July
25,
2002
Gary
Timm
U.
S.
Environmental
Protection
Agency
EPA
East
Building
1201
Constitution
Avenue,
N.
W.
Room
4106
L,
Mail
Code
7201M
Washington,
DC
20004
Dear
Gary:

Work
Assignment
2­
5,
Task
15
Study
Plan
to
Optimize
the
Sliced
Testis
Steroidogenesis
Assay
Attached
is
the
draft
version
of
the
Study
Plan
to
Optimize
the
Sliced
Testis
Steroidogenesis
Assay.
If
you
have
questions,
please
contact
me
at
(
614)
424­
3564.

Sincerely,

David
P.
Houchens,
Ph.
D.
Program
Manager
DPH:
lnl
Attachment
cc:
Robert
G.
Krumhansl,
EPA
CO
(
letter
only)
L.
Greg
Schweer,
EPA
Project
Officer
DRAFT
LETTER
REPORT
ON
PHASE
I
OPTIMIZATION
OF
THE
SLICED
TESTIS
STEROIDOGENESIS
ASSAY
EPA
Contract
Number
68­
W­
01­
023
WA
2­
27
May
19,
2003
PREPARED
FOR
GARY
E.
TIMM
WORK
ASSIGNMENT
MANAGER
U.
S.
ENVIRONMENTAL
PROTECTION
AGENCY
ENDOCRINE
DISRUPTOR
SCREENING
PROGRAM
WASHINGTON,
D.
C.

BATTELLE
505
KING
AVENUE
COLUMBUS,
OH
43201
Battelle
Report
i
July,
2002
TABLE
OF
CONTENTS
Page
1.0
ENDPOINT
MEASUREMENTS/
BACKGROUND
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
2
1.1
STEROIDOGENSIS
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
2
1.1.1
Signal
Transduction
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
2
1.1.2
Cholesterol
Synthesis
and
Transport
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
3
1.1.3
Enzymatic
Conversions
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
3
1.2
SLICED
TESTIS
STEROIDOGENESIS
ASSAY
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
5
1.2.1
Basis
for
Selection
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
5
1.2.2
Basis
for
Optimization
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
7
1.2.3
Prototype
of
the
Sliced
Testis
Steroidogenesis
Assay
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
8
2.0
PROTOCOL
RESOLUTION
ISSUES
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
10
2.1
INTRODUCTION
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
10
2.2
ANALYTICAL
ASSAY
VERIFICATION
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
10
2.2.1
Analytical
Assay
Validation
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
10
2.2.2
Analytical
Assay
Validation
Parameters
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
11
2.3
TESTOSTERONE
RIA
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
12
2.4
LACTATE
DEHYDROGENASE
(
LDH)
ASSAY
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
14
3.0
STUDY
DESIGN
(
OPTIMIZATION
OF
EXPERIMENTAL
FACTORS)
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
15
3.1
INCUBATION
FACTORS
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
15
3.1.1
Incubation
Media
Type
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
16
3.1.2
Incubation
Gaseous
Atmosphere
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
16
3.1.3
Incubation
Temperature
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
17
3.1.4
Incubation
Vessel
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
17
3.1.5
Incubation
Shaker
Speed
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
17
3.1.6
Incubation
Media
Volume
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
18
3.1.7
hCG
Concentrations
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
18
3.2
TESTICULAR
PARENCHYMA
FACTORS
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
18
3.2.1
Parenchymal
Fragment
Size
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
18
3.2.2
Preparation
Time
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
19
3.2.3
Preparation
Techniques
of
Whole
Testis
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
19
3.2.4
Aliquot
Volume
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
20
3.3
SAMPLE
STABILITY
FACTORS
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
20
3.3.1
Sample
Storage
Container
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
20
3.3.2
Sample
Storage
Time
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
21
3.3.3
Sample
Storage
Temperature
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
21
3.4
SAMPLE
COLLECTION
INTERVALS
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
21
3.5
CHARACTERIZATION
OF
VIABILITY
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
22
3.5.1
LDH
­
Evaluation
as
a
Marker
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
22
3.5.2
Equilibration
Period
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
22
3.5.3
Vehicle
Type
and
Concentration
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
23
4.0
RECOMMENDED
TEST
SUBSTANCES
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
23
5.0
STUDY
PROTOCOL
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
23
6.0
STATISTICAL
METHODS
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
24
6.1
INTRODUCTION
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
24
6.2
ASSUMPTIONS
AND
RESPONSE
ASSESSMENT
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
26
6.3
PRELIMINARY
EXPERIMENTAL
PHASE
(
PHASE
I)
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
26
Battelle
Report
ii
July,
2002
6.4
PRIMARY
EXPERIMENTAL
PHASE
(
PHASE
II)
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
27
6.4.1
Optimization
of
Incubation
Conditions
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
27
6.4.2
Optimization
of
Testis
Preparation
Factors
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
28
6.4.3
Optimization
of
Sample
Factors
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
29
6.4.4
Optimization
of
Sampling
Time
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
31
6.4.5
Sensitivity
Analysis
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
31
6.4.6
Characterization
of
Viability
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
32
6.5
STATISTICAL
ANALYSIS
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
32
7.0
REFERENCES
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
33
LIST
OF
TABLES
Table
1.
Representative
Studies
Using
the
Sliced
Testis
Assay
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
6
Table
2.
Cumulative
Mean
Concentration
and
Standard
Error
of
the
Mean
by
Hour
and
Study
.
.
7
Table
3.
Standard
Deviations
of
Between­
Study
and
Within­
Study
Components
of
Variance
by
Hour
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
8
Table
4.
Characteristics
of
a
Radioimmunoassay
Validated
for
Determination
of
Testosterone
in
Adult
Male
Sprague
Dawley
Rat
Plasma
and
Testicular
Fluid
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
13
Table
5.
Summary
of
Experimental
Incubation
Factors
for
Optimization
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
27
Table
6.
Summary
of
Experimental
Testis
Preparation
Factors
for
Optimization
.
.
.
.
.
.
.
.
.
.
.
29
Table
7.
Summary
of
the
Experimental
Sample
Stability
Factors
for
Optimization
.
.
.
.
.
.
.
.
.
.
30
LIST
OF
FIGURES
Figure
1.
Intracellular
Biochemical
Steroidogenic
Pathway
Following
Trophic
Hormone
Stimulation
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
2
Figure
2.
Enzymatic
Conversions
of
Cholesterol
and
Intermediate/
End­
Product
Hormones
.
.
.
.
.
4
Figure
3.
Technical
Flow
Illustration
of
the
Sliced
Testis
Steroidogenesis
Assay
.
.
.
.
.
.
.
.
.
.
.
.
.
9
Figure
4.
Sliced
Testis
Steroidogenesis
Assay
Experimental
Design
Organizational
Diagram
.
.
24
Battelle
Report
1
July,
2002
Endocrine
Disruptor
Screening
Program
Contract
No.
68­
W­
01­
023
Work
Assignment
2­
5,
Task
15
Study
Plan
to
Optimize
the
Sliced
Testis
Steroidogenesis
Assay
INTRODUCTION
In
1996,
the
Food
Quality
Protection
Act
(
FQPA)
amendments
were
enacted
by
Congress
to
authorize
the
EPA
to
implement
an
Endocrine
Disruptor
Screening
Program
(
EDSP)
on
pesticides
and
other
substances
found
in
food
or
water
sources
for
endocrine
effects
in
humans
(
FQPA,
1996).
In
this
program,
comprehensive
toxicological
and
ecotoxicological
screens
and
tests
are
being
developed
for
identifying
and
characterizing
the
endocrine
effects
of
various
environmental
contaminants,
industrial
substances,
and
pesticides.
A
two­
tiered
approach
will
be
utilized.
Tier
1
employs
a
combination
of
in
vivo
and
in
vitro
screens,
and
Tier
2
involves
in
vivo
testing
methods
using
two­
generation
reproductive
studies.
A
steroidogenesis
assay
is
proposed
as
one
of
the
Tier
1
screening
battery
assays.

A
detailed
review
paper
(
DRP)
about
steroidogenesis
was
prepared
(
EPA
2002).
The
DRP
(
1)
summarized
the
state
of
the
science
of
the
in
vivo,
ex
vivo,
and
in
vitro
methodologies
available
for
measuring
gonadal
steroidogenesis;
(
2)
for
each
methodology,
presented
a
review
of
the
individual
assays
and
representative
data
generated
by
investigators
who
used
the
assay
to
evaluate
a
substance
for
steroidogenic­
altering
activity;
(
3)
provided
an
evaluation
of
the
various
methodologies
and
the
assays
as
tools
for
screening
substances
with
suspected
steroidogenic
activity;
(
4)
recommended
a
particular
screening
method
and
assay
as
a
screening
tool;
and
(
5)
described
the
strengths,
weaknesses,
and
implications
for
further
research
associated
with
the
recommended
screening
assay.
This
information
is
summarized
and
presented
in
the
following
sections
of
this
study
plan
narrative.

Although
a
promising
tool,
the
sliced
testis
assay
remains
to
be
fully
tested
as
an
assay
that
can
meet
all
the
demands
of
an
endocrine
disruptor
screening
tool.
Concerns
raised
by
the
EPA
and
EDMVS
during
discussions
on
June
11,
2002,
and
thereafter
suggested
that
experiments
be
conducted
to
ensure
the
optimization
of
the
assay
prior
to
more
rigorous
pre­
validation
and
validation
testing.
The
most
notable
concerns
were
associated
with
1)
various
incubation
variables,
2)
variables
that
affect
optimal
hCG
stimulation,
3)
characterization
of
the
parenchymal
post­
slicing
equilibration
time,
and
4)
parenchymal
viability.
In
addition
to
these
most
notable
concerns,
other
factors
that
could
potentially
affect
the
optimal
performance
of
this
assay
were
identified.
Thus,
the
present
study
plan
and
protocol
describe
in
detail
the
experiments
intended
to
provide
data
for
setting
in
place
the
procedures
and
parameters
that
will
optimize
the
performance
of
this
assay.

This
study
plan
covers
the
following
issues:

°
Endpoint
measurements/
background
information
Battelle
Report
2
July,
2002
°
Protocol
issues
needing
resolution
°
Study
design
to
address
the
protocol
issues
°
Recommended
test
substances
°
The
detailed
study
protocol
°
Statistical
methods
for
comparing
the
performance
of
the
assay.

1.0
ENDPOINT
MEASUREMENTS/
BACKGROUND
Steroidogenesis
is
a
specific
pathway
of
chemical
reactions
that
result
in
the
production
of
gonadal
intermediary
and
end­
product
hormones.
The
pathway
(
1)
begins
with
intracellular
signal
transduction,
(
2)
continues
with
cholesterol
production
in
the
cytoplasm
and
transport
to
the
mitochodrial
inner
membrane,
and
(
3)
ends
with
a
set
of
multi­
step
enzymatic
conversions
from
cholesterol
to
the
end­
product
hormones.

1.1
STEROIDOGENESIS
1.1.1
Signal
Transduction
Signal
transduction
describes
the
intracellular
biochemical
reactions
that
occur
after
stimulation
of
the
luteinizing
hormone
(
LH)
membrane
bound
receptor
and
up
to
initiation
of
cholesterol
transport
to
the
mitochondria.
The
LH
receptor
is
coupled
with
a
G­
protein
and,
when
stimulated
by
LH
or
human
chorionic
gonadotropin
(
hCG),
interacts
with
adenylate
cyclase
to
form
cyclic
adenosine
3',
5'­
cyclic
monophosphate
(
cAMP).
Increased
cAMP,
the
second
messenger,
stimulates
protein
kinase
A,
which
initiates
cholesterol
biosynthesis
and
cholesterol
transport
protein
synthesis
(
Cooke,
1996;
Stocco,
1999).
The
receptor­
mediated
biochemical
reactions
are
illustrated
in
Figure
1
(
Stocco,
1999).

Figure
1.
Intracellular
Biochemical
Steroidogenic
Pathway
Following
Trophic
Hormone
Stimulation
Battelle
Report
3
July,
2002
Calcium
(
Ca2+)
is
involved
in
the
signal
transduction
of
the
steroidogenic
pathway
(
Janszen
et
al.,
1976).
In
order
for
the
maximal
stimulation
of
steroidogenesis
to
occur,
intracellular
calcium
levels
must
increase
following
LH
binding.
The
calcium­
mediated
reactions
also
involve
calmodulin,
a
calcium
binding
protein
(
Hall
et
al.,
1981).
Chloride
(
Cl­)
and
arachidonic
acid
have
also
been
implicated
in
steroidogenic
signal
transduction
(
Choi
and
Cooke,
1990;
Naor,
1991;
Cooke,
1996).
Arachidonic
acid
appears
to
produce
a
direct
inhibitory
effect
and
an
indirect
stimulatory
effect
on
steroidogenesis.
Steroid
hormone
production
is
inhibited
when
arachidonic
acid
activates
protein
kinase
C.
However,
metabolism
of
arachidonic
acid
to
its
metabolites,
e.
g.,
leukotrienes,
stimulates
cholesterol
transport
into
the
mitochondria,
thereby
enhancing
steroid
hormone
production.
Other
intracellular
substances
shown
to
affect
steroidogenesis
include
free
radicals,
i.
e.,
superoxide
anion
and
hydroxyl
free
radical,
as
well
as
hydrogen
peroxide
and
nitric
oxide
(
Clark
et
al.,
1994;
Davidoff
et
al.,
1995).

1.1.2
Cholesterol
Synthesis
and
Transport
Cholesterol
is
the
common
precursor
to
the
formation
of
all
gonadal
steroid
hormones.
The
primary
source
of
cellular
cholesterol
is
the
serum.
Cholesterol
is
transported
to
the
cell
via
serum
protein
carriers,
e.
g.,
high­
or
low­
density
lipoprotein
(
HDL
or
LDL).
Once
inside
the
cell,
cholesterol
is
immediately
utilized,
or
it
can
be
stored,
e.
g.,
in
lipid
droplets.
A
second,
minor
source
of
cholesterol
is
de
novo
synthesis,
which
increases
following
hormone
stimulation.
Upon
LH­
induced
stimulation,
mobilization
of
newly
synthesized
and
stored
cholesterol
(
enzymatic
hydrolysis
of
cholesterol
esters)
in
lipid
droplets
occurs.
Cholesterol
is
transported
out
of
the
cytoplasm
and
into
the
mitochondria.
In
the
mitochondria,
cholesterol
is
transported
from
the
outer
to
the
inner
membrane,
which
is
the
rate­
limiting
step
in
steroidogenesis.
The
transport
of
cholesterol
from
the
outer
to
the
inner
mitochondrial
membrane
requires
a
transport
protein.
LH
stimulation
of
steroidogenic
cells
activates
de
novo
production
of
the
cholesterol
transport
protein.
This
protein
is
essential
for
steroidogenesis
and,
since
it
mediates
the
rate­
limiting
step
of
steroid
hormone
production,
it
is
referred
to
as
the
steroid
acute
regulatory
(
StAR)
protein.
In
the
mitochondria,
StAR
protein
transports
cholesterol
to
the
inner
mitochondrial
membrane,
where
the
side­
chain
cleavage
enzyme
(
P450
SCC)
catalyzes
cholesterol
into
pregnenolone.

