore
differs
from
other
colonies
of
the
same
origin
(
Kacew
and
Festing,
1996).
Genetic
drift
occurs
when
the
frequency
of
alleles
in
a
population
change
due
to
chance,
i.
e.
sampling,
rather
than
by
natural
or
artificial
selection.
The
"
founder
effect"
is
an
extreme
form
of
genetic
drift,
where
gene
frequency
in
a
small
founding
subset
of
a
population
is
different
than
the
larger
population
from
which
it
was
derived.
For
commercially
available
outbred
strains,
genetic
drift
and
inbreeding
are
likely
greatest
in
the
generations
with
small
effective
population
size,
at
which
Caesarean
derivations
were
performed
to
establish
each
outbred
strain
and
each
outbred
substrain.
In
contrast,
a
large
number
of
breeders
(
typically
several
hundred
to
a
thousand)
are
selected
for
continuing
the
line
in
most
generations
of
breeding
in
most
outbred
mice
and
rat
strains.
During
these
generations
(
with
large
effective
population
sizes),
if
mating
is
random,
the
theoretical
rate
of
inbreeding
is
likely
to
be
quite
low
in
the
0.05%
to
0.5%
range
per
generation.
Nevertheless,
genetic
quality
control
data
comparing
outbred
strain
subpopulations
showed
evidence
for
considerable
genetic
drift
between
colonies
or
subpopulations.
For
example,
Major
Histocompatibility
Complex
RT1
haplotypes
showed
considerable
genetic
drift
between
colonies
of
Crl:
CD
®
(
SD)
BR
rats
(
Rodent
Genetics
and
Genetic
Quality
Control
for
Inbred
and
F1
Hybrid
Strains,
Part
II,
Winter
1992,
Table
8:
http://
www.
criver.
com/
techdocs/
rodent2.
html).

DNA
fingerprinting
analysis
showed
that
the
diversity
within
outbred
rat
strains
was
much
less
than
that
found
among
ten
commonly
available
inbred
rat
strains
(
Festing
1995).
Genetic
and
biochemical
marker
typings
in
63
rat
inbred
laboratory
strains
and
214
substrains
showed
an
average
polymorphism
of
53%
between
strains,
with
wild­
derived
Brown
Norway
(
BN)
strain
rats
showing
the
greatest
genetic
divergence
(
Canzian
1997).
While
much
more
diverse
than
within
any
inbred
strain,
the
genetic
diversity
within
outbred
rat
strains
is
far
lower
than
that
found
between
commonly
available
inbred
strains
and
more
closely
represents
the
level
of
diversity
found
in
island
populations
(
Festing
1995).
For
many
traits,
these
commercial
outbred
strains
show
much
lower
variability
than
that
found
in
genetically
heterogeneous
populations,
such
as
an
F2
cross
between
inbred
strains.
The
limited
genetic
diversity
of
common
commercial
outbreds,
such
as
the
CD
rat,
should
not
be
surprising,
given
its
12
extremely
narrow
genetic
base,
early
inbreeding
to
purge
deleterious
recessive
genes,
and
subsequent
history
of
selection
mainly
for
large
litter
size
and
vigor.

Migration/
Outbreeding:

The
amount
of
genetic
variability
in
a
strain
or
population
also
depends
on
the
frequency
of
migration
or
outbreeding
from
other
populations
and
the
difference
in
gene
frequencies
between
such
populations
(
Pirchner
1969;
Falconer
1989).

Genetic
Monitoring:

Genetic
monitoring
of
outbred
strains
can
be
used
to
confirm
a
strain's
genetic
integrity.
The
main
purpose
of
genetic
quality
control
is
to
preserve
isogenicity
(
http://
www.
criver.
com/
techdocs/
rodent1.
html).
The
loss
of
isogenicity,
i.
e.,
substrain
or
subline
divergence,
has
three
main
causes:
mutation,
drift
in
residual
heterozygosity,
and
genetic
contamination
caused
by
an
unintended
outcross.
Genetic
drift
resulting
from
mutation
or
residual
heterozygosity
is
difficult
to
detect
and
control
and
has
minimal
impact
on
most
research.
The
greatest
cause
of
subline
divergence
affecting
the
usefulness
of
inbred
strains
is
genetic
contamination.
Genetic
quality
control
procedures
are
aimed
at
preventing
and
detecting
genetic
contamination
by
strict
colony
management
and
routine
genetic
monitoring
(
http://
www.
criver.
com/
techdocs/
rodent1.
html).

Genetic
and
Environmental
Sources
of
Variation:

Selection
has
defined
the
available
outbred
strains
of
mice
and
rats.
While
some
of
this
genetic
divergence
between
outbred
line
substrains
is
likely
due
to
genetic
drift,
some
of
the
divergence
between
strains
and
substrains
is
also
very
likely
to
be
due
to
selection.
Early
in
their
development,
outbred
strains
(
including
Wistar,
SD,
and
Long
Evans
rats)
were
selected,
at
least
in
part,
for
docility
and
increased
reproductive
performance
(
Gray
1977).
For
economic
and
productivity
reasons,
most
outbred
strains
continued
to
be
selected
by
commercial
breeders
for
large
litter
size
and
vigor.
For
example,
following
caesarean
derivation
from
the
SD
line,
CRL
selected
the
CD
rat
strain
from
1950
until
1991,
i.
e.,
for
approximately
80­
100
generations,
mainly
for
large
litter
size
in
their
2nd
to
5th
litters
and
to
a
lesser
degree
for
increased
vigor
(
Charlie
Parady
and
Patricia
Mirley,
CRL,
personal
communication).
Following
selection
of
large
litters
at
birth,
litters
were
reduced
to
13
pups
per
litter,
and
vigorous
pups
from
larger
litters
were
selected
at
weaning
as
breeders
for
the
next
generation
of
the
line.
While
pedigrees
were
not
maintained,
selected
individuals
with
different
birth
dates
were
randomly
mated
to
avoid
inbreeding.
CRL
used
this
selection
criteria
and
mating
system
within
each
subpopulation
of
outbred
mice
and
rats.
While
much
of
the
selection
pressure
was
for
large
litter
size,
the
limited
selection
for
vigor
following
rearing
in
large
litters
(
mice
only
have
ten
teats)
may
have
also
resulted
in
some
selection
for
increased
lactational
yield
and
body
weight.

ICR
Swiss
outbred
mice
were
also
rigidly
selected
for
high
productivity
and
13
growth
rate
by
Dr.
TS.
Hauschka
at
the
Institute
for
Cancer
Research
(
www.
taconic.
com/
addinfo/
icrorigin.
htm).
Following
Caesarean
derivation
from
ICR
outbreds
in
1959,
the
CD­
1
mouse
was
again
selected
mainly
for
large
litter
size,
with
some
selection
for
vigor
through
1991
(
Charlie
Parady
and
Patricia
Mirley,
CRL,
personal
communication).
The
result
of
these
selective
breeding
and
random
mating
programs
has
been
large,
vigorous,
highly
prolific,
high
lactating
outbred
mouse
strains
that
have
been
widely
used
as
animal
models
in
biomedical
research.

Results
of
Controlled
Selection
Experiments:

Unfortunately,
unselected
controls
were
not
maintained
during
the
many
generations
of
selection
that
defined
these
commercial
outbred
laboratory
animal
strains.
Nevertheless,
several
selection
experiments
have
been
conducted
by
academic
researchers
that
did
maintain
appropriate,
randomly­
selected
control
lines
or
divergently­
selected
control
lines.
These
selection
experiments
in
outbred
stocks
showed
dramatic
responses
to
selection
for
large
litter
size,
growth
rate,
and
litter
weight
gain
(
Bradford
1968;
Bradford
1969;
Land
and
Falconer
1969;
Eisen
et
al.
1970;
Land
1970;
Eisen
1972;
Land
et
al.
1974;
Eisen
1975;
Bradford
1979;
Eisen
and
Durrant
1980).
Long­
term
selection
for
large
litter
size
results
in
increased
litter
size,
with
limited
effects
on
body
weight
(
Bradford
1968).
In
contrast,
selection
for
increased
growth
rate
increases
growth
rate
and
mature
weight
but
decreases
longevity
(
Eklund
and
Bradford
1977).
Many
of
the
reproductive
endpoints
of
interest,
including
puberty,
litter
size,
and
cyclicity,
are
threshold
traits,
invoking
need
to
consider
threshold
and
scale
effects
(
Falconer
1989).

Selection
for
large
litter
size
increases
ovulation
rate
and
embryo­
fetal
survival
(
Bradford
1968;
Bradford
et
al.
1979;
Durrant
et
al.
1980;
Eisen
and
Durrant
1980;
Spearow
and
Bradford
1983;
Spearow
1985;
Pomp
et
al.
1988).
Selection
for
growth
rate
also
increases
ovulation
rate,
but
through
different
physiological
mechanisms
than
selection
for
large
litter
size
(
Durrant
et
al.
1980;
Eisen
and
Durrant
1980;
Spearow
and
Bradford
1983;
Spearow
1985;
Pomp
et
al.
1988).
While
the
physiological
genetic
mechanisms
by
which
genetic
selection
for
increased
prolificacy
in
rodents
are
not
fully
understood,
they
involve
changes
in
the
regulation
of
the
hypothalamic­
pituitary­
gonadal
axis,
serum
gonadotropin
levels,
and
follicle
populations,
as
well
as
changes
in
sensitivity
to
gonadotropins,
estrogens
and
estrogen
negative
feedback,
the
induction
of
gonadal
LH
receptors,
the
induction
of
gonadal
steroidogenesis,
the
induction
of
follicular
growth,
and
follicular
atresia
(
Bradford
et
al.
1979;
Durrant
et
al.
1980;
Spearow
and
Bradford
1983;
Spearow
1985;
Spearow
1986;
Pomp
et
al.
1988;
Lubritz
et
al.,
1991).
In
essence,
selection
for
increased
litter
size
has
dramatic
effects
on
the
endocrine
mechanisms
regulating
reproductive
endocrine
function
and
development
traits.

Comparison
of
reproductive
endocrine
traits
between
inbred
strains,
congenic
substrains,
and
recombinant
inbred
lines
clearly
show
major
genetic
differences
in
these
traits
between
strains
of
mice,
further
confirming
that
these
reproductive
endocrine
traits
have
a
genetic
basis.
This
includes
evidence
for
significant
to
highly
14
significant
differences
between
strains
of
mice
in:
hormone­
induced
ovulation
rate
(
Spearow
1988a;
Spearow
1988b;
Spearow
and
Barkley
1999),
the
hormonal
induction
of
follicle
maturation,
hormonal
control
of
follicular
atresia
and
follicle
number
(
Spearow
et
al.,
1991),
ovarian
steroidogenesis
and
aromatase
activity,
estrogen­
induced
immature
uterotrophic
weight
(
Griffith
et
al.
1997;
Roper
et
al.
1999),
estrogen­
induced
uterine
eosinophil
and
macrophage
numbers
(
Griffith
et
al.
1997;
Roper
et
al.
1999),
and
estrogen­
induced
susceptibility
to
vaginal
candida
infection
(
Parmar
et
al.
2003).
Strains
of
inbred
mice
and
rats
also
differ
in
normative
testes
and
seminal
vesicle
weights
(
Zidek
et
al.,
1998;
Zidek
et
al.,
1999).

SD
and
ACI
strain
rats
also
differ
dramatically
in
the
incidence
of
mammary
cancers
in
response
to
DES
(
Shellabarger
et
al.
1978),
as
reviewed
by
Festing
(
1987).
Whereas
DES
failed
to
cause
mammary
cancer
in
any
SD
rat
(
0/
33),
DES
increased
the
incidence
of
mammary
cancers
from
0%
to
53%
in
ACI
strain
rats.
In
contrast,
in
response
to
atrazine,
female
SD
rats
developed
mammary
cancer
while
F­
344
rats
did
not,
apparently
due
to
a
disruption
of
ovarian
function
leading
to
persistent
estrus
and
increased
E2
in
SD
but
not
in
F­
344
rats
(
Wetzel
et
al.
1994;
Stevens
et
al.
1999).

Additional
evidence
for
genetic
variation
in
reproductive
endocrine
function
has
also
been
demonstrated
in
Quantitative
Trait
Loci
(
QTL)
linkage
studies
that
have
mapped
genes
or
QTL
controlling
reproductive
endocrine
traits
to
specific
chromosomal
regions.
Genes
controlling
the
increased
natural
ovulation
rate
of
large­
litter
size
selected
Quackenbush­
Swiss
strain
mice
over
that
of
C57BL/
6J
strain
mice
map
to
regions
of
chromosome
(
Chr)
2
and
4
(
Kirkpatrick
et
al.,
1998).
These
regions
of
Chr
2
and
4
overlap
with
loci
controling
major
strain
differences
in
hormone­
induced
ovulation
rate
and
ovarian
steroidogenesis.
Markers
on
Chr
13
are
significantly
associated
with
strain
differences
in
testes
weight
in
the
mouse
BXD
recombinant
inbred
strain
set
(
Zidek
et
al.,
1998).

Several
genes
controling
E2­
induced
uterine
hypertrophy
and/
or
E2­
induced
uterine
leukocyte
responses
have
also
been
mapped
to
specific
chromosomal
regions
(
Roper
et
al.
1999).
An
interacting
genetic
factor
on
Chr
10
controls
E2­
induced
uterine
weight
as
well
as
E2­
induced
leukocyte
responses
(
Roper
et
al.
1999).
These
and
other
studies
in
genetically­
defined
inbred
strains
of
mice
clearly
confirm
that
differences
in
reproductive
endocrine
traits
and
sensitivity/
susceptibility
to
estrogenic
agents
found
among
selected
strains
have
a
genetic
basis.

Due
to
genetic
variation
in
reproductive
development
and
function,
selection
can
also
have
a
major
effect
on
reproductive
function
in
other
mammalian
species.
Lines
of
sheep
selected
for
large
litter
size
increased
the
frequency
of
alleles
with
major
effects
on
reproductive
development,
function,
and
ovulation
rate
(
Bindon
1984;
McNatty
et
al.
1985;
McNatty
et
al.,
1995).
QTL
linkage
mapping
and
positional
cloning
studies
showed
that
a
Chr
6
mutation
(
FecB)
in
the
intracellular
kinase
signaling
domain
of
bone
morphogenetic
protein
IB
(
BMP­
IB)
receptor
(
also
known
as
ALK­
6),
which
binds
members
of
the
transforming
growth
factor­
beta
(
TGF­
beta)
superfamily,
has
a
major
15
effect
on
reproductive
function,
development,
ovulation
rate,
and
litter
size
(
Wilson
et
al.
2001).

Sources
of
Variation
in
Traits:

The
phenotype
of
an
individual
can
be
considered
as
the
sum
of
its
genotypic
value
(
G),
the
environmental
effects
(
E),
and
the
genotype
x
environment
interaction,
i.
e.,
phenotype
=
G
+
E
+
GxE
interaction
(
Falconer
1989;
Lynch
and
Walsh
1998).
Geneticists
normally
consider
each
of
these
factors
as
variance
components
in
analyzing
trait
phenotypes.
Potential
endocrine­
disrupting
chemicals
represent
environmental
or
nongenetic
sources
of
variation
affecting
a
given
trait.
The
purpose
of
the
EDSP
is
to
determine
if
a
given
chemical,
i.
e.,
environmental
factor,
has
detrimental
effects
on
reproductive
development
and
functional
phenotypes.
However,
it
is
not
just
the
occurrence
of
genetic
differences
in
reproductive
phenotypes,
but
also
the
potential
for
genotypes
to
interact
in
a
nonparallel
manner
with
environmental
factors
that
is
of
concern
in
designing
EDSP
assays.
If
G
and
GxE
interactions
are
not
important,
any
strain
of
animals
could
be
used
for
testing
chemicals
for
endocrine
disruptor
activity.
However,
if
the
genetic
variance
and
especially
if
the
genetic
x
environmental
variance
are
significant,
care
must
be
taken
to
avoid
screening
chemicals
with
resistant
strains,
since
the
effects
on
sensitive
animal
strains
would
be
underestimated
(
Narotsky
et
al.
2001;
Spearow
and
Barkley
2001).
The
ultimate
question
is,
of
course,
the
effects,
if
any,
on
humans
and
other
species
of
concern.
The
best
approach
would
be
the
use
of
the
most
relevant
animal
model,
if
known;
the
default
approach
is
the
use
of
the
most
sensitive
animal
model.

The
genetic
variance
in
a
trait
can
be
further
considered
broken
down
into
its
components,
including
the
additive
genetic
variance,
dominance
genetic
variance,
and
the
epistasis
genetic
variance,
e.
g.,
Total
Genetic
=
Additive
+
Dominance
+
Epistasis
(
Falconer
1989;
Lynch
and
Walsh
1998).
In
biological
terms,
the
additive
genetic
variance
is
the
variation
in
the
average
additive
effects
of
alleles.
Alleles
at
additiveacting
loci
behave
in
an
step­
wise
or
additive
manner,
with
each
+
allele
increasing
the
phenotype
in
an
additive
manner.
Conventional
mass
selection
for
a
trait
acts
mainly
on
additive
genetic
variance
to
increase
the
frequency
of
alleles,
which
on
average
have
a
beneficial
effect
on
the
trait(
s)
under
selection.
An
animal's
breeding
value
is
the
sum
of
the
additive
effects
of
its
genes.
Dominance
is
defined
as
a
nonadditive
interaction
between
alleles
at
a
given
locus.
While
loci
that
show
dominance
for
a
trait
can
also
have
an
additive
genetic
component,
they
show
a
deviation,
e.
g.
dominance
deviation,
from
the
regression
line
between
the
phenotypic
means
of
the
low
homozygote
and
the
high
homozygote.
For
example,
consider
a
trait
showing
complete
dominance
where
the
phenotype
of
individuals
that
are
aa
=
0,
Aa=
2
and
AA=
2.
While
the
average
of
the
two
homozygotes
=
1,
the
heterozygote
(
Aa)
has
a
mean
of
2,
and
thus
shows
a
dominance
deviation.
Epistasis
describes
the
nonadditivity
of
effects
between
loci.

The
effects
of
selection,
inbreeding,
and
crossing
are
very
different
on
traits
controlled
by
additive
versus
dominance
genetic
variation.
Selection
can
utilize
additive
16
genetic
variance
but
not
the
dominance
genetic
variance
in
a
population
to
improve
a
trait.
Traits
controlled
by
additively
acting
loci
do
not
show
inbreeding
depression,
since
the
number
of
loci
proportional
to
the
initial
gene
frequency
will
fix
for
the
­
versus
the
+
acting
alleles.
In
contrast,
loci
showing
dominance
generally
decline
with
inbreeding
due
to
the
fixation
of
less
desirable
recessive
alleles.
Finally,
F1
crosses
generally
show
heterosis
or
hybrid
vigor
(
e.
g.,
increased
phenotypic
variance)
in
traits
showing
dominant
gene
action
but
not
in
traits
controlled
by
additive
gene
action
unless
there
is
also
complementarity
among
component
traits.

Relative
Importance
of
Historic
Inbreeding
Versus
Selection
for
High
Prolificacy
in
Defining
Currently
Available
Strains:

In
full
sib
(
brother­
sister)
mating
programs
during
the
development
of
inbred
strains,
the
largest
evolutionary
force
is
genetic
drift
and
random
fixation
of
alleles.
This
is
especially
true
of
reproductive
traits
with
medium
to
low
heritability.
There
will
be
an
inbreeding
depression
for
phenotypes
controlled
by
dominantly
acting
loci
and
perhaps
for
some
loci
controlled
by
dominant
epistatic
interactions
(
Lynch
and
Walsh,
1998).
In
contrast,
on
average,
inbreeding
without
selection
will
not
change
the
phenotypes
controlled
by
additively
acting
genes
since
as
many
+
as
­
acting
alleles
are
likely
to
fix
in
a
given
inbred
strain.
This
is
particularly
true
for
traits
controlled
by
a
large
number
of
additive
loci.
Nevertheless,
for
traits
controlled
by
a
very
small
number
of
additive
loci,
there
is
likely
substantial
genetic
drift
in
the
trait,
depending
on
whether
inbreeding
fixes
a
given
inbred
strain
for
a
+
or
­
acting
allele.
Without
selection,
the
expectation
is
for
no
net
change
in
the
number
of
+
or
­
acting
alleles
affecting
a
trait.

In
contrast,
long­
term
selection
mainly
for
a
single
trait
will
dramatically
increase
the
frequency
of
+
alleles,
even
for
medium
to
low
heritability
traits
like
litter
size
(
Bradford
1968;
Eklund
and
Bradford
1977).
Furthermore,
since
commercial
outbred
populations
were
selected
for
large
litter
size
in
large
populations,
it
is
likely
that
any
beneficial
mutations
which
occurred
in
the
population
would
also
be
utilized
to
increase
prolificacy
even
further.

Correlated
Trait
Responses:

Correlated
traits
are
generally
due
to
the
action
of
genes,
with
pleiotropic
effects
on
several
traits
or
physiological
processes
(
Falconer
1989).
Comparison
of
strains
selected
for
high
fecundity
with
randomly
selected
control
strains
revealed
correlated
trait
responses
in
several
male
characters
for
reproductive
development
function.
In
addition
to
changing
reproductive
function
in
females,
selection
for
large
litter
size
also
increases
the
weights
of
testes,
epididymides,
and
seminal
vesicles
(
Eisen
and
Johnson
1981;
Spearow
et
al.
1999;
Spearow
et
al.
2001).
Strains
of
mice
and
rats
selected
for
large
litter
size
are
more
resistant
than
unselected
strains
to
disruption
of
testes
weight,
accessory
gland
weights,
spermatogenesis,
and
steroidogenesis
by
estradiol
or
DES
(
Spearow
et
al.
1987;
Inano
et
al.
1996;
Spearow
et
al.
1999;
Spearow
et
al.
2001).
These
observations
suggest
that
one
of
the
correlated
responses
to
17
selection
for
large
litter
size
is
resistance
to
endocrine
disruption
by
estrogenic
agents
(
Spearow
et
al.
1999;
Spearow
and
Barkley
2001;
Spearow
et
al.
2001).

Some
of
the
genes
with
pleiotropic
effects
on
related
reproductive
traits
have
been
mapped,
suggesting
potential
genetic
mechanisms
mediating
correlated
trait
responses.
Markers
on
Chr
8
showed
a
significant
association
with
seminal
vesicle
mass
and
a
suggestive
association
with
litter
size
in
HXB
and
BXH
recombinant
inbred
strain
sets
derived
from
SHR
and
Brown
Norway
(
BN)
rat
strains
(
Zidek
et
al.
1999).
Since
litter
size
was
significantly
associated
with
seminal
vesicle
mass,
these
data
suggest
that
both
of
these
traits
are
under
the
control
of
the
same
QTL
or
tightly
linked
QTL
on
rat
Chr
8
(
Zidek
et
al.,
1999).
Thus,
this
rat
Chr
8
QTL
may
have
pleiotropic
effects
on
litter
size
and
seminal
vesicle
mass,
explaining,
at
least
in
part,
how
selection
for
large
litter
size
also
affects
male
reproductive
developmental
traits.

Considerable
alarm
was
raised
by
the
toxicology
community
when
it
was
noted
that
certain
outbred
stocks,
including
CD
rats,
were
showing
excessive
litter
size,
increased
mature
body
weights,
and
decreased
longevity
(
Pettersen
et
al.
1996)
(
CRL
reference
paper
Vol
11,
#
1,
1999).
When
raised
in
the
same
environment,
CRL:
CD
males
were
found
to
be
significantly
heavier
than
Taconic:
SD
males
which
were
heavier
than
Hsd:
SD
males
(
Klinger
et
al.,
1996).
While
it
is
unknown
whether
the
increased
body
weights
of
CD
rats
are
a
correlated
response
to
selection
for
vigor
or
unintended
selection
for
increased
body
weight,
the
study
of
Klinger
et
al.
(
1996)
suggests
the
substrain
differences
are
genetic.
Increased
body
weight
is
correlated
with
decreased
longevity
(
Eklund
and
Bradford
1977),
and
a
higher
proportion
of
CD
rats
failed
to
survive
to
the
age
required
in
two­
year
carcinogenicity
studies
(
Petterson
et
al.,
1996).
Thus,
large
commercial
suppliers
such
as
CRL
have
reconsidered
and
changed
their
long­
term
selection
criteria
and
mating
system
(
CRL
reference
paper
Vol
11,
#
1,
1999).

Genetic
Standardization
of
Genetic
Variability
in
Outbred
Strains:

Due
to
the
observed
increased
litter
size
and
body
weight,
as
well
as
decreased
longevity,
CRL
has
recently
initiated
an
effort
to
"
standardize"
certain
outbred
strains
such
as
the
CD
®
(
SD)
IGS
BR
rat,
Wistar
Han
IGS
rat,
and
CD­
1
(
CD­
1
®
(
ICR)
BR)
mouse.
For
example,
in
the
early
1990s,
100
pairs
of
CD
rat
breeders
were
selected
from
each
of
eight
diverse
CRL
CD
rat
colonies
world
wide
and
rederived
in
isolators
to
form
the
CD
IGS
reference
foundation
colony
in
Wilmington,
MA.
Selection
criteria
were
relaxed,
and
this
CD
IGS
(
international
genetic
standard)
rat
reference
population
was
then
managed
with
procedures
to
minimize
genetic
drift
to
establish
each
CD
IGS
colony
world
wide.
CRL
plans
to
further
minimize
genetic
drift
by
migrating
additional
breeders
in
both
directions
between
the
IGS
reference
population
and
isolated
colonies
over
time
(
CRL
reference
paper
Vol
11,
#
1,
1999).
By
also
using
a
pedigreed
mating
system
designed
to
minimize
inbreeding
and
by
improving
genetic
quality
control
monitoring,
CRL
is
anticipated
to
dramatically
reduce
genetic
drift
and
variation
between
outbred
IGS
colonies.
18
Genetic
variation
within
"
narrow
genetic
base
outbred
strains":
While
CRL
refers
to
this
as
the
International
Genetic
Standard
(
IGS)
program
and
has
the
aim
of
minimizing
genetic
drift
and
producing
a
"
standardized"
outbred
rat,
it
is
the
variability
of
the
population
that
is
standardized,
not
the
individual
Crl:
CD
®
(
SD)
IGSBR
rat
(
abbreviated
as
CD
IGS).
While
much
less
diverse
than
outbreds
of
essentially
any
other
mammalian
species
that
have
not
undergone
an
extreme
genetic
bottleneck
followed
by
long­
term
selection
mainly
for
increased
prolificacy,
the
SD
strain
and
the
SD
derived
CD
IGS
rat
strain
are
segregating
at
many
loci.
Even
with
a
narrow
genetic
base,
it
is
impossible
to
predict
or
repeat
the
genotype
of
any
given
CD
IGS
strain,
or
for
that
matter
any
other
outbred
strain.
Selection
for
increased
prolificacy
is
likely
to
have
also
increased
the
frequency,
or
fixed
alleles
conferring
resistance
to
endocrine
disruption
by
some
hormonally
active
agents,
but
the
genes
conferring
genetic
susceptibility
to
diverse
hormonally­
active
compounds
have
not
been
identified.
Thus,
the
use
of
such
outbred
animals
makes
it
impossible
to
replicate
the
susceptibility
genotypes
and
therefore
the
conditions
used
for
testing
any
toxicant
x
dose
combination
in
the
EDSP.
This
makes
replication
of
experiments
involving
outbred
animal
models
problematic,
regardless
of
whether
replicates
are
conducted
by
the
same
or
different
laboratories.
Such
unpredictable
genetic
variation
within
even
narrow
genetic
base
outbred
strains
will
greatly
complicate
and
limit
efforts
to
use
conventional
reproductive
toxicological,
genomic
bioinformatic,
microarray,
and
proteomic
approaches
to
identify
chemicals
with
endocrine
disrupting
activities,
as
well
as
future
efforts
to
characterize
their
mechanisms
of
action
and
loci
controlling
genetic
susceptibility.
Furthermore,
the
within
strain
genetic
variation
common
to
outbred
strains
is
also
likely
to
be
a
major
component
of
"
between
litter"
effects.

While
the
extremely
narrow
genetic
background,
early
inbreeding,
and
long­
term
history
of
selection,
mainly
for
high
prolificacy
and
vigor,
has
resulted
in
highly
robust
and
productive
CD
IGS
rats,
it
is
highly
doubtful
these
animals
are
representative
of
any
natural
mammalian
outbred
population.
The
fact
that
Robert
Dawley
backcrossed
daughters
of
a
hybrid
male
back
to
the
same
male
for
seven
consecutive
generations
indicates
that
well
over
90%
of
the
gene
pool
(
an
estimated
99.6
%
of
the
genes
minus
deleterious
alleles
purged
by
selection)
of
the
SD
strain
came
from
the
single
foundation
male
rat.
Thus,
the
SD
strain
and
any
substrains
derived
from
this
closed
population
represent
an
exceedingly
narrow
genetic
base.
The
seven
consecutive
generations
of
backcrossing
daughters
to
the
same
foundation
male
rat
is
also
likely
to
have
purged
most
deleterious
recessive
genes,
including
those
affecting
reproductive
development
and
function.
While
crossing
inbred
sire­
daughter
lines
generated
from
the
single
foundation
male
is
likely
to
have
provided
some
heterozygosity
at
certain
loci
in
the
SD
population,
most
of
the
genetic
diversity
in
this
strain
had
to
come
from
the
single
hybrid
foundation
male
that
was
backcrossed
for
seven
consecutive
generations
to
his
daughters.

Just
as
important,
the
IGS
program
does
very
little
to
eliminate
the
changes
in
gene
frequency
affecting
reproductive
and
correlated
traits
brought
about
by
over
80
generations
of
selection
for
increased
litter
size.
Since
selection
of
the
CD
rat
was
conducted
in
large,
narrow
genetic
base
outbred
populations,
any
additional
mutations
19
improving
litter
size
would
also
be
selected
for.
It
has
been
argued
that
outbred
laboratory
animal
populations
show
phenotypes
more
representative
of
wild
outbred
populations
and
more
phenotypic
diversity
than
inbred
strains.
This
may
be
true
for
unselected
traits
in
a
fully
outbred
laboratory
animal
strain,
but
it
is
clearly
not
true
regarding
traits
(
and
their
correlated
traits)
for
which
outbreds
have
undergone
long­
term
selection
(
Eklund
and
Bradford
1977).
Many
long­
term
selected
outbred
strains
show
phenotypic
means
and
distribution
for
traits
under
selection
and
their
correlated
traits,
which
are
well
beyond
the
normal
distribution
of
values
from
an
unselected
control
population.
While
long­
term
selection
for
large
litter
size
increases
the
mean
litter
size,
it
also
decreases
the
additive
genetic
variance
for
this
trait
(
Eklund
and
Bradford
1977)
and
is
thus
likely
to
decrease
the
additive
genetic
variance
in
traits
correlated
with
high
fecundity.
Since
selection
for
large
litter
size
was
relaxed
since
the
formation
of
the
CD
IGS
rat
population,
this
population
may
be
restored
to
its
original
litter
sizes,
but
not
necessarily
the
original
genotype.
However,
if
the
restoration
of
original
litter
size
resulted
from
changing
the
genes
which
control
litter
size,
these
genes
may
also
control
response
of
other
endocrine­
sensitive
endpoints
(
i.
e.
a
pleiotropic
effect,
whereby
a
single
gene
controls
a
number
of
parameters/
responses).
Therefore
recovery
of
original
litter
size
may
also
change
the
sensitivity
of
the
strain
to
the
pleitropically­
related
endpoints
(
e.
g.
FSH
or
LH
levels,
number
of
eggs
ovulated,
responsivity
to
E2,
or
estrogen­
like
compounds,
etc.)
back
to
where
it
was,
whether
or
not
the
original
genotype
was
recovered.

Data
from
Spearow's
laboratory
show
dramatic
differences
in
susceptibility
to
endocrine
disruption
by
estrogenic
agents
between
strains
of
mice,
and
that
CD­
1
strain
mice
selected
for
high
fecundity
are
highly
resistant
to
E2.
This
includes
approximately
16­
to
100­
fold
differences
in
sensitivity
between
strains
of
mice
in
susceptibility
to
the
disruption
of
testes
weight,
spermatogenesis,
epididymal
sperm
counts,
testicular
sulfotransferase
activity,
and
gestational
fetal
losses
(
Spearow
et
al.
1999;
Spearow
et
al.
2001).
Data
from
other
laboratories
also
show
that
the
SD
rat
and
the
SD­
derived
CD
rat
are
less
sensitive
than
other
strains
in
susceptibility
to
estrogenic
agents,
including
estrogen
and
the
xenoestrogen
Bisphenol
A
(
BPA)
(
Steinmetz
et
al.
1997;
Steinmetz
et
al.
1998;
Long
et
al.
2000).
SD
rats
are
also
much
more
resistant
than
Wistar/
MS
or
Fisher
344
rats
to
the
reduction
in
testis
and
seminal
vesicle
weights
by
DES
(
Inano
et
al.
1996),
and
are
much
more
resistant
than
several
other
strains,
including
F344,
to
estrogen­
and
DES­
induced
pituitary
tumors
(
Gregg
et
al.
1996;
Wendell
et
al.
2000).
Thus,
the
fact
that
outbred
strains
such
as
the
SD­
derived
CD
rat
and
the
Swiss
and
ICR
derived
CD­
1
mouse
have
been
previously
long­
term
selected
for
increased
fecundity
and
vigor
is
of
special
concern
in
EDSP
assays
due
to
the
correlated
trait
response
of
increased
resistance
to
endocrine
disruption
by
estrogenic
agents.

However,
resistance
or
sensitivity
to
endocrine
disruptors
is
not
uniform
across
test
chemicals
and
endpoints.
For
example,
SD
rats
are
more
sensitive
than
F344
rats
to
the
uterotrophic
effects,
including
increased
uterine
weight
and
epithelial
cell
height
effects
of
endocrine­
active
compounds
such
as
tamoxifen
(
Bailey
and
Nephew
2002).
There
are
many
other
examples
of
strain
differences
in
endpoint­
and
test
substance­
20
specific
responses
to
endocrine­
active
compounds
(
see
Tables
2
and
3
for
a
range
of
responses
in
different
rat
strains).

Even
though
the
"
outbred"
strains
come
from
a
relatively
narrow
genetic
base
and
may
not
represent
all
sensitive
rats
of
the
world,
there
is
a
wider
range
of
responses
in
outbred
strains,
due
to
genetic
heterogeneity,
than
in
inbred
strains.
The
selection
of
an
outbred
versus
an
inbred
strain
for
use
in
these
assays
depends
on
whether
one
can
select
the
most
sensitive
inbred
strain
for
an
assay
with
all
the
confounders
discussed
above
and
below,
or
whether
the
broader
response
of
an
outbred
strain
provides
a
significant
advantage
at
the
sensitive
end
of
the
curve.

Genetic
Variability
in
Toxicological
Assays:

Since
all
members
of
a
given
isogenic
strain
(
inbred
strains
and
F1
hybrids)
are
essentially
genetically
identical,
the
dramatically
reduced
genetic
variability
in
isogenic
strains
enhances
the
reproducibility
and
comparability
of
data
generated
in
these
stocks
(
Festing
1979;
Festing
1993).
The
crucial
characteristic
common
to
both
inbred
and
F1
hybrid
strains
is
isogenicity,
i.
e.,
the
fact
that
all
individuals
of
an
authentic
strain
are
genotypically
the
same
and
therefore
phenotypically
more
uniform
than
individuals
of
outbred
stocks
(
http://
www.
criver.
com/
techdocs/
rodent1.
html).
Isogenicity
of
such
inbred
stains
and
F1
crosses
leads
not
only
to
a
much
greater
genetic
and
phenotypic
uniformity
but
also
a
high,
long­
term
genetic
stability.
Thus,
it
has
been
suggested
(
Festing
1995)
that
toxicologists
should
treat
genetics
like
every
other
variable
and
control
it
by
utilizing
isogenic
strains
(
F1
hybrids
heterozygous
at
every
locus).
However,
in
any
study
requiring
generation
of
offspring,
such
as
the
two­
generation
reproductive
toxicity
study,
the
advantages
of
utilizing
isogenic
strains
is
lost
since
use
of
F1
parents
will
produce
F2
offspring
which
are
segregating
at
many
loci
with
differing
genotypes
and
phenotypes.

The
view
is
held
by
some
researchers
(
Spearow
et
al.
1999;
Spearow
et
al.
2001)
that
commercial
outbred
strains
are
resistant
to
endocrine
disruption
by
estrogenic
agents
at
some
endpoints,
most
likely
as
a
correlated
response
to
long­
term
selection
for
high
prolificacy.
However,
toxicity
testing
in
outbreds
represents
testing
the
effects
of
toxicants
on
a
sampling
of
outbred
strain
genotypes.
One
argument
is
that
the
segregation
of
any
genes
in
outbred
strains
controlling
traits
that
have
not
been
fixed
by
early
inbreeding
or
long­
term
selection
will
result
in
genetic
noise
and
increased
phenotypic
variability.
Since
toxicity
testing
usually
involves
the
calculation
of
dose­
response
curves,
the
use
of
phenotypically
variable
nonisogenic
stocks
reduces
the
precision
with
which
such
curves
can
be
estimated
and
therefore
the
sensitivity
of
the
assay
(
Festing,
1979).
The
counterargument
is
that
the
confidence
interval
for
an
outbred
strain
is
wider
due
to
the
variability,
so
that
use
of
an
outbred
strain
with
greater
variability
and
therefore
less
precision,
would
be
better
because
of
its
variability,
than
using
a
very
precise
isogenic
strain
if
it
were
not
the
most
sensitive
one
to
the
specific
chemical
or
for
the
specific
endpoints.

Multiple
Strain
Assays:
21
There
is
a
risk
in
using
a
single
strain
in
toxicological
safety
testing,
particularly
when
that
strain
is
known
to
be
highly
resistant
to
one
or
more
classes
of
chemicals
and/
or
endpoints
to
be
tested
in
the
EDSP.
Any
time
there
is
considerable
genetic
variation
in
a
susceptibility
trait,
a
single
isogenic
or
outbred
strain
may
be
resistant
to
the
compound
being
tested
or
the
endpoint
being
assessed.
If
so,
a
chemical
that
is
toxic
to
other
genotypes
may
be
judged
to
be
relatively
safe
(
Festing
1993).
As
Narotsky
et
al.
(
2001)
concluded,
"
Thus,
routine
toxicity
tests
that
use
only
a
single
strain
may
be
unreliable
since
the
outcome
may
hinge
on
choice
of
strain."
This
is
particularly
true
for
traits
showing
major
strain
differences
in
susceptibility
to
endocrine
disruption
by
estrogenic
agents.
As
an
alternative,
Michael
Festing
has
pointed
out
for
two
decades
the
benefits
of
toxicological
testing
with
multiple
divergent
isogenic
strains,
rather
than
a
single
strain,
to
better
ensure
that
all
the
test
animals
are
not
resistant
(
Festing
1979;
Festing
1987;
Festing
1993;
Festing
1995),
by
the
use
of
several
isogenic
strains
from
diverse
genetic
backgrounds
in
a
factorial
experimental
design
to
overcome
the
problem
of
testing
animals,
all
of
which
are
genetically
resistant
to
the
compound
to
be
tested
(
Festing
1995;
Festing
et
al.
2001;
Festing
and
Altman
2002).

An
advantage
of
testing
with
multiple
strains
is
that
identification
of
strain
differences
enables
an
additional
resource
to
determine
the
mechanisms
of
toxicity
(
Narotsky
et
al.
2001).
Once
parental
strains
are
shown
to
differ,
congenic
inbred,
consomic
inbred,
and
recombinant
inbred
strains
can
be
compared
and
strain
distribution
profile
phenotypes
used
to
map
and
characterize
genes
controling
susceptibility.
Congenic
inbred,
consomic
inbred,
and
recombinant
inbred
strains
are
highly
reproducible
strains
with
defined
genotypes
(
Silver
1995).
Once
a
strain
set
has
been
genotyped
at
molecular
markers
along
each
chromosome,
genes
controling
traits
differing
between
parental
strains
can
be
mapped
to
specific
chromosomal
regions,
following
scoring
biochemical
or
physiological
phenotypes
of
the
set
of
strains
(
Matin
et
al.
1999;
Cowley
et
al.
2001;
Liang
et
al.
2002)
with
appropriate
consomic
strains,
if
available.
Such
congenic
inbred,
consomic
inbred,
and
recombinant
inbred
strains
resources,
along
with
available
gene
mapping
software,
also
enable
simple
as
well
as
complex
multigenetic
physiology,
disease,
and
toxic
susceptibility
traits
to
be
broken
down
to
allow
the
identification
of
individual
susceptibility
genes
(
Manly
et
al.
2001;
Williams
et
al.
2001).
Thus,
the
use
of
several
highly
divergent,
genetically­
defined
inbred
parental
strains
in
endocrine
disruptor
assays
could
greatly
enhance
the
possibility
of
identifying
genes
controlling
susceptibility
to
endocrine
disruption.

2.3.2
Species
Selection
for
Endocrine
Disruption
Assays
and
Genetic
Variability
Variability
across
strains,
in
both
rats
and
mice
in
reproductive
parameters,
must
be
considered
in
the
selection
of
the
appropriate
strains/
species.
Historically,
the
most
common
strains/
species
selected
for
assays
of
endocrine
disruption
have
been
the
SD
rat
and
CD­
1
mouse.
Since
the
objective
of
the
EDSP
study
is
to
examine
the
effects
of
a
multitude
of
chemicals
on
reproductive
developmental
structural
and
functional
toxicity,
strain
variation
in
developmental
rates,
as
well
as
other
biochemical
endocrine
and
signal
transduction
mechanisms,
may
be
important
in
considering
the
range
of
22
susceptibilities
to
reproductive
and
developmental
toxicity,
as
are
evidenced
by
effects
on
endocrine
endpoints.

In
selecting
the
appropriate
species
for
EDSP
assays,
the
rat
has
an
advantage
over
the
mouse
due
to
larger
size,
which
allows
easier
analysis
of
serum
hormone
concentrations
and
certain
other
physiological
endpoints.
While
considerably
more
genetic
resources
are
available
in
the
mouse,
progress
is
currently
being
made
in
sequencing
the
genome
of
Brown
Norway
rat
strain.
If
recombinant
inbred,
congenic,
and/
or
chromosome­
specific
consomic
strains
derived
from
the
parental
inbred
strains
to
be
used
in
the
EDSP
were
available,
the
utility
of
the
rat
for
such
studies
would
be
further
improved.

At
present,
one
of
the
biggest
problems
with
species
or
strain
selection
is
that
only
a
small
number
of
studies
have
examined
genetic
differences
in
susceptibility
to
endocrine
disruption.
Since
most
reproductive
toxicology
studies
involve
only
a
single
stock
of
laboratory
animals,
we
do
not
know
whether
the
response
to
a
given
xenobiotic
is
under
genetic
control
(
Festing,
1987).
Furthermore,
an
even
smaller
number
of
studies
have
compared
genetically­
defined
isogenic
strains.

The
fecundity
of
a
strain
limits
the
number
of
offspring
available
following
gestational
or
lacational
exposures.
Strains
with
low
fecundity
are
not
recommended
in
OPPTS
reproductive
and
developmental
test
guidelines.
Since
SD
rats
have
been
bred
for
high
fecundity
and
have
the
largest
historical
database,
they
have
been
used
the
most
frequently
for
regulatory
reproductive
toxicity
studies.
Nevertheless,
provided
the
supplier
maintains
enough
breeders
to
provide
the
animal
needs
of
the
EDSP,
strains
with
moderate
fecundity
might
not
limit
the
number
of
animals
available
from
the
supplier
for
conducting
pubertal
or
adult
exposures.
Furthermore,
even
with
gestational
and
lactational
exposures,
a
strain
with
moderate
fecundity
would
not
limit
an
EDSP
assay
under
current
guidelines
of
retaining
and
examining
one
individual
of
each
sex
per
litter
at
adulthood.

Another
consideration
of
species
and
strain
selection
relates
to
the
maintenance
of
early
pregnancy
in
many
species,
which
is
dependent
on
the
gonadotropins
maintaining
corpus
luteum
(
CL)
progesterone
production.
In
humans,
the
maintenance
of
pregnancy
is
dependent
on
LH
modulation
of
CL
progesterone
production
prior
to
implantation,
and
hCG
modulating
CL
progesterone
production
following
implantation
(
see
discussion
of
Narotsky
et
al
2001
Section
2.5.2).
Human
epidemiological
evidence
indicates
that
exposures
to
low­
dose
bromodichloromethane
(
BDCM,
a
common
drinking
water
disinfection
by­
product)
are
associated
with
an
increased
incidence
of
pregnancy
loss
in
humans
(
Waller
et
al.
1998).
Following
exposure
to
75
mg/
kg
BDCM
on
days
6­
10
of
gestation,
62%
of
Hsd:
F344
litters
showed
full
litter
resorption
(
Bielmeier
et
al.
2001).
In
contrast,
0%
of
the
Hsd:
SD
litters
showed
full
litter
resorption
in
response
to
75
or
100
mg/
Kg
BDCM
(
Bielmeier
et
al.
2001).
Thus,
sensitive
strains
such
as
the
F344
might
be
more
appropriate
to
estimate
human
risk
of
BDCM­
induced
abortion.
While
utilizing
strains
with
low
fecundity
may
reduce
the
number
of
animals
available
for
the
multitude
of
assessments
required
in
reproductive
toxicity
testing
23
guidelines,
use
of
a
potentially
resistant
strain
of
animals
may
underestimate
the
risk
to
sensitive
genotypes.

A
discussion
of
assays
under
consideration
for
testing
guidelines,
a
detailed
comparison
of
strain
differences
in
endocrine
endpoints,
and
sensitivity
of
these
endpoints
to
endocrine­
disrupting
chemicals
which
follow
provides
a
summary
of
potential
problems
with
species/
strain
selection
in
reproductive
toxicity
studies.
Specific
examples
of
differential
sensitivity
of
endocrine
endpoints
in
different
species
and
strains
to
many
endocrine­
disrupting
chemicals
will
be
discussed
in
Section
3.0.

2.3.3
Confounders
Affecting
Comparisons
of
Reproductive
Toxicity
Data
The
focus
in
this
White
paper
is
on
intra­
species
and
inter­
species
comparisons
of
responses
to
EACs.
The
critical
papers
are
those
in
which
the
same
laboratory
evaluates
two
or
more
different
strains/
stocks
(
intra­
species)
or
the
two
or
more
different
species
at
the
same
time
under
the
same
laboratory
conditions.
Comparisons
of
intra­
and
inter­
species
differences
in
response
to
EACs
performed
at
different
times,
in
different
laboratories
under
different
laboratory
conditions,
are
more
difficult
to
interpret
due
to
the
following
likely
confounders
to
the
determination
that
the
strains/
stocks/
species,
per
se,
differ
in
their
responses:

Same
laboratory,
different
times.
Source
of
the
animals,
genetic
drift,
age
of
animals,
status
of
the
plastic
cages
and/
or
water
bottles
(
new
versus
damaged),
change
in
feed
lot,
bedding
lot
numbers,
water
supply,
change
in
technical
staff.

Different
laboratories
A.
Animals
°
Source/
supplier
(
local
closed
colony,
national/
international
commercial
supplier,
change
in
location
from
commercial
supplier,
etc.;
the
same
strain
from
different
suppliers,
will
most
likely
be
genetically
different)
°
Age,
weight,
and
health
status
B.
Husbandry
°
Housing:
Group
housing
versus
single
housing
impacts
on
many
endpoints
in
both
sexes
for
rats
and
mice.
When
male
mice
are
group
housed,
a
"
dominance
hierarchy"
is
established,
with
one
dominant,
aggressive
male
and
the
remaining
males
as
subordinates
(
Haemisch
et
al.
1994;
McKinney
and
Pasley,
1973;
Van
Loo
et
al.
2001).
The
subordinates
exhibit
reduced
circulating
testosterone
levels,
testes
weight,
epididymides,
accessory
sex
organs,
epididymal
sperm
numbers
and
motility
(
Koyama
and
Kamimura,
1999;
2000),
and
testicular
spermatid
head
counts.
The
subordinates
also
exhibit
signs
of
increased
stress
such
as
increased
circulating
stress
hormones
and
adrenal
gland
weights,
24
altered
nervous
system
and
neuroendocrine
functions
(
D'Arbe
et
al.
2002;
Karolewicz
and
Paul,
2001),
altered
immune
competence,
and
increased
incidence
of
tumors
(
Bartolomucci
et
al.
2001;
Grimm
et
al.
1996;
Haseman
et
al.
1994).
In
many
cases
(
incidence
varies
by
mouse
strain),
the
dominant
male
"
barbers"
the
subordinates
by
removing
their
whiskers
and
large
patches
of
fur
(
Strozik
and
Festing,
1981;
Reinhardt
and
Militzer,
1979;
Sarna
et
al.
2000;
Long,
1972).
Even
phenotypic
variance
is
affected
by
housing
in
C57BL/
6J
mice
(
Nagy
et
al.
2002).
Therefore,
a
major
source
of
differences
among
study
results
in
different
laboratories
may
be
due
to
the
major
confounder
of
housing
conditions,
if
they
vary
from
laboratory
to
laboratory.
Group
housing
will
also
result
in
inter­
and
intra­
cage
variability
and
therefore
intra­
group
variability,
confounding
detection
of
any
treatment­
related
inter­
group
differences.
Singly­
housed
mouse
males
provide
a
more
uniform
population
in
terms
of
the
endpoints
under
consideration
in
this
White
paper,
which
are
affected
by
group
housing.
Dominance
hierarchy
is
also
established
in
group­
housed
male
and
female
rats
(
Becker
and
Flaherty,
1968;
Westenbroek
et
al.,
2003;
Sharp
et
al.
2003).
The
group­
housed
subordinate
rats
exhibit
effects
on
delta­
opioid
receptor
function
(
Pohorecky
et
al.
1999),
immune
status
(
Stefanski
1998;
2000),
endocrine
status
(
Taylor
and
Costanzo,
1975;
Popova
and
Naumenko,
1972),
circadian
rhythms
(
Greco
et
al.,
1989),
and
behavioral
and
neuroendocrine
parameters
(
Blanchard
et
al.,
1993).
The
dominant
rats
display
aggressive
behavior
(
Taylor
and
Moore,
1975)
and
"
barber"
the
subordinate
rats
(
Bresnahan
et
al.,
1983).
The
subordinate
female
rats
also
exhibit
stress­
like
responses
in
group
housing
situations
(
Westenbroek
et
al.
2003;
Sharp
et
al.
2003).
Therefore,
the
same
confounders
may
exist
in
studies
in
rats
across
laboratories
if
housing
situations
vary.

°
Caging:
Polycarbonate
caging
is
transparent,
autoclavable,
and
in
common
use.
Recent
evidence
(
Hunt
et
al.
2003;
Koehler
et
al.
2003)
indicates
that
crazed,
cracked,
damaged
polycarbonate
caging
from
inappropriate
and
inadvertent
use
of
a
harsh
detergent
releases
BPA
(
a
weak
xenoestrogen),
which
can
cause
adverse
reproductive
effects
(
meiotic
aneuploidy)
in
certain
sensitive
mouse
strains
(
Hunt
et
al.
2003).
Meiotic
aneuploidies
are
associated
with
embryo
and
fetal
mortality
as
well
as
Down's
Syndrome
(
Trisomy
21)
in
humans.
If
such
damaged
cages
are
in
use
in
a
laboratory,
its
ability
to
detect
effects
of
a
potentially
estrogenic
chemical
(
intentionally
introduced)
may
be
compromised
by
the
inadvertent
presence
of
BPA
(
Hunt
et
al.
2003).
A
laboratory
can
carefully
check
their
polycarbonate
cages
and
discard
damaged
ones
or
switch
to
caging
which
is
not
made
of
polycarbonate
(
e.
g.,
polypropylene).
However,
polypropylene
caging
is
translucent
but
not
transparent.
25
°
Water
bottles:
Damaged
polycarbonate
water
bottles
can
also
release
BPA
(
Hunt
et
al.
2003).
A
laboratory
can
carefully
check
and
discard
damaged
water
bottles
or
switch
to
glass
water
bottles.

°
Feed:
Laboratory
animal
feed
can
be
categorized
into
semipurified,
purified,
certified,
and
standard
open
or
closed
formulas.
Different
lots
of
the
same
feed
from
the
same
vendor,
as
well
as
different
feeds
from
different
vendors,
may
differ
in
their
relative
content
of
nutrients,
pesticides,
other
contaminants,
and
phytoestrogens:
genistein,
diadzein,
formononetin,
and
biochanin
A
(
the
isoflavones),
as
well
as
coumestrol
(
the
coumestans)
from
soybeans,
flax,
wheat,
barley,
corn,
alfalfa,
and
oats
commonly
used
in
laboratory
animal
diets
(
Thigpen
et
al.,
1999).
Phytoestrogens
and
estrogenic
mycotoxins
from
contaminating
molds
and
fungi
can
bind
to
the
estrogen
receptor
(
although
they
are
much
weaker
than
the
endogenous
steroidal
estrogens
of
humans
and
animals)
and
induce
estrogen­
like
effects
in
animals,
humans,
and
cells
in
culture.
Phytoestrogens
can
also
affect
sex­
specific
behavior,
gonadotropin
function
(
Whitten
et
al.
1995),
and
postnatal
development
(
Lewis
et
al.
2003).
The
predominant
phytoestrogens
in
feed
are
genistein
and
daidzein
from
soybeans.
The
concentrations
of
these
phytoestrogens
vary
significantly
across
rodent
diets
and
within
rodent
diets
by
batch
(
Thigpen
et
al.
1989;
1999).
The
content
is
influenced
by
the
use
of
different
plant
species,
portion
of
the
plant
used,
geographic
location,
time
of
harvest,
and
method
of
processing
into
pellets
or
meal
(
Eldridge
and
Kwolek,
1983).

°
Water:
Different
laboratories
use
tap
water,
deionized
water,
deionized/
distilled
water,
distilled
water,
reverse
osmosis
(
R0)
water,
etc.
Separate
from
palatability
issues,
different
salts,
ions,
organic
contaminants,
pesticides,
disinfection
by­
products
(
DBPs),
etc.,
in
different
concentrations
are
present
in
each
type
of
water
and
will
vary
by
season,
by
geographic
location,
by
different
water
disinfectants,
by
where
the
disinfectant
is
added
in
the
water
purification
process,
and
by
where
the
water
is
sampled
in
the
supply
lines.
Certain
DBPs
have
been
shown
to
affect
reproduction,
especially
those
which
are
brominated;
e.
g.,
bromochloroacetic
acid
(
Klinefelter
et
al.
2003),
dibromoacetic
acid,
bromochloromethane,
bromodichloromethane
(
Narotsky
et
al.
1993;
Bielmeier
et
al.
2001),
etc.,
as
well
as
perchlorate
which
inhibits
iodide
symporters
in
the
thyroid,
reducing
import
of
organic
iodide
into
thyroid
cells,
and
thereby
reducing
synthesis,
export,
and
circulating
and
local
T3
and
T4
levels,
causing
hypothyroidism
in
adults
and
children
(
Lamm
and
Doemland,
1999).

°
Temperature
and
relative
humidity:
Are
these
parameters
continuously
recorded,
monitored,
and
adjusted?
Elevated
temperatures
can
affect
26
spermatogenesis
in
males.
Reduced
temperatures
can
cause
stress
responses.
Reduced
relative
humidity
can
affect
pup
survival.

°
Light
cycle:
12:
12
or
14:
10
can
make
a
difference
in
circulating
hormone
levels
for
those
hormones
keyed
to
light/
dark
cycles
(
i.
e.,
circadian
rhythms;
Greco
et
al.
1989).

°
Technician
skills
and
experience:
The
better
trained
and
more
experienced
the
technical
staff,
the
better
the
dissections
and
examination
of
animal
data,
and
the
more
uniform
the
evaluations
(
low
inter­
technician
variability).

°
Source
of
the
test
material:
The
source
and
purity
of
the
test
material,
amount,
and
identification
of
impurities,
storage
stability,
etc.,
will
affect
the
results
if
the
performing
laboratory
does
not
know
the
status
of
these
parameters
and
does
not
adjust
for
them,
if
necessary.

C.
Study
Design
°
Number
of
animals/
group;
number
of
dose
groups;
source,
species/
strain/
stock,
age
and
weight
of
the
animals
°
Route
of
administration,
identity
of
vehicle,
doses
(
in
mg/
kg/
day),
dosing
volume
(
in
ml/
kg),
chemical
concentration
(
in
mg/
ml);
adjustment
of
dosing
volume,
based
on
most
recent
body
weight:
the
same
dose
(
in
mg/
kg),
administered
at
a
different
concentration
and
dosing
volume,
can
result
in
different
results
°
Frequency
of
body
weights
to
adjust
dose
and
detect
effects,
clinical
observations,
feed
consumption
°
Duration
of
dosing
period
(
age
of
animals
at
end
of
dosing)

°
Technician
experience
and
expertise
in
dosing,
observation
of
clinical
signs;
are
the
technicians
"
blind
for
dose?"

°
Time
of
necropsy
(
age
of
animals)
 
Anesthesia/
euthanasia
method
 
Endpoints
evaluated
 
Experience
of
technicians
in
necropsy,
trimming
and
weighing
organs
(
fresh/
fixed
weight,
especially
for
small
organs
such
as
pituitary,
adrenal
glands,
thyroid
gland,
ovaries;
risk
of
organs
drying
out)
 
Choice
of
fixative
(
testes
in
Bouin's,
other
organs
in
buffered
10%
formalin)
27
 
Trimming
in
tissues,
fixation,
embedment
(
GMA
plastic
for
testes,
paraffin
for
other
organs),
section
location
and
thickness,
microscope
slide
preparation,
staining
(
PAS/
H
for
testes,
hematoxylin
and
eosin
for
other
organs),
coverslipping
 
Experience
of
pathologist
reading
slides
 
Blood
collection
(
location,
volume,
speed
[
hemolysis?])
 
Analysis
of
hormones,
validation
method,
intra­
and
inter­
assay
variability,
intertechnician
variablity
(
single
terminal
blood
sample
versus
longitudinal
evaluation
by
serial
tail
vein
sampling
or
from
cannulated
animals)
°
Summarization
and
statistical
analyses
of
study
data
 
Are
the
correct
statistical
tests
used
for
the
correct
parameters?
 
Does
the
laboratory
maximize
sensitivity
and
power
(
number
of
independent
entities,
"
n"
per
group)?
 
Does
the
laboratory
have
and
use
historical
control
data
(
to
interpret
concurrent
control
values
in
the
context
of
historical
control
values,
and
to
track
the
control
values
over
time
and
between
studies)?

2.4
Assays
Under
Consideration
of
the
EDSP
and
Associated
Endocrine
Endpoints
The
EPA
is
in
the
process
of
implementing
the
EDSP.
To
support
this
program,
the
EPA
has
contracted
with
Battelle
(
as
prime
and
RTI
International
as
subcontractor)
to
provide
comprehensive
toxicological
and
ecotoxicological
screening,
including
chemical,
analytical,
statistical,
and
quality
assurance/
quality
control
support
to
assist
the
EPA
in
developing,
standardizing,
and
validating
a
suite
of
in
vitro,
mammalian,
and
ecotoxicological
screens
and
assays
for
identifying
and
characterizing
endocrine
effects
through
exposure
to
pesticides,
industrial
chemicals,
and
environmental
contaminants.
The
studies
conducted
will
be
used
to
develop,
standardize,
and
validate
methods;
prepare
appropriate
guidance
documents
for
peer
review
of
the
methods;
and
develop
technical
guidance
and
test
guidelines
in
support
of
the
Office
of
Prevention,
Pesticides
and
Toxic
Substances
regulatory
programs.

In
vivo
mammalian
assays
under
consideration
by
the
EDSP
include:
(
1)
pubertal
male
and
female,
(
2)
in
utero/
lactational,
(
3)
adult
male
15­
day
study,
all
for
Tier
I,
and
(
4)
two­
generation
mammalian
reproductive
assay
(
Tier
2).
In
addition,
as
an
alternative
to
the
mammalian
two­
generation
assay,
a
one­
generation
assay
has
been
described
in
the
EDSTAC
Final
Report
(
1998).
Table
1
summarizes
the
assays
under
consideration
and
their
associated
endpoints.
A
discussion
of
each
proposed
study
follows
Table
1.
28
Table
1.
Assays
Under
Consideration
by
the
EDSP
and
Associated
Endocrine
Endpoints
Endocrine
Endpoints
One­
Generation*
Pubertal
Male
and
Female
In
Utero
Through
Lactation
Adult
Male
Two­
Generation
Period
of
exposure/
testing
F0:
2
wks
of
prebreed
exposure,
2
wks
of
mating;
3
weeks
of
gestation,
3
wks
of
lactation,
then
selected
F1
animals
exposed
for
minimum
of
10
wks
postwean
F1
offspring
dosing
from
pnd
23­
52/
53
(
males),
pnd
22­
42/
43
(
females)
F0
maternal
dosing
from
gd
6
to
pnd
21
F1
cohorts:
1)
F1
females,
1/
litter
for
utero­
trophic
assay,
sc
injection
pnd
21­
24
2)
F2
females,
2/
litter
for
pubertal
assessments,
oral
dosing
pnd
21­
42
3)
F1
males,
2/
litter,
for
pubertal
assessments,
oral
dosing
pnd
21­
70
Dosing
for
15
consecutive
days
in
adult
males
F0:
10
wks
of
prebreed
exposure,
2
wks
of
mating,
3
wks
of
gestation,
3
wks
of
lactation,
then
selected
F1
animals
exposed
for
minimum
of
10
wks
prebreed,
mating,
gestation,
and
lactation,
with
termination
at
weaning
(
pnd
21)
of
F2
offspring
Litter
size/
gest­
ational
indices
yes
no
yes
no
yes
Pup
viability
yes
no
yes
no
yes
Anogenital
distance
yes,
pnd
0,
21
and
95
(
males)
no
yes,
pnd
0,
21
and
at
necropsy
(
pnd
42
females,
pnd
70
males)
no
yes
(
triggered
in
F2
in
1998
OPPTS
Testing
Guidelines)

Nipple/
areolar
retention
yes,
pnd
11­
13
(
males
and
females),
21
and
95
(
males)
no
yes,
pnd
11­
13
no
yes
(
not
in
1998
OPPTS
Testing
Guidelines)

Vaginal
patency
yes
yes
yes
no
yes
Preputial
separation
yes
yes
yes
no
yes
Table
1.
(
continued)

Endocrine
Endpoints
One­
Generation*
Pubertal
Male
and
Female
In
Utero
Through
Lactation
Adult
Male
Two­
Generation
29
Hormone
levels
yes,
T
3/
T
4
(
TSH
triggered)
TSH,
and
T
4
TSH,
T
3,
T
4
and
E2
Serum
T,
E2,
DHT,
follicle
stimulating
hormone
(
FSH),
luteinizing
hormone
(
LH),
prolactin
(
PL),
thyroid
stimulating
hormone
(
TSH),
thyroxine
(
T
4),
and
triiodothy­
ronine
(
T
3)
not
in
1998
OPPTS
Testing
Guidelines;
likely
to
include
TSH,
and
T
4
for
the
EDSP
(
personal
communication
with
J.
Kariya)

Urethral
vaginal
distance
no
no
yes
no
no
Estrous
cyclicity
yes
yes
yes
no
yes
Uterine
Weight
no
no
yes
no
no
Reproductive
tract
develop­
ment
yes
yes
yes
no
yes
Testes
descent
yes
no
no
no
yes
Behavioral
evaluations
(
clinical
observations)
yes
yes
yes
yes
yes
Andrology
yes
no
yes
yes
yes
Reproductive
organ
weights
and
histopathology
yes
yes
(
reproduct­
ive
organs
and
thyroid)
yes
yes
yes
*
as
described
in
the
EDSTAC
(
1998)
Final
Report
as
an
alternative
to
the
mammalian
two­
generation
reproductive
toxicity
study.
30
2.4.1
One­
Generation
Assay
Two
assays
were
proposed
by
the
EDSTAC
(
FR,
1998)
as
an
alternative
to
the
mammalian
two­
generation
reproductive
toxicity
study,
a
one­
generation
assay
and
an
"
Alternative
Mammalian
Reproductive
Test"
(
AMRT).
The
proposed
one­
generation
assay
is
a
shortened,
scaled
down
version
of
the
OPPTS
guideline
for
reproductive
toxicity
testing,
and
was
designed
to
assess
the
reproductive
effects
and
developmental
effects
after
an
in
utero
lactational
and
post­
lactational
exposure.
During
continuous
exposure,
the
assay
is
designed
to
assess
onset
of
puberty
(
VO
and
PPS),
estrous
cyclicity,
and
andrological
parameters.
The
period
of
exposure
begins
two
weeks
prior
to
mating
and
continues
through
weaning
of
the
F1
offspring,
followed
by
a
ten­
week
postwean
exposure
in
the
F1
animals.
The
AMRT
differs
from
the
one­
generation
assay
in
that
is
doesn't
include
prebreed
exposure,
but
includes
mating
of
the
F1
offspring
and
evaluations
of
the
F2
offspring,
with
no
direct
exposure
in
the
F1
and
F2
animals.
Though
the
exposure
periods
and
durations
of
the
two
alternative
assays
differ,
the
endpoints
evaluated
are
essentially
the
same
(
and
are
similar
to
the
endpoints
evaluated
in
the
two­
generation
assay).

2.4.2
Pubertal
Male
and
Female
Assays
The
EDSTAC
also
recommended
the
use
of
an
intact
female
20­
day
pubertal
assay
to
evaluate
test
materials
for
effects
on
the
thyroid,
hypothalamic­
pituitarygonadal
(
HPG)
and
hypothalamic­
pituitary­
thyroid
(
HPT)
axes,
aromatase
and
estrogens
(
and/
or
other
test
materials)
that
are
only
effective
orally
or
after
a
dosing
duration
longer
than
that
used
in
the
uterotrophic
assay
(
EDSTAC
Report,
1998,
Vol.
1,
Chapter
5,
p.
5­
26).
EDSTAC
also
recommended,
as
an
alternate
assay
to
be
evaluated,
the
intact
male
20­
day
thyroid/
pubertal
assay
in
rodents
(
EDSTAC,
1998,
Vol.
1,
Chapter
5,
p.
5­
30).

The
EDSTAC
discussion
on
the
usefulness
of
the
female
pubertal
assay
and
its
endpoints
included
the
following:

"
The
determination
of
the
ages
at
"
puberty"
in
the
female
rat
is
an
endpoint
that
has
already
gained
acceptance
in
the
toxicology
community.
Vaginal
opening
(
VO)
in
the
female
is
a
required
endpoint
measured
in
the
new
EPA
two­
generation
reproductive
toxicity
test
guideline.
In
this
regard,
this
assay
would
be
easy
to
implement
because
these
endpoints
have
been
standardized
and
validated
and
VO
data
are
currently
being
collected
under
GLP
conditions
in
most
toxicology
laboratories.
In
addition,
VO
data
are
reported
in
m
any
recently
published
developmental
and
reproductive
toxicity
studies
(
i.
e.,
see
studies
from
Drs.
R.
E.
Peterson's,
R.
Chapin's
and
L.
E.
Gray's
laboratories
on
dioxins,
antiandrogens,
and
xenoestrogens).

In
the
pubertal
female
assay,
oral
dosing
is
initiated
in
weanling
rats
at
21
days
of
age
(
10
per
group,
selected
for
uniform
body
weights
at
weaning
to
reduce
variance).
The
animals
are
dosed
daily,
7
days
a
week,
and
examined
daily
for
vaginal
opening
(
one
could
also
check
for
age
at
first
estrous
and
onset
of
estrous
cyclicity).
Dosing
continues
until
VO
is
attained
in
all
fem
ales
(
typically
two
weeks
after
weaning,
unless
delayed).
Age
at
VO
is
also
determined
in
the
fem
ale
rat.
Rats
are
dosed
by
gavage
with
xenobiotic
and
examined
daily
for
VO.
The
advantage
over
the
uterotrophic
assay
is
that
one
test
31
detects
both
agonists
and
antagonists,
it
detects
xenoestrogens
like
methoxychlor
that
are
almost
inactive
via
sc
injection,
it
detects
aromatase
inhibitors,
altered
HPG
function,
and
unusual
chemicals
like
betasitosterol.
In
addition,
at
necropsy
one
should
weigh
the
ovary
(
increased
in
size
with
aromatase
inhibitors,
but
reduced
with
betasitosterol),
save
the
thyroid
for
histopathology,
take
serum
for
T4,
and
measure
TSH.