1.1.3
Enzymatic
Conversions
Enzymatic
conversion
of
cholesterol
to
pregnenolone
constitutes
the
initial
step
in
a
series
of
biochemical
reactions
that
culminate
in
end­
product
hormone
production.
Figure
2
illustrates
the
final
stage
of
the
steroidogenic
biosynthetic
pathway,
as
well
as
the
cell
types
for
males
and
females
and
the
intracellular
location
of
various
enzymatic
steps
of
the
steroidogenic
pathway.

The
first
enzyme
reaction
is
the
conversion
of
cholesterol
to
pregnenolone
by
the
cytochrome
P450
cholesterol
side­
chain
cleavage
enzyme
(
P450
SCC).
P450
SCC
activity
is
also
considered
to
be
a
rate­
limiting
step
in
the
production
of
gonadal
steroid
hormones
(
Kagawa
&
Waterman,
1995).
Battelle
Report
4
July,
2002


Corpus
Luteum
Figure
2.
Enzymatic
Conversions
of
Cholesterol
and
Int
ermediate/
End­
Product
Hormones
The
second
enzymatic
reaction
results
in
the
conversion
of
pregnenolone
to
progesterone
by
the
enzyme
3 ­
hydroxysteroid
dehydrogenase/

5
­

4
isomerase
(
3 ­
HSD).
At
this
point,
the
steroidogenic
pathway
bifurcates
into
a

5­
hydroxysteroid
pathway
(
starting
with
pregnenolone)
and
a

4­
ketosteroid
pathway
(
starting
with
progesterone)
and,
even
though
the
same
enzymes
use
different
substrates
along
the
parallel
pathways,
both
pathways
eventually
converge,
culminating
in
the
production
of
androstenedione.

The
third
enzymatic
reaction
involves
cytochrome
P450
17 ­
hydroxylase/
C
17
­
20
lyase
(
P450c17).
For
the

5­
hydroxysteroid
pathway,
P450c17
initially
catalyzes
the
conversion
of
pregnenolone
to
17 ­
hydroxypregnenolone,
which
is
then
converted
to
DHEA.
As
mentioned
above,
DHEA
is
converted
to
androstenedione
by
3 ­
HSD.
Likewise
for
the

4­
ketosteroids,
P450c17
converts
progesterone
to
17 ­
hydroxyprogesterone,
which
is
then
converted
to
androstenedione.

Theca
Cells
Leydig
Cells
Granulosa
Cells
Testis
and
Peripheral
Tissues
Intracellular
Compartment
(
Females
Only)

(
Males
Only)
Mitochondria
Cytoplasm
Testis
Ovary
Battelle
Report
5
July,
2002
The
next
enzymatic
reaction
involves
the
conversion
of
androstenedione
to
testosterone
by
17­
ketosteroid
reductase
(
17KSR),
which
is
also
referred
to
as17 ­
hydroxysteroid
dehydrogenase
(
17 ­
HSD).
The
production
of
testosterone
is
considered
an
end­
hormone
product.
A
second
possible
reaction
involving
androstenedione
occurs
in
the
female,
whereby
androstenedione
is
converted
to
estrone
by
aromatase.

In
the
male,
testosterone
is
converted
to
dihydrotestosterone
(
DHT)
by
5 ­
reductase.
DHT
is
significantly
more
potent
as
an
androgen
than
testosterone
and
is
also
considered
an
endproduct
hormone.
DHT
is
produced
primarily
in
peripheral
tissues,
although
it
is
also
found
in
the
testis.

The
last
enzyme
in
the
steroidogenic
pathway
is
aromatase.
Aromatase
converts
testosterone
into
estradiol
and,
in
the
female,
androstenedione
into
estrone.
In
short,
aromatase
converts
androgenic
substances
into
estrogenic
substances.
As
mentioned
above
for
testosterone
and
DHT,
estradiol
and
estrone
are
considered
end­
product
hormones
of
the
steroidogenic
pathway.
Aromatase
is
found
in
many
different
peripheral
tissues,
as
well
as
male
and
female
gonadal
tissue.

In
summary,
cholesterol
is
the
common
precursor
for
production
of
steroid
hormones.
A
series
of
biochemical
reactions
involving
different
enzymes
results
in
conversion
of
cholesterol
to
end­
hormone
products:
testosterone,
DHT,
estradiol,
and
estrone.
The
steroidogenic
pathway
is
regulated
by
gonadotropins
and
end­
product
hormones.
An
alteration
of
the
regulatory
mechanisms,
as
well
as
direct
effects
on
the
substrates
and
enzymes
of
the
steroidogenic
pathway,
can
affect
end­
hormone
product
formation,
thereby
possibly
resulting
in
reproductive
system
toxicity.

1.2
SLICED
TESTIS
STEROIDOGENESIS
ASSAY
1.2.1
Basis
for
Selection
Steroid
hormones
produced
by
the
gonads
affect
most
of
the
organs
in
the
body
including
bone,
muscle,
brain,
and
reproductive
organs.
It
is
for
this
reason
that
the
EDSTAC
recommended
that
an
assay
that
measures
steroidogenic
function
be
considered
as
a
component
of
the
Tier
1
Screening
(
T1S)
battery.
An
evaluation
of
steroidogenic
assays
and
the
criteria
for
a
screen
as
presented
in
the
DRP
on
steroidogenesis
resulted
in
selection
of
the
sliced
testis
assay.
The
objective
of
this
assay
is
to
detect
disruption
of
the
steroidogenic
pathway.
It
may:
(
1)
be
used
as
one
of
the
protocols
recommended
by
EDSTAC
for
the
Tier
1
screening
battery,
(
2)
serve
as
a
follow­
up
test
for
certain
substances
for
which
additional
data
are
required
or
desired,
and/
or
(
3)
predict
the
likelihood
that
steroidogenesis
and
downstream
biologically
dependent
processes
would
be
affected
by
the
same
or
similar
substances
in
vivo.
The
endpoint,
testosterone,
was
selected
for
its
potential
to
detect
toxicant­
induced
alterations
of
steroidogenesis
in
testicular
tissue.

This
in
vitro
assay
has
been
used
with
fetal,
neonatal,
and
adult
testes,
and
is
not
limited
to
mammalian
species,
having
been
used
to
assess
steroidogenesis
in
fish,
reptile,
avian,
and
Battelle
Report
6
July,
2002
amphibian
systems
as
well.
Thus,
the
steroidogenesis
bioassay
as
a
component
in
the
T1S
phase
should
be
broadly
understood
to
screen
for
any
disruption
of
the
overall
steroid
biosynthetic
pathway.
The
sliced
testis
steroidogenesis
assay
has
the
capacity
to
evaluate
simultaneously
all
of
the
processes
involved
with
gonadal
synthesis
of
steroid
hormones
(
signal
transduction,
transcription,
translation,
synthesis,
and
cellular
secretion
of
the
steroids).

The
sliced
testis
assay
has
been
used
to
identify
substances
that
alter
steroidogenesis.
Examples
of
experimental
studies
from
the
literature
that
used
the
sliced
testis
assay
for
measuring
steroidogenesis
are
summarized
in
Table
1.

Table
1.
Representative
Studies
Using
the
Sliced
Testis
Assay
Animal/
Type
of
Preparation
Treatment
&
Stimulant
Measured
Response
Reference
Adult
male
Long­
Evans
rats/
Testes
Slices
(
1/
4)
Ethane
dimethanesulfonate
@
0,
3,
10,
32,
100,
320,
1000,
or
3200

g/
mL
media/
ovine
LH
(
100
ng/
mL)

testosterone
production
Gray
et
al.,
1995
Male
Long­
Evans
Hooded
rats
(
3­
25
weeks
of
age)/
Testes
Slices
(
1/
4)
Vinclozolin
@
5
to
100
mg/
kg/
day,
gavage,
for
22
weeks/
hCG,
50
IU

basal
and
hCG
stimulated
testosterone
@
15
and
100
mg/
kg/
day
Fail
et
al.,
1995
Male
Long­
Evans
Hooded
rats
(
3­
14
weeks
of
age)/
Testes
Slices
(
1/
4)
Methoxychlor
@
50
or
200
mg/
kg/
day,
gavage,
for
11
weeks/
hCG,
50
IU

basal
testosterone
production
no
effect
on
HCG
stimulated
testosterone
production
Fail
et
al.,
1994
Adult
male
SD
rat/
Testes
Slices
(
1/
4)
Ethane
dimethanesulfonate
@
0,
500,
or
3000

M/
100
mIU/
mL
hCG

testosterone
production
Laskey
et
al.,
1994
Adult
male
OFA
rat/
Testis
Slices
(~
1/
4)
14C­
pregnenolone
(
50
mCi/
mmol;
200
nCi
­
a
tracer
amount)
~
70
and
15
percent
of
the
14Cradioactivity
was
testosterone
and
androstenedione
Gurtler
and
Donatsch,
1979
In
summary,
the
most
salient
features
of
this
assay
and
the
basis
for
its
consideration
are
that
it
identifies
substances
that
alter
steroid
hormone
production
and
can
be
conducted
at
a
minimal
cost,
quickly,
and
simply
with
standard
laboratory
equipment
and
basic
laboratory
training,
which
are
all
important
characteristics
of
an
assay
to
be
used
as
a
screen
for
identifying
substances
with
steroidogenic­
altering
activity.
The
sliced
testis
assay
is
stable
and
the
parenchyma
remains
viable
over
a
sufficient
time
period
to
measure
changes
in
end­
product
hormone
production.
In
addition,
the
assay
is
relatively
sensitive
and
specific;
uses
parenchyma
that
maintains
the
cytoarchitecture
of
the
organ;
uses
a
reduced
number
of
animals;
will
be
relatively
easy
to
standardize
(
by
optimization);
and
has
a
well­
defined
endpoint
in
testosterone,
which
can
be
modified
to
include
additional
intermediate
hormonal
endpoints
if
so
desired.
1.2.2
Basis
for
Optimization
Although
the
advantages
and
strengths,
as
described
above,
foster
the
decision
for
consideration
of
this
assay
as
the
screening
tool
for
substances
that
alter
steroidogenesis,
the
sliced
testis
assay
also
has
some
limitations,
which
prompted
this
study
plan
to
optimize
the
assay.
Some
limitations
of
the
assay
are
not
able
to
be
resolved.
For
example,
substances
that
require
metabolic
activation
will
not
be
identified
as
substances
that
alter
steroid
hormone
production.
Battelle
Report
7
July,
2002
Also,
substances
that
are
insoluble
in
the
media
or
cannot
be
formulated
in
a
soluble
vehicle
are
unable
to
be
tested.
Aside
from
these
limitations,
there
are
a
number
of
variables
that
impact
the
assay
and
that
can
be
tested
in
order
to
determine
their
degree
of
impact
and
the
setting
that
optimizes
the
analytical
characteristics
of
the
assay.
Sections
2
and
3
of
this
study
plan
describe
the
variables
that
have
been
selected
for
optimization
testing.

An
important
objective
of
the
optimization
experiments
is
to
determine
at
what
levels
to
set
various
factors
in
order
to
reduce
to
a
minimum
the
variability
of
the
assay.
The
variability
of
the
assay
can
be
roughly
estimated
using
data
from
published
research
papers.
To
assess
the
extent
of
consistency
of
results
across
studies,
comparable
experimental
data
were
extracted
from
various
studies
and
compared
among
one
another.
Four
studies
were
identified
as
including
data
that
could
be
compared
­
Laskey,
et
al.
(
1994),
Fail,
et
al.
(
1994),
Gray,
et
al.
(
1995),
and
Gurtler
and
Donatsch
(
1979).
Each
study
utilized
different
test
chemicals,
different
test
chemical
concentrations,
and
even
different
test
chemical
concentration
units.
Thus,
only
the
untreated
control
groups
were
compared.
Testosterone
concentrations
(
ngT/
gm
Testes)
were
assessed
at
various
sampling
times.
Cumulative
standard
errors
of
the
mean
were
calculated
under
the
assumption
that
the
incremental
values
were
independent.
For
each
hour,
where
more
than
one
study
reported
a
cumulative
mean
and
a
cumulative
standard
error
of
the
mean,
a
weighted
one
way
analysis
of
variance
test
was
carried
out.
The
weights
were
based
on
the
standard
errors
of
the
mean.
The
summary
control
results
from
the
studies
are
presented
in
Table
2.

Table
2.
Cumulative
Mean
Concentration
and
Standard
Error
of
the
Mean
by
Hour
and
Study
Hour
Referencea
n
Mean
SEM
F­
value
Approx
Degr
Frb
Approx
p­
value
0
F
12
130
25.0000
­­
­­
­­

1
L
4
510
45.0000
4.14
(
1,5)
0.10
F
12
400
30.0000
Gr
6
216
(
c)
­­
­­
­­

2
L
4
1030
63.6396
116.23
(
2,
5)
<
0.0005*
F
12
700
75.0000
Gu
5
80
25.0000
Gr
6
417
(
c)
­­
­­
­­

3
L
4
1550
77.9423
48.91
(
2,
5)
0.001
F
12
925
80.0000
Gr
6
595
57.0000
4
L
4
2100
84.6404
­­
­­
­­

5
L
4
2620
90.8460
­­
­­
­­

*
p
<
0.005
based
on
a
comparison
following
a
logarithmic
transformation
a.
F
=
Fail
et
al.,
(
1994);
Gr
=
Gray
et
al.,
(
1995);
Gu
=
Gurthler
and
Donatsch
(
1979);
and
L
=
Laskey
et
al.,
(
1994).
b.
The
degrees
of
freedom
associated
with
each
study
is
somewhere
between
n­
1
and
2n­
1,
depending
on
whether
one
or
two
testes
per
animal
were
used
and
the
degree
of
correlation
between
testes
from
the
same
animal.
We
conservatively
assumed
n­
1
degrees
of
freedom.
c.
No
SEM
reported.
Also,
not
used
for
ANOVA
calculation.
Battelle
Report
8
July,
2002
The
results
of
this
comparison
showed
that
for
each
hour
for
which
several
studies
reported
cumulative
testosterone
concentrations
there
was
wide
variation
of
mean
concentrations
across
studies.
In
each
case
there
was
strong
indication
of
significant
statistical
differences
among
studies,
particularly
after
2
hours
and
3
hours.

For
each
hour
where
more
than
one
study
reported
results,
the
total
variance
of
the
cumulative
mean
values
among
studies
was
divided
into
variance
between
studies
and
variance
within
studies.
The
variance
within
studies
was
estimated
as
the
average
of
the
squares
of
the
within­
study
standard
errors
of
the
mean,
based
on
those
studies
for
which
standard
errors
were
reported.
The
variance
between
studies
was
estimated
as
the
variance
of
the
mean
values
among
studies
minus
the
variance
within
studies.
Table
3
displays
the
standard
deviations
among
the
means
between
studies
and
the
standard
errors
of
the
means
within
studies
after
1,
2,
and
3
hours.
The
standard
deviation
of
the
means
between
studies
is
approximately
3.8
to
6.8
times
the
standard
errors
of
the
means
within
studies.
This
agrees
with
the
results
shown
in
Table
2.
The
variation
among
study
means
far
exceeds
that
which
is
due
to
within­
study
variation.

Table
3.
Standard
Deviations
of
Between­
Study
and
Within­
Study
Components
of
Variance
by
Hour
Hour
n
mean
n
sem
Std
Dev
of
Means
Between
Studies
Std
Err
of
Mean
Within
Studies
1
3
2
143.5
38.24
2
4
3
400.4
58.59
3
3
3
479.6
72.40
1.2.3
Prototype
of
the
Sliced
Testis
Steroidogenesis
Assay
As
can
be
deduced
by
the
previous
information,
the
sliced
testis
assay
has
been
used
by
many
researchers
over
the
past
couple
of
decades.
The
parameters
and
settings
that
are
used
can,
and
often
do,
vary
from
laboratory
to
laboratory
and
researcher
to
researcher.
As
a
starting
point,
so
that
the
experimental
factors
and
their
levels
that
may
affect
the
assay
can
be
studied,
the
prototypical
sliced
testis
assay
is
described
and
illustrated
below.

The
sliced
testis
assay
prototype
uses
a
15
week
old
Sprague­
Dawley
rat,
which
is
euthanized
and
its
testes
removed.
The
testes
are
decapsulated,
weighed,
and
placed
in
cold
(
4

C)
media.
The
media
is
medium­
199
(
Gibco)
that
has
added
0.71
g
sodium
bicarbonate,
2.1
g
HEPES,
1.0
g/
L
BSA,
and
0.025
g/
L
soybean
trypsin
inhibitor,
and
is
adjusted
to
a
pH
of
7.4.
The
time
from
removal
to
the
time
of
slicing
is
held
to
under
1
hour.
Each
testis
is
sliced
along
the
longitudinal
axis
into
4
slices.
Each
slice
is
placed
in
a
20
mL
borosilicate
scintillation
vial
(
loosely
capped)
that
contains
5
mL
of
media
alone
(
Figure
3).
The
vials
containing
the
testicular
sections
and
media
are
incubated
at
34

C
on
a
shaker
(
low
speed)
in
5
percent
CO
2/
95
percent
air.
After
the
first
period
of
incubation,
e.
g.,
1
hour,
the
media
is
removed
and
discarded.
Fresh
media
(
5
mL)
is
added
to
the
vial
and
an
aliquot
of
media
(
0.5
mL)
is
collected.
The
sample
is
transferred
and
stored
at
­
70

C
in
a
siliconized
plastic
container.
This
sample
is
the
baseline
sample.
Next,
one
half
of
the
vials
are
challenged
with
a
stimulant,
e.
g.,
hCG,
and
the
other
half
Battelle
Report
9
July,
2002
are
not.
The
final
hCG
concentration
is
0.1
ug/
mL.
Additional
media
samples
are
collected
from
the
vials
at
time
0
(
baseline)
and
1,
2,
3,
and
4
hours
(
post
challenge).
These
media
samples
are
also
stored
frozen
for
later
analysis.
Samples
are
analyzed
for
testosterone
using
an
RIA
method.

Figure
3.
Technical
Flow
Illustration
of
the
Sliced
Testis
Steroidogenesis
Assay
Battelle
Report
10
July,
2002
2.0
PROTOCOL
RESOLUTION
ISSUES
2.1
INTRODUCTION
There
are
no
issues
that
require
resolution
prior
to
initiation
of
the
protocol
included
in
this
study
plan.
The
protocol
is
believed
to
be
complete
and
ready
for
implementation.
However,
there
are
some
analytical
assay
experiments
that
require
verification,
which
are
included
in
the
protocol
and
experimental
design,
and
these
experiments
must
be
completed
prior
to
initiation
of
the
sliced
testis
assay
optimization
experiments.
More
specifically,
the
analytical
assays
planned
for
use
in
this
study
plan
are
the
testosterone
RIA
assay
and
lactate
dehydrogenase
(
LDH)
spectrophotometric
assay.
Response
characteristics
of
these
two
analytical
assays
will
be
verified
in
the
media
selected
for
use
in
the
sliced
testis
assay.
The
following
paragraphs
describe
the
basis/
method
for
verifying
the
analytical
assays,
as
well
as
providing
details
about
each
of
the
analytical
assays
that
are
planned
to
be
used
to
measure
testosterone
and
LDH.

2.2
ANALYTICAL
ASSAY
VERIFICATION
A
set
of
guidelines
published
in
1977
by
endocrine
scientists
is
considered
the
gold
standard
for
validation
of
an
RIA
(
Hafs
et
al.
1977).
These
guidelines
include
validation
recommendations
for
serum
and
plasma,
but
have
been
modified
here
for
use
with
in
vitro
culture
media.
The
basis
and
methods
for
validation
follow.

2.2.1
Analytical
Assay
Validation
It
is
good
laboratory
practice
and
science
to
validate
an
analytical
assay's
performance
characteristics,
especially
when
measuring
the
analyte
in
a
new
or
different
matrix.
While
the
manufacturer
may
provide
some
information
about
the
analytical
assay,
many
times
the
supplier
characterizes
the
analytical
assay
in
a
particular
matrix,
e.
g.,
human
serum
or
plasma,
thereby
necessitating
further
characterization
by
a
user
who
uses
the
analytical
assay
in
a
different
matrix,
e.
g.,
animal
blood
or
culture
media.
Consequently,
each
laboratory
should
determine
the
sensitivity,
reliability,
repeatability,
and
robustness
of
the
analytical
assay
for
measuring
an
analyte
in
a
given
matrix.

Validation
of
an
analytical
assay
verifies
that
there
are
not
substances
that
interfere
with
the
assay.
For
example,
an
RIA
method
that
uses
an
antibody
for
testosterone
but
also
demonstrates
a
high
cross
reactivity
with
androstenedione
will
produce
erroneous
results.
The
antibody
characteristics
can
be
a
significant
source
of
inter­
laboratory
variation
in
determining
concentrations
of
the
target
hormone.
There
are
many
possible
interfering
substances.
Examples
include
other
tissue
products,
media
components,
the
test
substance,
and
the
solvent
or
vehicle
used
to
dissolve
the
test
substance,
to
name
just
a
few.
Examples
of
interfering
substances
in
the
present
study
include:

°
Serum
proteins.
These
proteins
are
present
to
some
extent
in
the
testes,
which
are
synthesized
by
the
testis
or
transported
to
the
testis
by
body
fluids
(
blood,
lymph).
The
major
serum
proteins
are
albumin,
sex
steroid­
binding
globulin,
and
androgen­
Battelle
Report
11
July,
2002
binding
protein.
The
type
and
concentration
of
these
proteins
differ
among
species.
Furthermore,
the
affinity
of
these
proteins
for
a
given
steroid
varies
with
sex,
age,
reproductive
state
(
i.
e.,
pregnancy,
anestrous,
stage
of
cycle,
age,
and
presence
of
other
steroids).
The
testis
also
produces
a
large
number
of
additional
proteins,
which
are
likely
to
be
released
into
the
media
during
the
incubation
period.
Thus,
results
of
experiments
with
and
without
testes
fragments
in
the
media
after
various
time
periods
will
be
evaluated
for
evidence
of
interference
to
the
testosterone
RIA
and
the
LDH
assay.

!
Media
components.
The
pH
of
the
media,
amino
acid
content,
or
ionic
concentration
characteristics,
steroid
metabolites,
and/
or
lipid
content
are
factors
that
could
alter
or
interfere
with
the
assay.

!
Organ
response.
Treatment
with
toxic
substances
may
alter
organ
responses,
resulting
in
varying
secretion
patterns
of
proteins
that
otherwise
would
not
interfere
with
the
assay.

For
these
reasons,
both
the
testosterone
RIA
kit
and
the
LDH
assays
will
be
validated.
The
validation
will
initially
be
conducted
using
the
prototype
media,
i.
e.,
medium­
199
(
Gibco),
0.71
g
sodium
bicarbonate,
2.1
g
HEPES,
1.0
g/
L
BSA,
and
0.025
g/
L
soybean
trypsin
inhibitor,
adjusted
to
pH
7.4.
Upon
completion
of
the
sliced
testis
media
optimization
experiment,
the
analytical
assays
may
need
to
be
re­
validated
using
the
optimal
media.
The
parameters
that
will
be
determined
in
the
validation
of
the
analytical
assay
are
described
below.

2.2.2
Analytical
Assay
Validation
Parameters
Parameters
that
will
be
determined,
as
appropriate,
for
the
analytical
assays
are
listed
below.

°
Accuracy.
Accuracy
is
a
measure
of
the
true
value
of
the
analyte
and
will
be
defined
as
the
compared
agreement
between
the
measured
and
true
analyte
values.
The
criterion
used
for
an
analytical
assay
with
acceptable
accuracy
is
±
10
percent.
The
accuracy
of
the
RIA
will
be
estimated
using
indirect
assessment
methods
­
recovery
and
parallelism.

Recovery.
Recovery
is
the
ability
of
the
assay
to
measure
spiked
analyte
in
the
matrix,
i.
e.,
media.
A
known
amount
of
nonlabeled
analyte
(
A)
will
be
added
to
the
media
(
M)
and
the
recovery
will
be
calculated
as
a
concentration
(
C)
using
the
following
formula:

(
C
­
M)/
A
x
100
Parallelism.
Parallelism
refers
to
the
dilutions
of
the
samples
and
the
accuracy
is
related
to
a
comparison
of
the
measured
and
true
values
at
several
volumes
of
fluid.
When
calculated
quantities
are
converted
to
concentrations
(
ng/
ml),
all
volumes
tested
Battelle
Report
12
July,
2002
should
contain
similar
concentrations
(
unit
per
milliliter)
of
testosterone.
Lack
of
parallelism
can
indicate
the
presence
of
interfering
substances.
Lack
of
parallelism
will
be
followed
up
using
extraction
or
chromatography
in
order
to
isolate
the
testosterone
from
the
media.

!
Sensitivity.
Sensitivity
is
the
capacity
of
the
assay
to
measure
the
smallest
amount
of
analyte.
The
smallest
quantity
of
analyte
that
can
be
quantified
reliably
will
be
determined
and
expressed
as
the
limit
of
detection.
Media
samples
containing
concentrations
below
the
level
of
sensitivity
of
the
assay
will
be
rerun
using
larger
sample
volumes,
assuming
that
a
sufficient
sample
volume
is
available.

!
Precision.
Precision
(
also
referred
to
as
reproducibility)
is
a
measure
of
the
variation
in
analyte
measurements
following
repeated
determination.
Precision
of
a
given
analytical
assay
will
be
evaluated
by
determining
the
mean,
standard
deviation
(
sd)
and
coefficient
of
variation
(
CV).
The
precision
of
the
assay
will
be
determined
within
runs
and
between
runs
from
day
to
day.

!
Cross
Reactivity.
Cross
reactivity
(
also
referred
to
as
specificity)
occurs
when
the
assay
measures
more
than
only
the
required
analyte.
The
antibody
will
be
characterized
for
binding
activity
with
related
compounds.

!
Identity
of
Unknowns.
Identity
of
an
unknown
will
not
be
performed,
but
demonstration
of
its
removal
from
the
assay
will
be
attempted.
Chromatographic
isolation
of
the
analyte
by
HPLC
(
Darney
et
al.,
1983)
or
other
methods
(
differential
extraction),
followed
by
RIA
or
EIA,
should
yield
values
identical
to
concentrations
found
using
RIA
or
EIA
without
chromatographic
separation.
If
not,
then
an
assessment
of
whether
the
unknown
"
steroid"
is
recognized
by
the
antibody
and
whether
it
is
present
in
similar
concentrations
in
all
samples.
Otherwise,
samples
analysis
may
involve
chromatographic
isolation
or
differential
extraction
of
the
principal
steroid.

!
Quality
Control.
Interassay
variation
will
be
calculated
from
values
of
one
or
more
samples
included
on
a
series
of
assays
(
standard
sera).
These
standard
or
"
control"
sera
will
be
aliquoted
and
frozen
prior
to
the
first
assay
in
the
series.
An
aliquot
of
these
control
sera
will
be
included
two
or
more
times
within
each
assay
and
on
each
subsequent
assay.
Internal
quality
control
measures
will
include,
where
appropriate,
high
and
low
control
sera
and
a
solvent
blank.
From
these
standards,
precision
of
the
assay
(
inter­
and
intra­
assay
coefficients
of
variation)
will
be
established.

2.3
TESTOSTERONE
RIA
The
end­
point
analyte
in
the
sliced
testis
assay
is
testosterone
and
it
will
be
measured
using
an
RIA
method.
In
this
method,
the
antibody
is
affixed
to
the
assay
tube,
and
any
bound
antigen
(
testosterone),
whether
labeled
or
unlabeled,
is
recognized
by
the
antibody.
The
labeled
testosterone
will
be
125I­
testosterone
and,
the
kit
that
will
be
used
includes
polypropylene
tubes
Battelle
Report
13
July,
2002
coated
with
a
testosterone­
specific
antibody.
Testosterone
will
be
used
to
prepare
the
standard
curve
(
Sigma,
St.
Louis,
MO;
T­
1500).
Procedural
controls
(
internal
controls)
for
each
assay
will
use
media
controls
(
two
each
with
high
or
low
testosterone
concentrations)
and
reagent
blanks.
Separation
of
bound
and
free
testosterone
will
be
accomplished
by
decanting.
The
bound
fraction
(
which
remains
in
the
tube)
will
be
counted
using
gamma
counter
technology.