Exposure
of
weanling
female
rats
to
environmental
estrogens
can
result
in
alterations
of
pubertal
development
(
Ramirez
and
Sawyer,
1964).
Exposure
to
a
weakly
estrogenic
pesticide
after
weaning
and
through
puberty
induces
pseudoprecocious
puberty
(
accelerated
vaginal
opening
without
an
effect
on
the
onset
of
es
trous
cyclicity)
after
only
a
few
days
of
exposure
(
Gray
et
al.,
1989),
or
precocious
pubery
with
both
accelerated
VP
and
onset
of
estrous
cyclicity.
Pubertal
alterations
also
result
in
girls
exposed
to
estrogen­
containing
creams
or
drugs
which
induce
pseudoprecocious
puberty
and
alterations
of
bone
developm
ent
(
Hannon
et
al.,
1978).

Several
examples
of
estrogenic
chemicals
affecting
vaginal
opening
in
rodents
are
known
and
include
methoxychlor
(
Gray
et
al.,
1989),
nonylphenol,
and
octylphenol
(
Gray
and
Ostby,
1998).
This
endpoint
appears
to
be
almost
as
sensitive
as
the
uterine
weight
bioassay,
but
the
evaluation
is
easier
to
conduct
and
does
not
require
that
the
animals
be
euthanized,
so
they
can
be
used
for
additional
evaluations.
For
example,
treatment
with
methoxychlor
at
weaning
(
6
mg/
kg/
day
or
higher)
caused
pseudoprecocious
puberty
in
female
rats.
Vaginal
opening
occurs
from
two
to
seven
days
earlier
in
treated
anim
als
than
controls,
in
a
dose­
related
fashion,
but
methoxychlor
did
not
alter
estrous
cyclicity
at
the
low
dosage
levels,
indicating
a
direct
estrogenic
effect
of
methoxychlor
on
vaginal
epithelial
cell
function
without
an
effect
on
hypothalamic­
pituitary
maturation.
Similar
effects
have
been
achieved
with
chlordecone,
another
weakly
estrogenic
pesticide,
and
octylphenol.
Chlordecone
also
induces
neurotoxic
effects
(
hyperactivity
to
handling
and
tremors).
In
addition
to
estrogens,
the
age
at
vaginal
opening
and
uterine
growth
can
be
affected
by
alteration
of
several
other
endocrine
mechanisms,
including
alterations
of
the
hypothalam
ic­
pituitary­
gonadal
axis
(
Shaban
and
Terranova,
1986;
Gonzalez
et
al.,
1983).
In
rats,
this
event
can
also
be
induced
by
androgens
(
Salamon,
1938)
and
EGF
(
Nelson
et
al.,
1991).
In
the
last
20
years,
there
have
been
over
200
publications
which
demonstrate
the
broad
utility
of
this
assay
to
identify
altered
estrogen
synthesis,
ER
action,
growth
hormone,
prolac
tin,
FSH
or
LH
secretion,
or
CNS."
(
EDSTAC
Report,
1998,
Vol.
1,
Chapter
5,
pp.
5­
26
­
5­
27)

Based
on
the
EDSTAC's
recommendations,
one
of
the
assays
that
the
EPA
has
proposed
to
include
in
an
EDSP
is
a
female
pubertal
assay
(
see
FR
Vol.
63,
No.
248,
pp.
71541­
71568,
December
28,
1998).
This
assay
is
the
most
comprehensive
femalespecific
assay
in
the
proposed
Tier
1
battery
of
assays,
as
it
is
capable
of
detecting
substances
that
alter
thyroid
function,
that
are
aromatase
inhibitors,
estrogenic,
antiestrogenic,
or
are
agents
which
interfere
with
the
HPG
and
HPT
axes.
Results
from
other
shorter
assays
and/
or
with
the
use
of
ip
injection
as
the
route
of
administration
have
also
been
reported
(
O'Connor
et
al.
1996,
1999).

The
EPA
is
also
pursuing
the
validation
of
a
male
pubertal
assay
as
a
potential
alternative
to
other
assays
in
the
Tier
1
battery.
The
EDSTAC
discussion
on
the
usefulness
of
the
male
pubertal
assay
and
its
endpoints
included
the
following:

"
This
assay
detects
androgens
and
antiandrogens
in
vivo
in
a
single
stage
apical
test.
"
Puberty"
is
m
easured
in
male
rats
by
determ
ining
age
at
PPS
(
preputial
separation).
Animals
are
dosed
by
gavage
beginning
one
week
before
puberty
(
which
occurs
at
about
40
days
of
age)
and
PPS
is
measured.
Androgens
will
accelerate
and
antiandrogens
and
32
estrogens
will
delay
PPS.
The
assay
takes
about
3
weeks,
and
allows
for
com
prehensive
assessment
of
the
entire
endocrine
system
in
one
study
(
10
per
group,
selected
for
uniform
body
weights
to
reduce
variance).
The
animals
are
dosed
daily,
seven
days
a
week,
and
exam
ined
daily
for
PPS.
Dosing
continues
until
53
days
of
age;
the
males
are
then
necropsied.
The
body,
heart
(
thyroid),
adrenal,
testis,
seminal
vesicle
plus
coagulating
glands
(
with
fluid),
ventral
prostate,
and
levator
ani
plus
bulbocavernosus
muscles
(
as
a
unit)
are
weighed.
The
thyroid
is
retained
for
histopathology
and
serum
is
taken
for
T4,
T3,
and
TSH.
Testosterone,
LH,
prolactin,
and
dihydrotestosterone
analyses
are
optional.
These
endpoints
take
several
weeks
to
evaluate
and
are
affected
not
only
by
estrogens
but
by
environmental
antiandrogens,
drugs
that
affect
the
hypothalamic­
pituitary
axis
(
Hostetter
and
Piacsek,
1977;
Ramaley
and
Phares,
1983),
and
by
prenatal
exposure
to
TCDD
(
Gray
et
al.,
1995a;
Bjerke
and
Peterson,
1994)
or
dioxin­
like
PCBs
(
Gray
et
al.,
1995b).
In
contrast
to
these
other
mechanisms,
only
peripubertal
estrogen
administration
accelerates
this
process
in
the
female
and
delays
it
in
the
m
ale.
Preputial
separation
in
the
male
rodent
is
easy
to
measure
and
this
is
not
a
terminal
measure
(
Korenbrot
et
al.,
1977).

Age
and
weight
at
puberty,
reproductive
organ
weights,
and
serum
hormone
levels
can
also
be
measured.
Delays
in
male
puberty
results
form
exposure
to
both
estrogenic
and
antiandrogenic
chemicals
including
methoxychlor
(
Gray
et
al.,
1989),
vinclozolin
(
Anderson
et
al.,
1995a),
and
p,
p'DDE
(
Kelce
et
al.,
1995).
Exposing
weanling
male
rats
to
the
antiandrogenic
pesticides
p,
p'DDE
or
vinclozolin
delays
pubertal
development
in
weanling
male
rats
as
indicated
by
delayed
preputial
separation
and
increased
body
weight
(
because
they
are
older
and
larger)
at
puberty.
In
contrast
to
the
delays
associated
with
exposure
to
estrogenic
substances,
antiandrogens
do
not
inhibit
food
consumption
or
retard
growth
(
Anderson
et
al.,
1995b).
Antiandrogens
cause
a
delay
in
preputial
separation
and
affect
a
number
of
endocrine
and
morphological
parameters
including
reduced
seminal
vesicle,
ventral
prostate,
and
epididymal
weights.
It
is
apparent
that
PPS
is
more
sensitive
than
are
organ
weights
in
this
assay.
In
addition,
responses
of
the
HPG
are
variable.
In
studies
of
vinclozolin,
increases
in
serum
LH
were
a
sensitive
response
to
this
antiandrogen,
whereas
serum
LH
is
not
increased
in
m
ales
exposed
to
p,
p'DDE
during
puberty
(
Kelce
et
al.,
1997).
Furthermore,
a
systematic
review
of
the
literature
indicates
that
the
sex
accessory
glands
of
the
immature
intact
m
ale
rat
are
consistently
m
ore
affected
than
in
the
adult
intact
m
ale
rat.

In
summary,
preputial
separation
and
sex
accessory
gland
weights
are
sensitive
endpoints.
However,
a
delay
in
preputial
separation
is
not
pathognomonic
for
antiandrogens.
Pubertal
alterations
result
from
chemicals
that
disrupt
hypothalamic­
pituitary
function
(
Huhtaniem
i
et
al.,
1986)
and,
for
this
reason,
additional
in
vivo
and
in
vitro
tests
are
needed
to
identify
the
mechanism
of
action
responsible
for
the
pubertal
alterations.
For
example,
alterations
of
prolactin,
growth
hormone,
gonadotrophin
(
LH
and
FSH)
secretion,
or
hypothalamic
lesions
alter
the
rate
of
pubertal
maturation
in
weanling
rats.

As
indicated
above,
the
determination
of
the
age
at
"
puberty"
in
the
male
rat
are
endpoints
that
already
have
gained
acceptance
in
the
toxicology
community.
Preputial
separation
in
the
m
ale
is
a
required
endpoint
in
the
new
EPA
two­
generation
reproductive
toxicity
test
guideline.
In
this
regard,
this
assay
would
be
easy
to
implement
because
these
endpoints
have
been
standardized
and
validated
and
PPS
data
are
currently
being
collected
under
GLP
conditions
in
most
toxicology
laboratories.
In
addition,
PPS
data
are
reported
in
many
recently
published
developmental
and
reproductive
toxicity
studies
(
i.
e.,
see
studies
from
R.
E.
Peterson's,
J.
Ashby's,
R.
Chapin's
and
L.
E.
Gray's
laboratories
on
dioxins,
PCBs,
antiandrogens,
and
xenoestrogens).

Sex
accessory
gland
weights
in
intact
adult
male
rats
also
can
be
affected
directly
or
indirectly
by
toxicant
exposure.
The
HPG
axis
in
an
intact
animal
is
able
to
compensate
for
the
action
of
antiandrogens
by
increasing
hormone
production,
which
counteracts
the
effect
of
the
antiandrogen
on
the
tract
(
Raynaud
et
al.,
1984;
Edgren,
1984;
Hershberger,
1953)."
(
EDSTAC,
1998,
Vol.
1,
Chapter
5,
pp.
5­
30
through
5­
32).
33
Based
on
the
EDSTAC's
recommendations,
one
of
the
assays
that
the
EPA
has
also
proposed
to
validate
as
a
potential
alternative
for
other
assays
in
the
Tier
1
battery
in
an
endocrine
disruptor
screening
program
is
a
male
pubertal
assay
(
see
FR
Vol.
63,
No.
248,
pp.
71541­
71568,
December
28,
1998).
This
assay
is
the
most
comprehensive
male­
specific
assay
in
the
proposed
Tier
1
battery
of
assays,
as
it
is
capable
of
detecting
substances
that
alter
thyroid
function,
that
are
aromatase
inhibitors,
androgenic,
anti­
androgenic,
or
that
are
agents
which
interfere
with
the
HPG
and
HPT
axes.
Results
from
other
shorter
assays,
and/
or
with
the
use
of
ip
injection
as
the
route
of
administration,
have
also
been
reported
(
O'Connor
et
al.,
1996,
1999).

2.4.3
In
Utero
Through
Lactation
Assay
The
proposed
protocol
has
been
identified
by
the
EPA
as
the
"
In
Utero/
Lactational
Exposure
Testing
Protocol"
and
has
been
assigned
for
development
under
the
EDSP.
The
objective
of
this
bioassay
is
to
detect
effects
mediated
by
alterations
in
the
estrogen,
androgen,
and
thyroid­
signaling
pathways
as
a
consequence
of
exposure
during
pre­
and
postnatal
development
in
the
laboratory
rat.
The
treatment
paradigm
allows
for
an
evaluation
of
effects
on
organogenesis,
sexual
differentiation,
and
puberty.
In
using
a
developing
system
as
the
basis
for
the
assay,
it
is
understood
that
modes
of
action,
other
than
those
of
the
estrogen,
androgen,
and
thyroid­
signaling
pathways,
may
be
involved
in
the
induction
of
toxicity.
As
such,
any
observed
effects
will
have
to
be
interpreted
in
light
of
the
overall
weight
of
the
evidence
that
they
are
endocrine
dependent.

2.4.4
Adult
Male
Assay
One
of
the
assays
considered
by
EDSTAC
as
an
alternate
assay
was
a
shortterm
screen
in
an
intact
adult
male
with
assessment
of
levels
of
various
circulating
hormones
at
necropsy
(
see
Table
1).
The
adult
male
assay
was
developed
to
detect
effects
on
male
reproductive
organs
that
are
sensitive
to
antiandrogens
and
agents
that
inhibit
testosterone
synthesis
or
inhibit
5­
alpha­
reductase
(
see
EDSTAC
FR
Vol.
I,
p.
5­
30,
August,
1998).
Results
from
this
assay
and/
or
with
the
use
of
ip
injection
as
the
route
of
administration,
and
other
assays
with
a
similar
purpose,
have
been
reported
(
O'Connor
et
al.
1996,
1999,
2002a,
b).

Based
on
the
EDSTAC's
recommendations,
one
of
the
assays
that
the
EPA
has
proposed
to
validate
as
a
potential
alternative
for
other
assays
in
the
Tier
1
battery,
in
an
endocrine
disruptor
screening
program,
is
an
adult
male
in
vivo
assay
(
see
FR
Vol.
63,
No.
248,
pp.
71541­
71568,
December
28,
1998).
The
utility
of
this
battery
for
screening
unknown
compounds
for
endocrine
activity
will
be
evaluated.
Endocrine
endpoints
for
this
study
are
listed
in
Table
1.

2.4.5
Two­
generation
Assay
For
the
Tier
2
battery,
EDSTAC
recommended
a
mammalian
two­
generation
reproductive
toxicity
study.
The
two­
generation
reproductive
toxicity
study
in
rats
is
34
designed
to
evaluate
the
health
effects
of
chemicals
on
reproduction
and
viability
through
two
generations
as
performed
in
accordance
with
EPA
Guideline
OPPTS
870.3800
(
1998),
and
OECD
Guideline
416
(
2001).
Endocrine
endpoints
for
this
study
are
listed
in
Table
1.
In
the
two­
generation
reproductive
toxicity
assay,
potential
endocrine­
disrupting
effects
can
be
detected
through
behavior,
fertility,
gestational
duration,
litter
size,
sex
ratio,
viability
of
the
offspring,
developmental
landmarks,
and
reproductive
development
(
histopathology
of
reproductive
organs,
onset
of
puberty
[
acquisition
of
preputial
separation
and
vaginal
opening],
and
estrous
cyclicity)
in
the
F1
offspring
exposed
initially
as
gametes
(
from
exposed
F0
parents),
their
gestation
(
in
exposed
F0
females),
lactation
(
nursed
by
exposed
F0
females),
and
directly
through
adulthood
and
reproduction
to
produce
the
F2
generation.
This
study
is
intended
to
evaluate
the
effects
of
chemicals
on
sensitive
life
stages
of
reproduction
and
development
in
a
transgenerational
design.
Proposed
endocrine
endpoints
for
this
study
are
listed
in
Table
1.
Table
1
indicates
what
is
tested
and
what
is
proposed
with
no
separation.

2.5
Endocrine
Endpoints
Under
Consideration
for
EDSP
Assays
and
Intraspecies
Variability
Study
designs
for
use
in
risk
assessment
require
endpoints
that
have
been
shown
to
be
robust,
reproducible,
appropriately
sensitive,
biologically
plausible,
and
relevant
to
the
adverse
outcomes
of
concern.
Definitions
of
the
attributes
of
such
endpoints
are
as
follows:

Reproducible:
These
endpoints
must
be
reliable;
i.
e.
the
same
findings
occur
under
the
same
conditions
within
the
initial
reporting
laboratory
(
intra­
laboratory)
and
among
other
laboratories
(
inter­
laboratory).
If
the
results
from
endpoints
are
not
reproducible,
they
cannot
form
the
basis
for
future
research
and
are
most
likely
not
useful
for
risk
assessment.

Robust:
These
endpoints
must
significant
enough
to
be
present
after
comparable
routes
of
exposure,
(
e.
g.,
dosed
feed
or
dosed
water),
at
the
same
doses
over
time.
Different
effects,
both
quantitative
and
even
qualitative,
may
be
observed
when
different
routes
of
administration
are
used.
The
use
of
oral
gavage,
a
bolus
dose
once/
day,
may
result
in
exacerbation
of
the
endpoint
if
the
parent
material
is
the
proximate
toxicant
and
is
metabolized
to
a
nontoxic
metabolite,
if
bolus
dosing
overwhelms
the
metabolic
capacity
of
the
organism
or
preparation,
or
it
may
result
in
diminution
or
loss
of
the
endpoint
if
the
parent
compound
must
be
metabolized
to
the
active
form.
The
use
of
nonoral
routes,
such
as
inhalation,
topical
application,
injection,
etc.,
will
also
likely
result
in
different
effects,
since
these
routes
bypass
"
first­
pass"
metabolism
by
the
liver.
The
findings
from
routes
unrelated
to
human
or
environmental
exposures
may
not
be
as
useful
for
risk
assessment.
35
Sensitive:
These
endpoints
should
not
be
so
sensitive
that
they
are
dependent
on
unique
conditions
(
e.
g.,
intrauterine
position
[
IUP],
etc.),
especially
those
which
are
not
relevant
to
the
species
at
risk.
Sensitivity
is
the
ability
of
an
endpoint
to
detect
small
differences
reliably.
These
endpoints
should
not
exhibit
high
variability
(
insensitive)
or
be
greatly
affected
by
confounders
(
too
sensitive).

Relevant:
These
endpoints
must
be
biologically
plausible
and
related
to
adverse
effects
of
interest/
concern.
If
there
are
no
adverse
consequences
at
the
dose/
duration/
route
evaluated,
these
endpoints
should
be
predictive
of
other
adverse
effects
at
higher
doses,
after
longer
exposure
duration,
and/
or
by
different
routes,
etc.

Consistent:
These
endpoints
should
occur
in
the
presence
of
effects
in
other
related,
relevant
endpoints,
if
possible,
at
the
same
dose,
timing,
duration,
routes
of
exposure,
etc.
(
i.
e.,
characteristic
syndrome
of
effects).

Individual
endocrine
endpoints
are
discussed
below.

2.5.1
Fertility
and
Gestational
Indices
Fertility
and
gestational
indices
and
litter
size
parameters
are
used
as
measures
of
reproductive
performance.
Examples
of
these
indices
and
parameters
follow.

Gestational
parameters:
No.
of
mating
pairs
No.
(%)
females
sperm/
plug
positive
Mating
index
=

Precoital
interval
(
time
in
days
from
pairing
to
evidence
of
copulation)

No.
(%)
females
pregnant
Pregnancy
index
=

Gestational
length
in
days
No.
(%)
females
with
live
litters
Gestational
index
=
36
Litter
size
parameters:

No.
ovarian
corpora
lutea/
dam
No.
uterine
implants/
dam
No.
(%)
preimplantation
loss
=

No.
resorbed
implants/
litter
No.
dead
fetuses/
litter
No.
nonlive
(
resorbed
and
dead)
implants/
litter
No.
(%)
litters
with
>
1:
resorptions,
dead
fetuses,
and
nonlive
No.
(%)
postimplantation
loss
=

No.
malformed
implants/
litter
No.
affected
implants
(
nonlive
plus
malformed)

No.
live
pups/
fetuses/
litter
2.5.2
Survival
and
Growth
Indices
Survival
and
growth
indices
are
used
as
measures
of
pup
viability.
Pups
are
evaluated
on
pnd
0
for
the
following:

Live
birth
index
=

Stillbirth
index
=
37
No.
live
pups/
litter:
total
and
by
sex
No.
dead
pups/
litter
total
and
by
sex
Mean
pup
body
weight/
litter:
total
and
by
sex
Anogenital
distance
(
absolute
in
mm,
relative
to
body
weight,
or
adjusted
with
body
weight
as
the
covariate
)
by
sex/
litter
Pups
are
evaluated
during
lactation
for
the
following:

Sex
ratio
(%
male
offspring/
litter)

4­
Day
survival
index
=

7­
Day
survival
index
=

14­
Day
survival
index
=

21­
Day
survival
index
=

Lactation
index
=

*
If
the
litters
are
standardized
to
a
fixed
number
(
normally
eight
or
ten)
on
pnd
4.

Several
studies
have
investigated
the
effects
of
strain
differences
on
implantation
and
pregnancy
outcomes.
One
study
by
Cummings
et
al.
(
2000)
compared
the
effects
of
atrazine
on
implantation
and
early
pregnancy
in
four
strains
of
rats:
Holtzman,
SD,
LE,
and
F344.
Since
atrazine
has
been
known
to
affect
prolactin
surge,
which
is
important
in
the
initiation
of
pregnancy
(
two
surges,
one
diurnal
and
one
nocturnal,
occur
daily
in
the
first
ten
days
of
pregnancy),
the
rats
were
dosed
on
the
first
eight
gestational
days
(
gd)
either
diurnally
or
nocturnally
with
0,
50,
100,
or
200
mg/
kg/
day
with
atrazine.
In
F344
rats,
atrazine
(
at
the
top
two
doses)
increased
the
percent
preimplantation
loss
only
after
nocturnal
dosing.
Holtzman
rats
showed
a
trend
toward
increasing
preimplantation
loss,
while
SD
and
LE
rats
were
not
significantly
affected
at
this
endpoint.
Percent
postimplantation
loss
was
significantly
higher
in
Holtzman
rats
only.
Clearly,
the
F344
and
Holtzman
strains
were
most
sensitive
to
the
effects
of
atrazine
on
early
pregnancy.

The
effects
of
atrazine
on
full
litter
resorption
and
pregancy
outcome
were
investigated
in
three
rat
strains:
F344,
SD,
and
LE
(
Narotsky
et
al.,
2001).
The
dams
were
dosed
from
gd
6­
10
with
0,
50,
100,
and
200
mg/
kg/
day
with
atrazine
and
then
38
allowed
to
deliver
their
pups,
thus
allowing
for
an
assessment
of
pup
viability.
At
the
highest
dose
(
200
mg/
kg/
day),
atrazine
caused
similar
rates
of
full
litter
resorption
in
all
three
strains.
Prenatal
loss
was
significantly
increased
in
F344
dams,
resulting
in
reduced
litter
sizes
for
dams
with
live
litters.
At
lower
doses
(
e.
g.
50
and
100
mg/
kg/
day),
atrazine
caused
pregnancy
loss
in
only
F344
rats,
while
SD
and
LE
litters
were
unaffected.
The
period
of
sensitivity
to
atrazine­
induced
pregnancy
loss
coincided
with
the
period
of
LH/
prolactin
dependency
on
the
maintenance
of
pregnancy.
While
gestational
loss
was
induced
in
the
sensitive
F­
344
strain
by
50
mg/
kg
atrazine
administered
on
gd
6­
10,
even
the
highest
dose
of
atrazine
(
200
mg/
kg)
was
without
effect
when
administered
after
the
LH
dependent
period
on
gd
11­
15.
The
authors
concluded
that
F344
rats
were
more
sensitive
than
SD
and
LE
rats
to
the
reproductive
effects
of
atrazine,
and
that
maternal
toxicity,
which
occurred
in
all
three
strains
at
higher
doses,
was
not
predictive
of
full
litter
resorption.

In
another
study,
the
effects
of
a
drinking
water
disinfection­
by­
product,
BDCM,
on
pregnancy
loss
were
studied
in
two
rat
strains,
F344
(
dosed
with
0
and
75
mg/
kg/
day
based
on
previous
studies),
and
SD
(
dosed
with
0,
75
and
100
mg/
kg/
day)
(
Bielmeier
et
al.
2001).
Daily
dosing
with
BDCM
(
75
mg/
kg)
from
gd
6
to
10
produced
a
62%
incidence
of
full
litter
resorption
(
pregnancy
loss)
in
F344
rats
and
no
effect
on
pregnancy
in
SD
rats.
Since
body
weights
were
significantly
reduced
in
both
strains,
it
is
possible
that
toxicokinetic
differences
in
the
strains
may
not
be
responsible
for
the
differential
sensitivities
of
the
strains
to
BDCM­
induced
pregancy
loss.
Thus,
the
data
of
Bielmeier
et
al.
(
2001)
suggest
that
F344
rats
are
genetically
more
sensitive
than
SD
rats
to
BDCM­
induced
diminishment
of
luteal
cell
responsiveness
to
LH
(
or
perhaps
BDCM
induced
luteolysis).

Though
these
studies
were
not
performed
in
the
same
laboratory,
Liberati
et
al.
(
2002)
compared
reproductive
and
litter
parameters
in
Wistar
Hannover
rats
with
historical
CD
rat
data
from
the
same
laboratory
and
from
CD
rat
data
provided
by
CRL.
Pregnant
female
Wistar
Hannover
rats
were
dosed
daily
with
distilled
water
from
gd
6
to
15.
Wistar
Hannover
rats
were
found
to
have
lower
pregnancy
rates
and
smaller
litter
sizes.
In
addition,
they
were
found
to
have
higher
percentages
of
preimplantation
loss,
postimplantation
loss,
and
resorptions
versus
CD
rats
(
14.1,
7.4
and
7.2%
versus
5.9,
5.6,
and
5.1
%,
respectively).
Thus,
it
appears
likely
that
fundamental
differences
in
reproductive
parameters
occur
between
outbred
stocks
of
rats.

2.5.3
Reproductive
Tract
Development
Reproductive
development
involves
both
morphological
and
hormonal
aspects,
which
operate
together
to
result
in
correctly
formed,
functional,
and
responsive
reproductive
systems
in
both
males
and
females.
In
mammals,
gonadal
origins
begin
early
in
embryonic
development,
prior
to
sexual
differentiation
(
Schardein,
1999).
Initial
stages
are
the
same
for
both
male
and
female.
Sexual
differentiation
and
maturation
are
under
hormonal
control.
Thus,
both
physical
and
hormonal
indicators
of
39
reproductive
development
can
be
monitored
to
detect
the
presence
of
endocrinedisrupting
activity.

 
Wolffian
duct
(
male
development).
Initially,
the
gonads
appear
as
a
pair
of
longitudinal
undifferentiated
genital
ridges
in
the
dorsal
abdominal
cavity
of
the
embryo.
The
primordial
germ
cells
migrate
into
the
genital
ridges
from
the
extra­
embryonic
yolk
sac
at
about
gd
10­
12
in
the
rat.
Concomitantly,
the
genital
ridges
form
primitive
sex
cords,
which
are
indistinguishable,
male
from
female.
These
small
primitive
indifferent
gonads
are
held
in
place
in
the
abdominal
cavity
by
cranial
suspensory
ligaments
(
from
gonad
cephala
to
diaphragm)
and
by
gubernacular
cords
(
from
gonad
caudally
to
the
base
of
the
abdominal
cavity).
In
the
male,
under
the
initiation
of
the
sry
gene
on
the
Y­
chromosome,
the
primitive
sex
cords
continue
to
proliferate
and
form
the
testis,
including
the
interstitial
Leydig
and
the
intratubular
Sertoli
cells.
The
Leydig
cells
begin
to
produce
T.
The
epididymides,
vas
deferens,
ventral
prostate,
and
seminal
vesicles
are
formed
from
the
embryonic
structures
known
as
the
Wolffian
ducts
in
the
presence
of
fetal
T.
The
secondary
sex
cords
characteristic
of
female
development
(
Müllerian
ducts)
regress
in
the
presence
of
testosterone
and
of
Müllerian
Inhibitory
Substance
(
MIS;
inhibin)
produced
by
fetal
Sertoli
cells,
as
male
sexual
differentiation
proceeds.
DHT
(
produced
locally
by
the
Leydig
cells
from
conversion
of
testosterone
by
the
enzyme
5­
alpha­
reductase),
directs
the
differentiation
of
the
male
genital
tubercle
into
the
external
genitalia
and
the
urogenital
sinus
into
the
prostate
at
Cowper's
(
bulbourethral)
glands;
DHT
also
causes
regression
of
nipple
anlagen
in
the
fetal
male
rodent.
In
the
presence
of
T,
the
cranial
suspensory
ligaments
regress,
as
the
gubernacular
cords
thicken
(
also
under
the
control
of
the
INSL3
gene)
to
cause
the
testes
to
descend
to
the
inguinal
ring
in
utero
(
in
rodents)
and
then
into
the
scrotal
sacs,
i.
e.
testes
descent
during
late
lactation
(
in
rodents).

 
Müllerian
duct
(
female
development).
After
formation
of
the
primitive
sex
cords,
genetically
female
embryos
undergo
differentiation
as
the
primitive
sex
cords
proliferate
to
form
the
ovaries,
whereas
the
secondary
sex
cords
(
Mdllerian
ducts)
form
the
uterus,
oviducts,
and
upper
end
of
the
vagina.
The
lower
end
of
the
vagina
and
external
genitalia
are
formed
from
the
female
genital
tubercle.
The
Wolffian
ducts
regress
in
female
fetuses
in
the
absence
of
T.
Also
in
the
absence
of
T,
the
gubernacular
cords
regress,
while
the
cranial
suspensory
ligaments
are
retained
to
keep
the
ovaries
held
abdominally
just
below
the
kidneys.

 
Pre­
and
postnatal
development.
Visual
examination
of
the
reproductive
tracts
of
both
males
and
females
at
birth
and
during
the
postnatal
period
provides
a
measure
of
both
pre­
and
postnatal
development
as
described
above
and
below.
40
 
Puberty.
Acquisition
of
puberty,
identified
by
the
age
(
in
days)
of
acquisition
of
vaginal
patency
(
VP)
in
offspring
females
and
the
age
(
in
days)
of
acquisition
of
balanopreputial
separation
(
preputial
separation,
PPS)
in
males,
can
be
used
to
compare
the
relative
effects
of
a
compound
on
male
and
female
reproductive
development.
In
the
authors'
laboratory,
the
age
at
acquisition
of
these
indicators
of
puberty
is
consistent,
with
very
tight
variances
intra­
and
inter­
studies.
Acquisition
of
puberty
is
a
critical
endpoint
in
endocrine
disruptor
assays.
Section
2.5.7
contains
a
more
detailed
discussion
of
pubertal
endpoints.

2.5.4
Anogenital
Distance
The
sex
differences
in
anogenital
distance
(
AGD)
at
birth
and
beyond
(
male
AGD
is
approximately
twice
as
long
as
female
AGD
in
rats,
mice,
and
newborn
humans)
are
under
androgen
control,
specifically
dihydrotestosterone
(
DHT)
(
Gray
et
al.
1998;
Gray
and
Ostby,
1998)
and
do
not
appear
to
be
affected
by
estrogens
(
Biegel
et
al.
1998a)
but
are
affected
by
pup
body
weights
(
Ashby
et
al.
1997).
Data
(
Gallavan
et
al.
1998)
from
1501
control
CD
®
(
SD)
rat
pups
indicated
that
a
1
gm
increase
in
body
weight
results
in
a
0.19
mm
increase
in
AGD.
Very
small
(
but
statistically
significant)
increases
in
female
AGD
(
with
no
effects
on
males)
on
pnd
0
have
been
reported
for
dietary
p­
tertoctylphenol
(
OP;
Tyl
et
al.
1999)
and
BPA
(
Tyl
et
al.
2002)
in
a
number
of
dose
groups,
not
dietary
dose
related,
with
no
developmental
or
reproductive
sequelae.
If
AGD
values
are
shorter
in
either
sex
(
especially
if
in
both
sexes)
in
a
treatment
group
with
reduced
pup
body
weights,
it
is
highly
likely
that
the
AGD
effect
is
due
to
the
body
weight
effect,
and
can
be
teased
out
by
analyses
of
covariance
(
ANCOVA)
with
body
weight
as
the
covariate.
The
precision
with
which
laboratories
measure
AGD
on
newborns
ranges
from
use
of
a
dissecting
microscope
with
an
ocular
micrometer
and
eyepiece
grid
or
a
vernier
caliper
(
and
the
pup
flat
on
the
microscopic
platform)
to
hand­
held
pups
and
a
ruler.
Obviously,
the
accuracy
and
variance
of
the
values
will
differ,
depending
on
the
method.
Precise
methods
result
in
very
tight
values,
which
may
result
in
very
similar
statistically
significantly
differences
in
group
means,
for
which
the
biological
significance
and
relevance,
if
any,
are
unknown.
AGD
has
also
been
shown
to
be
significantly
reduced
in
CD
®
(
SD)
newborn
rats
whose
dams
were
on
50%
feed
restriction
from
gd
7
(
Holsapple
et
al.
1998;
Carney
et
al.
1998).

AGD
is
DHT­
mediated,
and
the
endocrine­
mediated
effects
persist
into
adulthood.
However,
since
it
is
confounded
by
body
weight,
the
current
practice
is
to
present
the
data
as
mm,
mm/
cube
root
of
the
body
weight,
and/
or
to
analyze
the
data
by
ANCOVA
(
analysis
of
covariance),
with
the
body
weight
at
measurement
(
birth,
weaning,
etc.)
as
the
covariate.
These
procedures
help
to
account
for
differences
in
body
weight
(
especially
in
groups
where
there
is
systemic
toxicity,
expressed
as
reduced
parental
and
offspring
body
weights).
41
2.5.5
Urethral
Vaginal
Distance
(
UVD)

The
measurement
of
UVD
in
female
rodents
has
been
proposed
as
an
endpoint,
possibly
sensitive
to
levels
of
E2;
analogous
to
AGD
under
DHT
control.
It
is
currently
under
evaluation
in
the
authors'
laboratory.

In
a
study
which
investigated
the
effects
of
gestational
exposure
to
TCDD
on
reproductive
development
of
female
rat
offspring
in
two
strains
of
rats,
LE
and
Holtzmann
(
Gray
and
Ostby,
1995),
administration
of
TCDD
on
gd
15
(
1
:
g/
kg)
produced
malformations
of
the
external
female
genitalia
and
vaginal
orifice,
a
delay
in
puberty,
and
significantly
increased
UVD
in
both
rat
strains.