Data
are
presented
below
from
a
study
that
used
a
specific
testosterone
RIA.
This
RIA
was
developed
for
human
blood
(
Diagnostic
Products
Corporation)
and
was
validated
for
use
with
plasma
and
intra­
testicular
(
ITT)
fluid
testosterone
analysis
(
Fail
et
al.,
1992;
1995;
1996).
Table
4
summarizes
data
for
rat
plasma
and
ITT
fluid.
In
this
study,
only
one
assay
was
done
so
no
inter­
assay
data
were
generated.
In
another
study,
estimates
of
variation
were
calculated
using
the
internal
controls
from
nine
testosterone
RIAs
completed
during
a
five­
month
period.
Testosterone
mean
and
SEM
values
were
11.37
±
0.57
ng/
ml
across
assays
for
an
inter­
assay
variation
(
CV)
of
14.96%.
The
testosterone
values
(
mean
±
SEM)
for
the
internal
controls
in
these
nine
assays
ranged
between
9.65
±
0.13
and
14.64
±
0.62
ng/
ml
within
assays
(
CV
­
7.5%).

Table
4.
Characteristics
of
a
Radioimmunoassay
Validated
for
Determination
of
Testosterone
in
Adult
Male
Sprague
Dawley
Rat
Plasma
and
Testicular
Fluid
[
Personal
Communication,
Dr.
P.
Fail,
2002*]

Hormone
Assay
Parameter
Plasma
Testosterone
(
ng/
ml)
ITT
Testosterone
(
ng/
ml)
Sensitivity
3.5
pg
3.5
pg
Intra­
assay
Variationa
blank
0/
8.7%
0/
7.6%

mass
added
2/
11.8%
12.5/
5.4%
8/
5.2%
25/
4.4%
50/
4.2%
Interassay
Variationa
NA
NA
%
Recovery
of
Added
Massb
2/
92.5%
8/
94.3%
12.5/
60.7%
25/
68.8%
50/
88.5%

Index
of
Parallelismc
127.1%
112.6%

Test
article
cross
reactivity
(
28
µ
g/
ml)
0%
NT
a
Numbers
are
mass
added/
percentage
variation.
For
interassay
variation,
only
one
run
was
performed.
NT
=
not
tested.
b
Numbers
are
mass
added/
percentage
recovered.
c
Index
of
parallelism
=
concentration
of
low
volume
÷
concentration
of
high
volume
x
100.
*
The
table
values
are
most
likely
fluid
and
species
specific.
Thus,
they
should
be
used
as
representative
of
what
may
be
found
in
characterization
of
other
assays.
Battelle
Report
14
July,
2002
2.4
LACTATE
DEHYDROGENASE
(
LDH)
ASSAY
The
enzyme,
lactate
dehydrogenase
(
LDH),
is
found
in
the
cytoplasm
of
all
cells.
Cellular
LDH
concentrations
are
approximately
500
fold
higher
than
that
found
in
the
serum.
The
magnitude
of
this
gradient
makes
it
possible
to
use
LDH
as
a
marker
for
cellular
viability.
Although
a
small
amount
of
LDH
leaks
from
the
cells
under
normal
conditions,
high
concentrations
of
LDH
are
observed
when
cellular
damage
is
sufficiently
detrimental
that
the
cell
ruptures
and
its
contents
are
released
into
the
extracellular
environment.
The
normal
serum
LDH
concentration
ranges
from
50
to
233
U/
L
(
normal
healthy
adult
human).
In
rat
serum,
historical
levels
are
reported
to
range
from
150
to
300
U/
L
(
Jacobs
et
al.,
1988;
Young,
1990).
In
humans,
elevation
in
serum
LDH
occurs
following
myocardial
infarction,
liver
disease,
pernicious
and
megaloblastic
anemias,
pulmonary
emboli,
malignancies,
and
muscular
dystrophy,
to
name
just
a
few
conditions
and
diseases
(
Tietz,
1986).
For
these
reasons,
LDH
was
selected
as
a
potentially
useful
marker
in
the
sliced
testis
assay
in
order
to
provide
an
indication
of
the
extent
of
cellular
damage
induced
by
dissection
trauma,
incubation
conditions,
and/
or
chemical
substance
testing.
Repeated
LDH
measurements
may
assist
in
the
characterization
of
damage
over
time
and
in
determining
whether
cellular
damage
might
be
associated
with
testis
fragment
size,
e.
g.,
more
extensive
damage
occurring
with
smaller
testicular
fragments.
In
addition,
this
information
may
be
helpful
in
establishing
the
duration
of
the
testis
fragment
equilibration
period,
i.
e.,
the
time
in
culture
before
treatment.
Thus,
the
LDH
assay
is
being
tested
as
a
marker
to
determine
whether
it
can
be
used
to
monitor
cellular/
tissue
viability.

The
LDH
assay
that
will
be
tested
for
its
use
in
the
sliced
testis
assay
is
based
on
the
bioreactivity
of
the
enzyme.
The
assay
is
commercially
available
as
a
kit
and,
and
one
such
kit
(
LD
KINETIC
PROCEDURE,
ThermoDMA,
Arlington,
Texas)
is
representative
of
the
type
of
assay
that
will
be
used
in
the
present
study
design.
In
the
cell,
LDH
specifically
catalyzes
the
oxidation
of
lactate
to
pyruvate
with
the
simultaneous
reduction
of
NAD
to
NADH.
This
same
reaction
can
be
used
to
measure
the
amount
of
LDH
in
a
sample
from
a
biological
matrix.
The
rate
at
which
NADH
is
formed
is
measured
at
340
nm
using
a
spectrophotometer,
thereby
providing
a
stoichiometric
relationship
that
allows
for
the
determination
of
LDH
(
Duhl
and
Jackson,
1978).
The
assay
requires
attention
to
several
factors
that
can
affect
the
level
of
variability.
For
example,
after
a
30
second
incubation
time,
the
change
in
absorbance
per
minute
remains
stable
for
2
minutes.
In
addition,
although
the
assay
has
been
characterized
at
30
°
C
and
37
°
C,
the
assay
should
be
consistently
run
at
the
same
temperature
because
the
LDH
values
change
with
fluctuations
in
temperature.
Furthermore,
substances
and
conditions
in
the
animal
have
been
shown
to
interfere
with
the
assay.
Examples
include:

°
Hemolysis
(
false
elevation)
°
Oxalate
and
ascorbic
acid
(
decrease
LDH
levels)
°
Drugs
(
increase
or
decrease
serum
LDH
levels)
(
Young,
1990).

As
a
means
to
monitor
the
performance
of
the
LDH
assay,
quality
control
procedures
will
be
implemented
in
those
experiments
that
begin
to
characterize
LDH
as
a
monitor
for
cellular
viability.
Normal
and
abnormal
control
sera
of
known
LDH
activities
will
be
analyzed
on
a
routine
basis.
Quality
control
samples
are
commercially
available
(
DMA's
Data­
Trol
N
and
Data­
Battelle
Report
15
July,
2002
Trol
A;
Cat.
No.
1902­
605
and
1901­
605).
In
addition
to
the
quality
control
procedures,
the
experimental
design
as
described
in
the
protocol
includes
evaluating
the
assay
in
the
media
selected.
Furthermore,
the
laboratory
will
establish
its
own
range
of
expected
values,
since
differences
exist
among
instruments,
laboratories,
and
local
populations.
Finally,
the
accuracy,
precision,
and
variability
of
the
assay
will
be
determined.

3.0
STUDY
DESIGN
(
OPTIMIZATION
OF
EXPERIMENTAL
FACTORS)

The
planned
purpose
for
the
sliced
testis
assay
is
to
develop
a
screening
tool
that
will
assess
the
steroidogenic
pathway
capacity
of
the
Leydig
cell
so
as
to
identify
substances
with
toxic
effects
on
hormone
production.
The
objective
of
this
study
plan
is
to
present
optimization
experiments
that
are
designed
to
optimize
the
environmental,
chemical,
and
biological
components
of
the
in
vitro
sliced
testis
steroidogenesis
assay.
The
study
plan
includes
experiments
using
single
factor
and
factorial
design
approaches
so
as
to
identify
factors
that
affect
assay
performance,
as
well
as
to
test
for
interactions
of
the
factors,
which
may
be
integral
in
uncovering
those
conditions
essential
in
minimizing
assay
variability.
This
section
describes
the
factors
to
be
tested
and
the
basis
for
the
levels
selected
for
each
factor.

In
order
to
try
to
make
an
in
vitro
assay
simulate
the
in
vivo
environment,
the
culture
conditions
must
mimic
the
cellular
environment
as
closely
as
possible.
The
assay
factors
believed
most
important
in
accomplishing
this
simulation
were
selected
for
optimization
in
the
present
study
plan.
The
designers
of
these
experiments
kept
in
mind
that
in
order
to
make
the
assay
usable
for
many
different
laboratories,
under
varied
environments,
and
sometimes
difficult
situations,
the
study
plan
would
need
to
test
factors
at
levels
that
may
or
may
not
be
best
for
the
assay
but
are
easily
achieved
at
most
laboratories,
e.
g.,
room
temperature,
air
for
the
atmosphere,
etc.
In
that
way
the
robustness
of
the
assay
can
be
determined.
In
addition,
the
factors
and
levels
selected
for
testing
will
provide
information
that
will
evaluate
how
to
perform
the
assay
as
simply,
economically,
rapidly,
and
efficiently
as
possible
without
compromising
the
performance
of
the
assay.
In
a
related
vein,
some
of
the
factors
selected
and
the
levels
to
be
tested
were
based
on
their
use
by
previous
researchers.
By
including
factors
and
levels
used
previously
with
those
that
have
been
used
primarily
by
researcher
consensus,
this
study
plan
provides
the
opportunity
for
the
assay
to
be
optimized
without
the
confounding
results
that
come
from
multiple
researchers
and
laboratories.
Thus,
multiple
factors
and
various
levels
have
been
selected
for
testing
so
as
to
identify
optimal
assay
settings
and
conditions,
as
well
as
to
define
the
boundaries
and
limitations
of
the
assay.

Biological
and
chemical
parameters
of
the
assay
were
given
consideration
in
identification
and
selection
of
factors
for
testing.
The
factors
were
categorized
as
those
that
affect
incubation,
the
testicular
parenchyma,
parenchymal
viability,
and
sample
analysis.
Each
of
the
factors
in
these
categories
and
the
basis
for
the
levels
selected
for
testing
are
described
in
the
following
subsections.

3.1
INCUBATION
FACTORS
The
sliced
testis
assay
involves
incubation
of
the
testicular
parenchyma
in
media.
The
Battelle
Report
16
July,
2002
incubation
factors
selected
for
testing
include
the
media
type,
gaseous
atmosphere
type,
temperature,
vessel
type,
shaker
speed,
media
volume,
and
hCG
stimulant
concentration.
Each
of
these
factors
and
their
levels
selected
for
testing
are
described
in
further
detail
below.

3.1.1
Incubation
Media
Type
A
review
of
the
literature
indicates
that
several
different
types
of
media
with
supplemental
components
have
been
used
in
the
conduct
of
the
sliced
testis
assay.
These
media
and
components
include:

°
medium­
199
(
Gibco)
with
0.71
g
Na
bicarbonate,
2.1
g
HEPES,
1.0
g/
L
BSA,
and
0.025
g/
L
soybean
trypsin
inhibitor.
Adjusted
to
pH
7.4.

Refs.
Laskey
et
al.,
1994;
Klinefelter
et
al.,
1994;
Fail
et
al.,
1994;
1995;
Gray
et
al,
1995
°
Eagles
MEM
Ref.
Wilker
et
al.,
1995
°
RPMI­
1640
medium
(
without
phenol
red)
with
10%
FCS
and
50
ug/
mL
soybean
trypsin
inhibitor.

Ref.
Powlin
et
al,
1998
As
is
apparent
above,
the
majority
of
the
researchers
have
used
medium­
199
that
is
supplemented
with
buffer
and
reagents
that
reduce
cellular
instability
and
degradation.
Of
these
researchers,
none
have
investigated
the
possible
effect
of
media
type
on
assay
performance.
However,
Powlin
et
al.
(
1998)
did
examine
the
effect
of
conducting
the
sliced
testis
assay
with
and
without
fetal
calf
serum
(
FCS),
bovine
serum
albumin
(
BSA),
and
soybean
trypsin
inhibitor.
Although
their
results
indicated
that
the
presence
or
absence
of
each
of
these
supplemental
components
had
no
effect
on
testosterone
production,
the
incubation
media
that
they
used
for
follow­
on
experiments
included
all
three
components
at
the
concentrations
listed
above.
Since
these
components
were
not
considered
critical
to
the
assay
performance,
the
present
study
plan
does
not
include
testing
all
combinations
of
media
and
supplemental
components.
The
ready
availability
of
the
supplemental
components
and
ease
of
combining
them
with
the
media,
together
with
the
number
of
experiments
needed
to
test
all
the
combinations
and
the
results
reported
by
Powlin
et
al
(
1998),
precluded
inclusion
of
such
experiments
in
the
present
study
plan.
Thus,
a
decision
was
made
in
the
design
of
the
present
study
plan
to
compare
the
effect
of
the
in
toto
media
used
by
the
different
investigators,
rather
than
effects
of
individual
reagents,
on
the
performance
of
the
assay.

3.1.2
Incubation
Gaseous
Atmosphere
The
types
of
gaseous
atmosphere
that
have
been
and
are
used
in
this
assay
are
limited.
Battelle
Report
17
July,
2002
The
most
commonly
used
atmosphere
is
5%
CO
2/
95%
air.
However,
a
commonly
used
atmosphere
for
cell
culture
systems
is
5%
CO
2/
95%
O
2.
In
addition,
it
is
possible
that
the
performance
of
the
assay
is
not
affected
by
the
atmosphere
as
long
as
it
is
similar
to
the
ambient
air,
which,
if
correct,
would
simplify
the
requirements
of
performing
the
assay.
For
these
reasons,
the
following
three
atmospheres
were
selected
for
testing
­
air
(
three
gases);
5%
CO
2/
95%
air;
and
5%
CO
2/
95%
O
2.

3.1.3
Incubation
Temperature
The
sliced
testis
assay
is
most
commonly
conducted
using
an
incubation
temperature
of
34

C.
This
temperature
is
used
because
it
is
believed
to
be
the
temperature
of
the
testis
in
the
scrotum.
Testicular
temperature
is
lower
than
body
temperature
to
allow
spermatogenesis
to
occur.
No
report
was
found
that
determined
by
experimentation
the
temperature
of
the
rat
testes
by
insertion
of
thermal
probes.
It
is
assumed
that
most
investigators
use
34

C
based
on
bull
and
human
studies.
In
addition
to
testing
the
assay's
performance
at
34

C,
a
temperature
of
25

C
was
selected
to
test
whether
room
temperature
had
an
effect.
If
not,
then
the
assay
could
be
performed
more
efficiently.
A
third
temperature,
37

C,
was
chosen
to
determine
whether
whole
body
temperature
had
an
effect
on
assay
performance.
Finally,
40

C
was
chosen
as
an
extreme
level
because
temperatures
up
to
42

C
have
been
used
to
simulate
cryptorchidism
in
rats.
Thus,
the
incubation
temperatures
selected
for
testing
are
25,
34,
37,
and
40

C.

3.1.4
Incubation
Vessel
The
incubation
vessel
used
during
the
incubation
period
will
be
tested
for
its
effect
on
assay
performance.
This
experimental
factor
is
included
in
the
study
plan
because
of
the
work
reported
by
Powlin
et
al
(
1998).
The
sliced
testis
assay
was
conducted
using
a
20
mL
scintillation
vial
or
a
16
x
100
mm
test
tube;
both
were
made
of
borosilicate
glass.
Their
results
showed
a
4­
and
9­
fold
increase
in
testosterone
production
when
the
scintillation
vial
was
used
relative
to
the
test
tube.
They
attributed
this
finding
to
greater
physical
dispersion
of
the
parenchymal
fragment
in
the
scintillation
vial.
Based
on
these
results
and
explanation
for
the
effect,
incubation
vessels
with
varying
sizes
of
bottom
surface
area
were
selected
for
testing.
The
levels
of
this
factor,
in
increasing
order
of
bottom
surface
area,
that
will
be
tested
are:

°
16
x
100
mm
test
tube
°
20
mL
scintillation
vial
°
20
mL
Erlenmeyer
flask
3.1.5
Incubation
Shaker
Speed
The
sliced
testis
assay
includes
shaking
of
the
parenchyma
in
the
media
during
the
incubation
period.
The
speed
of
the
shaker
is
not
an
experimental
factor
that
has
been
uniformly
decided
upon
by
researchers,
nor
has
it
been
tested
for
its
affect
on
the
assay.
Most
articles
refer
to
using
a
shaker
during
incubation
but
the
speed
of
the
shaker
is
rarely
reported,
if
known.
Powlin
et
al.
(
1998)
used
a
shaker
speed
of
175
rpm,
which
is
considered
a
high
(
or
vigorous)
speed.
Since
shaking
allows
the
parenchyma
to
have
greater
exposure
to
the
culture
components
Battelle
Report
18
July,
2002
such
as
media
and
gases,
it
is
possible
that
shaking
speeds
may
provide
varying
degrees
of
exposure
between
the
media
components
and
gaseous
atmospheres.
For
these
reasons,
the
shaking
speed
was
selected
as
a
factor
for
testing
and
the
levels
to
be
tested
are:
no
shaking,
and
low
and
high
speeds,
where
the
low
speed
will
be
less
than
60
rpm
and
the
high
speed
approximately
175
rpm.

3.1.6
Incubation
Media
Volume
The
sliced
testis
assay
includes
bathing
the
parenchyma
in
media
during
the
incubation
period.
The
volume
of
media
used
by
most
researchers
is
generally
5
mL.
This
volume
was
arrived
at
by
consensus
rather
than
experimentation.
The
volume
of
media
used
must
be
balanced
between
two
off­
setting
considerations.
First,
sufficient
media
must
be
used
to
keep
the
parenchyma
in
an
aqueous
environment,
as
well
as
to
provide
a
means
to
deliver
substances
for
toxicity
testing.
Varying
the
incubation
media
volume
will
result
in
different
degrees
of
exposure
to
the
parenchyma,
as
well
as
to
the
gaseous
atmosphere.
Second,
the
media
must
be
kept
to
a
minimum
in
order
to
preclude
diluting
testosterone
concentrations
below
the
level
of
detection
of
the
analytical
method.
Thus,
the
media
volume
is
a
factor
that
is
included
in
the
study
design
and
the
levels
to
be
tested
are
2.5,
5,
and
10
mL.

3.1.7
hCG
Concentrations
Human
chorionic
gonadotropin
(
hCG)
is
used
in
the
sliced
testis
assay
to
stimulate
steroidogenesis.
Concentrations
used
in
previous
studies
include100
mIU/
mL
(
Laskey
et
al.,
1994)
and
1
IU/
mL
(
Fail
et
al.,
1994;
Powlin
et
al.,
1998).
A
dose­
response
experiment
is
necessary
to
determine
the
concentrations
that
stimulate
maximal
testosterone
production.
From
such
an
experiment,
the
optimal
concentration
can
be
selected
based
on
the
other
conditions
selected
to
conduct
the
assay.
The
optimal
concentration
is
not
necessarily
the
concentration
that
produces
a
maximal
response.
The
goal
of
the
study
plan
is
to
test
several
concentrations
of
hCG
in
order
to
identify
a
concentration
that
does
not
overstimulate
to
exhaustion
the
parenchymal
fragment
but
that
does
provide
a
steady
linear
production
of
testosterone
over
the
maximal
possible
incubation
period.
The
levels
selected
for
testing
range
from
a
concentration
that
is
expected
to
have
no
effect
to
one
that
would
likely
cause
exhaustion
of
the
cells
from
overstimulation.
The
levels
selected
for
testing
are
0.001,
0.01,
0.1,
1,
and
10
IU/
mL.

3.2
TESTICULAR
PARENCHYMA
FACTORS
The
sliced
testis
assay
involves
incubation
of
testicular
parenchyma.
The
experimental
factors
selected
for
testing
include
parenchymal
fragment
size,
time
to
prepare
whole
testis
to
fragment,
media/
temperature
used
during
collection
of
whole
testes,
and
effect
of
sample
aliquot
volume
on
media:
tissue
ratio.
Each
of
these
factors
and
their
levels
selected
for
testing
are
described
in
further
detail
below.

3.2.1
Parenchymal
Fragment
Size
The
sliced
testis
assay
uses
a
fragment
of
the
whole
testis.
Obviously,
the
size
of
the
testis
Battelle
Report
19
July,
2002
slice
used
is
directly
proportional
with
the
number
of
Leydig
cells
placed
in
culture.
However,
the
size
of
the
parenchymal
fragment
used
for
testing
has
not
been
uniformly
accepted.
Most
researchers
have
used
quartered
testis
(
Laskey
et
al.,
1994;
Gray
et
al.,
1995;
Fail
et
al.,
1994,
1995)
but
50
mg
fragments
have
also
been
used
(
Powlin
et
al.,
1998).
Personal
communications
with
researchers
in
this
area
have
indicated
that
as
the
fragment
size
is
decreased
the
variability
of
the
assay
increases.
However,
documentation
of
the
relationship
between
fragment
size
and
assay
variability
is
not
available.
Another
issue
that
affects
the
fragment
size
is
the
composition
of
the
testis.
Consistent
fragment
sizes
are
difficult
to
obtain
due
to
the
gelatin­
like
consistency
of
the
parenchyma.
Once
the
testis
is
sliced
in
half,
quartered,
etc.,
small
adjustments
to
the
fragment
size
to
be
tested
are
not
feasible.
In
addition,
the
parenchymal
composition
is
such
that
slicing
the
testis
to
smaller
and
smaller
sizes
is
no
longer
possible
beyond
1/
8
slices
(~
125
mg).
Use
of
fragments
less
than
approximately
125
mg
is
possible
but
these
fragments
are
obtained
by
"
scooping"
the
desired
amount
of
parenchyma
rather
than
transferring
a
slice
of
the
testis
to
the
incubation
vessel.
Regardless
of
the
advantages
associated
with
using
larger
fragments,
it
is
important
to
determine
the
smallest
possible
amount
of
parenchyma
that
can
be
used
without
compromising
assay
performance
so
that
the
maximum
number
of
assays
can
be
conducted
using
a
minimal
number
of
animals,
which
will
also
reduce
biological
variability.
Therefore,
based
on
this
information,
the
fragment
sizes
selected
for
testing
are:

°
½
testis
slice
(~
500
mg)
°
¼
testis
slice
(~
250
mg)
°
c
testis
slice
(~
125
mg)
°
50
mg
parenchyma
°
25
mg
parenchyma
3.2.2
Preparation
Time
The
sliced
testis
assay
involves
isolation
and
removal
of
the
testis
prior
to
processing
it
into
fragments
for
testing
and
incubation.
The
duration
of
time
that
it
takes
to
remove
the
testis
and
slice
it
into
fragments
may
affect
the
viability
of
the
parenchyma,
thereby
compromising
steroidogenesis
and
assay
performance.
This
experimental
factor
is
exacerbated
when
several
testes
are
used
for
multiple
fragment
preparation
and
testing.
Since
the
size
of
the
present
study
plan
will
require
many
days
involving
multiple
fragment
preparations,
the
delay
time
in
processing
the
testis
was
considered
to
be
an
important
experimental
design
factor
for
testing.
Experimental
data
that
define
the
limits
of
the
testis
processing
procedure
will
allow
laboratories
to
maximize
their
efforts
while
at
the
same
time
ensure
that
the
organs
are
viable
for
testing.
Therefore,
the
amounts
of
time
that
can
elapse
before
the
testis
must
be
processed
without
having
an
effect
on
the
assay's
performance
are
0.5,
1,
and
2
hours.

3.2.3
Preparation
Techniques
of
Whole
Testis
Following
removal
of
the
testis
and
prior
to
processing
into
fragments,
the
testis
is
bathed
in
media.
However,
a
variety
of
preparation
techniques
have
been
used
to
keep
the
testis
fresh.
The
different
procedures
used
by
researchers
primarily
differ
with
respect
to
the
bathing
media
and
temperature.
Few
articles
include
a
description
of
this
procedure
but,
for
those
that
do,
the
Battelle
Report
20
July,
2002
testis
are
bathed
in
cold
phosphate­
buffered
saline
(
Gray
et
al.,
1995)
or
cold
saline
(
Powlin
et
al.,
1998).
Some
papers
imply
that
media
at
room
or
incubation
temperature
was
used
(
Klinefelter
et
al.,
1994).
Thus,
the
following
levels
were
selected
for
testing
based
upon
the
general
or
specific
information
provided
by
in
the
literature:

°
Cold
(
4

C)
phosphate
buffered
saline
(
pH
7.4)

°
Warm
(
25

C)
phosphate
buffered
saline
(
pH
7.4)

°
Cold
(
4

C)
media
°
Place
encapsulated
testis
in
labeled
cassette,
moisten
with
cold
physiological
saline,
place
on
ice,
and
cover
with
saline­
moistened
paper
towels
(
Powlin
et
al.,
1998).

3.2.4
Aliquot
Volume
Aliquots
of
the
media
are
collected
during
the
incubation
period
for
analyte
analysis.
The
aliquot
sample
size
will
affect
the
volume
of
the
culture
system
since
the
media
is
not
replaced
after
the
baseline
sample
is
collected
and
the
incubation
period
for
testing
begins.
As
the
aliquots
are
removed
during
the
incubation
period,
different
tissue:
media
ratios
exist.
In
addition,
the
size
of
the
aliquot
may
have
a
large
bearing
on
the
tissue:
media
ratio,
e.
g.,
smaller
aliquots
taken
over
time
allow
the
tissue:
media
ratio
to
remain
more
constant
than
removal
of
larger
aliquots.
For
this
reason,
the
study
plan
includes
testing
of
different
aliquot
volumes.
The
selections
included
aliquot
volumes
that
are
typically
used
(
0.25
and
0.50
mL)
and
a
minimum
volume
(
0.1
mL),
which
may
be
insufficient
to
conduct
the
RIA
based
on
the
testosterone
concentration
in
the
media.

3.3
SAMPLE
STABILITY
FACTORS
The
sliced
testis
assay
involves
collection
of
samples
from
the
media
for
analysis
of
the
endpoint.
For
this
assay,
the
endpoint
will
be
testosterone.
Several
experimental
factors
could
affect
the
measurement
of
testosterone;
not
because
of
the
assay
conditions
but
as
a
result
of
the
storage
conditions.
These
experiments
will
not
require
conducting
the
sliced
testis
assay
to
evaluate
the
effect
of
the
factors
on
the
assay.
Rather,
these
factors
will
be
experimentally
determined
by
adding
a
known
amount
of
the
analyte,
testosterone,
to
the
media
and
comparing
the
actual
concentrations
to
the
target
concentration.
The
concentrations
to
be
tested
will
be
selected
from
the
results
of
the
RIA
verification
experiments.
A
concentration
near
the
level
of
detection
or
the
low
end
range
of
the
expected
assay
results
will
be
used
in
order
to
test
the
limitations
of
the
analytical
method.
The
experimental
factors
selected
for
testing
sample
stability
include
storage
container
type,
storage
time,
and
storage
temperature.
Each
of
these
factors
and
their
levels
selected
for
testing
are
described
in
further
detail
below.

3.3.1
Sample
Storage
Container
The
storage
container
is
a
factor
that
can
alter
the
analyte
measurement.
The
composition
of
the
container
may
react
with
the
analyte
and,
if
adsorption
or
leaching
of
a
reactive
component
Battelle
Report
21
July,
2002
from
the
container
occurs,
then
the
analyte
concentration
may
be
compromised.
For
this
reason,
the
study
plan
includes
examining
sample
storage
containers
that
are
plastic
with
a
siliconized
or
non­
siliconized
surface
to
determine
whether
testosterone
concentrations
are
disparate.