2.5.6
Retention
of
Nipples
in
Preweanling
Males
This
is
evaluated
usually
in
male
rats
on
pnd
11­
13
(
and
in
male
mice
on
pnd
9­
11)
and
is
DHT­
mediated.
Effects
may
persist
into
adulthood.
In
the
authors'
laboratory,
retained
nipples
have
never
been
observed
in
control
preweanling
CD
®
(
SD)
males,
although
areolae
have
been
observed
in
our
laboratory
in
0­
7.5%
of
control
males
on
pnd
11­
13
(
based
on
examination
of
over
5000
males
in
toto).
This
is
a
sensitive
indicator
of
altered
testosterone
and/
or
DHT
levels
(
effects
on
synthesis,
degradation,
receptor
binding,
transcriptional
activation,
etc.).
Male
pups
with
retained
nipples
(
especially
as
weanlings
and/
or
adults)
are
more
likely
to
exhibit
reproductive
system
malformations,
but
the
correlation
is
not
perfect
(
i.
e.,
some
males
with
nipples
exhibit
no
malformations,
some
males
with
no
nipples
do
exhibit
malformations).
Retention
of
nipples
is
also
a
reasonable,
but
not
infallible,
predictor
of
male
reproductive
malformations
caused
by
perinatal
exposures
at
similar
or
higher
doses
(
McIntyre
et
al.
2001;
McIntyre
et
al.
2002).

One
study
by
You
et
al.
(
1998)
was
designed
to
compare
male
sexual
development
in
two
strains
of
rats
after
gestational
exposure
to
an
antiandrogenic
compound.
In
LE
and
SD
rats
exposed
in
utero
on
gd
14­
18
to
p,
pN­
DDE
(
a
metabolite
of
DDT)
at
0,
10,
or
100
mg/
kg/
day,
the
high
dose
produced
a
significant
14%
decrease
in
AGD
in
male
LE,
and
a
7.8%
decrease
(
not
statistically
significant)
in
SD
rats
on
pnd
2,
with
no
effect
on
AGD
in
females
of
either
strain.
On
pnd
13,
males
from
both
the
low
and
high­
dose
groups
in
the
SD
rats
and
males
from
the
high­
dose
group
only
in
the
LE
rats
had
retained
nipples.
Preputial
separation
occurred
in
control
SD
and
LE
rats
at
about
the
same
time,
but
vaginal
opening
occurred
earlier
in
control
LE
versus
SD
rats.
Regardless,
neither
preputial
separation
nor
vaginal
opening
in
either
strain
were
affected
by
developmental
exposure
to
p,
pN­
DDE,
and
growth
of
male
reproductive
organs
was
also
not
affected.
Flutamide
(
an
androgen
receptor
antagonist),
which
was
used
as
a
positive
control
in
this
study,
produced
decreases
in
AGD,
nipple
retention,
and
changes
in
male
reproductive
organ
growth
(
decrease
in
testis,
ventral
prostate
and
epididymis
weight
in
SD
rats,
and
a
decrease
in
seminal
vesicle
weight
in
LE
rats).
Both
strains
of
rat
had
a
differential
sensitivity
to
the
effects
of
p,
pN­
DDE.
In
LE
but
not
SD
rats,
AGD
was
decreased
by
100
mg/
kg/
day
of
p,
pN­
DDE
and
in
SD
rats,
nipple
retention
was
produced
at
a
lower
dose
(
10
mg/
kg/
day)
than
in
LE
rats.
In
response
to
the
100
42
mg/
kg
dose,
LE
rats
showed
6­
to
8­
fold
higher
serum
concentrations
of
p,
p'­
DDE
than
SD
rats.
One
explanation
for
the
differential
effects
with
p,
pN­
DDE
may
be
due
to
different
tissue
levels
of
p,
pN­
DDE
from
potentially
different
pharmacokinetic
characteristics
in
the
two
strains.

In
a
study
comparing
the
effects
of
developmental
exposure
to
VIN
(
an
antiandrogen
which
has
metabolites
that
bind
to
the
androgen
receptor)
in
Wistar
and
LE
rats,
similarities
and
differences
were
reported
(
Hellwig
et
al.
2000).
Exposure
to
200
mg/
kg/
day
from
gd
14
to
pnd
3
produced
similar
effects
on
male
offspring
of
both
strains,
including
reduced
AGD,
nipple
and
areolae
retention
lasting
into
adulthood,
hypospadias,
penile
hypoplasia
or
development
of
vaginal
pouch,
transient
paraphimosis
(
penile
edema),
and
reduced
function
and
chronic
inflammation
of
the
epididymides,
prostate,
seminal
vesicles,
and
coagulating
glands.
In
adults,
LE
had
testis
atrophy
and
chronic
inflammation
of
the
urinary
bladder,
which
were
not
observed
in
Wistar
offspring.
Exposure
to
12
mg/
kg/
day
produced
only
transient
nipple/
areolae
retention
in
male
offspring
of
both
strains,
but
produced
nipple/
areolae
retention
persisting
into
adulthood
in
a
few
LE
but
no
Wistar
males.
In
addition,
adult
LE
but
not
Wistar
exposed
to
12
mg/
kg/
day
had
slightly
reduced
prostate,
seminal
vesicle
and
coagulating
gland
weights.
Overall
there
were
more
similarities
than
differences
in
the
effects
of
VIN
in
both
strains,
and
the
NOAEL
was
12
and
6
mg/
kg
in
Wistar
and
LE
rats,
respectively.

2.5.7
Puberty
Acquisition
of
puberty
can
be
determined
in
both
females
and
males
by
a
number
of
physical
changes.
For
females,
vaginal
patency
and
age
of
first
estrus
are
most
often
used,
whereas
in
males,
preputial
separation
is
most
often
monitored.
In
both
sexes,
acquisition
of
puberty
is
affected
by
body
weight,
so
the
current
approach
is
to
covary
the
age
at
acquisition
by
the
body
weight
at
acquisition,
at
an
arbitrary
age
during
the
time
of
acquisition,
or
by
some
measure
of
weight
gain
during
the
postlactational
or
prepubertal
period
(
the
selection
of
the
end
date
for
weight
gain
is
problematic).
Small
changes
in
acquisition
(#
3
days)
may
indicate
body
weight­
related
delays
in
development;
large
changes
(
accelerations
or
delays
of
$
4
days)
most
likely
indicate
effects
from
endocrine
disruption,
especially
in
the
absence
of
body
weight
effects.
Minor
delays/
accelerations
in
puberty
in
the
BPA
rat
study
(
Tyl
et
al.
2002)
were
presented
and
analyzed
as
absolute
values,
and
as
values
covaried
by
body
weight
at
acquisition
and
at
an
arbitrary
age.

Statistically
significant
differences
in
age
at
acquisition
of
puberty
may
indicate
endocrine­
mediated
effects,
especially
if
the
effects
are
different
for
the
sexes
(
e.
g.,
VP
is
delayed
and
PPS
is
accelerated
or
unchanged,
VP
is
accelerated
and
PPS
is
delayed
or
unchanged,
etc.)
and
if
the
effects
are
profound
(
acceleration
or
delay
of
many
days
versus
only
a
few
days).
However,
acquisition
of
developmental
landmarks
is
dependent
on
both
age
and
weight
(
i.
e.,
heavier
animals
acquire
the
landmark
earlier,
while
lighter
animals
acquire
the
landmark
later),
but
lighter
animals
do
acquire
the
landmark
(
unless
there
is
another
cause
for
the
delay)
and
in
many
cases
acquire
the
43
landmark
at
a
lighter
weight
than
the
heavier
animals.
This
observation
is
consistent
with
the
recognition
by
the
EPA
(
1996,
p.
56295)
that
"
body
weight
at
puberty
may
provide
a
means
to
separate
specific
delays
in
puberty
from
those
that
are
related
to
general
delays
in
development."
The
significance
(
i.
e.,
the
consequence,
if
any)
and
"
the
biologic
relevance
of
a
change
in
these
measures
of
a
day
or
two
is
unknown"
(
EPA,
1996,
p.
56295).

The
recognition
that
body
weight
is
important
in
analyzing
and
understanding
acquisition
of
puberty
is
strengthened
by
the
work
of
Kennedy
and
Mitra
(
1963)
and
Carney
et
al.
(
1998)
who
showed
that
body
weight
and
food
intake
are
factors
in
the
initiation
of
puberty
in
the
rat,
and
by
the
work
of
Holsapple
et
al.
(
1998)
who
put
groups
of
26
timed­
mated
SD
rats
on
standard
diets
at
100%
(
control),
70%,
or
50%
of
historical
control
feed
intake
levels
from
gd
7
through
weaning
on
pnd
21.
Selected
weanlings
were
continued
on
feed
restriction
until
ten
weeks
(
with
100%
feed
from
ten
to
20
weeks
of
recovery)
or
until
20
weeks
of
age,
with
necropsy
of
all
offspring
at
20
weeks
of
age.
Feed
restriction
resulted
in
reduced
weight
gains
for
dams
and
pups
related
to
the
degree
of
restriction.
In
both
the
50%
and
70%
feed
restriction
groups,
gestation
length
was
significantly
increased,
and
age
at
VP
and
PPS
was
also
delayed
(
by
one
day
at
70%
restriction
and
by
six
days
at
50%
restriction
for
both
parameters).
AGD
at
birth
was
significantly
reduced
in
both
sexes
in
the
50%
restriction
group,
but
AGD:
body
weight
ratios
were
essentially
identical
across
groups,
indicating
that
smaller
(
low
body
weight)
pups
had
shorter
AGDs
and
that
the
effects
were
proportional.
The
authors
conclude
that
"
these
results
show
that
certain
reproductive
and
developmental
endpoints
are
altered
by
feed
restriction
in
the
range
relevant
to
common
testing
scenarios"
(
Holsapple
et
al.
1998).

2.5.7.1
Vaginal
Patency
in
Females
In
females,
acquisition
of
puberty
is
indicated
by
vaginal
opening
or
patency
(
VP),
dependent
on
E2
and
resulting
from
E2­
dependent
cornification
of
the
vaginal
seam.
In
control
CD
®
(
SD)
rats
in
the
authors'
laboratory,
the
grand
mean
age
at
VP
is
31.1
days
(
based
on
16
studies
from
1996
to
2000).
VP
may
be
observed
first
as
the
appearance
of
a
small
"
pin
hole(
s)"
or
perforations
but
is
recorded
as
acquired
when
vaginal
opening
is
complete.
Vaginal
threads
across
the
vaginal
opening
may
be
temporary
or
persistent
(
Wolf
et
al.
1999;
Flaws
et
al.
1997).

Vaginal
opening
may
be
advanced
by
estrogenic
compounds,
and
estrogen
receptor
modulators
either
advanced
or
delayed
with
various
environmental
chemicals
(
see
Review
by
Goldman
et
al.
2000).
In
a
study
of
the
effects
of
BPA
on
sexual
development
in
two
strains
of
rats,
SD
and
Alderley
Park
(
AP),
Tinwell
et
al.
(
2002)
found
strain­
related
differences
in
VP.
Vaginal
opening
was
significantly
delayed
in
AP
rats
and
not
SD
rats
exposed
to
BPA.
There
was
no
effect
on
age
of
first
estrus
(
an
explanation
of
this
endpoint
follows).
44
2.5.7.2
Age
of
First
Estrus
in
Females
On
or
within
a
few
days
of
VP,
the
female
exhibits
her
first
estrus,
so
age
at
first
estrus
(
absolute
age
and/
or
interval
from
VP
to
first
estrus)
is
also
useful.
Late
follicular
growth
of
the
first
ovulatory
cells
is
stimulated
about
the
time
of
vaginal
opening,
although
there
is
some
variation
in
the
initial
release
of
oocytes.
Following
vaginal
opening,
daily
vaginal
smears
are
monitored
to
determine
the
age
of
first
estrus
and/
or
first
vaginal
cycle.
Irregular
estrous
cycles
are
often
seen
in
the
immediate
postpubertal
period
(
Goldman
et
al.
2000).

2.5.7.3
Preputial
Separation
in
Males
Acquisition
of
puberty
in
males
is
indicated
by
preputial
separation
(
PPS;
balanopreputial
separation)
or
separation
of
the
foreskin
of
the
penis
from
the
glans.
PPS
is
dependent
on
androgens.
PPS
is
a
process
that
leads
to
the
cleavage
of
the
epithelium
through
cornification,
forming
the
squamous
lining
of
the
prepuce
of
the
penis
(
Goldman
et
al.
2000).
As
a
sign
of
puberty
and
an
essential
prerequisite
for
further
development
of
the
ejaculatory
process,
PPS
has
been
used
as
a
reliable,
noninvasive
endpoint
by
which
to
monitor
rodent
pubertal
development
and
perturbations
of
this
process.
This
landmark
of
acquisition
generally
occurs
during
the
peripubertal
period
(
pnd
36­
55
or
60;
Stoker
et
al.
2000).
In
control
CD
®
(
SD)
rats
in
the
authors'
laboratory,
the
grand
mean
age
at
PPS
is
41.9
days
(
based
on
16
studies
from
1996
to
2000).

Estrogenic
and
anti­
androgenic
compounds
have
been
shown
to
delay
PPS,
while
androgen
receptor
agonists
accelerate
PPS
(
see
review
by
Stoker
et
al.
2000).
Tinwell
et
al.
(
2002)
found
that
BPA
had
no
effect
on
PPS
in
two
strains
of
rats
(
AP
and
SD)
at
a
dose
that
delayed
vaginal
opening
in
female
AP
rats
only.
In
AP
rats,
vaginal
opening
was
at
33.8
±
0.8
days
in
control
animals,
compared
to
35.4
±
0.6
days
in
rats
exposed
to
50
mg/
kg/
day
BPA.
Sensitivity
to
BPA
was
found
to
be
not
only
strain
related
but
sex
related.

2.5.8
Estrous
Cyclicity
and
Ovulation
Rate
in
Postpubertal
Females
After
the
initial
release
of
ova,
female
rats
begin
to
exhibit
four­
to
five­
day
estrous
cycles,
with
accompanying
changes
in
vaginal
cytology
and
circulating
hormones.
The
acquisition
of
estrous
cyclicity
results
from
shifts
in
the
hypothalamic­
pituitary­
ovarian
endocrine
axis
and
is
the
culmination
of
the
maturation
of
reproductive
processes
that
began
prenatally.
As
indicated
above,
irregular
estrous
cycles
are
more
common
in
the
first
weeks
after
acquisition
of
puberty.

Ovulation
rate
is
affected
by
the
dose
and
ratio
of
FSH
to
LH,
stage
of
the
cycle,
and
age
of
the
female.
Large
genetic
differences
in
ovulation
rate
exist
between
strains
of
mice
and
in
response
to
exogenous
gonadotropins
(
Spearow
et
al.
1999;
Spearow
and
Barkley,
1999).
45
Ovulation
rate
(
the
number
of
eggs
ovulated
per
female)
is
not
included
in
the
endpoints
discussed
in
this
White
Paper
because
the
protocols
of
the
proposed
EDSP
assays
preclude
measurement
of
ovulation
rate
in
order
to
measure
other
relevant
endpoints.
Ovulation
rate
is
based
on
the
number
of
corpora
lutea
counted
on
the
ovaries.
These
are
postovulation
ruptured
follicles
(
one
per
ovulated
ovum)
producing
large
amounts
of
P4
and
lesser
amounts
of
E2,
to
prepare
the
uterus
for
implantation
of
the
conceptuses.
However,
maternal
ovarian
corpora
lutea
involute
beginning
at
delivery
of
offspring
(
with
involution
completed
on
or
about
pnd
4)
to
become
corpora
albicans,
indistinguishable
from
corpora
albicans
from
previous
ovulation
cycles.
All
of
the
studies
in
Tiers
I
and
II
that
involve
production
of
offspring
require
that
the
dams
remain
with
their
pups
through
lactation
to
weaning.
The
parental
females
are
necropsied
at
the
weaning
of
their
litters
on
pnd
21,
when
the
corpora
lutea
are
no
longer
present
on
the
ovaries.
The
inability
to
collect
ovarian
corpora
lutea
counts
in
these
studies
also
precludes
calculation
of
percent
preimplantation
loss:

What
can
be
ascertained,
and
is
therefore
included
in
the
list
of
endpoints
to
be
discussed,
is
percent
postimplantation
loss,
which
is
based
on
the
number
of
uterine
implantation
sites
(
i.
e.,
the
number
of
conceptuses
implanted;
these
"
nidation
scars"
persist
at
least
40
days
after
delivery)
and
the
number
of
total
pups
delivered.
Both
of
these
parameters
are
present
and
recorded
in
the
Tier
I
and
II
studies
involving
generation
of
offspring.
The
calculation
for
percent
postimplantation
loss
is:

In
studies
comparing
estrous
cycles
across
rat
strains,
there
were
strain­
related
differences
in
estrous
cyclicity
in
response
to
food
deprivation
(
Tropp
and
Markus,
2001).
Prior
to
food
restriction,
Brown
Norway
rats
had
irregular
estrous
cycle
patterns
while
SD,
LE,
and
F344
rats
had
regular
estrous
cycle
patterns.
By
day
5
of
food
deprivation,
75%
of
F344
rats
and
100%
of
Brown
Norway
rats
stopped
cycling
and
SD
and
LE
rats
were
unaffected
(
the
animals'
weights
were
reduced
to
85%
of
ad
libitum
body
weight).
It
is
possible
that
SD
and
LE
rats,
which
have
generally
larger
body
masses
may
have
more
energy
store
and
therefore
be
less
sensitive
to
changes
in
body
weight.
Another
possibility
is
that
sensitivity
to
food
deprivation
is
higher
in
inbred
versus
outbred
strains.
However,
given
the
fact
that
food
deprivation
schedules
were
adjusted
to
account
for
differences
in
initial
body
weight,
it
would
seem
unlikely
that
simple
strain
differences
in
body
weight
account
for
the
results.
These
data
suggest
outbred
strains
selected
for
larger
litter
size
are
relatively
resistant
to
the
disruption
of
estrous
cyclicity
by
dietary
restriction.
46
In
a
study
comparing
estrous
cycles
in
Lewis
and
F344
rats,
by
obtaining
vaginal
smears
and
quantitating
E2,
P,
FSH,
and
LH
levels
at
different
phases
of
the
cycle,
Smith
et
al.
(
1994)
reported
that
metestrus
was
significantly
longer,
while
diestrus
and
estrus
were
significantly
shorter
in
Lewis
rats
compared
to
F344
rats.
Proestrus
was
similar
in
both
strains.
During
estrus,
E2
levels
were
significantly
higher
in
Lewis
compared
to
F344
rats,
and
P
levels
were
significantly
higher
in
all
stages
of
the
estrous
cycle
in
Lewis
compared
to
F344
rats.
LH
and
FSH
levels
did
not
differ
between
strains
at
any
stage
of
the
estrous
cycle.
The
authors
suggest
that
elevated
E2
and
P
levels
may
affect
corticosterone
levels
which
could
affect
hypothalamic­
pituitary­
adrenal
axis
responsiveness.

In
response
to
endocrine­
disrupting
chemicals,
strain
differences
in
the
ovarian
cycle
have
been
reported.
Cooper
et
al.
(
2000)
reported
that
LE
rats
were
more
sensitive
than
SD
rats
to
atrazine­
induced
disruption
of
the
ovarian
cycle.
In
addition,
Ando­
Lu
et
al.
(
1998)
found
that
in
aging
Donryu
rats,
estrous
cycle
abnormalities
(
e.
g.
persistent
estrus)
were
more
common
than
in
F344
rats.
Finally,
in
a
study
by
Eldridge
et
al.
(
1994),
atrazine
administration
to
SD
and
F344
rats
for
up
to
12
months
produced
changes
in
estrous
cyclicity
in
SD
rats
(
increased
the
number
of
days
of
vaginal
estrus),
increased
E2,
decreased
P,
and
increased
incidence
of
mammary
tumors
in
SD
rats
only,
with
no
significant
treatment­
related
effects
in
F344
rats.

BPA,
an
environmental
estrogen,
has
been
found
to
stimulate
Prl
secretion
in
F344
but
not
SD
rats
(
Steinmetz
et
al.
1997).
More
recently,
Long
et
al.
(
2000)
found
that
BPA
increased
DNA
synthesis
and
cell
proliferation
in
the
vaginal
epithelium
of
F344
rats
but
not
SD
rats.
Thus,
the
rat
vagina,
an
estrogen
target
tissue,
is
more
sensitive
to
the
effect
of
BPA
in
a
strain­
specific
manner.
Long
et
al.
(
2000)
also
showed
that
F344
and
SD
rats
showed
no
difference
in
clearance
of
3H­
BPA
from
the
blood,
concentration
or
affinity
of
estradiol
receptor,
or
induction
of
early
gene
c­
fos
in
response
to
BPA.
Since
BPA
increased
vaginal
cell
proliferation
and
DNA
synthesis
in
F344
but
not
in
SD,
these
data
show
that
strains
differ
in
the
intermediate
effects
of
these
xenoestrogens
downstream
of
the
ER.

Differences
in
estrous
cyclicity
have
been
reported
in
outbred
strains
of
mice
which
have
been
selected
for
large
litter
size,
high
embryo
survival,
or
small
litter
size
(
Barkley
and
Bradford,
1981;
DeLeon
and
Barkely,
1987).
Selection
for
large
litter
size
(
Line
S1)
and
high
embryo
survival
(
Line
E)
increased
the
regularity
of
estrous
cycles,
and
selection
for
small
litter
size
(
CN)
dramatically
decreased
the
regularity
of
estrous
cycles.
Therefore,
the
BN
rat,
F344
rat,
or
the
CN
mouse
may
provide
better
animal
models
than
strains
that
have
been
bred
for
large
litter
size.

2.5.9
Andrology
Depending
on
the
age
of
the
male
rodent
when
sampled,
the
cauda
epididymis
(>
80
days
old)
or
the
entire
epididymis
(
65­
80
days
old)
is
evaluated
for
total
number
of
sperm
per
cauda
or
per
gram
cauda,
motility
and
progressive
motility
(
as
percent
of
total
sperm
examined.
This
evaluation
must
be
done
within
two
minutes
of
animal's
demise
47
with
microscope
slide
and
buffer
kept
at
37
°
C).
Percent
malformed
sperm
should
be
examined
usually
manually
by
microscopic
examination
of
200
fixed
and
stained
(
Eosin
Y)
sperm
per
male.
In
addition,
one
testis
at
necropsy
should
be
frozen
and
subsequently
homogenized
in
buffer
and
evaluated
for
homogenization­
resistant
spermatid
head
counts
(
SHC)
to
calculate
daily
sperm
production
(
DSP)
and
efficiency
of
DSP.

For
andrological
studies,
epididymal
sperm
counts
and
testicular
homogenizationresistant
spermatid
head
counts
are
found
to
be
good
markers
for
altered
spermatogenesis.
Wilkinson
et
al.
(
2000)
compared
the
outbred
strains
of
Wistar
and
SD
with
the
inbred
strain
Dark
Agouti
(
DA).
While
a
small
number
of
SD
were
used
in
this
study,
the
DA
rat
has
lower
absolute
and
relative
(%
body
weight)
testes
weight,
and
more
variability
in
sperm
counts
but
there
was
no
significant
difference
in
testicular
histology,
sperm
count
per
gram
of
testis,
or
epididymal
sperm
count.
There
were
also
no
differences
in
weights
(
relative
to
body
weight)
of
the
epididymis,
seminal
vesicles,
or
ventral
prostate
or
of
testosterone
values
for
whole
blood.
DA
rats
are
deficient
in
CYP2D1
activity,
and
several
P450
cytochromes
may
also
be
absent.

In
the
Tinwell
et
al.(
2002)
study
that
found
that
the
weak
xenoestrogen
BPA
had
no
effect
on
PPS
in
two
strains
of
rats
(
AP
and
SD),
they
reported
that
50
mg/
kg
BPA
decreased
total
sperm
count
and
daily
sperm
count
in
AP
rats
but
not
in
SD
rats.
Thus,
there
were
strain­
related
differences
in
the
effects
of
BPA
in
rats.

Apostoli
et
al.
(
1998),
in
a
review
article
on
the
toxicology
of
lead,
stated
that
SD
rats
appeared
to
be
relatively
resistant
to
the
toxicological
effects
of
lead.
However,
in
general,
lead
impaired
spermatogenesis
and
decreased
androgens
in
other
rat
strains
(
e.
g.
Wistar
and
Charles
Foster
rats).
Concentrations
of
blood
lead
>
40
:
l/
dl
were
associated
with
decreased
sperm
counts,
volume,
motility,
morphology
and
endocrine
effects.

In
mice,
strain
differences
in
andrological
parameters
have
been
observed.
For
example,
CD­
1
mice
have
been
shown
to
be
much
greater
than
16­
fold
more
resistant
than
C57Bl/
6J
(
B6)
or
C17/
Jls
strain
mice
to
the
inhibition
of
spermatogenesis
by
pubertal
exposure
to
estradiol
(
Spearow
et
al.
1999).
Additional
studies
in
Spearow's
laboratory
have
confirmed
these
observations
and
have
also
shown
CD­
1
mice
to
be
much
more
resistant
to
the
inhibition
of
testes
weight,
elongated
spermatids
per
seminiferous
tubule
crosssection
and
epididymal
sperm
counts
than
outbred
wild­
derived
Mus
spretus/
RP/
Jls
mice.

2.5.10
Organ
Weights
and
Histopathology
°
Reproductive
(
including
accessory
sex
organ
weights).
Reproductive
organ
weights
should
be
obtained
at
adulthood
and
should
include:
(
a)
ovaries
and
uterus
for
females
and
(
b)
testes,
epididymides
(
total
and
separated
into
caput,
corpus
and
cauda),
prostate
(
whole,
and
dorsolateral
and
ventral
lobes
separately;
dissection
may
be
postfixation),
seminal
48
vesicles,
coagulating
glands,
preputial
glands,
bulbourethral
(
Cowper's)
glands,
and
levator
ani/
bulbocavernosus
(
LABC)
complex
for
males.

°
Thyroid.
Thyroid
hormones
(
T3
and
T4)
are
necessary
for
normal
growth,
development,
differentiation,
and
regulation
of
most
organ
systems
(
Goldman
et
al.
2000;
Stoker
et
al.
2000).
Disruption
of
the
feedback
control
of
thyroid
function
may
result
in
either
a
hypertrophic
(
goiter)
or
hypotrophic
thyroid,
depending
on
the
mechanism
of
disruption.
These
changes
would
be
evident
in
the
weight
of
the
thyroid
gland.
Since
the
thyroid
gland
surrounds
the
trachea,
the
thyroid
plus
embedded
trachea
is
fixed
and
the
trachea
dissected
away
post
fixation.
The
thyroid
can
then
be
weighed
with
little
or
no
damage
to
the
organ
for
subsequent
histopathology.

°
Systemic
(
liver,
kidneys,
brain,
etc.).
Systemic
organ
weights
should
be
obtained
at
adulthood
in
both
sexes
and
should
include
liver,
kidneys,
adrenal
glands,
pituitary,
brain
(
regions),
etc.
Comparison
of
the
effect
of
the
test
compound
on
these
organ
weights
(
absolute
and
relative)
to
effects
on
reproductive
organ
weights
will
provide
a
more
complete
characterization
of
toxicity
and
suggest
whether
observed
toxicity
is
more
or
less
targeted
to
the
endocrine
system.

°
Absolute
and
relative
to
body
weight
(
and
brain
weight).
Organ
weights
(
both
reproductive
and
systemic)
should
be
presented
as
absolute
and
relative
to
terminal
body
weight
and
brain
weight.
Relative
organ
weights
will
correct
for
effects
on
body
weights
(
i.
e.,
systemic
toxicity).
Brain
weight
is
generally
considered
more
stable
than
body
weight
after
exposure
to
exogenous
compounds
and
provides
a
basis
for
determination
if
changes
in
organ
weights
are
primary
or
secondary
to
altered
body
weights.

In
a
study
by
Putz
et
al.
(
2001),
the
estrogenic
effects
of
neonatal
exposure
to
ß­
estradiol­
3­
benzoate
(
EB)
were
studied
in
two
rat
strains,
SD
and
F344.
Neonatal
rats
were
injected
with
EB
(
over
a
7­
log
range
of
doses
from
0.015
:
g/
kg/
day
to
15
mg/
kg/
day
in
SD
and
a
5­
log
range
of
doses
from
0.15
:
g/
kg/
day
to
1.5
mg/
kg/
day
in
F344)
on
pnd
1,
3,
and
5.
While
F344
were
not
examined
on
pnd
35,
SD
male
rats
on
pnd
35,
exhibited
significant
increases
in
absolute
and
relative
testis
and
epididymis
weights
at
the
low
dose,
0.015
:
g/
kg/
day,
and
significant
reductions
at
higher
doses
(
1.5
and
15
mg/
kg/
day).
Since
hepatic
testosterone
hydroxylase
activity
was
increased
in
the
low­
dose
animals,
it
may
have
advanced
puberty,
therefore
resulting
in
increased
organ
weights.
On
pnd
90,
in
SD
males
exposed
neonatally
to
the
highest
dose
used
in
both
strains
(
1.5
mg/
kg/
day),
there
were
significant
reductions
in
absolute
and
relative
seminal
vesicle,
and
coagulating
gland
weights,
but
not
in
testis
or
epididymis
weights.
SD
rats
on
pnd
90
also
showed
an
increase
in
testis
and
epididymis
weights
at
the
lowest
dose
(
at
one
order
of
magnitude
lower
than
the
increase
observed
on
pnd
35).
This
dose
was
not
tested
in
F344
rats.
In
F344
males,
the
reduction
in
male
reproductive
organ
weights
(
absolute
and
relative
to
body
weight)
at
pnd
90
was
greater
at
the
highest
dose
(
1.5
mg/
kg/
day)
than
in
SD
rats.
Whereas
relative
testicular
weights
49
were
44%
of
controls
in
F344
rats,
they
were
67%
of
controls
in
SD
rats.
Similarly,
1.5
mg/
kg
EB
reduced
epididymal
weights
to
36%
of
controls
in
F344
versus
87%
of
controls
in
SD
rats.
Pnd
90
testis
and
epididymal
weights
were
reduced
much
more
by
1.5
mg/
kg
EB
in
F344
than
at
a
10­
fold
higher
dose
(
15
mg/
kg)
in
SD
rats.
Thus,
SD
rats
were
greater
than
10­
fold
more
resistant
than
F344
to
the
inhibition
of
testes
weight
by
EB.

Strain­
related
differences
in
absolute
pituitary
weights
have
been
reported
in
ovariectomized
rats
exposed
to
E2
(
silastic
implants)
for
10
or
20
days
(
Schechter
et
al.
1987).
Pituitary
weights
were
dramatically
increased
in
F344
rats,
with
comparatively
minimal
effects
in
SD
rats,
and
Prl
levels
were
dramatically
increased
in
F344
rats
($
1000
fold),
while
only
moderately
increased
in
SD
rats
(
100
fold).
In
addition,
E2
implants
in
F344
strain
rats
produced
a
dramatic
hyperplasia
of
anterior
pituitary
lactotropes,
activation
of
phagocytic
folliculo­
stellate
cells
(
FS),
increase
of
cells
positive
for
basic
fibroblast
growth
factor,
and
reorganization
of
the
blood
supply
from
vessels
in
the
adjacent
meninges.
Estradiol­
treated
SD
rats
did
not
show
comparable
responses
(
Schechter
and
Weiner
1991).
Pituitary
weights
were
also
different
across
strains
in
ovariectomized
rats
exposed
to
E2
(
10
mg
s.
c.
pellet)
for
four
weeks
(
Yin
et
al.
2001).
Control
pituitary
gland
weights
were
the
lowest
in
Brown­
Norway
rats
(
4.4
±
0.2
mg),
and
more
than
two­
fold
higher
in
Wistar
rats
(
13.0
±
2.1
mg);
F344
control
pituitary
weights
were
7.5
±
0.1
mg,
and
Donryu
10.7
±
1.0
mg.
After
exposure
for
four
weeks
to
E2,
there
was
a
significant
>
3
fold
increase
in
pituitary
weights
in
F344
rats,
a
significant
>
0.5
fold
increase
in
pituiary
weights
in
Brown­
Norway
rats,
and
no
difference
in
pituitary
weights
of
Wistar
and
Donryu
rats.
The
F344
strain
was
the
most
susceptible
to
estrogen
induction
of
pituitary
tumorigenesis,
followed
by
Wistar
and
Brown­
Norway.
The
work
of
Schechter
et
al.
and
Yin
et
al.
demonstrates
that
the
pituitary
gland
of
F344
rats
is
more
sensitive
to
the
effects
of
E2.

Differential
effects
of
DES
in
particular
rat
strains
have
been
demonstrated
in
studies
by
Gorski
et
al.,
who
showed
strain
differences
in
estrogen
dependent
pituitary
mass
(
Edpm)
and
pituitary
tumor
growth
(
Wendell
et
al.
1996;
Wendell
and
Gorski
1997;
Chun
et
al.
1998;
Wendell
et
al.
2000).
While
F344
are
highly
susceptible
to
DESinduced
pituitary
growth/
tumors,
Brown
Norway
(
BN)
and
SD
rats
are
highly
resistant.
Following
DES
treatment,
F344
strain
rats
and
F344
x
BN
F2
rats
with
largest
pituitary
tumors
showed
a
reduction
in
retinoblastoma
susceptibility
gene
product
(
pRb)
(
Chun
et
al.
1998).
QTL
linkage
analysis
in
a
F344
x
BN
F2
mapped
several
additive
and
epistatic
loci
controlling
Edpm,
including
susceptibility
alleles
from
F344
and
from
BN
(
Wendell
and
Gorski
1997).
Through
QTL
mapping
in
a
(
F344
x
BN)
x
F344
backcross
Wendell
et.
al.
(
2000)
showed
that
several
QTL
including
Edpm2­
1,
Edpm3,
Edpm5,
and
Edpm9­
2
all
had
significant
effects
on
pituitary
mass.
While
Edpm2­
1
and
Edpm9­
2
primarily
affected
DNA
content,
Edpm5
primarily
affected
hemoglobin/
DNA
ratio,
and
Edpm3
affected
all
of
these
traits
equally
(
Daun
et
al.
2000).
These
data
defining
genes
controlling
susceptibility
to
estrogenic
agent­
induced
tumors
among
genetically
defined
strains
provide
a
powerful
tool
for
understanding
genetic
differences
in
susceptibility
to
endocrine
disruption
by
estrogenic
agents.
These
data
also
have
value
as
historic
controls.
Since
these
studies
used
genetically­
defined
isogenic
parental
strains,
they
50
can
easily
be
repeated
and
enhance
the
identification
of
genes
controling
susceptibility
to
environmentally­
induced
disease
in
humans
as
well.

In
a
review
by
Kacew
et
al.
(
1995),
strain­
related
differences
in
mammary
tumorigenesis
were
summarized.
SD
rats
are
more
susceptible
to
mammary
tumorigenesis
after
exposures
to
2­
acetylaminofluorene,
1,4­
bis(
4­
fluorophenyl)­
2­
propynyl­
N­
cyclooctyl
carbamate,
and
atrazine
than
were
F344
rats.
In
addition,
Wistar
and
SD
rats
were
more
susceptible
than
Copenhagen
or
LE
rats
to
the
effects
of
DMBA
(
7,12­
dimethylbenz
(
a)
anthracene),
while
Wistar
were
more
sensitive
than
LE
to
the
effects
of
2­
acetylaminofluorene.
Therefore,
there
is
an
inherent
difference
in
mammary
tissue
sensitivity
among
rat
strains.
In
males,
the
sensitivity
of
the
tumorigenic
response
in
the
prostate
of
F344,
ACI,
Lewis,
CD
and
Wistar
rats
to
3,2'­
dimethyl­
4­
aminobiphenyl
(
DMAB)
was
ordered
as
follows:
F344>
ACI>
Lewis>
CD;
the
Wistar
rats
were
insensitive
(
Shirai
et
al.
1990).

Strain
differences
in
susceptibility
to
the
effect
of
chemicals
on
testis
weight
in
mice
have
been
reported.
Oishi
(
1993)
found
that
administration
of
di­
2­
ethylhexyl
phthalate
(
DEHP)
(
0,
0.1,
0.2,
0.4,
and
0.8%
in
feed,
for
two
weeks)
to
two
strains
of
mice
(
Jcl:
ICR
and
CD­
1)
caused
significant
increases
in
absolute
and
relative
liver
weights
in
both
strains
at
the
highest
doses
and
reduced
testicular
weights
in
CD­
1
mice
only
(
at
a
dose
as
low
at
0.2%).
DEHP
was
associated
with
testicular
atrophy
in
CD­
1
mice
only,
at
the
doses
administered.

In
another
report
demonstrating
strain
differences
in
mice,
Nagao
et
al.
(
2002)
exposed
male
C57BL/
6N
and
ICR
mice
to
BPA
at
0,
2,
20
or
200
:
g/
kg/
day
for
various
periods
encompassing
adulthood,
the
juvenile
period
(
just
after
weaning),
and
the
embryo/
fetal
period.
Though
BPA
did
not
affect
male
reproductive
organ
weights
during
any
dose/
exposure
period,
E2
(
10
:
g/
kg
from
pnd
27
to
48,
as
a
positive
control)
produced
significant
decreases
in
absolute
and
relative
testes,
epididymides,
and
seminal
vesicle
weights
(
as
low
as
55%
of
control
values)
compared
to
controls
in
C57BL/
6N
mice,
while
ICR
mice
were
unaffected.
Histopathology
showed
that
10
:
g/
kg
E2
was
without
effect
on
ICR
males,
while
B6
males
showed
slight
to
severe
effects
on
elongated
spermatids,
decreased
epididymal
sperm,
and
seminal
vesicle
atrophy.
Thus,
C57BL/
6N
mice
were
more
sensitive
than
ICR
mice
to
the
effects
of
E2.
These
data
are
consistent
with
data
presented
by
Spearow
et
al.
(
1999).

In
male
mice,
strain­
related
differences
in
susceptibility
to
endocrine
disruption
by
endocrine­
active
chemicals
have
been
reported
(
Spearow
et
al.
1999;
Spearow
et
al.
2001).
Mouse
strains
included
B6
(
an
inbred
strain),
CD­
1
(
outbred,
with
larger
litter
size),
C17/
Jls
(
bred
randomly,
then
inbred),
and
S15
(
bred
for
large
litters,
then
inbred).
In
control
mice,
testicular
weight
(
absolute
and
relative
to
body
weight)
was
higher
in
CD­
1
and
S15
strains
selected
for
larger
litter
sizes.
In
juvenile
male
mice
exposed
for
three
weeks
to
E2
(
at
0,
2.5,
10,
20
or
40
:
g
in
silastic
implants),
B6
and
C17/
Jls
were
sensitive
to
E2,
showing
a
maximal
suppression
of
testis
weight
and
spermatogenesis
even
at
the
lowest
dose
of
E2
(
2.5
:
g),
with
no
effect
on
testis
weight
or
spermatogenesis
in
CD­
1
or
S15
at
any
dose
up
to
10
:
g
E2.
Thus,
Spearow
et
al.
51
(
2001)
demonstrated
genetic
differences
in
sensitivity
to
estrogen
that
may
be
related
to
breeding
animals
for
high
fecundity.

Additional
studies
exposing
juvenile
male
mice
from
3
to
7
weeks
of
age
to
0,
0.625,
2.5,
10,
40
and
160
:
g
E2
in
silastic
implants
showed
a
dramatic
strain
difference
in
susceptibility
to
endocrine
disruption
(
Spearow
et
al.
2002;
Spearow
et
al.
2003).
CD­
1
mice
were
greater
than
195­
fold
more
resistant
than
B6
mice
to
the
disruption
by
E2
of
testes
weight,
number
of
elongated
spermatids
per
tubule
and
Spermatogenic
Index
(
SI).
CD­
1
strain
mice
were
also
>
41
times
more
resistant
than
B6
strain
mice
to
the
inhibition
by
E2
of
epididymal
sperm
counts,
and
were
more
resistant
than
outbred
wild­
derived
Mus
spretus
mice
to
the
disruption
of
testes
weight
and
spermatogenesis
by
E2.

In
a
separate
experiment,
immature
B6
males,
outbred
CD­
1,
CD­
1
derived
inbred
strains
CD10
and
CD3,
and
F1
crosses
were
implanted
subcutaneously
at
3
weeks
of
age
with
silastic
implants
containing
0,
2.5,
or
40
:
g
E2
(
Spearow
et
al.
2003).
Susceptibility
to
endocrine
disruption
by
estrogenic
agents
(
SEDE)
was
evaluated
4
weeks
later
by
determining
testicular
weight,
histology
and
epididymal
sperm
counts.
The
effects
of
Strain,
Dose
of
E2
and
the
Strain
x
E2
Dose
interaction
were
all
highly
significant
on
testes
weight
(
TW),
seminiferous
tubule
diameter,
elongated
spermatids
per
tubule,
spermatogenic
index
and
epididymal
sperm
counts
(
P<
0.0003)
(
Spearow
et
al.
2003).
Resistance
of
mouse
strains
to
disruption
of
testes
weight
by
E2
ranked:
B6
<<
CD3
<
CD10
<
CD­
1.
While
CD10
x
B6
F1
mice
showed
limited
hybrid
vigor
or
heterosis
for
resistance
to
the
disruption
of
testes
weight
by
E2,
the
CD10
x
CD3
F1
showed
a
large
amount
of
heterosis
in
this
trait.
The
data
suggest
that
susceptibility
to
the
disruption
of
testes
weight
by
estrogen
is
controlled
by
additively
and
non­
additively
acting
genes.
Thus
the
observed
>
16­
fold
to
>
195­
fold
strain
differences
in
susceptibility
to
the
disruption
of
spermatogenesis
and
testes
weight
between
strains
questions
the
adequacy
of
the
standard
10­
fold
within­
species
safety
factor
if
only
genetically
resistant
strain(
s)
are
used
for
toxicological
safety
testing.

2.5.11
Behavioral
Assessments/
Clinical
Observations
Courtship
and
mating
behaviors
in
both
sexes,
and
maternal
and
neonatal
behaviors
involving
nesting,
pup
retrieval,
and
nursing
are
also
under
the
control
of
the
endocrine
system.
Qualitative
evaluation
of
these
behaviors,
as
they
affect
viability
and
ability
to
thrive,
provides
another
measure
of
possible
endocrine­
disrupting
activity
of
a
test
compound.
Strain­
related
differences
in
lordotic
behavior
have
been
reported.
In
LE
rats
exposed
gestationally
to
1,4,6­
androstatriene­
3,17­
dione,
high
levels
of
lordotic
behavior
are
observed
in
male
adult
offspring
treated
with
estrogen
and
progesterone,
while
SD
rats
only
showed
slight
effects
(
Whalen
et
al.
1986).
In
an
earlier
study
by
Emery
and
Larsson
(
1979),
Wistar
males
retained
copulatory
behavior
longer
than
SD
males
following
castration
and
systemic
para­
chlorophenylalanine
treatment
(
which
facilitates
copulatory
behavior).
The
castrated
Wistar
males
also
were
more
responsive
to
androgen
replacement
than
SD
males.
In
ovariectomized
females,
Wistar
females
were
behaviorally
more
sensitive
to
estrogen
than
SD
females.
52
2.5.12
Hormonal
Controls
The
endocrine
system
(
also
referred
to
as
the
hormone
system)
is
made
up
of
glands
located
throughout
the
body,
hormones
that
are
synthesized
and
secreted
by
the
glands
into
the
bloodstream,
hormone
carrier
proteins
(
e.
g.
steroid
hormone
binding
proteins,
globulin
and
albumin,
"­
fetoprotein),
receptors
in
the
cell
membranes,
cytosol
and
nucleus
of
the
cells
of
various
target
organs,
and
tissues
that
recognize
and
respond
to
the
hormones.
The
function
of
the
system
is
to
regulate
a
wide
range
of
biological
processes,
including
control
of
blood
sugar
(
through
the
hormone
insulin
from
the
pancreas),
growth
and
function
of
reproductive
systems
(
through
the
hormones
T
and
estrogen
and
related
components
from
the
testes
and
ovaries),
regulation
of
metabolism
(
through
the
hormones
cortisol
from
the
adrenal
glands
and
thyroxin
from
the
thyroid
gland),
development
of
the
brain
and
the
rest
of
the
nervous
system
(
estrogen
and
thyroid
hormones),
and
development
of
an
organism
from
conception
through
adulthood
and
old
age.
Normal
functioning
of
the
endocrine
system,
therefore,
contributes
to
homeostasis
(
the
body's
ability
to
maintain
itself
in
the
presence
of
external
and
internal
changes)
and
to
the
body's
ability
to
control
and
regulate
reproduction,
development,
and/
or
behavior.
An
endocrine
system
is
found
in
nearly
all
animals,
including
mammals,
nonmammalian
vertebrates
(
e.
g.,
fish,
amphibians,
reptiles,
and
birds),
and
invertebrates
(
e.
g.,
snails,
lobsters,
insects,
and
other
species).
In
humans,
the
system
comprises
more
than
50
different
hormones,
and
the
complexity
in
other
species
appears
to
be
comparable.

Puberty,
the
period
in
which
sexual
maturation
occurs,
begins
in
the
hypothalamic­
pituitary­
gonadal
(
HPG)
axis
and
leads
to
the
development
of
secondary
sex
characteristics
and
fertility
in
both
males
and
females
(
Stoker
et
al.,
2000;
Goldman
et
al.,
2000).
Within
the
hypothalamus,
gonadotropin­
releasing
hormone
(
GnRH)
from
neurosecretory
neurons
act
as
the
primary
controller,
whereas
in
the
anterior
lobe
of
the
pituitary,
gonadotropes,
which
secrete
luteinizing
hormone
(
LH)
and
follicle­
stimulating
hormone
(
FSH),
and
lactrotropes,
which
secrete
prolactin
(
Prl),
serve
the
controller
function.
The
primary
gonadotropin­
responsive
elements
in
males
are
the
Leydig
and
Sertoli
cells
in
the
testes,
whereas
in
the
female,
the
thecal
and
granulosa
cells
in
the
ovarian
follicle
respond.

°
Hypothalamus
(
GnRH).
It
is
generally
believed
that
the
CNS
is
the
trigger
point
for
initiation
of
sexual
maturation
in
the
male
and
female
rat
(
Goldman
et
al.,
2000;
Stoker
et
al.,
2000).
GnRH
is
present
in
the
fetal
brain
and
slowly
increases
until
the
second
postnatal
week
in
females
and
the
third
postnatal
week
in
males.
At
that
point,
GnRH
increases
steeply
and
remains
elevated
until
puberty.
At
puberty,
the
GnRH
neurons
undergo
a
morphological
change,
developing
spiny­
like
processes
that
may
be
related
to
an
increase
in
synapses
on
the
cells.
It
has
been
shown
that
at
puberty,
the
GnRH
neurons
become
more
responsive
to
neurotransmitter
(
norepinephrine
and
dopamine)
stimulation.
GnRH
is
released
in
a
pulsatile
manner
in
both
male
and
female
animals,
which
induces
a
similar
pattern
of
LH
and
FSH
secretion
from
the
53
anterior
pituitary.
GnRH
levels
can
be
viewed
as
an
indicator
of
initiation
of
sexual
maturation.

°
Pituitary
(
FSH,
LH,
Prl,
TSH).
The
gonadotropins
FSH,
LH,
and
Prl,
secreted
by
the
anterior
pituitary,
are
essential
in
the
process
of
sexual
maturation.
In
the
male,
LH
stimulates
T
secretion
by
direct
action
on
the
Leydig
cells
in
the
testis,
and
FSH
binds
to
the
Sertoli
cells
within
the
seminiferous
tubules
to
aid
spermatogenesis.
FSH
also
increases
the
number
of
LH
receptors
in
the
testis,
which
in
turn
increases
T
production
and
testis
growth.
Increased
prolactin
is
associated
with
growth
of
the
prostate
and
seminal
vesicle
glands.
In
the
female,
FSH
and
LH
act
on
the
ovarian
follicular
granulosa
and
thecal
cells,
respectively,
to
stimulate
production
of
E2,
follicular/
oocyte
maturation
and
ovulation.
An
increase
in
prolactin
levels
is
essential
in
the
acquisition
of
vaginal
opening
and
the
transition
to
sexual
maturity.
Thyroid
stimulating
hormone
(
TSH)
also
from
the
anterior
pituitary,
is
the
trigger
for
the
release
of
T3
and
T4
from
the
thyroid
gland
(
though
T3
is
also
produced
locally
in
target
organs)
,
and
is
essential
in
the
regulation
of
thyroid
activity
(
see
below).

Endocrine­
disrupting
chemicals
have
been
shown
to
alter
levels
of
pituitary
hormones,
such
as
prolactin
and
LH.
Cummings
et
al.
(
2000)
dosed
four
strains
of
rats,
either
diurnally
or
nocturnally
with
atrazine
(
0,
50,
100,
or
200
mg/
kg/
day
on
the
first
eight
days
of
pregnancy).
There
were
reductions
in
LH
levels
in
Holtzman
rats
and
LE
rats
but
not
in
SD
or
F344
rats
after
diurnal
dosing,
and
reductions
in
LH
levels
in
LE
and
F344
at
the
highest
dose
(
200
mg/
kg)
after
nocturnal
dosing,
with
no
effect
on
SD
or
Holtzman
rats.
There
were
strain­
related
differences
in
control
levels
of
LH.
For
example,
serum
LH
levels
were
low
in
Holtzman
and
F344
rats
and
significantly
higher
in
SD
and
LE
rats.
Control
progesterone
levels
tended
to
be
higher
in
F344
but
the
ranking
of
other
strains
differed
according
to
time
of
collection.
Thus
basal
levels
of
pituitary
hormones
may
contribute
to
the
sensitivity
of
certain
strains
to
endocrine­
disrupting
chemicals.

In
a
study
by
Cooper
et
al.
(
2000),
50­
300
mg/
kg/
day
of
atrazine
was
administered
to
ovariectomized
SD
and
LE
rats
for
1,
3,
or
21
days,
and
surges
of
LH
and
Prl
induced
by
estrogen
were
examined.
After
one
or
three
doses
of
atrazine
(
300
mg/
kg),
LH
and
Prl
were
suppressed
in
ovariectomized
LE
but
not
SD
rats.
After
21
doses,
LH
and
Prl
were
suppressed
in
both
rat
strains
in
a
dose­
dependent
manner.
Therefore,
although
a
longer
exposure
resulted
in
similar
effects
in
both
strains,
LE
rats
were
more
sensitive
to
shorter
exposures
of
atrazine.

The
differential
effects
of
an
environmental
estrogen,
bisphenol
A
(
BPA)
were
studied
in
female
F344
and
SD
rats
by
Steinmetz
et
al.
(
1997).
Basal
levels
of
Prl
were
40
and
25
ng/
ml
in
F344
and
SD
rats
respectively.
Within
three
days,
E2
(
in
silastic
capsules
inserted
s.
c.)
significantly
increased
Prl
levels
10­
fold
in
F344
rats
and
only
3­
fold
in
SD
rats;
while
BPA
significantly
increased
Prl
levels
7­
8­
fold
in
F344
rats,
with
no
effect
on
SD
rats.
Interestingly,
E2
increased
anterior
pituitary
weight
in
F344
rats,
but
not
in
SD
rats,
while
BPA
had
no
significant
effect
on
pituitary
weight
in
either
strain.
54
While
the
authors
speculated
that
genetic
differences
in
estrogen
receptors
may
be
involved
in
strain­
related
sensitivities,
subsequent
studies
in
uteri
and
vagina
showed
that
F344
and
SD
rats
differ
in
the
intermediate
effects
of
xenoestrogens
downstream
of
the
estrogen
receptor
(
Long
et
al.
2000).

°
Gonads
(
T,
DHT,
E2,
P).
Androgens
are
essential
in
the
development
of
the
male
reproductive
tract,
as
well
as
for
feedback
regulation
of
the
hypothalamic­
pituitary
axis,
sex
accessory
organ
development
and
maintenance,
and
spermatogenesis
(
Goldman
et
al.,
2000;
Stoker
et
al.,
2000).
T
and
DHT
are
the
two
most
active
androgens.
Testes
descent
and
development;
maturation
of
the
epididymides,
vas
deferens,
seminal
vesicles,
coagulating
glands,
prostate,
Cowper's
glands,
levator
ani/
bulbocavernosus,
and
other
aspects
of
the
male
reproductive
tract
are
dependent
upon
T,
whereas
DHT
is
responsible
for
male
AGD,
and
external
genitalia.
DHT
is
key
in
the
development
and
maintenance
of
the
external
genitalia,
prostate,
and
urethra.
E2
and
progesterone
(
P)
serve
similar
developmental
and
maintenance
functions
in
the
female.

In
four
strains
of
rats,
Holtzman,
LE,
SD
and
F344,
dosed
with
atrazine
(
0,
50,
100
or
200
mg/
kg/
day,
for
the
first
eight
days
of
pregnancy)
one
group
diurnally
and
the
other
nocturnally,
Holtzman
rats
were
the
only
strain
to
show
a
significant
postimplantation
loss
and
decreased
P
levels
(
Cummings
et
al.,
2000).
In
the
same
study,
serum
E2
was
increased
only
in
SD
rats
dosed
diurnally
with
the
high
dose
of
atrazine
(
200
mg/
kg).
In
control
animals,
there
were
strain­
related
differences
in
both
P
and
E2
levels.
For
example,
serum
P
levels
on
day
9
controls
were
significantly
higher
in
F344
and
Holtzman
rats
than
in
SD
or
LE
rats.
At
the
same
time,
E2
levels
in
F344
rats
were
significantly
lower
than
those
of
the
other
three
strains.
Thus,
there
are
strain
differences
in
control
gonadal
hormone
levels,
as
well
as
in
response
to
atrazine.

In
19.5
day
old
male
rat
fetuses,
gestational
exposure
to
TCDD
(
0,
0.5,
0.1,
0.5
or
1.0
:
g/
kg)
was
associated
with
increases
in
prenatal
T
and
pituitary
LH
production
in
Han/
Wistar
but
not
LE
rats
(
Haavisto
et.
al.,
2001).
The
lowered
sensitivity
of
fetal
LE
rats
may
be
associated
with
prenatal
T
levels
which
are
only
15%
of
those
in
Han/
Wistar
rats.
Conversely,
adult
LE
rats
are
1000
times
more
sensitive
to
TCDD
compared
to
Han/
Wistar
rats
(
Pohjanvirta
et
al.,
1988).
Interestingly,
strain­
related
differences
in
sensitivity
to
TCDD,
are
also
age­
dependent.
In
the
same
report,
gestational
exposure
to
DES
(
100
:
g/
kg
in
Han/
Wistar
rats;
100,
200
or
300
:
g/
kg
in
SD
rats)
significantly
decreased
prenatal
T
production
in
SD
and
Han/
Wistar
male
rats,
thus
the
authors
concluded
that
both
TCDD
and
DES
exposure
in
utero
may
interfere
with
the
timing
of
the
prenatal
T
surge.

°
Thyroid
(
T3,
T4).
TSH
released
from
the
pituitary
gland
stimulates
the
thyroid
to
secrete
triiodothyronine,
T3,
and
thyroxine,
T4.
T4
is
more
prevalent
in
the
blood
(
98%)
than
is
T3
(
2%).
T3
is
predominantly
produced
locally
in
target
tissues.
Prenatally,
maternal
T4
is
essential
for
normal
offspring
development.
Thyroid
hormones
are
well
known
to
play
essential
roles
in
vertebrate
55
development
(
Dussault
and
Ruel,
1981;
Myant,
1971;
Porterfield
and
Hendrich,
1993;
Porterfield
and
Stein,
1994;
Timiras
and
Nzekwe,
1989).
Experimental
work
focused
on
the
effects
of
thyroid
hormone
on
brain
development
in
the
neonatal
rat
supports
the
concept
of
a
"
critical
period"
during
which
thyroid
hormone
must
be
present
to
avoid
irreversible
damage
(
Timiras
and
Nzekwe,
1989).
Though
the
duration
of
this
critical
period
may
be
different
for
different
thyroid
hormone
effects,
the
general
view
has
developed
that
the
period
of
maximal
developmental
sensitivity
to
thyroid
hormone
occurs
during
the
lactational
period
in
the
rat
(
Oppenheimer
et
al.,
1994;
Timiras
and
Nzekwe,
1989).
Although
thyroid
hormone
receptors
are
expressed
in
fetal
rat
brains
(
Bradley
et
al.,
1989;
Strait
et
al.,
1990)
and
thyroid
hormone
can
exert
effects
on
the
fetal
brain
(
Escobar
et
al.,
1990;
Escobar
et
al.,
1987;
Escobar
et
al.,
1988;
Porterfield,
1994;
Porterfield
and
Hendrich,
1992,
1993;
Porterfield
and
Stein,
1994),
the
lactational
period
represents
a
stage
of
rapid
expansion
of
the
thyroid
hormone
receptors
(
Perez­
Castillo
et
al.,
1985)
and
an
increase
in
the
number
of
demonstrated
effects
of
changed
levels
of
thyroid
hormone
on
brain
development.

In
a
study
of
the
effects
of
2,3,7,8­
Tetrachlorodibenzo­
p­
dioxin
(
TCDD)
on
two
strains
of
adult
rats,
LE,
and
Han/
Wistar,
there
were
strain­
related
differences
in
control
and
treated
rats
(
Pohjanvirta
et
al.,
1989).
Control
T3
values
in
both
strains
were
not
significantly
different,
but
T4
levels
were
about
1.2
times
higher
and
TSH
levels
twice
as
high
in
Han/
Wistar
compared
to
LE
rats.
Rats
were
injected
once
i.
p.
with
0,
5,
50,
or
500
(
Han/
Wistar
only)
:
g/
kg
of
TCDD,
and
tissues
and
hormones
were
collected
at
1,
4,
8
and
16
days
post
treatment.
TCDD
decreased
T4
levels
slightly
more
in
LE
(
59%)
compared
to
Han/
Wistar
rats
(
43%)
after
4
days.
After
16
days,
T4
levels
had
returned
to
basal
levels
in
the
two
highest
dose
groups
in
the
Han/
Wistar
rats,
but
in
LE
rats,
T4
levels
remained
more
than
two
fold
lower
than
controls
with
no
sign
of
recovery.
By
day
16,
TCDD
increased
T3
levels
in
the
two
highest
dose
groups
in
the
Han/
Wistar
rats
and
had
decreased
T3
levels
by
one­
half
in
LE
rats.
Thus,
LE
rats
exhibited
a
greater
sensitivity
to
TCDD
with
respect
to
thyroid
hormone
levels.
The
greater
sensitivity
of
LE
rats
to
TCDD
has
been
confirmed
in
more
recent
work
by
Pohjanvirta
et
al.
(
1999)
who
showed
that
LE
rats
are
1000
times
more
sensitive
to
the
acute
lethal
effects
of
TCDD
than
are
Han/
Wistar
rats.

Another
study
compared
the
effects
on
levels
of
TSH,
T3
and
T4
after
an
endocrine
challenge
test
[
with
thyrotropin­
releasing
hormone
(
TRH),
and
TSH],
on
two
strains
of
male
rats,
SD
and
F344
(
Fail
et
al.,
1999).
Both
strains
of
rats
responded
to
the
challenge
with
increases
in
TSH
levels.
In
F344
rats,
there
were
significant
increases
in
levels
of
T3
and
T4,
while
in
SD
rats,
there
were
only
increases
in
T4.

Strain­
related
differences
in
various
hormone
levels
and
organ
weights
were
reported
after
exposure
to
the
weak
antiandrogen,
p,
p'­
DDE,
by
O'Connor
et
al.
(
1999).
After
exposure
of
adult
male
LE
and
CD
rats
for
15
days
to
p,
p'­
DDE
(
0,
100,
200,
300
or
350
mg/
kg/
day
in
CD
rats;
0,
200
or
300
mg/
kg/
day
in
LE
rats),
the
following
effects
were
reported.
56
CD
LE
8
relative
liver
weight,
9absolute
epididymis
weight
8relative
liver
weight,
8absolute
epididymis
and
relative
accessory
organ
weight
8E2
($
200
mg/
kg/
day),
no
change
in
T,
DHT
8E2
(
300
mg/
kg/
day),
T,
DHT($
200
mg/
kg/
day)

9FSH
($
200
mg/
kg/
day)
,
no
change
in
Prl,
LH
no
change
in
FSH,
Prl
or
LH
9T4
($
100
mg/
kg/
day),
no
change
in
TSH
8TSH
and
9T4
($
200
m
g/
kg/
day)

These
data
demonstrate
strain­
sensitive
differences
in
response
to
an
endocrinedisrupting
chemical.
CD
rats
were
much
less
sensitive
to
the
effects
of
p,
p'­
DDE
than
were
LE
rats.

There
is
also
a
report
by
Dhaher
et.
al,
(
2000)
of
intraspecies
differences
in
estrogen
receptor
number
and
binding
affinity
between
Balb/
c
strain
mice
and
a
strain
used
as
a
systemic
lupus
erthematosus
(
SLE)
model,
MRL/
MP­
lpr/
lpr
(
Dhaher,
Greenstein
et
al.
2000).
The
MRL
mice
showed
significantly
higher
affinity
for
E2
binding
(
using
3H­
moxestrol
as
ligand)
than
did
the
Balb/
c
mice,
which
may
be
related
to
the
exacerbation
by
E2
of
SLE
in
the
mouse
model.

2.5.13
Uterine
Weight
Uterine
weight
is
sensitive
to
estrogenic
compounds.
The
increase
in
uterine
weight
after
estrogenic
compound
exposure
may
be
due
to
increased
fluid
uptake
(
imbibition),
increased
cell
size
(
hypertrophy),
and/
or
increased
cell
number
(
hyperplasia).
Although
the
uterotrophic
assay
is
being
standardized
and
validated
by
an
OECD/
EPA
initiative,
the
measurement
of
uterine
weight
is
a
sensitive
parameter
for
inclusion
in
many
of
the
in
vivo
screens
involving
females.
The
uterotrophic
assay
was
designed
to
identify
chemicals
that
act
as
estrogen
receptor
agonists
or
antagonists,
directly
on
the
uterus
in
ovariectomized
females
(
since
the
HPG
axis
is
not
intact).
An
alternative
is
to
use
prepubertal
intact
females
as
the
animal
model.
In
this
case,
the
test
chemical
may
affect
any
point
along
the
HPG
axis,
and
the
end
organs
such
as
the
uterus,
and
the
chemical
may
not
be
an
ER
agonist
or
antagonist
since
it
does
not
require
ER
binding
in
the
uterus.
A
description
of
processes
and
endpoints
to
evaluate
chemical
effects
on
uterine
weight
follows.

1.
Ovariectomized
adult
females
are
exposed
to
test
material
for
3­
5
days
(
po,
sc,
etc.)
and
are
evaluated
for
the
following:

A.
Increased
uterine
wet/
blotted
weight,
6
to
24
(
hypertrophy
and
hyperplasia,
respectively)
hours
after
test
dose;
must
be
due
to
a
uterine
57
estrogen
receptor­
mediated
response
(
since
gonad
is
missing,
HPG
axis
is
not
intact);
an
estrogen
receptor
agonist
will
be
detected.

B.
Administration
of
authentic
E2,
the
potent
endogenous
estrogen
plus
the
test
material;
if
uterine
wet
and
blotted
weights
increased
with
E2,
but
to
lesser
extent
from
E2
plus
test
chemical,
then
test
material
is
an
estrogen
antagonist.

2.
Intact
prepubertal
females
are
exposed
to
test
material
for
3
days
(
po,
sc,
etc.)
prior
to
normal
onset
of
puberty
and
are
evaluated
for
the
following:

A.
Increased
uterine
wet/
blotted
weight
(
can
occur
through
HPG
axis
since
it
is
intact;
if
uterine
weight
is
increased,
the
test
material
is
an
estrogenmimic
or
estrogen­
like
(
not
necessarily
an
estrogen
receptor
agonist).

B.
Adminstration
of
E2
and
test
material
for
detection
of
anti­
estrogens
(
need
not
be
an
estrogen
receptor
antagonist).

A
three­
day
uterotrophic
assay
for
detecting
the
estrogenic
activity
of
octylphenol,
nonylphenol,
methoxychlor,
and
bisphenol
A
in
prepubertal
LE
rats
was
found
to
be
the
most
accurate
method
of
detecting
estrogenic
activity
when
compared
to
age
of
VO
and
estrous
cyclicity
(
Laws
et
al.,
2000),
and
may
provide
a
sensitive
endpoint
for
detection
of
endocrine
disrupting
chemicals
that
act
via
estrogen
receptor
binding
(
ovariectomized
female),
or
via
interaction
with
the
HPG
axis
(
intact
prepubertal
female).

In
studies
performed
in
ovariectomized
Wistar,
Da/
Han,
and
SD
rats,
dosed
two
weeks
after
surgery
with
increasing
doses
of
genistein
(
25,
50
and
100
mg/
kg/
day),
p­
tert­
octylphenol
(
5,
50
and
200
mg/
kg/
day),
bisphenol
A
(
0,
5,
50
and
200
mg/
kg/
day),
and
as
a
positive
control
100
:
g/
kg
EE,
there
was
a
strong
stimulation
by
EE
in
uterine
weight
in
the
Wistar
and
Da/
Han
rats
and
less
pronounced
response
in
the
SD
rats
(
Diel
et
al.,
2001).
All
strains
showed
comparable
slight
uterotrophic
responses
to
50
and
100
mg/
kg
genistein
and
comparable
moderate
uterotrophic
response
to
200
mg/
kg
p­
tert­
octylphenol.
No
doses
of
bisphenol
A
applied
stimulated
uterine
wet
weight
in
Wistar
or
Sprague
Dawley
rats,
whereas
in
the
Da/
Han
rats
a
slight
stimulation
was
detected
in
the
highest
dose
(
200
mg/
kg
BW).
These
studies
demonstrated
a
strain­
and
chemical­
specific
sensitivity
in
the
uterotrophic
assay
with
SD
rats
less
sensitive
than
DA/
Han
rats
to
EE
and
BPA.
While
Wistar
rats
were
more
sensitive
than
SD
to
EE,
both
Wistar
and
SD
rats
were
resistant
to
BPA.

Similarly,
EE
(
1,3,10
or
30
:
g/
kg/
day),
DES
(
0.5,
1.5,
5
and
15
:
g/
kg/
day)
and
a
weak
phytoestrogen,
coumestrol
(
CE)
(
10,
35,
75
and
150
mg/
kg/
day),
as
positive
controls,
produced
increases
in
uterine
weight
in
SD
and
F344
rats
(
McKim
et
al.,
2001).
However,
in
response
to
the
chemical
D4
(
0,10,50,100,
250,
500
and
1000
mg/
kg/
day),
the
maximal
uterine
weight
was
increased
160%
relative
to
control
values
in
the
SD
rats
and
only
86%
relative
to
control
values
in
F344
rats.
Thus,
SD
rats
were
more
sensitive
to
the
effects
of
octamethylcyclotetrasiloxane
(
D4)
in
the
uterotrophic
assay.
McKim
et
58
al.
(
2001)
suggested
that
the
metabolism
of
D4
may
be
slower
in
SD
versus
F344
rats
based
on
pharmacokinetic
data.

In
a
study
by
Christian
et
al.
(
1998),
the
uterotrophic
assay
was
compared
across
three
rat
strains,
Wistar­
Chbb:
THOM­
SPF,
and
Wistar­
CRL:(
WI)
BR,
and
SD
(
Charles
River).
When
administered
DES,
all
three
strains
exhibited
a
positive
uterine
response,
with
a
similar
response
in
both
Wistar
strains
and
a
slightly
lower
response
in
SD
rats.
Variability
of
responses
was
associated
with
background
spontaneous
incidences
of
abnormally
high
relative
uterine
weights
possibly
due
to
fluctuations
in
estrogen
occurring
between
pnd
21
to
25.
Due
to
differences
in
control
mean
uterine
weights
between
rat
strains
(
i.
e.
means
were
lower
in
Wistar­
Chbb:
THM­
SPF
versus
Wistar­
CRL
and
SD),
criteria
for
biological
outliers
were
different
in
Wistar­
Chbb:
THOM­
SPF
($
0.15%)
versus
Wistar­
CRL
and
SD
rats
($
0.20%).
Christian
et
al.
(
1998)
demonstrated
the
importance
of
historical
control
data
in
the
determination
of
statistically
significant
effects
in
subsequent
studies.

Sometimes
genetic
differences
have
no
observable
effect
on
an
endocrine
endpoint.
A
study
by
Odum
et
al.,
1999a,
investigated
the
effects
of
p­
Nonylphenol
(
NP)
(
0
to
250
mg/
kg/
day
orally
for
3
or
11
days;
and
0
to
7.12
mg/
kg/
day
via
mini­
pumps
implanted
s.
c.,
for
11
days)
in
the
uterotrophic
assay
in
Alderley
Park
(
Wistar­
derived)
and
SD
rats.
Results
were
similar
in
both
strains
with
a
positive
response
to
both
DES
(
0.01
mg/
kg/
day)
and
NP
(
250
mg/
kg/
day),
which
were
of
the
same
magnitude
as
in
previous
studies
performed
in
Noble
rats
(
Odum
et
al.,
1999b).
The
uterotrophic
effects
of
NP
and
DES
were
found
to
be
independent
of
rat
strain.