3.3.2
Sample
Storage
Time
The
length
of
sample
storage
will
be
tested
to
determine
if
there
is
deterioration
of
the
testosterone
over
time.
Spiked
samples
will
be
stored
and
removed
for
analysis
in
order
to
compare
the
concentrations
over
time.
The
samples
will
be
stored
in
different
containers
so
that
no
samples
will
go
through
freeze/
thaw
cycles.
The
storage
sample
times
to
be
tested
are
the
next
day,
1
week
and
1
and
3
months.
The
3
month
time
period
was
selected
because
this
was
believed
to
be
the
longest
time
period
needed
before
analysis
could
be
completed
for
all
samples
from
a
given
set
of
experiments.

3.3.3
Sample
Storage
Temperature
Some
substances
are
known
to
deteriorate
at
different
storage
temperatures.
For
this
reason,
the
effect
of
sample
storage
temperature
on
stability
of
the
samples
will
be
tested.
Not
all
laboratories
have
access
to
­
70

C
storage
facilities.
Therefore,
the
temperatures
selected
for
testing
are
­
20
and
­
70

C.
The
effect
of
temperature
on
sample
stability
will
only
be
determined
on
the
samples
stored
for
1
and
3
months.

3.4
SAMPLE
COLLECTION
INTERVALS
Sample
collection
interval
refers
to
the
spacing
of
time
between
removing
aliquots
of
media
for
testosterone
analysis.
In
general,
most
researchers
have
taken
hourly
samples
for
4
to
5
hours
after
the
completion
of
the
initial
equilibration
period
(
Laskey
et
al.,
1994;
Gray
et
al.,
1995;
Fail
et
al.,
1995;
1995;
Powlin
et
al.,
1998).
Previous
investigators
have
shown
the
sliced
testis
assay,
when
stimulated
with
hCG,
will
produce
testosterone
at
a
rate
of
approximately
200
ng/
g
testis/
hour
for
up
to
3
hours
(
Gray
et
al.,
1995).
However,
there
has
been
no
systematic
investigation
conducted
to
determine
whether
hourly
collection
intervals
are
the
optimal
time
points
for
measuring
the
production
of
testosterone,
nor
if
the
number
of
time
points
collected
are
actually
necessary.
Perhaps
fewer
samples
at
specific
time
points
will
give
the
same
information
with
increased
efficiency.

In
addition
to
experimentally
determining
the
optimal
sample
collection
intervals,
the
duration
of
the
sliced
testis
assay
may
be
able
to
be
taken
beyond
4
to
6
hours.
Previous
studies
have
shown
that
substances
can
have
a
delayed
effect
on
alteration
of
steroidogenesis.
Thoreux­
Manlay
and
co­
workers
(
1995)
showed
that
the
effect
of
lead
is
delayed
and
may
not
become
apparent
for
a
few
hours
after
adding
it
to
the
incubation
media.
For
this
reason,
determination
of
the
latest
period
of
time
that
the
sliced
testis
assay
produces
testosterone
at
the
established
linear
rate
will
be
useful
for
testing
those
substances
with
possible
delayed
effects.

The
study
plan
includes
experiments
that
will
determine
the
optimal
sample
collection
intervals
and
the
duration
of
time
that
the
assay
can
be
used.
The
time
points
to
be
tested
(
after
equilibration)
are
0.5,
1,
2,
3,
4,
8,
12,
and
24
hours.
Battelle
Report
22
July,
2002
3.5
CHARACTERIZATION
OF
VIABILITY
The
viability
of
the
parenchymal
fragment
used
in
the
sliced
testis
assay
is
an
important
consideration
if
it
is
used
to
evaluate
steroidogenic
altering
effects
of
a
substance.
Viability
refers
to
the
capacity
of
the
parenchymal
fragment
to
simulate
its
in
vivo
function
in
an
in
vitro
environment.
In
regard
to
the
sliced
testis
assay,
a
distinction
must
be
made
between
testosterone
released
as
a
result
of
basal
operation
or
hCG
stimulation
versus
testosterone
release
due
to
cellular
damage
and
death.
A
marker
is
needed
that
will
allow
the
integrity
of
the
parenchyma
to
be
monitored.
Several
techniques
have
been
used
in
in
vitro
assays
to
evaluate
viability.
Klinefelter
et
al.,
(
1987)
described
how
to
stain
for
3 ­
HSD
activity,
and
many
investigators
now
use
this
technique
to
demonstrate
viability
of
Leydig
cell
cultures.
Other
techniques
that
assess
cell
viability
include
Trypan
blue,
which
is
excluded
by
intact
cells,
histological
examination
of
the
cells
(
Klinefelter
et
al.,
1991),
quantification
of
[
35S]
 
methionine
incorporation
into
proteins
synthesized
de
novo
(
Kelce
et
al.,
1991),
and
a
colorometric
assay
using
tetrazolium
salt
MTT,
which
is
reduced
by
succinate
dehydrogenase
(
a
mitochondrial
enzyme)
to
formazan
(
Thoreux­
Manlay
et
al.,
1995).
Other
viability
markers
include
cytokine
release
and
the
ATP
bioluminescence
assay
(
Dr.
Jerome
Goldman,
personal
communication).
Lactate
dehydrogenase
(
LDH),
a
cytoplasmic
enzyme,
is
also
used
as
a
marker
for
cell
damage
(
as
described
in
Section
2).
In
the
present
study,
LDH
will
be
evaluated
as
a
marker
of
parenchymal
damage
in
the
sliced
testis
assay.
The
LDH
assay
was
selected
because
it
is
inexpensive
to
perform,
uses
equipment
commonly
found
in
laboratories,
has
a
fast
through­
put,
and
is
well
established.

Characterization
of
the
marker
assay
is
the
objective
of
the
experiments
included
in
this
part
of
the
study
plan.
These
experiments
will
be
conducted
using
the
optimized
assay.
The
experiment
will
evaluate
the
viability
of
the
parenchyma
and
capacity
of
LDH
to
detect
damage
and
potential
factors
that
may
affect
viability,
i.
e.,
vehicle
type
and
concentration.
In
addition,
the
LDH
marker
may
be
useful
for
characterizing
the
actual
time
needed
for
parenchymal
equilibration.
Each
of
these
experiments
and
the
basis
for
their
selection
are
described
in
further
detail
below.

3.5.1
LDH
­
Evaluation
as
a
Marker
Experiments
will
be
conducted
to
evaluate
LDH
as
a
marker
for
parenchymal
fragment
viability.
Viability
will
have
to
be
compromised
in
order
to
show
that
LDH
is
released
and
can
be
measured
as
a
result
of
parenchymal
fragment
damage.
The
study
plan
will
use
trauma
(
blunt
instrument)
and/
or
heat
(
45

C
for
the
duration
of
the
incubation
sampling
period)
and/
or
chemicals
(
EDS
­
ethane
dimethanesulfonate
at
500
or
3000

M)
as
sources
of
potential
cellular
damage.
Initially,
the
intent
will
be
to
inflict
sufficient
damage
to
the
fragment
such
that
a
clear
LDH
response
can
be
demonstrated.
Later,
a
gradation
of
the
damage
will
be
examined
in
order
to
assess
the
extent
of
damage
needed
in
order
to
elicit
a
response.
The
endpoints
during
these
viability
tests
will
include
testosterone
and
LDH.
In
that
way,
the
parenchymal
responsiveness
based
on
its
viability
can
be
compared
with
the
change
in
LDH
concentrations
over
time.

3.5.2
Equilibration
Period
Battelle
Report
23
July,
2002
The
sliced
testis
assay
includes
an
equilibration
period.
This
period
begins
when
the
parenchymal
fragment
is
placed
in
the
incubation
vessel
and
media
is
added.
In
general,
most
researchers
use
a
1
hour
equilibration
period.
Equilibration
is
needed
in
order
for
the
parenchymal
fragment
to
recover
after
being
transferred
from
an
in
vivo
to
an
in
vitro
environment.
It
is
also
during
this
time
that
any
release
of
testosterone
and
other
intracellular
components
are
discharged
into
the
media.
The
media
used
during
the
equilibration
period
is
discarded
at
the
conclusion
of
the
equilibration
period.
New
media
is
added
and
it
is
generally
agreed
upon
that
only
basal
or
hCG
stimulated
testosterone
is
released
into
the
media
from
the
Leydig
cells
after
the
equilibration
period
is
over.

The
objective
of
these
experiments
will
be
to
characterize
parenchymal
viability
during
the
equilibration
period.
LDH
will
be
measured
at
various
time
points
to
determine
whether
a
1­
hour
equilibration
period
is
sufficient
or
necessary.
The
information
obtained
from
this
experiment
will
provide
a
basis
for
enhancing
assay
performance,
e.
g.,
longer
time
period
is
needed,
or
increased
efficiency,
e.
g.,
a
shorter
time
period
is
adequate.
The
sliced
testis
assay
will
be
conducted
by
taking
samples
during
the
equilibration
period.
The
sampling
times
selected
are
15,
30,
and
45
minutes
and
1,
1.5,
and
2
hours.

3.5.3
Vehicle
Type
and
Concentration
The
sliced
testis
assay
is
being
optimized
for
use
as
a
screening
tool
of
chemicals
with
possible
steroidogenic
altering
effects.
The
vehicle
represents
a
component
that
is
added
to
the
media,
which
has
the
potential
to
cause
damage
to
the
integrity
of
the
parenchymal
fragment.
These
experiments
will
test
the
effect
of
vehicles
and
concentrations
of
the
vehicles
that
are
commonly
used
to
deliver
to
the
incubation
vessel
the
substances
being
tested.
A
wide
range
of
concentrations
of
the
vehicles
will
be
evaluated
because
solubility
of
the
test
substance
can
often
be
a
limiting
factor
in
using
in
vitro
assays.
These
experiments
will
provide
the
information
needed
about
the
limitations
of
the
vehicles
as
it
pertains
to
parenchymal
viability.
Both
testosterone
and
LDH
will
be
measured
at
specified
time
points
after
the
equilibration
period.
The
vehicles
and
the
concentrations
selected
for
testing
are
ethanol,
DMSO,
and
Tween
20
at
0.5,
1,
5,
and
10
percent
(
v/
v).

4.0
RECOMMENDED
TEST
SUBSTANCES
The
purpose
of
this
study
plan
is
to
design
experiments
that
will
evaluate
factors
for
the
optimal
performance
of
the
sliced
testis
assay.
For
this
reason,
there
will
be
no
substance
testing
included
in
this
study
plan.
Once
the
optimization
experiments
are
completed,
the
data
evaluated,
and
the
assay
further
reviewed
for
its
potential
to
meet
the
objectives
as
a
screening
tool,
then
a
pre­
validation
study
plan
will
be
developed.
In
the
pre­
validation
study
plan,
the
optimized
sliced
testis
assay
would
be
evaluated
for
its
capacity
to
detect
an
effect
of
a
substance
on
steroidogenesis.
Battelle
Report
24
July,
2002
Figure
4.
Sliced
Testis
Steroidogenesis
Assay
Experimental
Design
Organizational
Diagram
5.0
STUDY
PROTOCOL
The
study
protocol
is
included
with
this
study
plan
as
an
attachment.

6.0
STATISTICAL
METHODS
6.1
INTRODUCTION
The
study
plan
for
testing
the
factors
described
in
the
previous
sections
involves
two
phases
and
utilizes
single
factor
and
factorial
experimental
designs.
A
diagram
of
the
experimental
design
for
this
study
plan
is
illustrated
in
Figure
4
Battelle
Report
25
July,
2002
The
study
plan
is
divided
into
Phases
1
and
2.

In
Phase
1,
the
Preliminary
Experimental
Phase,
the
analytical
assays
planned
for
use
will
be
verified
and
three
factors
that
may
affect
the
performance
of
the
assay
will
be
tested.
The
reasoning
for
including
these
three
factors
in
the
preliminary
phase
was
to
establish
early
whether
a
given
level
of
each
factor
was
going
to
affect
assay
performance.
Although
any
factor
listed
in
the
study
plan
could
be
rationalized
to
fit
such
a
criteria,
inclusion
in
the
preliminary
phase
also
required
that
the
factor
be
unlikely
to
have
an
interaction,
or
at
best
a
minimal
interaction,
with
another
experimental
factor.
Although
subjective,
these
three
factors
were
believed
to
best
fit
these
criteria.
Furthermore,
it
was
believed
essential
to
establish
the
optimal
level
for
each
of
these
factors
before
proceeding
with
the
factorial
experiments
since
an
effect
of
one
of
these
would
require
additional
verification
experiments
after
sensitivity
analysis.
Finally,
by
establishing
the
media
type
early
on
in
the
experiment,
the
analytical
assay
verification
testing
(
Phase
1)
and
Optimization
of
Sample
Testing
(
Phase
II)
could
be
initiated
earlier
in
the
study
milestone
schedule.

In
Phase
II,
Primary
Experimental
Phase,
the
majority
of
the
experiments
designed
to
evaluate
the
factors
that
may
affect
assay
performance
will
be
tested.
Phase
II
is
divided
into
six
different
sections.
Each
section
is
comprised
of
those
factors
that
may
produce
interactions.
The
groupings
allow
the
interactions
to
be
identified
and
estimated.
It
is
not
possible
to
test
all
factors
and
their
levels
for
all
possible
interactions,
nor
is
it
prudent.
For
example,
some
factors
will
absolutely
not
result
in
an
interaction,
e.
g.,
incubation
vessel
type
and
sample
storage
container
type.
Others
are
unlikely
to
result
in
an
interaction,
e.
g.,
incubation
temperature
and
time
taken
to
process
the
testis
into
fragments.
Others
were
sufficiently
likely
to
interact
that
it
was
believed
necessary
to
include
them
in
two
different
optimization
experiments,
e.
g.,
hCG
concentration
testing
with
both
the
incubation
condition
and
the
testis
preparation
factors.
Thus,
an
attempt
was
made
to
categorize
factors
according
to
type
and
whether
a
possible
interaction
needed
to
be
identified.
These
experiments
comprise
the
first
four
sections
of
Phase
II.

Sensitivity
analysis
is
the
fifth
section
of
Phase
II.
Briefly,
this
analysis
involves
the
development
of
a
response
surface
model
based
on
the
results
of
all
the
optimization
experiments.
The
response
surface
model
describes
the
variation
in
response
level
as
a
function
of
the
variation
in
the
totality
of
experimental
factors.
This
analysis
is
described
in
further
detail
later
in
this
section.
Upon
completion
of
this
analysis,
selected
verification
experiments
may
be
needed
to
evaluate
any
untested
interactions
and
verify
the
model
predictions.

Upon
completion
of
the
sensitivity
analysis,
the
sliced
testis
assay
will
be
sufficiently
characterized
to
describe
the
optimal
settings
of
factors
and
levels.
The
assay
will
then
be
used
to
characterize
whether
LDH
can
be
used
as
a
marker
to
assess
viability
of
the
parenchymal
fragment.
In
addition,
other
factors
will
be
tested
during
this
stage
of
Phase
II
that
may
affect
the
viability
of
the
fragment.

The
statistical
methods
used
in
the
design
of
this
study
plan
are
described
below.
Battelle
Report
26
July,
2002
6.2
ASSUMPTIONS
AND
RESPONSE
ASSESSMENT
Assumptions
used
in
the
experimental
design
include:

°
The
prototypical
design
of
the
sliced
assay
is
considered
a
good
assay.
Slight
modifications
of
this
design
will
serve
as
the
the
initial
reference
design
and
the
point
of
comparison.
°
All
combinations
of
specified
factor
levels
are
a
priori
sensible.
°
While
it
is
believed
that
the
optimum
conditions
will
be
in
the
vicinity
of
the
prototypical
conditions,
it
is
also
possible
that
any
combination
of
the
specified
factor
levels
may
be
better.
Thus,
the
entire
response
region
needs
to
be
explored.
°
All
factors
can
be
varied
with
equal
ease.
Thus,
the
order
of
the
tests
(
within
experimental
sections)
should
be
completely
randomized.
°
Approximately
20
runs
can
be
performed
a
day
and
approximately
60
testosterone
RIA
samples
can
be
analyzed
a
day.

The
principal
response
that
will
be
measured
is
the
rate
of
production
of
testosterone
(
ng
T/
mg
testis/
hr).
This
will
be
measured
both
in
unenhanced
medium
(
baseline)
and
in
medium
stimulated
with
hCG.
Responses
will
be
assessed
at
1,
2,
3,
and
4
hours
following
parenchymal
equilibration.
From
these
measurements,
the
objective
will
be
to
determine
the
combination(
s)
of
factors
that
result
in
baseline
and
hCG­
enhanced
testosterone
production
response
curves
that
are
°
Individually
linear
°
Individually
steep
°
Well
separated
from
one
another.

These
objectives
will
be
evaluated
by
determining
the
slope
and
the
curvature
of
the
baseline
trend
and
the
hCG­
enhanced
trend
at
each
set
of
experimental
conditions.
At
the
optimum
conditions
the
slopes
are
large
and
different
and
the
curvature
is
small.

6.3
PRELIMINARY
EXPERIMENTAL
PHASE
(
PHASE
I)

The
conduct
of
the
optimization
experiments
in
this
phase
will
involve
setting
all
factors
to
their
prototypical
levels
(
See
Section
1.2.3).
The
three
factors
to
be
tested
will
be
varied
one
at
a
time.
It
is
assumed
that
effects
of
variation
of
these
factors
will
not
interact
among
one
another
or
with
the
remaining
experimental
factors.
For
example,
if
a
particular
incubation
atmosphere
produces
the
best
results
with
the
remaining
factors
set
at
the
prototypical
conditions,
then
it
will
also
produce
the
best
results
for
all
combinations
of
conditions
in
the
region
of
the
optimum.

The
preliminary
experiments
include
testing:

1.
Media
Type
(
3
levels)
2.
Incubation
atmospheres
(
3
levels)
3.
Rat
age
(
3
levels)
4.
Testosterone
RIA
­
verify
claims
of
the
commercial
supplier
Battelle
Report
27
July,
2002
5.
LDH
assay
­
verify
claims
of
the
commercial
supplier.

Each
level
of
these
preliminary
experiments
above,
i.
e.
1,
2,
and
3,
will
be
repeated
three
times
with
true
replicates
(
with
independently
generated
media
and
testis
combinations).
These
replicate
determinations
will
provide
some
initial
indication
of
whether
the
differences
in
responses
at
the
various
factor
levels
represent
real
effects
or
may
just
be
due
to
experimental
variation.

6.4
PRIMARY
EXPERIMENTAL
PHASE
(
PHASE
II)

6.4.1
Optimization
of
Incubation
Conditions
The
incubation
factors
experiment
includes
six
factors,
each
varied
across
three,
four,
or
five
levels.
It
is
assumed
that
these
factors
may
interact
among
one
another
but
will
not
interact
with
the
factors
discussed
in
the
other
parts
of
the
design.
All
factors
not
included
in
this
part
of
the
design
will
be
held
constant
at
their
prototypical
levels.
The
incubation
factors
to
be
studied
and
their
ranges
of
variation
are
as
follows:

Table
5.
Summary
of
Experimental
Incubation
Factors
for
Optimization
Experimental
Levels
Factor
Identification
Units
Factor
Designation
1
2
3
4
5
Incubation
Temperature
0C
X1
25
34*
37
40
­­

Incubation
Vessel
Type
­­
X2
­­
scintillation*
vial
Erlenmeye
r
flask
test
tube
­­

Incubation
Shaker
Speed
­­
X3
­­
none
low
high*
­­

Incubation
Media
Volume
ml
X4
­­
2.5
5*
10
­­

hCG
Concentration
IU/
ml
X5
0.001
0.01
0.1*
1
10
Fragment
Size
mg
X6
25
50
125
250*
500
*
The
central
levels
of
the
factors.

The
principal
portion
of
the
design
will
consist
of
two
26­
1
factorial
parts
in
the
"
central"
levels
of
the
factors
(
i.
e.,
levels
2,3,4
­
except
for
incubation
temperature,
for
which
levels
1,2,
and
4
will
be
run).
The
prototypical
level
of
each
factor,
i.
e.,
indicated
with
an
asterisk,
will
be
included
in
both
parts.
The
other
central
levels
will
be
included
in
just
one.
Thus,
more
information
will
be
obtained
about
the
prototypical
levels
of
the
factors
than
about
perturbations
Battelle
Report
28
July,
2002
about
these
levels.
The
principal
portion
of
the
design
will
permit
estimation
of
linear
main
effects
and
linear
by
linear
two
factor
interactions
for
each
of
the
six
variables.
Quadratic
main
effects
will
be
estimable
under
the
assumption
that
they
do
not
interact
with
other
factors.

In
addition,
a
center
point
will
be
run,
axial
points
will
be
run
for
variables
1,
5,
and
6,
and
9
replicate
points
will
be
spaced
over
the
design
space.
All
variables,
other
than
those
that
are
included
in
the
primary
incubation
factor
experiment
will
be
set
to
the
prototypical
settings.

The
experimental
matrix
is
divided
into
five
parts.

°
Part
1.
26­
1
fractional
factorial
design.
32
runs.
This
includes
the
prototypical
level
of
each
factor
and
one
of
the
central
factor
levels.
This
permits
estimation
of
all
linear
main
effects
and
linear
by
linear
interactions
among
the
levels
of
the
factors
included
in
this
portion
of
the
design.

°
Part
2.
26­
1
fractional
factorial
design.
32
runs.
This
includes
the
prototypical
level
of
each
factor
and
the
central
factor
level
that
was
not
included
in
Part
1.
This
permits
estimation
of
all
linear
main
effects
and
linear
by
linear
interactions
among
the
levels
of
the
factors
included
in
this
portion
of
the
design.
Part
2,
in
conjunction
with
Part
1,
permits
estimation
of
quadratic
main
effects
in
each
of
the
factors,
under
the
assumption
that
the
quadratic
effects
do
not
interact
with
the
other
factors.

°
Part
3.
Center
Point.
This
estimates
the
response
when
all
factors
are
set
to
their
prototypical
levels.
This
does
not
require
any
additional
runs
since
Part
1,
Run
19,
is
already
at
the
center
point.

°
Part
4.
Axial
Points.
5
runs.
These
points
explore
additional
factor
levels
for
variables
1,
5,
and
6.
They
provide
additional
estimates
of
the
nature
and
extent
of
nonlinearity
in
the
main
effects
of
these
variables.

°
Part
5.
Replicates.
9
runs.
These
points
provide
estimates
of
the
reproducibility
of
the
response
at
the
center
point
of
the
design
and
at
various
extremes
of
the
design
space.

The
total
number
of
runs
is
78,
which
will
be
performed
for
both
the
non­
stimulated
and
the
hCG
stimulated
assays,
which
is
a
grand
total
of
156
runs.
The
order
of
the
runs
will
be
randomized.

6.4.2
Optimization
of
Testis
Preparation
Factors
The
testis
preparation
experiment
includes
four
factors,
each
varied
across
three
or
four
levels.
Interactions
of
these
factors
among
one
another
will
be
studied;
however,
interactions
of
these
factors
included
in
other
portions
of
the
design
will
not.
All
factors
not
included
in
this
part
of
the
design
will
be
held
constant
at
their
prototypical
values.
The
testis
preparation
factors
to
be
studied
and
their
ranges
of
variation
are
as
follows:
Battelle
Report
29
July,
2002
Table
6.
Summary
of
Experimental
Testis
Preparation
Factors
for
Optimization
Experimental
Levels
Factor
Identification
Units
Factor
Designation
1
2
3
4
hCG
Concentration
IU/
ml
X1
0.01
0.1*
1
­­

Time
Delay
hr
X2
0.5
1*
2
­­

Organ
Preparation
Techniques
­­
X3
Cold
buffered
saline
Warm
buffered
saline
Cold
media*
Cold
physiological
saline
Sample
Aliquot
Volume
ml
X4
0.1
0.25
0.5*
­­

*
The
central
levels
of
the
factors.

The
design
will
consist
of
four
parts,
with
a
total
of
46
runs.
Both
non­
stimulated
and
hCG
stimulated
will
be
performed
for
a
grand
total
of
92
runs.
The
order
of
the
92
runs
will
be
randomized.

°
Part
1.
34­
1
fractional
factorial
design.
27
runs.
Three
of
the
four
levels
of
Organ
Preparation
Techniques
­
1,
2,
3
­
will
be
included.
This
portion
of
the
design
permits
estimation
of
linear
and
quadratic
main
effects
of
factors
X
1,
X
2,
X
4
independent
of
2
factor
interactions;
contrasts
among
levels
1,
2,
3
of
factor
X
3
independent
of
2
factor
interactions;
and
all
pairwise
interactions
among
factors
X
1,
X
2,
X
3.
Interactions
with
Sample
Aliquot
Volume
cannot
be
estimated.

°
Part
2.
33­
1
fractional
factorial
design.
9
runs.
Part
2
will
be
run
at
level
4
of
Organ
Preparation
Techniques,
which
was
omitted
from
Part
1.
This
portion
of
the
design
permits
estimation
of
the
linear
and
quadratic
main
effects
of
factors
X
1,
X
2,
X
4
at
level
4
of
Organ
Preparation
Techniques,
independent
of
2
factor
interactions.
This,
in
conjunction
with
Part
1
permits
estimation
of
pairwise
interactions
among
factors
X
1,
X
2,
and
all
levels
of
X
3.

°
Part
3.
Center
Point.
1
run.
This
estimates
the
response
when
all
factors
are
set
to
their
"
conventional"
levels.

°
Part
4.
Replicates.
9
runs.
These
points
provide
estimates
of
the
reproducibility
of
the
response
at
the
center
point
of
the
design
and
at
various
extremes
of
the
design
space.

6.4.3
Optimization
of
Sample
Factors
Battelle
Report
30
July,
2002
The
sample
factors
experiment
is
a
stability
test
of
the
constituency
of
the
medium
or
testosterone
in
the
medium.
Stability
is
assessed
as
a
function
of
sample
handling
factors.
The
incubation
medium
used
will
be
that
which
was
found
to
be
best
in
the
preliminary
experiments
(
Phase
I).
It
will
be
spiked
with
testosterone
to
known
concentration(
s),
to
be
specified.
The
testosterone
concentration
at
the
end
of
the
test
run
will
be
compared
to
the
known
spike
level
and
the
response
will
be
the
percentage
recovery.

The
sample
factors
experiment
includes
four
factors,
each
varied
over
multiple
levels.
The
factors
to
be
studied
and
their
ranges
of
values
are
as
follows:

Table
7.
Summary
of
the
Experimental
Sample
Stability
Factors
for
Optimization
Experimental
Levels
Factor
Identification
Units
Factor
Designation
1
2
3
4
Sample
Storage
Container
­­
X1
Siliconized
plastic*
Nonsiliconized
plastic
­­
­­

Sample
Storage
Time
X2
1
day
1
week
1
month*
3
months
Sample
Storage
Temperature
(
1
and/
or
3
months
only)
0C
X3
­­
­­
­
200C
­
70
0
C*

*
The
central
levels
of
the
factors.

Since
sample
storage
time
and
sample
storage
temperature
are
not
fully
crossed,
they
are
expressed
as
a
single
variable
with
six
levels,
namely:

Experimental
Levels
Factor
Identification
Factor
Designation
1
2
3
4
5
6
Sample
Storage
Time/
Temp
X2/
3
1
day
1
week
1
month/
­
200C
1
month/
­
700C*
3
months/
­
200C
3
months/
­
700C
The
design
will
consist
of
a
full
2
X
6
factorial
layout
in
12
runs
and
9
replicate
points,
for
a
total
of
21
runs.
Both
the
non­
stimulated
and
hCG
stimulated
assays
will
be
run
for
a
grand
total
of
42
runs.
The
order
of
the
runs
will
be
randomized.
Battelle
Report
31
July,
2002
6.4.4
Optimization
of
Sampling
Time
In
the
tests
discussed
above
(
Sections
6.4.1
to
6.4.3)
testosterone
concentrations
will
be
determined
for
the
baseline
(
Time
0)
and
at
1,
2,
3,
and
4
hours
after
equilibration.
Tests
will
be
carried
out
at
the
optimum
conditions
to
determine
whether
the
sampling
times
can
be
shortened
or
lengthened.
Shortening
the
sampling
times
will
improve
assay
efficiency.
Lengthening
the
sampling
times
will
permit
application
of
the
assay
to
test
substances
that
require
a
relatively
lengthy
time
period
to
display
their
effects
on
testosterone
production.

Tests
will
be
carried
out
at
the
previously
determined
optimum
conditions.
The
medium
will
be
sampled
for
testosterone
concentration
at
a
succession
of
sampling
times,
from
short
to
long,
e.
g.
0.5,
1,
2,
3,
4,
8,
12,
and
24
hours.
Five
independent
replicate
determinations
will
be
made
at
each
time
point.
The
baseline
concentration
and
the
hCG­
induced
concentration
will
be
determined
at
each
time
point.
Average
concentration
and
coefficient
of
variation
will
be
determined
as
a
function
of
sampling
time.
Of
interest
is
how
long
the
trend
in
testosterone
concentration
remains
linear
and
how
soon
the
coefficient
of
variation
settles
down
to
an
approximately
constant
value.