3.0
Interspecies
Similarities
and
Differences
in
Endocrine
Endpoints
Few
studies
have
been
conducted
in
a
single
laboratory
comparing
the
effects
of
endocrine­
disrupting
chemicals
in
more
than
one
strain
within
a
species,
and
even
fewer
studies
have
been
conducted
in
a
single
laboratory
comparing
the
effects
of
endocrinedisrupting
chemicals
in
more
than
one
species.
Therefore
the
criterion
that
only
studies
performed
in
the
same
laboratory
across
species
would
be
included
in
this
white
paper
could
not
be
applied.
One
review
paper
which
looked
at
many
studies
on
many
chemicals
across
many
laboratories
in
mice
versus
rats
was
included
as
well
as
other
studies
comparing
endocrine
endpoints
across
species
in
response
to
endocrinedisrupting
chemicals.

There
are
many
reviews
on
species
differences
in
reproductive
and
developmental
toxicology
studies.
Among
these
is
a
comparison
of
reproductive
organ
weights,
sperm
parameters,
and
vaginal
cytology
from
fifty
13­
week
studies
involving
24
chemicals
in
seven
different
laboratories
(
and
four
routes
of
exposure
for
the
National
Toxicology
Program
in
B6C3F1
mice
and
F344
rats
(
Morrissey
et
al.,
1988).
Considerable
interlaboratory
variability
was
demonstrated,
but
overall,
it
was
concluded
that
there
were
no
differences
in
types
of
sperm
head
abnormalities
between
control
and
treated
rats
and
mice,
and
that
testis,
epididymis,
and
cauda
epididymis
weights
and
sperm
motility
were
the
most
statistically
powerful
endpoints
evaluated.
Of
all
the
59
chemicals
tested,
only
one,
methylphenidate
in
the
rat,
produced
an
increase
in
abnormal
sperm
without
effects
on
any
other
male
endpoint.
The
agreement
in
results
in
these
endocrine
endpoints
in
response
to
reproductive
toxicants
between
rats
and
mice
was
about
58%.
A
combination
of
confounding
factors
and
species
differences
may
have
accounted
for
disparity
in
toxicological
data.

Several
reports
have
focused
on
differences
in
endocrine
endpoints
between
rats
and
mice.
In
the
uterotrophic
assay,
the
estrogenic
activity
of
parabens
was
assessed
in
B6D2F1
mice
and
Wistar
rats
(
Hossaini
et
al.,
2000).
The
parabens
tested
were
methyl­,
ethyl­,
propyl­
and
butyl
p­
hydroxybenzoate,
and
p­
hydroxybenzoic
acid,
which
were
administered
either
orally
or
subcutaneously
for
three
days
at
doses
up
to
1000
mg/
kg/
day,
and
E2
(
0.1
mg/
kg)
was
used
as
a
positive
control.
In
the
mouse
uterotrophic
assay,
there
was
no
significant
effect
on
uterus
weight
in
doses
up
to
1000
mg/
kg/
day
for
all
parabens.
In
the
rat
uterotrophic
assay,
600
mg/
kg/
day
of
butyl­
paraben
produced
a
positive
response.
Thus
the
estrogenic
activity
of
parabens
was
found
to
be
weak
in
rats
and
was
not
observed
in
mice.

Two
separate
studies
reported
species
differences
in
the
ovarian
toxicity
of
reproductive
toxicants.
In
one
study
(
Doerr
et
al.,
1996),
1,3­
butadiene
epoxides
(
butadiene
monoepoxide,
BMO;
and
butadiene
diepoxide,
BDE)
were
administered
at
doses
up
to
1.43
mmol/
kg/
day
intraperitoneally
for
30
days,
to
young
SD
rats
and
B6C3F1
mice.
BMO
was
ovotoxic
in
mice,
producing
decreases
in
follicle
counts
and
reproductive
organ
weights,
with
no
effects
in
rats
at
the
doses
tested.
BDE
was
ovotoxic
in
both
rats
and
mice,
with
a
greater
sensitivity
to
BDE
in
mice,
resulting
in
reductions
in
uterine
and
ovarian
weights
in
mice
at
lower
concentration
than
in
rats.
In
addition,
follicle
counts
were
greatly
reduced
in
mice
at
lower
doses
of
BDE
than
in
rats.
The
authors
speculate
that
metabolic
differences
affecting
the
conversion
of
BMO
to
BDE
may
be
responsible.
In
another
study
by
Takizawa
et
al.,
1985,
intraovarian
injection
of
increasing
concentrations
of
benzo(
a)
pyrene,
a
polycyclic
aromatic
hydrocarbon,
reduction
in
small
oocytes
occurred
in
a
dose­
dependent
fashion
at
doses
ranging
from
0.01
to
30
:
g/
ovary
in
C57BL/
6N
and
DBA/
2N
mice
and
0.8
to
240
:
g/
ovary
in
SD
rats.
Thus
effects
of
benzo(
a)
pyrene
on
small
oocyte
number
were
present
in
both
SD
rats
and
two
strains
of
mice.

Cadmium
has
also
been
found
to
induce
ovarian
toxicity
in
animals.
In
a
study
by
Rehm
and
Waalkes
(
1988),
the
effects
of
cadmium
were
assessed
in
immature
and
mature
female
Syrian
hamsters,
four
mouse
strains
(
BALB/
cAnNCr,
DBA/
2NCr,
C57BL/
6NCr,
NFS/
NCr)
and
two
rat
strains
[
F344
and
Wistar­
Furth
(
WF)]
after
a
single
injection
(
sc)
of
a
dose
ranging
from
20
to
47.5
:
mol/
kg,
and
the
reproductive
tracts
were
examined
by
light
microscopy.
Syrian
hamsters
were
the
most
sensitive
to
cadmium­
induced
ovarian
hemorrhagic
necrosis,
in
particular,
shortly
before
ovulation.
In
mice,
only
the
DBA/
2NCr
strain
showed
significant
cadmium­
induced
ovarian
hemorrhagic
necrosis,
and
uterine
lesions
in
any
of
the
mouse
strains
were
rare.
Though
rats
showed
dose
and
age­
dependent
toxicity
of
the
ovaries,
uterus,
cervix,
and
liver,
cadmium
induced
uterine
lesions
only
in
mature
F344
rats,
not
WF
rats.
Thus
species
and
strain
differences
in
cadmium­
induced
reproductive
toxicity
were
reported.
60
As
female
mice
were
more
sensitive
than
female
rats
to
the
ovotoxic
effects
of
1,3­
butadiene
(
Doerr
et
a.,
1996),
male
mice
were
found
to
be
more
sensitive
to
the
reproductive
effects
of
1,3­
butadiene,
while
the
same
doses
produced
no
effects
in
rats.
Anderson
et
al.
(
1998)
compared
the
effects
of
exposure
by
inhalation
to
1,3­
butadiene
(
up
to
1250
ppm
in
rats
and
up
to
130
ppm
in
mice)
for
10
weeks
prior
to
mating,
in
male
CD­
1
mice
and
male
SD
rats.
Exposure
in
mice
resulted
in
F1
abnormalities
and
increases
in
early
deaths.
None
of
these
effects
were
observed
at
the
same
exposure
concentrations
in
SD
rats.

In
an
earlier
study
by
Brinkworth,
Anderson,
and
McLean
(
1992),
dietary
restriction
in
CD­
1
mice
was
found
to
increase
abnormal
sperm
which
may
have
been
related
to
a
decrease
in
calories,
and
to
decrease
epididymal
sperm
counts,
which
may
have
been
due
to
a
lack
of
protein.
In
CD
rats,
dietary
restriction
only
reduced
epididymal
sperm
count.
Thus
dietary
changes
had
different
impacts
on
spermatogenesis
in
mice
compared
to
rats.

A
differential
sensitivity
to
the
developmental
toxicity
of
BPA
was
reported
by
Morrissey
et
al.
(
1987)
in
CD­
1
mice
compared
to
CD
rats.
Mice
and
rats
were
dosed
daily
from
gd
6
to
15,
with
0­
640
mg/
kg/
day
BPA
(
rats)
and
0­
1250
mg/
kg/
day
BPA
(
mice).
Maternal
toxicity
(
as
evidenced
by
significant
decreases
in
maternal
weight
gain)
occurred
in
rats
at
$
160
mg/
kg/
day
and
at
a
much
higher
dose
(
1250
mg/
kg/
day)
in
mice.
No
fetal
toxicity
occurred
at
doses
up
to
640
mg/
kg/
day
in
rats,
and
in
mice,
at
1250
mg/
kg/
day,
there
were
significant
increases
in
the
number
of
resorptions,
and
a
decreases
in
average
fetal
body
weight.
Maternal
mice
were
more
sensitive
to
toxic
doses
of
BPA
than
maternal
rats,
and
fetal
malformations
were
not
observed
in
either
rats
or
mice.

Species
differences
in
endocrine
receptor
binding
characteristics
have
been
reported
in
in
vitro
experiments.
Mathews
et
al.
(
2000)
examined
the
ability
of
several
natural
and
synthetic
chemicals
to
compete
with
3H­
E2
for
binding
to
bacterially
expressed
estrogen
receptor
alpha
D,
E
and
F
domains
from
five
different
species
(
human,
mouse,
chicken,
anole,
and
rainbow
trout)
fused
to
glutathione
S­
transferase
(
GST).
While
all
of
these
fusion
proteins
displayed
high
affinity
for
E2
(
Kd
of
0.3
to
0.9
nM),
species
differed
in
affinity
of
binding
other
estrogenic
chemicals.
With,
for
example,
rainbow
trout,
ER
fusion
proteins
showing
much
higher
relative
binding
affinity
for
alpha­
zearalenol,
Bisphenol
A,
octylphenol,
o,
p'
DDT,
Methoxychlor,
p,
p'­
DDT,
o,
p'­
DDE,
p,
p'­
DDE,
alpha­
endosulfan
and
dieldrin
than
ERs
from
the
human,
mouse,
chicken
and
green
anole
(
Matthews,
Celius
et
al.
2000).
Thus
sequence
variation
in
the
ER
ligand
binding
domain
between
strains
and
between
species
could
be
a
major
source
of
variation
in
EDSP
assays,
in
evaluating
compound
which
act
via
the
ER.

4.0
Summary
and
Conclusions
of
Intraspecies
and
Interspecies
Similarities
and
Differences
in
Endocrine
Endpoints
and
Conclusions
Endocrine­
mediated
toxicity
of
chemicals
varies
among
strains
of
animals
of
the
same
species
and
among
different
species.
Endocrine
endpoints
vary
in
sensitivity
to
61
chemicals
across
strains
and
species.
The
sensitivity
to
endocrine­
active
chemicals
is
obviously
dependent
on
the
endpoint
evaluated,
the
chemical
administered,
and
the
genotype
of
the
animal
model.
Genetic
variability
which
exists
within
outbred
strains,
between
inbred
and
outbred
strains
as
well
as
in
the
animal
and
human
populations
to
which
the
results
will
be
applied
has
the
potential
to
confound
the
detection
and
interpretation
of
reproductive
toxicity.
The
following
table
summarizes
strain­
related
similarities
and
differences
in
endocrine
endpoints
in
many
key
research
articles
designed
to
investigate
the
effects
of
endocrine
disruptors.
The
focus
of
the
comparison
is
on
rat
strains,
with
some
examples
in
mouse
strains.
62
Table
2.
Intraspecies
Comparisons
of
Endocrine
Endpoints
Endocrine
Endpoints
Strains
Chemical
Similarities
Differences
Key
References
Uterine
Weight
Wistar,
SD,
Da/
Han
Alderley
Park,
SD
SD,
F344
SD,
F344
Wistar,
SD
SD,
F344
Ethinyl
Estradiol
(
EE),
BPA
NP
E2,
BPA
EE,
DES
and
octamethylcyclot
etrasiloxane
(
D4)

DES
E2,4­
OH
tamoxifen
(
4­
OHT)
Wistar
and
SD
less
sensitive
to
BPA
than
Da/
Han
Results
were
similar
in
both
strains
with
a
positive
response
to
NP
(
250
mg/
kg/
day),
which
were
of
the
same
magnitude
as
previous
studies
perform
ed
in
Noble
rats
(
Odum
et
al.,
1999b)

Both
strains
showed
increased
uterine
weight
in
response
to
E2
EE,
DES
and
D4
produced
similar
positive
uterine
response
in
SD
and
F344
rats.

SD
and
W
istar(
CRL)
had
similar
control
values
for
uterine
weight;
sim
ilar
uterotrophic
response
in
both
Wistar
strains.

F344
and
SD
showed
similar
uterine
weight
responses
to
E2
SD
less
sensitive
than
Da/
Han
to
EE
and
BPA,
W
istar
more
sensitive
than
SD
to
EE.

250
m
g/
kg
NP
resulted
in
slightly
greater
uterotrophic
response
in
Alderlery
Park
(
1.84­
fold)
than
in
SD
(
1.55­
fold).

F344
more
sensitive
to
BPA
and
E2
Maximal
uterine
response
to
D4
was
two­
fold
higher
in
SD
than
F344
rats.
F344
m
ore
sensitive
to
EE.

Wistar(
Chbb­
THOM­
SPF)
had
lower
mean
uterine
weights
in
controls
versus
SD
and
Wistar(
CRL),
less
variability
in
response
to
DES
F344
more
sensitive
than
SD
to
E2
induced
uterine
epithelial
cell
height.
SD
more
sensitive
than
F344
to
induction
of
uterine
weight
and
epithelial
cell
height
by
4­
OHT
1Diel
et
al.,
2001
2Odum
et
a.,
1999a
3Steinmetz
et
al.,
1998
4McKim
et
al.,
2001
5Christian
et
al.,
1998
6Bailey
et
al.,
2002
Table
2.
(
continued)

Endocrine
Endpoints
Strains
Chemical
Similarities
Differences
Key
References
63
Male
and
female
sexual
development
(
AGD,
PPS,
VO)
SD,
LE
SD,
F344
p,
p'­
DDE;
flutamide
$­
E2­
3­
benzoate
PPS
same
time
in
LE
and
SD
controls;
p,
p'­
DDE
had
no
effect
on
PPS
or
VO
in
either
strain;
flutamide
decreased
AGD
and
caused
nipple
retention
and
changes
in
male
reproductive
organ
weights
in
both
strains
neonatal
exposure
to
1.5
mg/
kg/
day
decreased
male
reproductive
organ
weights
in
both
F344
and
SD
males
(
pnd
90),
and
delayed
PPS
p,
p'­
DDE
produced
significant
decrease
in
AGD
in
male
LE
rats,
not
in
SD;
p,
p'­
DDE
produced
nipple
retention
at
a
lower
dose
in
SD
than
LE;
VO
earlier
in
LE
controls
versus
SD
controls
At
pnd
90,
in
rats
exposed
to
1.5
mg/
kg/
day,
there
were
greater
reductions
in
reproductive
organ
weights
in
F344
rats
than
in
SD
rats
(
greater
responsiveness
in
F344,
higher
sensitivity
in
SD)
7You
et
al.,
1998
8Putz
et
al.,
2001
Female
reproductive
tract
SD,
Alderley
Park
F344,
SD
LE,
Holtzmann
BPA
BPA
TCDD
There
was
no
effect
on
age
of
first
estrus
in
either
strain.

Metabolic
clearance
of
BPA
is
same
1
:
g/
kg
on
gd
15
produced
malformations
of
F1
offspring
female
external
genitalia
and
increased
UVD
in
both
strains.
VO
delayed
in
Alderley
Park,
but
not
in
SD
rats
BPA
increased
DNA
synthesis
and
cell
proliferation
in
the
vaginal
epithelium
of
F344
rats
but
not
of
SD
rats
9Tinwell
et
al.,
2002
10Long
et
al.,
2000
11Gray
and
Ostby,
1995
Table
2.
(
continued)

Endocrine
Endpoints
Strains
Chemical
Similarities
Differences
Key
References
64
Male
reproductive
tract/
andrology
F344,
ACI,
Lewis,
SD,
Wistar
LE,
SD
LE,
Wistar
3,2'­
dimethyl
4­
amino­
biphenyl
(
DMAB)

p,
p'­
DDE
Vinclozolin
In
both
SD
and
LE
rats,
there
were
increases
in
liver
weight,
and
in
E2;
decreases
in
T4;
and
no
change
in
Prl
or
LH
200
mg/
kg/
day
from
gd
14
to
pnd
3
produced
malformations
of
m
ale
external
genitalia,
nipple
retention
lasting
into
adulthood,
and
increased
inflammation
of
epididymides,
prostate,
seminal
vesicles
and
coagulating
glands
in
both
strains.
12
mg/
kg
produced
transient
retention
of
nipples/
areolae
in
preweanling
males
of
both
strains.
Similar
NOAEL
in
both
strains,
12
and
6
mg/
kg
bod
wt.
In
Wistar
and
LE
respectively.
tumorigenic
response
in
the
prostate
of
F344,
ACI,
Lewis,
CD
and
W
istar
rats
to
DMAB
was
more
sensitive
in
F344>
ACI>
Lewis>
CD
and
W
istar
rats
were
insensitive
In
SD
rats
there
was
a
decrease
in
epididymis
weight,
no
change
in
T,
DHT,
or
TSH,
and
a
decrease
in
FSH.
In
LE
rats,
there
was
an
increase
in
epididymis
weight,
increases
in
T,
DHT
and
TSH,
and
no
change
in
FSH.

200
mg/
kg/
day
from
gd
14
to
pnd
3
reduced
maternal
body
weights
and
pup
weights
in
Wistar
but
not
LE.
Chronic
inflammation
of
urinary
bladder,
and
testis
atrophy
were
noted
in
LE
and
not
Wistar.
At
12
mg/
kg/
day,
nipple/
areolae
retention
persisted
into
adulthood
in
LE,
not
in
Wistar.
Adult
offspringLE
also
had
reduced
prostate,
seminal
vesicle,
and
coagulating
gland
weights;
these
effects
were
not
seen
in
W
istar
adult
offspring..
12Shirai
et
al.,
1990
13O'Connor
et
al.,
1999
14Hellwig
et
al.,
2000
Table
2.
(
continued)

Endocrine
Endpoints
Strains
Chemical
Similarities
Differences
Key
References
65
AP,
SD
BPA
No
difference
in
PPS
.
50
mg/
Kg
BPA
decreased
total
sperm
count
and
daily
sperm
production
in
AP
(
Wistar­
derived)
rats
but
not
SD.
15Tinwell
et
al.,
2000
Wistar,
SD
and
Dark
Agouti
(
DA)

SD
versus
other
strains
(
review)

F344,
SD
none
lead
Neonatal
E2­
3­
Benzo­
ate
(
EB)
DA
exhibited
lower
absolute
testis
weight
than
the
other
two
strains,
no
difference
in
sperm
count
among
the
three
strains.

SD
more
sensitive
than
other
strains
to
testicular
toxicity
of
lead.

F344
more
sensitive
and
responsive
than
SD
to
reduction
in
TW,
Epididymis,
Seminal
Vessicle
and
Coagulating
gland
weights.
16Wilkinson
et
al.,
2000
17Apostoli
et
al.,
1998
18Putz
et
al.,
2001
Table
2.
(
continued)

Endocrine
Endpoints
Strains
Chemical
Similarities
Differences
Key
References
66
Estrous
cycle/
ovulation
F344,
LE,
SD
Donryu,
F344
LE,
SD
SD,
F344
diet
(
feed
restriction)

none
atrazine
atrazine
Prior
to
food
restriction,
Brown
Norway
rats
had
irregular
estrous
cycle
patterns
while
SD,
LE
and
F344
rats
had
regular
estrous
cycle
patterns.
By
day
5
of
food
restriction,
75%
of
F344
rats
and
100%
of
Brown
Norway
rats
stopped
cycling
and
SD
and
LE
rats
were
unaffected
In
aging
Donryu
rats,
estrous
cycle
abnormalities
(
e.
g.
persistent
estrus)
were
more
common
than
in
F344
rats
LE
rats
were
more
sensitive
than
SD
rats
to
atrazine­
induced
disruption
of
the
ovarian
cycle.

Atrazine
administration
to
SD
and
F344
rats
for
up
to
12
months
produced
changes
in
estrous
cyclicity
in
SD
rats
(
increased
the
number
of
days
of
vaginal
estrus),
increased
E2,
decreased
P,
and
increased
incidence
of
mam­
mary
tumors
in
SD
rats
only,
with
no
effect
in
F344
rats.
19Tropp
et
al.,
2001
20Ando­
Lu
et
al.,
1998
21Cooper
et
al.,
2000
22Eldridge
et
al.
(
1994)
Table
2.
(
continued)

Endocrine
Endpoints
Strains
Chemical
Similarities
Differences
Key
References
67
Gonadal/
Pituitary
Hormone
levels
Han/
Wistar,
SD,
LE
SD,
LE
SD,
F344
SD,
LE,
Holtzman,
F344
TCDD,
DES
atrazine
E2,
BPA
atrazine
gestational
exposure
to
DES
(
100
:
g/
kg
in
Han/
Wistar
rats;
100,
200
or
300
:
g/
kg
in
SD
rats)
significantly
decreased
prenatal
T
production
in
SD
and
Han/
Wistar
male
rats
After
21
doses
of
atrazine
LH
and
Prl
were
suppressed
in
SD
and
LE
rats.
BPA
had
no
effect
on
pituitary
weight
in
either
strain.
In
19.5
day
old
male
rat
fetuses,
gestational
exposure
to
TCDD
(
0,
0.5,
0.1,
0.5
or
1.0
:
g/
kg)
was
associated
with
increases
in
prenatal
T
and
pituitary
LH
production
in
Han/
Wistar
but
not
LE
rats.

LH
and
Prl
were
suppressed
in
LE
but
not
SD
rats
after
1
and
3
doses
of
atrazine.

After
3
days
of
E2
exposure,
Prl
was
increased
10X
in
F344,
only
3X
in
SD;
increased
pituitary
weight
in
F344,
not
in
SD.
BPA
increased
Prl
7­
8X
in
F344,
no
effect
on
SD
rats.

Serum
E2
increased
by
atrazine
in
SD
rats,
O
nly
Holtzman
strain
with
decreased
P
levels.
Control
and
treated
levels
of
E2
much
lower
and
non­
detectable
in
Dirunal
F344
rats.
Control
levels
of
E2
higher
nocturnally
in
LE
rats.
Diurnal
atrazine
reduced
LH
in
Holtzman
and
LE
but
not
in
SD
or
F344.
Nocturnal
atrazine
reduced
LH
in
LE
and
F344
but
not
in
SD
or
Holtzman.
23Haavisto
et
al.,
2001
24Cooper
et
al.,
2000
25Steinmetz
et
al.,
1997
26Cummings
et
al,
2000
Table
2.
(
continued)

Endocrine
Endpoints
Strains
Chemical
Similarities
Differences
Key
References
68
Thyroid
hormone
levels
LE,
Han/
Wistar
SD,
F344
TCDD
TRH,
TSH
T3
levels
in
controls
were
comparable
T4
and
TSH
levels
were
higher
in
Han/
Wistar
controls;
TCDD
produced
a
greater
reduction
in
T4
levels
in
LE
than
in
Han/
Wistar
rats.

Challenge
with
TSH
and
TRH
increases
levels
of
T3
and
T4
in
F344
rats,
and
increased
T4
only
in
SD
rats.
27Pohjanvirta
et
al.,
1989
28Fail
et
al.,
1999
Fertility/
reproductive
parameters
SD,
LE,
Holtzman,
F344
SD,
LE,
F344
F344,
SD
Wistar
Hannover,
CD
atrazine
atrazine
BDCM
none
no
effect
of
atrazine
on
pre­
or
post­
implantation
loss
in
SD
and
LE
rats
full
litter
resorption
at
highest
dose
(
200
mg/
kg/
day)
in
all
three
strains.

75
mg/
kg/
day
significantly
reduced
body
weights
in
both
strains
Increase
in
%
preimplantation
loss
in
F344;
SD
and
LE
not
affected.
Increase
in
%
postimplantation
loss
only
in
Holtzman
rats
only.

At
lower
doses
(
50
and
100
mg/
kd/
day)
pregnancy
loss
in
F344,
not
in
SD
or
LE
75
mg/
kg/
day
BDCM
produced
62%
litter
resorption
in
F344,
no
effect
on
SD
Control
W
istar
Hannover
have
lower
pregnancy
rates
and
litter
sizes;
higher
%
pre­
and
postimplantation
loss
and
resorptions
versus
control
CD
®
.
29Cummings
et
al.,
2000
30Narotsk
y
et
al.,
2001
31Bielm
eier
et
al.,
2001
32Liberati
et
al.,
2002
Table
2.
(
continued)

Endocrine
Endpoints
Strains
Chemical
Similarities
Differences
Key
References
69
organ
weights,
histopath­
ology
(
rats)
SD,
F344
Brown­
Norway,
F344,
W
istar,
Donryu
F344,
BN,
SD
F344,
W
F
E2
E2
DES
Cadmium
No
difference
in
pituitary
weights
of
Wistar
and
Donryu
rats,
after
E2
exposure.
In
ovariectomized
rats
exposed
to
E2
(
silastic
implants)
for
10
or
20
days,
pituitary
weights
and
Prl
levels
were
dramatically
increased
in
F344,
minimal
effects
in
SD
rats.

In
ovariectomized
rats,
control
pituitary
gland
weights
were
the
lowest
in
Brown­
Norway
>
W
istar
rats;
After
E2
treatment
(
10
mg
s.
c.
pellet
for
4
weeks),
there
was
a
significant
>
3
fold
increase
in
pituitary
weights
in
F344
rats,
a
significant
>
0.5
fold
increase
in
pituitary
weights
in
Brown­
Norway
rats
F344
sensitive
to
DES­
induced
pituitary
tumors;
SD
and
BN
resistant
to
effects
of
DES
on
pituitary
gland.

Cadmium­
induced
toxicity
of
the
ovaries,
uterus,
cervix,
and
liver,
cadmium
induced
uterine
lesions
in
F344
not
W
F
33Schechter
et
al.,
1987
34Yin
et
al.,
2001
35Wendell
et
al.,
1996,
1997,
1998,
2000;
Chun
et
al.,
1998
36Rehm
and
Waalkes,
1988
Table
2.
(
continued)

Endocrine
Endpoints
Strains
Chemical
Similarities
Differences
Key
References
70
Organ
weights,
histopathology
spermatogenesis
(
mice)
Jcl:
ICR
and
CD­
1
mice
C57BL/
6N
and
ICR
mice
C57BL/
6J
(
B6),
CD­
1,
C17/
Jls,
and
S15
BALB/
cAnNCrD
BA/
2NCr,
C57BL/
6NCr,
NFS/
NCr
di­
2­
ethylhexylphthalate
(
DEHP)

BPA,
E2
E2
Cadmium
Increase
in
liver
weight,
both
strains
BPA
did
not
affect
m
ale
reproductive
organ
weights
during
any
dose/
exposure
period
DEHP
decreased
testicular
weight
in
CD­
1
mice
not
Jcl:
ICR
E2
(
10
:
g/
kg
from
pnd
27
to
48,
as
a
positive
control)
produced
significant
decreases
in
absolute
and
relative
testes,
epididymides,
and
sem
inal
vesicle
weights
compared
to
controls
in
C57BL/
6N
mice,
while
ICR
m
ice
were
unaffected.

B6
and
C17/
Jls
were
sensitive
to
E2
showing
a
maximal
suppression
of
testis
weight
and
spermatogenesis
even
at
the
lowest
dose
of
E2
(
2.5
:
g),
with
no
effect
on
testis
weight
or
spermatogenesis
in
CD­
1
or
S15
up
to
10
:
g
E2.

cadmium­
induced
ovarian
hemorrhagic
necrosis
only
observed
in
DBA/
2NCr
mice
37Oishi
et
al.,
1993
38Nagao
et
al.,
2002
39Spearow
et
al.,
1999;
2001
40Rehm
and
Waalkes,
1988
1­
40Footnotes
are
used
for
the
identification
of
references
in
Table
3
71
Thus
after
conducting
this
literature
review,
strain­
related
differences
in
effects
on
endocrine­
mediated
endpoints
in
response
to
a
wide
variety
of
endocrine­
disrupting
chemicals
was
obvious.
The
sensitivities
of
various
strains
to
chemicals
producing
effects
on
various
endocrine
endpoints
are
summarized
in
Table
3.
Since
effects
on
many
of
the
endpoints
were
also
chemical
specific,
the
chemicals
are
also
included
in
Table
3.

Table
3
Summary
of
Agent­
and
Endpoint­
Specific
Intraspecies
Differences
Endocrine
Endpoint
Chemical
Sensitive*
Strains
Less
Sensitive/
Insensitive
Strains
References
(
from
Table
2)

Uterine
Weight
EE
Wistar,
Da/
Han
SD
1
BPA
Da/
Han
Wistar,
SD
1
NP
AP>
SD
2
EE,
DES
SD,
F344
3
D4
SD
F344
4
E2
SDïF344
6
tamoxifen
SD
F344
6
AGD
p,
p'­
DDE
LE
SD
7
flutamide
SD,
LE
7
Nipple
retention
p,
p'­
DDE
SD
LE
7
flutamide
SD,
LE
7
vinclozolin
LE
>
W
istar
14
PPS
E2
F344,
SD
8
p,
p'­
DDE
SD,
LE
7
VO
p,
p'­
DDE
SD,
LE
7
BPA
AP
SD
9
Male
reproductive
organ
wts.
flutamide
LE,
SD
7
E2
F344,
SD
8
low
dose
E2
SD
F344
8
vinclozolin
LE
Wistar
14
BPA
C57BL/
6N,
ICR
38
E2
C57BL/
6N
38
E2
B6,
C17/
Jls
ICR,
CD­
1,
S15
39
DEHP
CD­
1
Jcl:
ICR
37
Table
3.
(
continued)

Endocrine
Endpoint
Chemical
Sensitive*
Strains
Less
Sensitive/
Insensitive
Strains
References
(
from
Table
2)

72
Estrous
cycle/
ovulation
feed
restriction
F344,
BN
SD,
LE
18
atrazine
LE
SD
21
atrazine
SD
F344
22
Fertility/
gestational
effects
atrazine
Holtzman
F344,
SD,
LE
29
atrazine
F344
SD,
LE
30
BDCM
F344
SD
31
Andrology
BPA
AP
SD
15
lead
SD
17
E2
B6,
C17/
Jls
CD­
1,
S15
39
Hormone
Levels
p,
p'DDE
SD
(
FSH,
E2,
T4)
LE
(
FSH,
Prl,
LH)
13
Hormone
Levels
(
continued)
p,
p'DDE
LE
(
E2,
T4,
T,
DHT,
TSH)
SD
(
Prl,
LH,
T,
DHT,
TSH)
13
E2
SD
(
Prl)
F344
(
Prl)
22
TCDD
Han/
W
istar
(
T,
LH)
LE
(
T,
LH)
23
atrazine
LE
(
LH,
Prl)
SD
(
LH,
Prl)
24
atrazine
Holtzman
(
P)
SD
(
E2,
P)
26
E2
F344
(
Prl)
SD
(
Prl)
25
BPA
F344
(
Prl)
SD
(
Prl)
25
TCDD
LE
(
T4)
27
TSH,
TRH
SD,
F344
(
T4)
SD
(
T3)
28
TSH,
TRH
F344
(
T3)
28
Pituitary
Weights
E2
F344
SD
33
E2
F344>
BN
Wistar,
Donryu
34
DES
F344
SD,
BN
35
Histopathology
(
reproduct­
ive
organs)
BPA
F344
(
females)
SD
(
females)
10
DMAB
F344>
ACI>
Lewis>
CD
(
males)
Wistar
(
males)
12
vinclozolin
LE
(
males)
Wistar
(
males)
14
Table
3.
(
continued)

Endocrine
Endpoint
Chemical
Sensitive*
Strains
Less
Sensitive/
Insensitive
Strains
References
(
from
Table
2)

73
cadmium
F344
(
females)
WF
(
cadmium)
40
Histopathology
(
continued)
atrazine
SD
(
females)
F344
(
females)
22
*"
Sensitive"
refers
to
a
greater
response,
and
may
not
reflect
dose
level
Table
3
summarizes
the
intraspecies
comparisons
of
the
effects
of
endocrinedisrupting
chemicals
on
endocrine
endpoints
in
EDSP
assays.
In
the
uterotrophic
assay,
the
SD
rat
is
sensitive
to
many
endocrine
disruptors,
except
for
EE
and
BPA
in
one
study,
in
addition
to
many
other
strains.
Less
sensitive
strains
were
F344
in
response
to
D4,
and
Wistar
in
response
to
BPA.
LE
and
SD
rats
were
differentially
sensitive
to
the
effects
of
flutamide
and
p
p'­
DDE
on
AGD
and
nipple
retention.
SD
and
F344
rats
showed
sensitivity
to
PPS
in
response
to
E2,
while
SD
and
LE
were
insensitive
to
p,
p'­
DDE.
AP
rats
were
sensitive
to
BPA
on
VO,
while
SD
and
LE
rats
were
less
sensitive
to
BPA.
Sensitivity
to
the
effects
on
male
reproductive
organ
weights
was
found
in
most
rat
strains
and
chemicals
studies
except
for
F344
(
low
dose
E2),
and
Wistar
(
vinclozolin).
Different
strains
of
mice
were
differentially
sensitive
to
the
effects
of
E2
on
male
reproductive
organ
weights
and
andrology.
Different
strains
of
rats
were
differentially
sensitive
to
the
effects
of
BPA,
atrazine,
BDCM,
p,
p'­
DDE,
TCDD,
E2,
lead,
cadmium,
DES,
vinclozolin,
and
DMAB,
on
estrous
cycle,
fertility,
pregnancy
loss,
andrology,
hormone
levels
and
histopathology
of
the
reproductive
organs.
Pituitary
weights
were
affected
in
F344
rats
after
exposure
to
DES
and
E2
while
in
other
strains
pituitary
weights
were
unaffected.
Overall,
there
was
no
one
strain
which
was
sensitive
to
all
of
the
endocrine­
disrupting
chemicals
at
most
of
the
endocrine
endpoints.
Effects
on
endocrine
endpoints
were
dependent
on
strain
and
chemical,
and
effects
on
strains
were
dependent
on
the
endpoints
and
chemicals.
There
were
no
clear
patterns
indicating
the
optimal
strain
for
detection
of
effects
due
to
the
majority
of
endocrineactive
compounds
tested.
In
addition,
it
was
not
clear
that
inbred
strains
or
outbred
strains
would
be
the
better
choice
in
species/
strain
selection.
In
selection
of
the
appropriate
species/
strain
for
EDSP
assays,
it
may
be
important
to
consider
the
endocrine
endpoints
assessed
and
the
test
chemical
employed.
Obviously,
it
would
be
more
thorough
to
conduct
multi­
strain
assays
to
increase
to
chances
of
detecting
endocrine
effects,
but
extremely
difficult
to
determine
which
strains
should
be
chosen
considering
the
considerable
variation
across
strains,
even
under
the
same
experimental
conditions
(
which
would
minimize
confounders).
74
The
five
example
assays
under
consideration
by
the
EDSP
(
Table
1)
were
compared
against
strain
sensitivities
(
Table
3).
In
the
one­
generation,
two­
generation,
and
in
utero
lacational
assays,
a
similar
array
of
endpoints
are
assessed,
from
gestational
indices,
reproductive
development,
onset
of
puberty,
estrous
cyclicity,
hormone
levels,
andrology,
organ
weights,
and
histopathology.
Selecting
a
rat
strain
that
is
sensitive
at
the
greatest
number
of
endpoints
is
difficult,
due
to
the
different
strain
sensitivities
observed
with
different
chemicals.
For
example,
if
selecting
one
chemical,
in
this
case,
p,
pNDDE,
SD
rats
are
sensitive
at
about
half
of
the
endpoints,
and
LE
rats
are
sensitive
at
the
other
half
of
the
endpoints.
These
data
support
the
case
for
performing
reproductive
toxicity
assays
in
more
than
one
strain
to
maximize
the
probability
of
detecting
an
effect
at
an
endocrine
endpoint.
Uterine
weight,
as
assessed
in
the
uterotrophic
assay,
which
is
currently
in
the
proposed
in
utero
lactational
assay,
is
an
endpoint
which
appears
relatively
sensitive
across
strains.
In
a
review
of
uterotrophic
assays
performed
in
19
different
laboratories
and
two
model
systems
(
ovariectomized
and
immature
rats),
the
response
to
EE
was
robust,
reproducible
and
sensitive
across
laboratories,
regardless
of
differences
in
strain,
diet,
bedding,
housing,
and
vehicle
(
Kanno
et
al.,
2001).
In
endpoints
like
fertility
and
gestational
parameters,
the
SD
rat
appears
less
sensitive
than
the
F344
rat
to
several
chemicals,
suggesting
that
the
F344
rat
may
be
a
better
strain
for
assessing
the
effects
of
chemicals
at
these
endpoints.
However,
the
F344
has
a
small
litter
size,
thus
reducing
the
number
of
animals
available
for
multiple
evaluations,
and
has
a
high
incidence
of
spontaneous
testicular
tumors,
which
may
confound
potential
effects
on
male
reproductive
organs.
Both
the
adult
male
and
pubertal
male
and
female
assays
include
assessments
of
hormone
levels.
Chemical
effects
on
hormone
levels
were
highly
strain
dependent,
with
no
one
strain
insensitive
to
changes
in
every
hormone
level.
With
inclusion
of
enough
different
hormone
level
determinations,
the
possibility
detecting
a
chemical­
induced
change
in
one
strain
would
be
enhanced.
Inclusion
of
an
additional
rat
strain
(
e.
g.
the
LE
rat)
in
the
pre­
validation
of
EDSP
assays
may
provide
more
information
on
the
strain
sensitivity
of
the
various
endpoints
and
assays,
in
addition
to
providing
more
flexibility
to
laboratories
in
selecting
strains
for
performance
of
the
assays.

Since
interspecies
comparisons
performed
under
the
same
experimental
conditions
are
few,
and
strain
is
obviously
also
a
confounder
in
comparing
speciesrelated
differences
in
sensitivity
to
endocrine­
disrupting
chemicals,
conclusions
about
comparisons
are
limited.
There
are
differences
in
the
response
of
the
rat
and
mouse
reproductive
tracts
to
various
reproductive
toxicants.
In
some
cases,
mice
appear
more
sensitive
and
in
other
cases
rats
are
more
sensitive.
In
a
comparison
of
reproductive
organ
weights,
sperm
parameters,
and
vaginal
cytology
from
fifty
13­
week
studies
involving
24
chemicals
in
seven
different
laboratories
(
and
four
routes
of
exposure)
for
the
National
Toxicology
Program
in
B6C3F1
mice
and
F344
rats
(
Morrissey
et
al.,
1988),
there
was
considerable
interlaboratory
variability
due
to
confounding
factors
(
such
as
different
suppliers
and
environmental
conditions)
resulting
in
only
a
58%
agreement
in
endocrine
endpoints
(
i.
e.
either
no
response
in
either
species,
or
at
least
one
endpoint
affected
in
both
species)
in
response
to
reproductive
toxicants.
It
is
possible
that
EDSP
assays
involving
only
one
species
and
strain,
namely
the
SD
rat,
may
not
detect
effects
in
endocrine
endpoints
that
occur
in
mice
(
and
vice
versa).
However,
a
major
question
75
that
cannot
be
answered
is
which
animal
model
will
provide
the
most
appropriate
data
on
the
ability
of
the
test
chemical
to
interact
with
the
endocrine
system,
in
order
to
predict
the
effects
of
endocrine­
active
chemicals
in
humans,
and/
or
other
species
of
concern.
Since
we
cannot
identify
the
most
relevant
or
sensitive
animal
model
with
the
existing
data,
because
the
sensitivity
depends
on
the
endpoint(
s)
chosen,
the
chemicals
evaluated,
the
timing,
duration
and
route
of
exposure,
the
dose
levels,
and
is
confounded
by
the
varying
genotype
by
environment
interactions,
the
main
issue
is
whether
to
use
inbred
versus
outbred
strains.
Inbred
strains
are
homogeneous
at
all
loci,
and
have
a
limited
range
of
responses
(
less
variability,
but
an
effect
may
be
missed),
though
using
several
genetically­
defined
inbred
strains
in
endocrine
screens
may
provide
a
wider
spectrum
of
responsivity.
If
selecting
a
single
strain
for
endocrine
screens,
outbred
strains
have
more
genetic
variability,
exhibit
a
broader
range
of
responsivity
(
with
a
greater
likelihood
of
detecting
an
effect),
and
may
be
more
appropriate.
76
5.0
References
Adham,
I.
M.,
Steding,
G.,
Thamm,
T.,
Bullesbach,
E.
E.,
Schwabe,
C.,
Paprotta,
I.
And
Engel,
W.
2002.
The
overexpression
of
the
insl3
in
female
mice
causes
descent
of
the
ovaries.
Mol.
Endocrinol.
16,
244­
262.

Anderson,
D.,
Hughes,
J.
A.,
Edwards,
A.
J.,
and
Brinkworth,
M.
H.
(
1998).
A
comparison
of
male­
mediated
effects
in
rats
and
mice
exposed
to
1,3­
butadiene.
Mutation
Research
397,
77­
84.

Ando­
Lu,
J.,
Sasahara,
K.,
Nishiyama,
K.,
Takano,
Satoshi,
Takahashi,
M.,
Yoshida,
M.,
and
Maekawa,
A.
(
1998).
Strain­
differences
in
proliferative
activity
of
uterine
endometrial
cells
in
Donryu
and
Fischer
344
rats.
Exp.
Toxic.
Pathol.
50
(
3),
185­
190.

Apostoli,
P.,
Kiss,
P.,
Porru,
S.,
Bonde,
J.
P.,
Vanhoorne,
M.,
and
the
ASCLEPIOS
study
group.
(
1998).
Review:
Male
reproductive
toxicity
of
lead
in
animals
and
humans.
Occup.
Environ.
Med.
55,
364­
374.

Ashby,
J.,
Odum,
J.,
and
Foster,
J.
R.
(
1997).
Activity
of
raloxifene
in
immature
and
ovariectomized
rat
uterotrophic
assays.
Fundamental
and
Applied
Toxicology
25(
3),
226­
31.

Bailey,
J.
A.
and
Nephew,
K.
P.
(
2002).
Strain
differences
in
tamoxifen
sensitivity
of
Sprague­
Dawley
and
Fischer
344
rats.
Anticancer
Drugs
13(
9):
939­
47.

Barkley,
M.
S.
and
Bradford,
G.
E.
(
1981).
Estrous
cycle
dynamics
in
different
strains
of
mice.
Proc.
Soc.
Exp.
Biol.
Med.
167,
70­
77.

Bartolomucci,
A.,
Palanza,
P.,
Gaspani,
L.,
Limiroli,
E.,
Panerai,
A.
E.,
Cerzsini,
G.,
Poli,
M.
D.,
and
Parmigiani,
S.
(
2001).
Social
status
in
mice:
behavioral,
endocrine,
and
immune
changes
are
context
dependent.
Physiol.
Behav.
73(
3),
401­
410.

Becker,
G.,
and
Flaherty,
T.
B.
(
1968).
Group
size
as
a
determinant
of
dominancehierarchy
stability
in
the
rat.
J.
Comp.
Physiol.
Psychol.
66(
2),
473­
476.

Biegel,
L.
B.,
J.
C.
Cook,
M.
E.
Hurtt,
and
J.
C.
O'Connor
(
1998).
Effects
of
17
betaestradiol
on
serum
hormone
concentrations
and
estrous
cycle
in
female
Crl:
CD
BR
rats:
effects
on
parental
and
first
generation
rats.
Toxicol.
Sci.
44,
143­
154.

Bielmeier,
S.
R.,
Best,
D.
S.,
Guidici,
D.
C.,
and
Narotsky,
M.
G.
(
2001).
Pregnancy
loss
in
the
rat
caused
by
bromodichloromethane.
Tox.
Sci.
59(
2),
309­
315.
77
Bindon,
B.
M.
(
1984).
Reproductive
biology
of
the
Booroola
Merino
sheep.
Aust
J
Biol
Sci
37(
3):
163­
89.

Blanchard,
D.
C.,
Sakai,
R.
R.,
McEwen,
B.,
and
Weiss,
S.
M.
(
1993).
Subordination
stress:
behavioral,
brain
and
neuroendocrine
correlates.
Behav.
Brain
Res.
58(
1­
2),
113­
121.

Bradford,
G.
E.
(
1968).
Selection
for
litter
size
in
mice
in
the
presence
and
absence
of
gonadotropin
treatment.
Genetics
58:
283­
295.

Bradford,
G.
E.
(
1969).
Genetic
control
of
ovulation
rate
and
embryo
survival
in
mice.
I.
Response
to
selection.
Genetics,
Princeton
61:
905­
921.

Bradford,
G.
E.
(
1979).
Genetic
variation
in
prenatal
survival
and
litter
size.
J
Anim
Sci
49(
Suppl
2):
66­
74.

Bradford,
G.
E.,
Barkely,
M.
S.
et
al.
(
1979).
Physiological
effects
of
selection
for
aspects
for
efficiency
of
reproduction.
Symposium
on
Selection,
E.
A.
A.
P.,
Harrowgate,
England,
Commonwealth
Bureau.

Bradford,
G.
E.
1969.
Genetic
control
of
ovulation
rate
and
embryo
survival
in
mice.
I.
Response
to
selection.
Genetics,
Princeton.
61,
905­
921.

Bradley,
D.
J.,
Young,
W.
S.,
and
Weinberger,
I.
C..
(
1989).
Differential
expression
of
alpha
and
beta
thyroid
hormone
receptor
genes
in
rat
brain
and
pituitary.
Proc
Natl
Acad
Sci
USA.
(
86),
7250­
7254.

Bresnahan,
J.
F.,
Kitchell,
B.
B.,
and
Wildman,
M.
F.
(
1983).
Facial
hair
barbering
in
rats.
Lab.
Anim.
Sci.
33(
3),
290­
291.

Brinkworth,
M.
H.,
Anderson,
D.,
and
McLean,
A.
E.
M.
(
1992).
Effects
of
dietary
imbalances
on
spermatogenesis
in
CD­
1
mice
and
CD
rats.
Food
Chem.
Toxic.
30(
1),
29­
35.

Brown,
S.
L.,
Brett,
S.
M.,
Gough,
M.,
Rodericks,
J.
V.,
Tardiff,
R.
G.,
And
Turnbull,
D.
1988.
Review
of
interspecies
risk
comparisons.
Regulatory
Toxicology
and
Pharmacology
8,
191­
206.

Canzian,
F.
(
1997).
Phylogenetics
of
the
laboratory
rat
Rattus
norvegicus.
Genome
Res.
7(
3):
262­
7.
78
Carney,
E.
W.,
Schortichini,
B.
S.,
and
Crissman,
J.
W.
(
1998)
Feed
restriction
during
in
utero
and
neonatal
life:
Effects
on
reproductive
and
developmental
endpoints
in
the
CD
rat.
The
Toxicologist
42(
Suppl.
1),
102­
103.

Chapin,
R.
E.,
Gulati,
D.
K.,
Barnes,
L.
H.,
and
Teague
J.
L.
1993.
The
effects
of
feed
restriction
on
reproductive
function
in
SD
rats.
Fundamental
and
Applied
Toxicology
20,
23­
29.

Christian,
M.
S.,
Hoberman,
A.
M.,
Bachmann,
S.,
and
Hellwig,
J.
(
1998).
Variability
in
the
uterotrophic
response
assay
(
an
in
vivo
estrogenic
response
assay)
in
untreated
control
and
positive
control
(
DES­
DP,
2.5
:
g/
kg,
BID)
Wistar
and
Sprague­
Dawley
rats.
Drug
and
Chemical
Toxicology
21(
Suppl.
1),
51­
100.

Chun,
T.
Y.,
D.
Wendell,
et
al.
(
1998).
Estrogen­
induced
rat
pituitary
tumor
is
associated
with
loss
of
retinoblastoma
susceptibility
gene
product.
Mol.
Cell.
Endocrinol.
146(
1­
2),
87­
92
Cooper,
R.
L.,
Stoker,
T.
E.,
Tyrey,
L.,
Goldman,
J.
M.,
and
McElroy,
W.
K.
(
2000).
Atrazine
disrupts
the
hypothalamic
control
of
pituitary­
ovarian
function.
Toxicol.
Sci.
53,
297­
307.

Cowley,
A.
W.,
Jr.,
Roman,
R.
J.,
et
al.
(
2001).
Brown
Norway
chromosome
13
confers
protection
from
high
salt
to
consomic
Dahl
S
rat.
Hypertension
37(
2
Part
2),
456­
61.

Cummings,
A.
M.,
Rhodes,
B.
E.
and
Cooper,
R.
L.
(
2000).
Effect
of
atrazine
on
implantation
and
early
pregnancy
in
4
strains
of
rats.
Toxicol.
Sci.
58,
135­
143.

D'Arbe,
M.,
Einstein,
R.,
and
Lavidis,
N.
A.
(
2002).
Stressful
animal
housing
conditions
and
their
potential
effect
on
sympathetic
neurotransmission
in
mice.
Am.
J.
Physiol.
Regul.
Integr.
Comp.
Physiol.
282(
5),
R1422­
1428.

De
Leon,
D.
D.
and
Barkley,
M.
S.
(
1987).
Male
and
female
genotype
mediate
pheromonal
regulation
of
the
mouse
estrous
cycle.
Biol.
Reprod.
37(
5),
1066­
74.

Dhaher,
Y.
Y.,
B.
Greenstein,
de
Fougerolles,
N.
E.,
Khamashta,
M.,
Hugh,
G.
R.
(
2000).
Strain
differences
in
binding
properties
of
estrogen
receptors
in
immature
and
adult
BALB/
c
and
MRL/
MP­
lpr/
lpr
mice,
a
model
of
systemic
lupus
erythematosus.
Int.
J.
Immunopharmacol.
22(
3),
247­
54
Diel,
P.,
Laudenbach,
U.,
Smolnikar,
K.,
Schulz,
T.,
and
Michna,
H.
(
2001)
Bisphenol
A:
morphological
and
molecular
uterine
and
mammary
gland
reactions
in
different
strains
of
the
rat
(
Wistar,
Sprague­
Dawley,
Da/
Han).
Toxicologist
60(
1),
296.
79
Durrant,
B.
S.,
Eisen,
E.
J.,
et
al.
(
1980).
Ovulation
rate,
embryo
survival
and
ovarian
sensitivity
to
gonadotropins
in
mice
selected
for
litter
size
and
body
weight.
J.
Reprod.
Fert.
59,
329­
339.

Dussault,
J.
H.,
and
Ruel,
J.
(
1987).
Thyroid
hormones
and
brain
development.
Annu
Rev
Physiol.
(
49),
321­
334.

EDSTAC
(
Endocrine
Disruptor
Screening
and
Testing
Advisory
Committee)
(
1998).
Final
Report:
Volume
I
and
II.

Eisen,
E.
J.,
Legates,
J.
E.
et
al.
(
1970).
Selection
for
12­
day
litter
weight
in
mice.
Genetics
64(
3):
511­
32.

Eisen,
E.
J.
(
1972).
Long­
term
selection
response
for
12­
day
litter
weight
in
mice.
Genetics
72(
1):
129­
42.

Eisen,
E.
J.
(
1975).
Population
size
and
selection
intensity
effects
on
long­
term
selection
response
in
mice.
Genetics
79(
2):
305­
23.

Eisen,
E.
J.
and
Durrant,
B.
S.
(
1980).
Genetic
and
maternal
environmental
factors
influencing
litter
size
and
reproductive
efficiency
in
mice.
J.
Anim.
Sci.
50(
3):
428­
41.

Eisen,
E.
J.
and
Johnson,
B.
H.
(
1981).
Correlated
responses
in
male
reproductive
traits
in
mice
selected
for
litter
size
and
body
weight.
Genetics
99(
3­
4):
513­
24.

Eklund,
J.
and
Bradford,
G.
E.
(
1977).
Genetic
analysis
of
a
strain
of
mice
plateaued
for
litter
size.
Genetics
85(
3):
529­
42.

Eklund,
J.
and
Bradford,
G.
E.
(
1977).
Longeveity
and
lifetime
body
weight
in
mice
selected
for
rapid
growth.
Nature
265(
5589):
48­
9.

Eldridge,
A.
C.,
and
Kwolek,
W.
F.
(
1983).
Soybean
isoflavones:
effect
of
environment
and
variety
on
composition.
J.
Agric.
Food
Chem.
31,
394­
396.

Eldridge,
A.
C.,
Tennant,
M.
K.,
Wetzel,
L.
T.,
Breckenridge,
C.
B.,
and
Stevens,
J.
T.
(
1994).
Factors
affecting
mammary
tumor
incidence
in
chlortriazine­
treated
female
rats:
hormonal
properties,
dosage,
and
animal
strain.
Environmental
Health
Perspectives
102(
Suppl.
11),
29­
36.

Emery,
D.
E
and
Larsson,
K.
(
1979).
Rat
strain
differences
in
copulatory
behavior
after
para­
chlorophenylalanine
and
hormone
treatment.
J.
Comp.
Physiol.
Psychol.
93(
6),
1067­
1084.

Escobar,
G.,
Obregon,
M.
J.,
and
Rey,
F.
(
1990).
Contribution
of
maternal
thyroxine
to
fetal
thyroxine
pools
in
normal
rats
near
term.
Endocrinology
(
126),
2765­
2767.
80
Escobar,
G.,
Obregon,
M.
J.,
and
Rey,
F.
(
1987).
Fetal
and
maternal
thyroid
hormones.
Hormone
Research.
(
26),
12­
27.

Escobar,
G.,
Obregon,
M.
J.,
and
Rey,
F.
(
1988).
Transfer
of
Thyroid
hormones
from
the
mother
of
the
fetus.
In
Delang,
F.,
Fisher,
D.
A.,
and
Glinoer,
D.
(
Eds.)
Research
in
Congenital
Hypothyroidism.
Plenum
Press,
New
York,
NY,
p
15­
28.

Fail,
P.
A.,
Anderson,
S.
A.,
and
Friedman,
M.
A.
(
1999).
Response
of
the
Pituitary
and
Thyroid
to
Tropic
Hormones
in
Sprague­
Dawley
versus
Fischer
344
Male
Rats.
Toxicol.
Sci.
52,
107­
121.

Falconer,
D.
S.
(
1971).
Improvement
of
litter
size
in
a
strain
of
mice
at
a
selection
limit.
Genet.
Res.
17(
3):
215­
35.

Falconer,
D.
S.
(
1989).
Introduction
to
Quantitative
Genetics.
Essex,
England,
Longman
Scientific
&
Technical.

Festing,
M.
F.
(
1979).
Properties
of
inbred
strains
and
outbred
stocks,
with
special
reference
to
toxicity
testing.
J.
Toxicol.
Environ.
Health
5(
1):
53­
68.

Festing,
M.
F.
W.
(
1986).
The
case
for
isogenic
strains
in
toxicological
screening.
Arch.
Toxicol.
Suppl.
9,
127­
137.

Festing,
M.
F.
(
1987).
Genetic
factors
in
toxicology:
implications
for
toxicological
screening.
Crit.
Rev.
Toxicol.
18(
1):
1­
26.

Festing,
M.
F.
W.
(
1991).
Genetic
factors
in
neurotoxicology
and
neuropharmacology:
A
critical
evaluation
of
the
use
of
genetics
as
a
research
tool.
Experientia
47,
990­
998.

Festing,
M.
F.
W.
(
1992).
The
scope
for
improving
the
design
of
laboratory
animal
experiments.
Laboratory
Animals
26,
256­
267.

Festing,
M.
F.
(
1993).
Genetic
variation
in
outbred
rats
and
mice
and
its
implications
for
toxicological
screening.
J.
Exp.
Anim.
Sci.
35(
5­
6):
210­
20.

Festing,
M.
F.
(
1995).
Use
of
a
multistrain
assay
could
improve
the
NTP
carcinogenesis
bioassay.
Environ.
Health
Perspect.
103(
1):
44­
52.

Festing,
M.
F.
and
Altman,
D.
G.
(
2002).
Guidelines
for
the
design
and
statistical
analysis
of
experiments
using
laboratory
animals.
Ilar
J.
43(
4):
244­
58.

Festing,
M.
F.,
Diamanti,
P.,
and
Turton,
J.
A.
(
2001).
Strain
differences
in
haematological
response
to
chloramphenicol
succinate
in
mice:
implications
for
toxicological
research.
Food.
Chem.
Toxicol.
39(
4):
375­
83.
81
Flaws,
J.
A.,
Sommer,
R.
J.,
Silbergeld,
E.
K.,
Peterson,
R.
E.,
Hirshfield,
A.
N.
1997.
In
utero
and
lactational
exposure
to
2,3,7,8­
tetrachlorodibenzo­
p­
dioxin
(
TCDD)
induces
genital
dysmorphogenesis
in
the
female
rat.
Toxicol.
Appl.
Pharmacol.
147(
2),
351­
362.

Gallavan,
R.
H.,
J.
F.
Holson,
J.
F.
Knapp,
V.
L.
Reynolds,
and
D.
G.
Stump
(
1998).
Anogenital
distance:
Potential
for
confounding
effects
of
progeny
body
weights
on
interpreting
toxicologic
significance.
Teratology
57(
4/
5),
245
(
Abstract
87).

Goldey,
E.
S.,
Kehn,
L.
S.,
Rehnberg,
G.
L.,
and
Crofton,
K.
M.
1995.
Effects
of
developmental
hypothyroidism
on
auditory
and
motor
function
in
the
rat.
Toxicol.
Appl.
Pharmacol.
135(
1),
67­
76.

Goldman,
J.
M.,
S.
C.
Laws,
S.
K.
Balchak,
R.
L.
Cooper,
and
R.
J.
Kavlock
(
2000).
Endocrine­
disrupting
chemicals:
prepubertal
exposures
and
effects
on
sexual
maturation
and
thyroid
activity
in
the
female
rat.
A
focus
on
the
EDSTAC
recommendations.
Crit.
Rev.
Toxicol.
30(
2),
135­
196.

Gray,
J.
E.
(
1977).
Chronic
progressive
nephrosis
in
the
albino
rat.
CRC
Crit.
Rev.
Toxicol.
5(
2):
115­
44.

Gray,
L.
E.,
and
Ostby,
J.
S.
(
1995).
In
utero
2,3,7,8­
tetrachlorodibenzo­
p­
dioxin
(
TCDD)
alters
reproductive
morphology
and
function
in
female
rat
offspring.
Toxicol.
Appl.
Pharmacol.
133,
285­
294.

Gray,
L.
E.
Jr.,
J.
Ostby,
C.
Wolf,
C.
Lambright,
and
W.
Kelce
(
1998).
Annual
Review.
The
value
of
mechanistic
studies
in
laboratory
animals
for
the
production
of
reproductive
effects
in
wildlife:
endocrine
effects
on
mammalian
sexual
differentiation.
Environm.
Toxicol.
Chem.
17(
1),
109­
118.

Gray,
L.
E.
Jr.,
and
J.
Ostby
(
1998).
Effects
of
pesticides
and
toxic
substances
on
behavioral
and
morphological
reproductive
development:
endocrine
versus
nonendocrine
mechanisms.
Toxicol.
and
Indust.
Health
14(
1/
2),
159­
184.

Greco,
A.
M.,
Gambardella,
P.,
Sticchi,
R.,
D'Aponte,
D.,
Di
Renzo,
G.,
and
DeFranciscis,
P.
(
1989).
Effects
of
individual
housing
on
circadian
rhythms
of
adult
rats.
Physiol.
Behav.
45(
2),
363­
366.

Gregg,
D.
W.,
Galkin,
M.,
et
al.
(
1996).
Effect
of
estrogen
on
the
expression
of
galanin
mRNA
in
pituitary
tumor­
sensitive
and
tumor­
resistant
rat
strains.
Steroids
61(
8):
468­
72.

Griffith,
J.
S.,
S.
M.
Jensen,
et
al.
(
1997).
Evidence
for
the
genetic
control
of
estradiol­
regulated
responses.
Implications
for
variation
in
normal
and
pathological
hormone­
dependent
phenotypes.
Am.
J.
Pathol.
150(
6):
2223­
30.
82
Grimm,
M.
S.,
Emerman,
J.
T.,
and
Weinberg,
J.
(
1996).
Effects
of
social
housing
condition
as
behavior
on
growth
of
the
Shionogi
mouse
mammary
carcinoma.
Physiol.
Behav.
59(
4­
5),
633­
642.

Haavisto,
T.,
Nurmela,
K.,
Pohjanvirta,
R.,
Huuskonen,
H.,
El­
Gehani,
F.,
and
Paranko,
J.
(
2001).
Prenatal
testosterone
and
luteinizing
hormone
levels
in
male
rats
exposed
during
pregnancy
to
2,3,7,8­
tetrachlorodibenzo­
p­
dioxin
and
diethylstilbestrol.
Molecular
and
Cellular
Endocrinology
178,
169­
179.

Haemisch,
A.,
Vass,
T.,
and
Gartner,
K.
(
1994).
Effects
of
environmental
enrichment
on
aggressive
behavior,
dominance
hierarchies,
and
endocrine
states
in
male
DBA/
2J
mice.
Physiol.
Behav.
56(
5),
1041­
1048.

Haseman,
J.
K.,
Bourbina,
J.,
and
Eustis,
S.
L.
(
1994).
Effect
of
individual
housing
and
other
experimental
design
factors
on
tumor
incidence
in
B6C3F1
mice.
Fundam.
Appl.
Toxicol.
23(
1),
44­
52.

Hayes,
T.,
K.
Haston,
M.
Tsui,
A.
Hoang,
C.
Haeffele
and
A.
Vonk
(
2003).
"
Atrazine­
Induced
Hermaphroditism
at
0.1
ppb
in
American
Leopard
Frogs
(
Rana
pipiens):
Laboratory
and
Field
Evidence."
Environ
Health
Perspect
111(
4):
568­
75.
http://
www.
ncbi.
nlm.
nih.
gov/
entrez/
query.
fcgi?
cmd=
Retrieve&
db=
PubMed&
dopt=
Citati
on&
list_
uids=
12676617
Hayes,
T.
B.,
A.
Collins,
M.
Lee,
M.
Mendoza,
N.
Noriega,
A.
A.
Stuart
and
A.
Vonk
(
2002).
"
Hermaphroditic,
demasculinized
frogs
after
exposure
to
the
herbicide
atrazine
at
low
ecologically
relevant
doses."
Proc
Natl
Acad
Sci
U
S
A
99(
8):
5476­
80.
http://
www.
ncbi.
nlm.
nih.
gov/
entrez/
query.
fcgi?
cmd=
Retrieve&
db=
PubMed&
dopt=
Citati
on&
list_
uids=
11960004
Hellwig,
J.,
van
Ravenzwaay,
B.,
Mayer,
M.,
and
Gembardt,
C.
(
2000).
Pre­
and
postnatal
oral
toxicity
of
vinclozolin
in
Wistar
and
Long­
Evans
rats.
Regul.
Toxicol.
Pharmacol.
32,
42­
50.

Holsapple,
M.,
Reynolds,
B.
Wiescinski,
C.,
Anderson,
P.,
Carney,
E.
1998.
Feed
restriction
during
in
utero
and
neonatal
life:
effects
on
immune
parameters
in
the
rat.
Toxicologist
42(
1­
S),
102.

Hossaini,
A.,
Larsen,
J.­
J.,
and
Larsen,
J.
C.
(
2000).
Lack
of
oestrogenic
effects
of
food
preservatives
(
parabens)
in
uterotrophic
assays.
Food
Chem.
Toxicol.
38,
319­
323.

Hunt,
P.
A.,
Koehler,
K.
E.,
Susiarjo,
J.,
Hodges,
C.
A.,
Ilagan,
A.,
Voight,
R.
C.,
Thomas,
S.,
Thomas,
B.
F.,
and
Hassold,
T.
J.
(
2003).
Bisphenol
A
exposure
causes
meiotic
aneuploidy
in
the
female
mouse.
Current
Biology
13,
546­
553.
83
Inano,
H.,
Suzuki,
K.,
et
al.
(
1996).
Relationship
between
induction
of
mammary
tumors
and
change
of
testicular
functions
in
male
rats
following
gamma­
ray
irradiation
and/
or
diethylstilbestrol.
Carcinogenesis
17(
2):
355­
60.

ILAR
Journal
Online
(
Fall
1992),
Volume
34(
4);
http://
dels.
nas.
edu/
ilar/
jour_
online/
34_
4/
definitionandnomenrat.
asp
Johnson,
R.
K.,
Eckardt,
G.
R.,
Rathje,
T.
A.,
Drudik,
D.
K.
(
1994).
Ten
generations
of
selection
for
predicted
weight
of
testes
in
swine:
direct
response
and
correlated
response
in
body
weight,
backfat,
age
at
puberty,
and
ovulation
rate.
J.
Anim.
Sci.
72(
8),
1978­
1988.

Kacew,
S.,
Ruben,
Z.,
and
McConnell,
R.
F.
(
1995).
Strain
as
a
determinant
factor
in
the
differential
responsiveness
of
rats
to
chemicals.
Toxicologic
Pathology
23(
6),
701­
714.

Kacew,
S.
and
Festing,
M.
F.
W.
(
1996).
Role
of
rat
strain
in
the
differential
sensitivity
to
pharmaceutical
agents
and
naturally
occurring
substances.
Journal
of
Toxicology
and
Environmental
Health
47,
1­
30.

Kanno,
J.,
Onyon,
L.,
Haseman,
J.,
Fenner­
Crisp,
P.,
Ashby,
J.,
and
Owens,
W.
(
2001).
The
OECD
program
to
validate
the
rat
uterotrophic
bioassay
to
screen
compounds
for
in
vivo
estrogenic
responses:
phase
1.
Environ.
Health
Perspect.
109(
8),
785­
794.

Karolewicz,
B.,
and
Paul,
I.
A.
(
2001).
Group
housing
of
mice
increases
immobility
and
antidepressant
sensitivity
in
the
forced
swim
and
tail
suspension
tests.
Europ.
J.
Pharmacol.
415(
2­
3),
197­
201.

Kennedy,
G.
C.,
and
Mitra,
J.
(
1963).
Body
weight
and
food
intake
as
initiating
factors
for
puberty
in
the
rat.
Physiology
166,
408­
418.

Kirkpatrick,
B.
W.,
Mengelt,
A.,
et
al.
(
1998).
Identification
of
quantitative
trait
loci
for
prolificacy
and
growth
in
mice.
Mammalian
Genome
9:
97­
102
Klinefelter,
G.
R.,
Strader,
L.
F.,
Suarez,
J.
D.,
and
Roberts,
N.
L.
(
2003).
Bromochloroacetic
acid
exerts
qualitative
effects
on
rat
sperm:
implications
for
a
novel
biomarker.
Toxicol.
Sci.
68(
1),
164­
173.

Klinger,
M.
M.,
MacCarter,
G.
D.,
et
al.
(
1996).
Body
weight
and
composition
in
the
Sprague
Dawley
rat:
comparison
of
three
outbred
sources.
Lab.
Anim.
Sci.
46(
1):
67­
70.

Koehler,
K.,
Voight,
R.,
Thomas,
S.,
Lamb,
B.,
Hassold,
T.,
and
Hunt,
P.
A.
(
2003).
When
disaster
strikes:
rethinking
polycarbonate
caging.
Lab.
Animal
32,
32­
36.
84
Koyama,
S.,
and
Kamimura,
S.
(
1999).
Lowered
sperm
motility
in
mice
of
subordinate
social
status.
Physiol.
Behav.
65(
4­
5),
665­
669.

Koyama,
S.,
and
Kaminura,
S.
(
2000).
Influence
of
social
dominance
and
female
odor
on
the
sperm
activity
of
male
mice.
Physiol.
Behav.
71(
3­
4),
415­
422.
Kumagai,
J.,
Hsu,
S.
Y.,
Matsumi,
H.,
Roh,
J.­
S.,
Fu,
P.,
Wade,
J.
D.,
Bathgate,
R.
A.
D.,
and
Hseuh,
A.
J.
W.
2002.
INSL3/
Leydig
Insulin­
like
Peptide
Activates
the
LGR8
Receptor
Important
in
Testis
Descent.
J.
Biol.
Chem.
277(
35),
31283­
31286.

Lamm,
S.
H.,
and
Doemland,
M.
(
1999).
Has
perchlorate
in
the
drinking
water
increased
the
rate
of
congenital
hypothyroidism?
J.
Occup.
Environ.
Med.
41(
5),
409­
411.

Land,
R.
B.
(
1970).
Genetic
and
phenotypic
relationships
between
ovulation
rate
and
body
weight
in
the
mouse.
Genet.
Res.
15:
171­
182.

Land,
R.
B.,
De
Reviers,
M.
M.,
et
al.
(
1974).
Quantitative
physiological
studies
of
genetic
variation
in
the
ovarian
activity
of
the
rat.
J.
Reprod.
Fertil.
38(
1):
29­
39.

Land,
R.
B.
and
Falconer,
D.
S.
(
1969).
Genetic
studies
of
ovulation
rate
in
the
mouse.
Genetical
Research
13:
25­
46.

Laws,
S.
C.,
Carey,
S.
A.,
Ferrell,
J.
M.,
Bodman,
G.
J.,
and
Cooper,
R.
L.
2000.
Estrogenic
activity
of
octylphenol,
nonylphenol,
bisphenol
A
and
methoxychlor
in
rats.
Toxicol.
Sci.
54(
1),
154­
167.

Lewis,
R.
W.,
Brooks,
N.,
Milburn,
G.
M.,
Soames,
A.,
Stone,
S.,
Hall,
M.,
and
Ashby
J.
(
2003).
The
effects
of
the
phytoestrogen
genistein
on
the
postnatal
development
of
the
rat.
Toxicol.
Sci.
71(
1),
74­
83.

Liang,
M.,
Yuan,
B.
et
al.
(
2002).
Renal
medullary
genes
in
salt­
sensitive
hypertension:
a
chromosomal
substitution
and
cDNA
microarray
study.
Physiol.
Genomics
8(
2):
139­
49.

Liberati,
T.
A.,
Roe,
B.
J.,
and
Feuston,
M.
H.
(
2002).
An
oral
gavage
control
embryo­
fetal
development
study
of
the
Wistar
Hannover
rat.
Drug
and
Chemical
Toxicology
25(
1),
109­
130.

Long,
S.
Y.
(
1972).
Hair
nibbling
and
whisker
trimming
as
indicators
of
social
hierarchy
in
mice.
Anim.
Behav.
20(
1),
10­
12.

Long,
X.,
Steinmetz,
R.,
Ben­
Jonathan,
N.,
Caperell­
Grant,
A.,
Young,
P.
C.
M.,
Nephew,
K.
P.,
and
Bigsby,
R.
M.
(
2000).
Strain
differences
in
vaginal
responses
to
the
xenoestrogen
Bisphenol
A.
Environmental
Health
Perspectives
108(
3),
243­
247.
85
Lubritz,
D.
L.,
Eisen,
E.
J.,
et
al.
(
1991).
Effect
of
selection
for
litter
size
and
body
weight
on
hormone­
induced
ovulation
rate
in
mice.
J.
Anim.
Sci.
69(
11):
4299­
4305.

Lynch,
M.
and
Walsh,
B.
(
1998).
Genetics
and
Analysis
of
Quantitative
Traits.
Sunderland,
MA
01375,
Sinauer
Associates,
Inc.

Manly,
K.
F.,
Cudmore,
R.
H.
Jr.,
et
al.
(
2001).
Map
Manager
QTX,
cross­
platform
software
for
genetic
mapping.
Mamm.
Genome
12(
12):
930­
2.

McIntyre,
B.
S.,
Barlow,
N.
J.,
Foster,
P.
M.
2001.
Androgen­
mediated
development
in
male
rat
offspring
exposed
to
flutamide
in
utero:
permanence
and
correlation
of
early
postnatal
changes
in
anogenital
distance
and
nipple
retention
with
malformations
in
androgen­
dependent
tissues.
Toxicol.
Sci.
62(
2),
236­
249.

McIntyre,
B.
S.,
Barlow,
N.
J.,
Foster,
P.
M.
2002.
Male
rats
exposed
to
linuron
in
utero
exhibit
permanent
changes
in
anogenital
distance,
nipple
retention,
and
epididymal
malformations
that
result
in
subsequent
testicular
atrophy.
Toxicol.
Sci.
65(
1),
62­
70.

McKim,
J.
M.,
Wilga,
P.
C.,
Breslin,
W.
J.,
Plotzke,
K.
P.,
Gallavan,
R.
H.
and
Meeks,
R.
G.
(
2001).
Potential
estrogenic
and
antiestrogenic
activity
of
the
cyclic
siloxane
octamethylcyclotetrasiloxane
(
D4)
and
the
linear
siloxane
hexamethyldisiloxane
(
HMDS)
in
immature
rats
using
the
uterotrophic
assay.
Toxicol.
Sci.
63,
37­
46.

McKinney,
T.
D.,
and
Pasley,
J.
N.
(
1973).
Effects
of
social
rank
and
social
disruption
in
adult
male
house
mice.
Gen.
Comp.
Endocrinol.
20(
3),
579­
583.

McNatty,
K.
P.,
Henderson,
K.
M.,
et
al.
(
1985).
Ovarian
activity
in
Booroola
X
Romney
ewes
which
have
a
major
gene
influencing
their
ovulation
rate.
J.
Reprod.
Fertil.
73(
1):
109­
20.

McNatty,
K.
P.,
Smith,
P.,
et
al.
(
1995).
Development
of
the
sheep
ovary
during
fetal
and
early
neonatal
life
and
the
effect
of
fecundity
genes.
J.
Reprod.
Fertil.
Suppl.
49:
123­
35.

Marshall,
K.
E.,
Godden,
E.
L.,
et
al.
(
2002).
In
silico
discovery
of
gene­
coding
variants
in
murine
quantitative
trait
loci
using
strain­
specific
genome
sequence
databases.
Genome
Biol.
3(
12):
RESEARCH0078.

Matthews,
J.,
T.
Celius,
et
al.
(
2000).
Differential
estrogen
receptor
binding
of
estrogenic
substances:
a
species
comparison.
J.
Steroid
Biochem.
Mol.
Biol.
74(
4),
223­
34.

Matin,
A.,
Collin,
G.
B.,
et
al.
(
1999).
Susceptibility
to
testicular
germ­
cell
tumours
in
a
129.
MOLF­
Chr
19
chromosome
substitution
strain.
Nat.
Genet.
23(
2):
237­
40.
86
Miller,
R.
A.,
Burke,
D.,
and
Austad,
S.
1999.
Exotic
mice
for
aging:
Authors'
reply.
Neurobiology
of
Aging
20,
245­
246.

Morrissey,
R.
E.,
George,
J.
D.,
Price,
C.
J.,
Tyl,
R.
W.,
Marr,
M.
C.
and
Kimmel,
C.
A.
(
1987).
The
developmental
toxicity
of
bisphenol
A
in
rats
and
mice.
Fundam.
Appl.
Toxicol.
8,
571­
582.

Morrissey,
R.
E.,
Schwetz,
B.
A.,
Lamb,
J.
C.
IV,
Ross,
M.
D.,
Teague,
J.
L.,
and
Morris,
R.
W.
(
1988).
Evaluation
of
rodent
sperm,
vaginal
cytology,
and
reproductive
organ
weight
data
from
National
Toxicology
Program
13­
week
studies.
Fundam.
Appl.
Toxicol.
11,
343­
358.

Myant,
N.
B.
(
1971).
The
role
of
thyroid
hormone
in
the
fetal
and
postnatal
development
of
mammals.
Hormones
in
Development.

Nadeau,
J.
H.,
Singer,
J.
B.
et
al.
(
2000).
Analysing
complex
genetic
traits
with
chromosome
substitution
strains.
Nat.
Genet.
24(
3):
221­
5.

Nagao,
T.,
Saito,
Y.,
Usumi,
K.,
Yoshimura,
S.,
and
Ono,
H.
(
2002).
Low­
dose
bisphenol
A
does
not
affect
reproductive
organs
in
estrogen­
sensitive
C57BL/
6N
mice
exposed
at
the
sexually
mature,
juvenile,
or
embryonic
stage.
Reproductive
Toxicology
16,
123­
130.

Nagy,
T.
R.,
Krzywanski,
D.,
Li,
J.,
Meleth,
S.,
and
Desmond,
R.
(
2002).
Effect
of
group
vs.
single
housing
on
phenotypic
variance
in
C57BL/
6J
mice.
Obes.
Res.
10(
5),
412­
415.

Narotsky,
M.
G.,
Best,
D.
S.,
Guidici,
D.
L.,
and
Cooper,
R.
L.
(
2001).
Strain
comparisons
of
atrazine­
induced
pregnancy
loss
in
the
rat.
Reproductive
Toxicology
15,
61­
69.

Narotsky,
M.
G.,
Hamby,
B.
T.,
Mitchell,
D.
S.,
and
Kavlock,
R.
J.
(
1993).
Evaluation
of
the
critical
period
of
bromodichloromethane­
induced
full­
litter
resorption
in
F­
344
rats.
Teratology
47(
5),
429
(
Abstract
No.
P58).

National
Toxicology
Program's
Report
of
the
Endocrine
Disruptors
Low
Dose
Peer
Review.
2000.
Organized
by
the
National
Institute
of
Environmental
Health
Sciences,
NIH,
National
Toxicology
Program.

O'Connor,
J.
C.,
J.
C.
Cook,
S.
C.
Craven,
C.
S.
Van
Pelt,
and
J.
P.
Obourn
(
1996).
An
in
vivo
battery
for
identifying
endocrine
modulators
that
are
estrogenic
or
dopamine
regulators.
Fundam.
Appl.
Toxicol.
33,
182­
195.

O'Connor,
J.
C.,
Frame,
S.
R.,
Davis,
L.
G.,
and
Cook,
J.
C.
(
1999).
Detection
of
the
environmental
antiandrogen
p,
p'­
DDE
in
CD
and
Long­
Evans
rats
using
a
Tier
1
Screening
Battery
and
a
Hershberger
Assay.
Toxicol.
Sci.
51,
44­
53.
87
O'Connor,
J.
C.,
S.
R.
Frame,
L.
G.
Davis,
and
J.
C.
Cook
(
1999).
Detection
of
thyroid
toxicants
in
a
tier
I
screening
battery
and
alterations
in
thyroid
endpoints
over
28
days
of
exposure.
Toxicol.
Sci.
51(
1),
54­
70.

O'Connor,
J.
C.,
S.
R.
Frame,
and
G.
S.
Ladics
(
2002a).
Evaluation
of
a
15­
day
screening
assay
using
intact
male
rats
for
identifying
steroid
biosynthesis
inhibitors
and
thyroid
modulators.
Toxicol.
Sci.
69,
79­
91.

O'Connor,
J.
C.,
S.
R.
Frame,
and
G.
S.
Ladics
(
2002b).
Evaluation
of
a
15­
day
screening
assay
using
intact
male
rats
for
identifying
antiandrogens.
Toxicol.
Sci.
69,
92­
108.

Odum,
J.,
Pyrah,
T.
G.,
Soames,
A.
R.,
Foster,
J.
R.,
Van
Miller,
J.
P.,
Joiner,
R.
L.,
and
Ashby,
J.
(
1999a).
Effects
of
p­
nonylphenol
(
NP)
and
diethylstilboestrol
(
DES)
on
the
alderley
park
(
Alpk)
rat:
comparison
of
mammary
gland
and
uterus
sensitivity
following
oral
gavage
or
implanted
mini­
pumps.
J.
Appl.
Toxicol.
19,
367­
368.

Odum,
J.,
Pyrah,
T.
G.,
Foster,
J.
R.,
Van
Miller,
J.
P.,
Joiner,
R.
L.,
and
Ashby,
J.
(
1999b).
Comparative
activities
of
p­
nonylphenol
and
diethylstilbestrol
in
Noble
rat
mammary
gland
and
uterotrophic
assays.
Regul.
Toxicol.
Pharmacol.,
29,
184­
195.

Oishi,
S.
(
1993).
Strain
differences
in
susceptibility
to
di­
2­
ethylhexyl
phthalate­
induced
testicular
atrophy
in
mice.
Toxicology
Letters
66,
47­
52.

Okwun,
O.
E.,
Igboeli,
G.,
Ford,
J.
J.,
Lunstra,
D.
D.,
Johnson,
L.
(
1996a).
Number
and
function
of
Sertoli
cells,
number
and
yield
of
spermatogonia,
and
daily
sperm
production
in
three
breeds
of
boar.
J.
Reprod.
Fertil.
107(
1),
137­
149.

Okwun,
O.
E.,
Igboeli,
G.,
et
al.
(
1996b).
Testicular
composition,
number
of
A
spermatogonia,
germ
cell
ratios,
and
number
of
spermatids
in
three
different
breeds
of
boars.
J.
Androl.
17(
3):
301­
9.

Oppenheimer,
J.
H.,
Schwartz,
H.
L.,
and
Strait,
K.
A.
(
1994).
Thyroid
hormone
action
1994:
The
plotthickens.
Eur
J
Endocrinol.
130,
15­
24.

Parmar,
R.,
Spearow,
J.,
et
al.
(
2003).
Genetic
Susceptibility
of
Mice
to
Candida
Albicans
Vaginitis
Correlates
With
Estrogen
Sensitivity.
IV
congress
of
the
International
Society
for
Human
and
Animal
Mycoses,
San
Antonio,
Texas.

Pirchner,
F.
(
1969).
Population
Genetics
in
Animal
Breeding.
San
Francisco,
W.
H.
Freeman
and
Company.

Pomp,
D.,
E.
J.
Eisen,
et
al.
(
1988).
LH
receptor
induction
and
ovulation
rate
in
mice
selected
for
litter
size
and
body
weight.
J.
Reprod.
Fertil.
84(
2):
601­
9.

Perez­
Castillo,
A.,
Bernal,
J.,
Ferreiro,
B.,
and
Pans,
T.
(
1985).
The
early
ontogenesis
of
thyroid
hormone
receptor
in
the
rat
fetus.
Endocinology.
(
117),
2457­
2461.
88
Pettersen,
J.
C.,
Morrissey,
R.
L.,
et
al.
(
1996).
A
2­
year
comparison
study
of
Crl:
CD
BR
and
Hsd:
Sprague­
Dawley
SD
rats.
Fundam.
Appl.
Toxicol.
33(
2):
196­
211.

Pohjanvirta,
R.,
Kulju,
T.,
Morselt,
A.
F.
W.,
Tuominen,
R.,
Juvonen,
R.,
Rozman,
K.,
Mannisto,
Pekka,
Collan,
Y.,
Sainio,
E.­
L.,
and
Tuomisto,
J.
(
1989).
Target
tissue
morphology
and
serum
biochemistry
following
2,3,7,8­
tetrachlorodibenzo­
p­
dioxin
(
TCDD)
exposure
in
a
TCDD­
susceptible
and
a
TCDD­
resistant
rat
strain.
Fundam.
Appl.
Toxicol.
12,
698­
712.

Pohjanvirta,
R.,
Viluksela,
M.,
Tuomisto,
J.
T.,
Unkila,
M.,
Karasinksa,
J.,
Franc,
M.
A.,
Holowenko,
M.,
Giannone,
J.
V.,
Harper,
P.
A.,
Tuomisto,
J.,
Okey,
A.
B.
(
1999).
Physicochemical
differences
in
the
AH
receptors
of
the
most
TCDD­
susceptible
and
the
most
TCDD­
resistant
rat
strains.
Toxicol.
Appl.
Pharmacol.
155(
1),
82­
95.

Pohorecky,
L.
A.,
Skiandos,
A.,
Zhang,
X.,
Rice,
K.
C.,
and
Benjamin,
D.
(
1999).
Effect
of
chronic
social
stress
on
delta­
opioid
receptor
function
in
the
rat.
J.
Pharmacol.
Exp.
Ther.
290(
1),
196­
206.

Popova,
N.
K.,
and
Naumenko,
E.
V.
(
1972).
Dominance
relations
and
the
pituitaryadrenal
system
in
rats.
Anim.
Behav.
20(
1),
108­
111.

Porterfield,
S.
P.
(
1994).
Vulnerability
of
the
developing
brain
to
thyroid
abnormalities:
Environmental
insults
to
the
thyroid
system.
Environmental
Health
Perspectives.
102(
2),
125­
130.

Porterfield,
S.
P.,
and
Stein,
S.
A.
(
1994).
Thyroid
hormones
and
neurological
development:
Update
1994.
Endocrine
Rev.
(
3),
357­
363.

Porterfield,
S.
P.,
and
Hendrich,
C.
E.
(
1993).
The
role
of
thyroid
hormones
in
prenatal
neonatal
neurological
development
 
current
perspectives.
Endocrine
Rev.
(
14),
94­
106.

Porterfield,
S.
P.,
and
Hendrich,
C.
E.
(
1992).
Tissue
iodothyroidine
levels
in
fetuses
of
control
and
hypothyroid
rats
at
13
and
16
days
gestation.
Endocrinol.
(
131),
195­
206.

Putz,
O.,
Schwartz,
C.
B.,
LeBlanc,
G.
A.,
Cooper,
R.
L.
and
Prins,
G.
S.
(
2001).
Neonatal
low­
and
high­
dose
exposure
to
estradiol
benzoate
in
the
male
rat:
II.
Effects
on
male
puberty
and
the
reproductive
tract.
Biology
of
Reproduction
65,
1506­
1517.

Rehm,
S.,
and
Waalkes,
M.
P.
(
1988).
Cadmium­
induced
ovarian
toxicity
in
hamsters,
mice
and
rats.
Fundam.
Appl.
Toxicol.
10,
635­
647.
89
Reinhardt,
W.,
and
Militzer,
K.
(
1979).
"
Schnurrbartbeissen"
und
sozialer
rang
bei
mausen
[
whisker
trimming
and
social
rank
in
mice].
Z.
Versuchstierkd.
21(
2),
83­
91.

Roper,
R.
J.,
Griffith,
J.
S.,
et
al.
(
1999).
Interacting
quantitative
trait
loci
control
phenotypic
variation
in
murine
estradiol­
regulated
responses.
Endocrinology
140(
2),
556­
61.

Sarna,
J.
R.,
Dyck,
R.
H.,
and
Wishaw,
I.
Q.
(
2000).
The
Dalila
effect:
C57BL6
mice
barber
whiskers
by
plucking.
Behav.
Brain
Res.
108(
1),
39­
45.
Schechter,
J.,
Ahmad,
N.,
Elias,
K.,
and
Weiner,
R.
(
1987).
Estrogen­
induced
tumors:
changes
in
the
vasculature
in
two
strains
of
rat.
Am.
J.
Anat.
179(
4),
315­
323.
Schechter,
J.
and
R.
Weiner
(
1991).
Changes
in
basic
fibroblast
growth
factor
coincident
with
estradiol­
induced
hyperplasia
of
the
anterior
pituitaries
of
Fischer
344
and
Sprague­
Dawley
rats.
Endocrinology
129(
5),
2400­
8.

Schardein,
J.
L.
(
1999).
Reproductive
Hazards.
In:
Product
Safety
Evaluation
Handbook.
(
S.
Gad,
ed.)
Marcel
Dekker,
N.
Y.
299­
380.

Sharp,
J.,
Zammit,
T.,
Azar,
T.,
and
Lawson,
D.
(
2003).
Stress­
like
responses
to
common
procedures
in
individually
and
group­
housed
female
rats.
Contemp.
Top.
Lab.
Animal
Sci.
42(
1),
9­
18.

Shellabarger,
C.
J.,
Stone,
J.
P.,
et
al.
(
1978).
Rat
differences
in
mammary
tumor
induction
with
estrogen
and
neutron
radiation.
J.
Natl.
Cancer
Inst.
61(
6):
1505­
8.

Silver,
L.
(
1995).
Mouse
Genetics:
Concepts
and
Applications,
Oxford
University
Press.

Shirai,
T.,
Nakamura,
A.,
Fukushima,
S.,
Yamamoto,
A.,
Tada,
M.,
and
Ito,
N.
(
1990).
Different
carcinogenic
responses
in
a
variety
of
organs,
including
the
prostate,
of
five
different
rat
strains
given
3,2'­
dimethyl­
4­
aminobiphenyl.
Carcinogenesis
11(
5),
793­
797.

Smith,
G.
C.,
Cizza,
G.,
Gomez,
M.,
Greibler,
C.,
Gold,
P.
W.,
and
Sternberg,
E.
M.
(
1994).
The
estrous
cycle
and
pituitary­
ovarian
function
in
Lewis
and
Fischer
rats.
Neuroimmunomodulation
1,
231­
235.

Spearow,
J.
L.
and
Bradford,
G.
E.
(
1983).
Genetic
variation
in
spontaneous
ovulation
rate
and
LH
receptor
induction
in
mice.
J.
Reprod.
Fertil.
69(
2):
529­
537.

Spearow,
J.
L.
(
1985).
The
mechanism
of
action
of
genes
controlling
reproduction.
Genetics
of
Reproduction
in
Sheep.
R.
B.
Land
and
D.
W.
Robinson:
Ch.
22,
pp.
203­
215.

Spearow,
J.
L.
(
1986).
Changes
in
the
kinetics
of
follicular
growth
in
response
to
selection
for
large
litter
size
in
mice.
Biol.
Reprod.
35:
1175­
1186.
90
Spearow,
J.
L.,
Turgai,
J.
T.,
et
al.
(
1987).
Genetic
variation
in
estradiol
negative
feedback
on
testes
and
vesicular
gland
weight
and
LH
in
mice.
J.
Animal
Sci.
65
Suppl.
1:
399A.

Spearow,
J.
L.
(
1988).
Characterization
of
genetic
differences
in
hormone­
induced
ovulation
rate
in
mice.
J.
Reprod.
Fert.
82:
799­
806.

Spearow,
J.
L.
(
1988).
Major
genes
control
hormone­
induced
ovulation
rate
in
mice.
J.
Reprod.
Fertil.
82:
787­
797.

Spearow,
J.
L.,
Erickson,
R.
P.,
et
al.
(
1991).
The
effect
of
H­
2
region
and
genetic
background
on
hormone­
induced
ovulation
rate,
puberty,
and
follicular
number
in
mice.
Genet.
Res.
57(
1):
41­
9.

Spearow,
J.
L.
and
M.
Barkley
(
1999).
Genetic
control
of
hormone­
induced
ovulation
rate
in
mice.
Biol.
Reprod.
61(
4):
851­
856.

Spearow,
J.
L.,
Doemeny,
P.,
Sera,
R.,
Leffler,
R.,
and
Barkley,
M.
(
1999).
Genetic
variation
in
susceptibility
to
endocrine
disruption
by
estrogen
in
mice.
Science
285,
1259­
1261.

Spearow,
J.
L.
and
Barkley,
M.
(
2001).
Reassessment
of
models
used
to
test
xenobiotics
for
oestrogenic
potency
is
overdue.
Hum.
Reprod.
16(
5):
1027­
9.

Spearow,
J.
L.,
O'Henley,
P.,
Doemeny,
P.,
Sera,
R.,
Leffler,
R.,
Sofos,
T.,
Barkley,
M.
(
2001).
Genetic
variation
in
physiological
sensitivity
to
estrogen
in
mice.
APMIS
109(
5),
356­
364.

Stefanski,
V.
(
1998).
Social
stress
in
loser
rats:
opposite
immunological
effects
in
submissive
and
subdominant
males.
Physiol.
Behav.
63(
4),
605­
613.

Stefanski,
V.
(
2000).
Social
stress
in
laboratory
rats:
hormonal
responses
and
immune
cell
distribution.
Psychoneuroendocrinology
25(
4),
389­
406.

Steinmetz,
R.,
Brown,
N.
G.,
Allen,
D.
L.,
Bigsby,
R.
M.,
and
Ben­
Jonathan,
N.
(
1997).
The
environmental
estrogen
bisphenol
A
stimulates
prolactin
release
in
vitro
and
in
vivo.
Endocrinology
138(
5),
1780­
1786.

Steinmetz,
R.,
Mitchner,
N.
A.,
et
al.
(
1998).
The
xenoestrogen
bisphenol
A
induces
growth,
differentiation,
and
c­
fos
gene
expression
in
the
female
reproductive
tract.
Endocrinology
139(
6):
2741­
7.

Stevens,
J.
T.,
C.
B.
Breckenridge,
et
al.
(
1999).
A
risk
characterization
for
atrazine:
oncogenicity
profile.
J.
Toxicol.
Environ.
Health
A
56(
2):
69­
109.
91
Stoker,
T.
E.,
Laws,
S.
C.,
Guidici,
D.
L.,
and
Cooper,
R.
L.
2000.
The
effect
of
atrazine
on
puberty
in
male
Wistar
rats:
An
evaluation
in
the
protocol
for
the
assessment
of
pubertal
development
and
thyroid
function.
Toxicological
Sciences
58(
N1),
50­
59.
(
Reprint).

Strait,
K.
A.,
Schwartz,
H.
L.,
Perez­
Castillo,
A.,
and
Oppenheimer,
J.
H.
(
1990).
Relationship
of
c­
erbA
content
to
tissue
triiodothyronine
nuclear
binding
capacity
and
function
in
developing
and
adult
rats.
J
Boil
Chem.
(
265),
10514­
10521.

Strozik,
K.
E.,
and
Festing,
M.
F.
(
1981).
Whisker
trimming
in
mice.
Lab.
Anim.
15(
4),
309­
312.

Takizawa,
K.,
Yagi,
H.,
Jerina,
D.
M.,
and
Mattison,
D.
R.
(
1985).
Experimental
ovarian
toxicity
following
intraovarian
injection
of
benzo(
a)
pyrene
or
its
metabolites
in
mice
and
rats.
Reproductive
Toxicology
ed.
R.
L
Dixon,
Raven
Press,
New
York.
69­
94.

Taylor,
L.
R.,
and
Costanzo,
D.
J.
(
1975).
Social
dominance,
adrenal
weight,
and
the
reticuloendothelial
system
in
rats.
Behav.
Biol.
13(
2),
167­
174.

Taylor,
G.
T.,
and
Moore,
S.
(
1975).
Social
position
and
competition
in
laboratory
rats.
J.
Comp.
Physiol.
Psychol.
88(
1),
424­
430.

Thigpen,
J.
E.,
Li,
L.­
A.,
Richter,
C.
B.,
Cebetkin,
E.
H.,
and
Jameson,
C.
W.
(
1989).
The
mouse
bioassay
for
the
detection
of
estrogenic
activity
in
rodent
diets.
II.
Comparative
estrogenic
activity
of
purified,
certified
and
standard
open
and
closed
formula
rodent
diets.
Lab.
Anim.
Sci.
37,
602­
605.

Thigpen,
J.
E.,
Setchell,
K.
D.
R.,
Ahlmark,
K.
B.,
Locklear,
J.,
Spahr,
T.,
Caviness,
G.
F.,
Goelz,
M.
F.,
Haseman,
J.
K.,
Newbold,
R.
R.,
and
Forsythe,
D.
B.
(
1999).
Phytoestrogen
content
of
Purina
open­
and
closed­
formula
laboratory
animal
diets.
Lab.
Anim.
Sci.
49(
5),
530­
534.

Timiras,
P.
S.,
and
Nzekwe,
E.
U.
(
1989).
Thyroid
hormones
and
nervous
system
development.
Biol
Neonate.
(
55),
376­
385.

Tinwell,
H.,
Haseman,
J.,
Lefevre,
P.
A.,
Wallis,
N.,
and
Ashby,
J.
(
2002).
Normal
sexual
development
of
two
strains
of
rat
exposed
in
utero
to
low
doses
of
Bisphenol
A.
Toxicol.
Sci.
68,
339­
348.

Tropp
J.,
and
Markus,
E.
J.
(
2001).
Effects
of
mild
food
deprivation
on
the
estrous
cycle
of
rats.
Physiol.
Behav.
73,
553­
559.
92
Tyl,
R.
W.,
C.
B.
Myers,
M.
C.
Marr,
D.
R.
Brine,
P.
A.
Fail,
J.
C.
Seely,
and
J.
P.
Van
Miller
(
1999).
Two­
generation
study
with
para­
tert­
octylphenol
in
rats.
Regulatory
Toxicol
Pharmacol.
30,
81­
95.
Tyl,
R.
W.,
C.
B.
Myers,
M.
C.
Marr,
T.­
Y.
Chang,
J.
C.
Seely,
D.
R.
Brine,
M.
M.
Veselica,
P.
A.
Fail,
R.
L.
Joiner,
J.
H.
Butala,
S.
S.
Dimond,
and
L.
R.
Harris
(
2000).
Threegeneration
reproductive
toxicity
evaluation
of
bisphenol
A
administered
in
the
feed
to
CD
®
(
Sprague­
Dawley)
rats.
Draft
final
report,
submitted
July
15,
2000
to
the
Society
of
the
Plastics
Industries,
Bisphenol
A
Task
Group,
Washington,
DC.

U.
S.
Environmental
Protection
Agency
(
EPA)
(
1985).
Toxic
Substances
Control
Act,
EPA
(
TSCA)
Section
798.4700,
Reproduction
and
Fertility
Effects.
Federal
Register
50
(
188),
39432­
39433.

U.
S.
EPA
(
1996).
Office
of
Prevention,
Pesticides
and
Toxic
Substances
(
OPPTS),
Health
Effects
Test
Guidelines,
OPPTS
870.3800,
Reproduction
and
Fertility
Effects
(
Public
Draft,
February
1996).

U.
S.
EPA
(
1998).
Office
of
Prevention,
Pesticides
and
Toxic
Substances
(
OPPTS),
Health
Effects
Test
Guidelines,
OPPTS
870.3800,
Reproduction
and
Fertility
Effects
(
Final
Guideline,
August
1998).

U.
S.
Environmental
Protection
Agency
(
EPA)
(
1996).
Part
II,
Environmental
Protection
Agency:
Reproductive
Toxicity
Risk
Assessment
Guidelines;
Notice.
Federal
Register
61
(
212),
56274­
56322
(
October
31,
1996).

Van
Loo,
P.
L.,
Mol,
J.
A.,
Koolhaas,
J.
M.,
Van
Zutphen,
B.
F.,
and
Baumans,
V.
(
2001).
Modulation
of
agression
in
male
mice:
influence
of
group
size
and
cage
size.
Physiol.
Behav.
72(
5),
675­
683.

Waller,
K.,
Swan,
S.
H.,
et
al.
(
1998).
Trihalomethanes
in
drinking
water
and
spontaneous
abortion.
Epidemiology
9(
2):
134­
40.

Wendell,
D.
L.,
Daun,
S.
B.,
et
al.
(
2000).
Different
functions
of
QTL
for
estrogen­
dependent
tumor
growth
of
the
rat
pituitary.
Mamm.
Genome
11(
10):
855­
61.

Wendell,
D.
L.
and
Gorski,
J.
(
1997).
Quantitative
trait
loci
for
estrogen­
dependent
pituitary
tumor
growth
in
the
rat.
Mamm.
Genome
8(
11),
823­
9.

Wendell,
D.
L.,
A.
Herman,
et
al.
(
1996).
Genetic
separation
of
tumor
growth
and
hemorrhagic
phenotypes
in
an
estrogen­
induced
tumor.
Proceedings
of
the
National
Academy
of
Sciences
93(
15),
8112­
6.

Westenbroek,
C.,
Ter
Horst,
G.
J.,
Roos,
M.
H.,
Kuipers,
S.
D.,
Trentani,
A.,
and
den
Boer,
J.
A.
(
2003).
Gender­
specific
effects
of
social
housing
in
rats
after
chronic
mild
stress
exposure.
Prog.
Neuropsychopharmacol.
Biol.
Psychiatry
27(
2),
21­
30.
93
Wetzel,
L.
T.,
Luempert,
G.
R.,
et
al.
(
1994).
Chronic
effects
of
atrazine
on
estrus
and
mammary
tumor
formation
in
female
Sprague­
Dawley
and
Fischer
344
rats.
J.
Toxicol.
Environ.
Health
43(
2):
169­
82.

Whalen,
R.
E.,
Gladue,
B.
A.,
and
Olsen,
K.
L.
(
1986).
Lordotic
behavior
in
male
rats:
genetic
and
hormonal
regulation
of
sexual
differentiation.
Horm.
Behav.,
20(
1),
73­
82.

Whitten,
P.
L.,
Lewis,
C.,
Russell,
E.,
and
Naftolin,
F.
(
1995).
Phytoestrogen
influences
on
the
development
of
behavior
and
gonadotropin
function.
Proc.
Soc.
Exp.
Biol.
Med.
208,
82­
86.

Wilkinson,
J.
M.,
Halley,
S.,
and
Towers,
P.
A.
(
2000).
Comparison
of
male
reproductive
parameters
in
three
rat
strains:
Dark
Agouti,
Sprague­
Dawley
and
Wistar.
Laboratory
Animals
34,
70­
75.

Williams,
R.
W.,
Gu,
J.,
et
al.
(
2001).
The
genetic
structure
of
recombinant
inbred
mice:
high­
resolution
consensus
maps
for
complex
trait
analysis.
Genome
Biol.
2(
11):
RESEARCH0046.

Wilson,
T.,
Wu,
X.
Y.,
et
al.
(
2001).
Highly
Prolific
Booroola
Sheep
have
a
Mutation
in
the
Intracellular
Kinase
Domain
of
Bone
Morphogenetic
Protein
1B
Receptor
(
ALK­
6)
That
is
Expressed
in
Both
Oocytes
and
Granulosa
Cells.
Biol.
Reprod.
64(
4):
1225­
1235.

Wolf,
C.
J.,
Ostby,
J.
S.,
Gray,
L.
E.
Jr.
1999.
Gestational
exposure
to
2,3,7,8­
tetrachlorodibenzo­
p­
dioxin
(
TCDD)
severely
alters
reproductive
function
of
female
hamster
offspring.
Toxicol.
Sci.
51(
2),
259­
264.

Yin,
H.,
Fujimoto,
N.,
Maruyama,
S.,
and
Asano,
K.
(
2001).
Strain
difference
in
regulation
of
pituitary
tumor
transforming
gene
(
PTTG)
in
estrogen­
induced
pituitary
tumorigenesis
in
rats.
Jpn.
J.
Cancer
Res.
92,
1034­
1040.

You,
L.,
Casanova,
M.,
Archibeque­
Engle,
S.,
Madhabananda,
S.,
Li­
Qun,
F.,
and
Heck,
H.
d'
A.
(
1998).
Impaired
male
sexual
development
in
perinatal
Sprague­
Dawley
and
Long­
Evans
hooded
rats
exposed
in
utero
and
lactationally
to
p,
p'­
DDE.
Toxicol.
Sci.
45,
162­
173.

Zidek,
V.,
Musilova,
A.,
et
al.
(
1998).
Genetic
dissection
of
testicular
weight
in
the
mouse
with
the
BXD
recombinant
inbred
strains.
Mamm.
Genome
9(
7):
503­
5.

Zidek,
V.,
Pintir,
J.,
et
al.
(
1999).
Mapping
of
quantitative
trait
loci
for
seminal
vesicle
mass
and
litter
size
to
rat
chromosome
8.
J.
Reprod.
Fertil.
116(
2):
329­
33.