6.4.5
Sensitivity
Analysis
The
experimental
runs
carried
out
in
Sections
6.4.1
through
6.4.4
will
be
used
to
develop
a
response
surface
model
that
describes
the
variation
in
response
level
as
a
function
of
the
variation
in
the
totality
of
experimental
factors.
Sensitivity
analyses
will
initially
be
carried
out
based
on
predictions
from
the
response
surface
model.
Changes
in
response
from
the
predicted
optimum
will
be
estimated
as
a
function
of
changes
in
experimental
variables
from
their
optimum
values.
The
linear
combination
of
experimental
factors
corresponding
to
the
eigenvector
associated
with
the
minimum,
i.
e.,
the
most
negative,
eigenvalue
of
the
matrix
of
second
partial
derivatives
at
the
maximum
will
be
studied
to
assess
the
effect
on
reaction
sensitivity
of
perturbations
in
experimental
conditions
in
that
direction.

Categorical
variables
will
be
varied
across
their
possible
levels.
Quantitative
variables
will
be
varied
by,
e.
g.,
±
5%,
±
10%,
±
20%,
and
±
30%
from
their
optimum
values.
Those
individual
variables
or
combinations
of
variables
that
have
the
greatest
influence
on
response
will
be
determined.
The
predicted
changes
in
sensitivity
due
to
changes
in
these
variables
will
be
verified
experimentally.
The
variables
that
are
verified
to
in
fact
to
be
the
most
influential
on
assay
efficacy
will
need
to
be
controlled
the
most
tightly.

Those
individual
variables
or
combinations
of
variables
that
have
the
least
influence
on
response
are
also
of
interest.
These
variables
can
be
changed
relatively
large
amounts
without
substantially
affecting
assay
efficacy.
Model
predictions
of
the
least
sensitive
individual
variables
or
combinations
of
variables
will
be
verified
experimentally.
The
variables
that
are
verified
in
fact
to
be
the
least
influential
on
assay
efficacy
may
suggest
ways
to
increase
assay
efficiency.
For
example,
if
sample
storage
at
­
200C
results
in
nearly
as
good
stability
as
sample
storage
at
­
700C,
then
there
is
a
way
to
simplify
sample
storage
without
much
impact
on
results.
Battelle
Report
32
July,
2002
6.4.6
Characterization
of
Viability
The
experiments
described
below
will
be
conducted
after
the
preceding
sections
are
completed,
evaluated,
and
an
"
optimized"
assay
is
identified.
These
experiments
will
be
single
factor
experiments.
Each
level
for
a
given
factor
will
be
run
in
triplicate.

1.
Effect
of
potential
solvents
that
will
be
used
as
carriers
or
vehicles
for
test
substances.
The
solvents
to
be
tested
include
ethanol,
DMSO,
and
tween
20.
In
addition,
each
of
the
solvents
will
be
tested
at
concentrations
of
0.5,
1,
5,
and
10
percent.

2.
Time
needed
to
achieve
parenchymal
equilibration
(
the
minimum
amount
of
time
needed
to
allow
the
parenchyma
to
stop
releasing
testosterone
from
damaged
cells).
The
times
tested
will
include
15,
30,
and
60
minutes.

3.
Use
of
LDH
to
characterize
cellular
damage.
The
assay
will
be
conducted
using
both
testosterone
and
LDH
as
endpoints.
It
will
include
sampling
before
and
after
damage
to
parenchyma.
Damage
will
be
induced
using
trauma
and/
or
heat.

4.
Variability
assessment.
Evaluate
optimized
assay
for
day­
to­
day
variability
(
the
same
technician
conducts
the
assay
on
different
days).

6.5
STATISTICAL
ANALYSIS
Statistical
analysis
will
be
based
on
multiple
regression
analysis.
Preliminary
graphical
displays
will
be
used
to
identify
the
nature
of
the
trends,
the
nature
of
the
response
variability,
and
the
need
for
transformations
of
the
response
or
of
the
primary
experimental
variables.
Full
quadratic
response
surface
models
will
be
fitted
to
the
data.
Residuals
from
the
model
will
be
examined
graphically
and
numerically
to
identify
outlying
observations,
heterogeneity
of
variability,
and
trend
departures
from
model
assumptions.

If
several
individual
factors
or
combinations
of
factors
exhibit
particularly
strong
influence
on
the
response
outcome,
consideration
will
be
given
to
augmenting
the
design
with
supplemental
runs
to
study
the
trends
or
interactions
in
these
directions
in
greater
detail.

The
final
response
surface
model
will
be
optimized
to
determine
the
experimental
conditions
associated
with
the
optimum
response.
A
confidence
region
will
be
constructed
around
the
optimum
by
considering
the
set
of
experimental
factors
whose
associated
responses
do
not
differ
significantly
from
that
at
the
maximum.
The
optimum
condition
may
occur
at
the
interior
of
the
design
space
or
at
a
boundary.
In
the
former
case
the
optimum
and
an
associated
confidence
region
will
be
reported.
In
the
latter
case
it
may
be
possible
to
further
improve
the
efficiency
of
the
reaction
by
extrapolating
outside
the
design
space.
Consideration
will
be
given
to
extending
the
experimental
region
in
the
direction
of
the
(
increasing)
response
gradient
to
determine
the
extent
of
possible
improvement
in
efficiency.
Battelle
Report
33
July,
2002
7.0
REFERENCES
Choi,
M.
S.
K.
and
Cooke,
B.
A.
(
1990).
Evidence
for
two
independent
pathways
in
the
stimulation
of
steroidogenesis
by
luteinizing
hormone
involving
chloride
channels
and
cyclic
AMP.
FEBS
Letters,
261:
402­
404.

Clark,
B.
J.,
Wells,
J.,
King,
S.
R.,
and
Stocco,
D.
M.
(
1994).
The
purification,
cloning,
and
expression
of
a
novel
luteinizing
hormone­
induced
mitochondrial
protein
in
MA­
10
mouse
Leydig
tumor
cells:
Characterization
of
a
steroidogenic
regulatory
protein
(
STAR).
J.
Biol.
Chem.
269(
45),
28314­
28322.

Cooke,
B.
A.
(
1996).
Leydig
cell
structure
and
function
during
aging.
In:
The
Leydig
Cell
(
eds.
Payne,
A.
H.,
Hardy,
M.
P.,
and
Russel,
L.
D.),
Cache
River
Press,
Vienna,
IL.

Darney,
K.
J.,
Jr.,
T­
Y
Wing
and
Ewing,
L.
J.
(
1983).
Simultaneous
measurement
of
four
testicular
 
4­
3­
Ketosteroids
by
isocratic
high­
pressure
liquid
chromatography
with
on­
line
ultraviolet
absorbance
detection,
J.
Chromatography
257:
81­
90.

Davidoff,
M.
S.,
Middendorff,
R.,
Mayer,
B.,
and
Holstein,
A.
F.
(
1995).
Nitric
oxide
synthase
(
NOS­
i)
in
Leydig
cells
of
the
human
testis.
Arch.
Histol.
Cytol.,
58:
17­
30.

Duhl,
S.
N.
and
Jackson,
R.
Y.
(
1978)
In:
Clinical
Chemistry.
P.
828.

EPA
2002.
Draft
Detailed
Review
Paper
on
Steroidogenesis
Screening
Assays
and
Endocrine
Disruptors,
prepared
by
Battelle
for
U.
S.
EPA
Endocrine
Disruptor
Screening
Program,
contract
68­
W­
01­
023,
Work
Assignment
2­
6.

Fail,
P.
A.
and
Gray,
L.
E.,
Jr.
(
1996).
Endocrine
toxicity
of
vinclozolin
in
Long­
Evans
hooded
male
rats:
in
vivo
and
in
vitro.
Presented
at
4th
Biennial
International
Symposium
on
"
Alternatives
in
the
Assessment
of
Toxicity:
Issues,
Progress,
and
Opportunities,"
June
12­
14,
1996,
Aberdeen
Proving
Ground,
MD,
U.
S.
Army,
technical
program
Abstract
14,
p.
26.

Fail,
P.
A.,
Pearce,
S.
W.,
Anderson,
S.
A.,
Tyl,
R.
W.,
and
Gray,
L.
E.,
Jr.
(
1995).
Endocrine
and
reproductive
toxicity
of
vinclozolin
(
VIN)
in
male
Long­
Evans
hooded
rats.
The
Toxicologist
15,
293
(
Abstract
1570).

Fail,
P.
A.,
Pearce,
S.
W.,
Anderson,
S.
A.,
and
Gray,
L.
E.,
Jr.
(
1994).
Methoxychlor
alters
testosterone
and
LH
response
to
human
chorionic
gonadotropin
(
hCG)
or
gonadotropin­
releasing
hormone
(
GNRH)
in
male
Long­
Evans
hooded
rats.
Biology
of
Reproduction
50(
1),
106
(
Abstract
206).

Fail,
P.
A.,
Sauls,
H.
R.,
Pearce,
S.
W.,
Izard,
M.
K.,
and
Anderson,
S.
A.
(
1992).
Measures
of
pituitary
and
testicular
function
evaluated
with
an
endocrine
challenge
test
(
ECT)
in
cannulated
male
rats.
The
Toxicologist
12(
1),
436
(
Abstract
1725).
Battelle
Report
34
July,
2002
FQPA
(
1996).
Food
Quality
Protection
Act
of
1996,
U.
S.
Public
Law
104­
170,
21
U.
S.
C.
46a(
p),
Section
408(
p),
110
STAT.
1489,
August
3,
1996.

Goldman,
Jerome
(
2002)
personal
communication
with
Jerry
D.
Johnson
of
Battelle,
Columbus,
OH.

Gray,
L.
E.,
Klinefelter,
G.,
Kelce,
W.,
Laskey,
J.,
Ostby,
J.,
and
Ewing,
L.
(
1995).
Hamster
Leydig
cells
are
less
sensitive
to
ethane
dimethanesulfonate
when
compared
to
rat
Leydig
cells
both
in
vivo
and
in
vitro.
Toxicol.
Appl.
Pharmacol.
130,
248­
256.

Gürtler,
J.
and
Donatsch,
P.
(
1979).
Effects
of
two
structurally
different
antispermatogenic
compounds
on
the
synthesis
of
steroids
in
rat
testes.
Arch.
Toxicol.
41(
2),
381­
385.

Hafs,
H.
D.,
Niswender,
G.
D.,
Malven,
P.
V.,
Kaltenbach,
C.
C.,
Zimbelman,
R.
G.
,
and
Condon,
R.
J.
(
1977).
Guidelines
for
hormone
radioimmunoassays,
J.
Animal
Science
46(
4):
927­
928.

Hall,
P.
F.,
Osawa,
S.,
and
Mrotek,
J.
(
1981).
The
influence
of
calmodulin
on
steroid
synthesis
in
Leydig
cells
from
rat
testis.
Endocrinology,
109:
1677­
1682.

Jacobs,
D.
S.,
Kasten,
B.
L.,
De
Mott,
W.
R.,
and
Wolfson,
W.
L.,
(
1988)
Laboratory
Test
Handbook,
Lexi­
Comp/
Mosby,
Cleveland,
149.

Kagawa,
N.
and
Waterman,
M.
R.
(
1995).
Regulation
of
steroidogenic
and
related
P450s.
In:
Cytochrome
P450:
Structure,
Mechanism,
and
Biochemistry,
Second
Edition
(
Ortiz
de
Montellano,
P.
R.,
ed),
Plenum
Press,
pp.
419­
442.

Kelce,
W.
R.,
Zirkin,
B.
R.,
and
Ewing,
L.
L.
(
1991).
Immature
rat
Leydig
cells
are
intrinsically
less
sensitive
than
adult
Leydig
cells
to
ethane
dimethanesulfonate.
Toxicol.
Appl.
Pharmacol.
111:
189­
200.

Klinefelter,
G.
R.
Hall,
P.
F.,
and
Ewing,
L.
L.
(
1987).
Effect
of
leuteinizing
hormone
deprivation
in
situ
on
steroidogenesis
of
rat
Leydig
cells
purified
by
a
multistep
procedure.
Biol.
Reprod.
36:
769­
783.

Klinefelter,
G.
R.,
Laskey,
J.
W.,
and
Roberts,
N.
L.
(
1991).
In
vitro/
in
vivo
effects
of
ethane
dimethanesulfonate
on
Leydig
cells
of
adult
rats.
Toxicol.
Appl.
Pharmacol.
107,
460­
471.

Klinefelter,
G.
R.,
Laskey,
J.
W.,
Kelce,
W.
R.,
Ferrell,
J.,
Roberts,
N.
L.,
Suarez,
J.
D.,
and
Slott,
V.
(
1994).
Chloroethylmethanesulfonate­
induced
effects
on
the
epididymis
seem
unrelated
to
altered
Leydig
cell
function.
Biol.
Reprod.
51,
82­
91.

Laskey,
J.
W.,
Klinefelter,
G.
R.,
Kelce,
W.
R.,
and
Ewing,
L.
L.
(
1994).
Effects
of
ethane
dimethanesulfonate
(
EDS)
on
adult
and
immature
rabbit
Leydig
cells:
comparison
with
EDS­
treated
rat
Leydig
cells.
Biol.
Reprod.
50,
1151­
1160.

Naor,
Z.
(
1991).
Is
arachidonic
acid
a
second
messenger
in
signal
transduction.
Mol.
Cell
Battelle
Report
35
July,
2002
Endocrinol.,
80:
C181­
C186.

Powlin,
S.
S.,
Cook,
J.
C.,
Novak,
S.,
and
O'Connor,
J.
C.
(
1998).
Ex
Vivo
and
in
Vitro
Testis
and
Ovary
Explants:
Utility
for
Identifying
Steroid
Biosynthesis
Inhibitors
and
Comparison
to
a
Tier
1
Screening
Battery.
Toxicological
Sciences
46:
61­
74.

Stocco,
D.
M.
(
1999).
Steroidogenic
acute
regulatory
protein.
Vitamins
and
Hormones.
55:
399­
441.

Thoreux­
Manlay,
A.,
Le
Goascogne,
C.,
Segretain,
D.,
Jégou,
B.,
Pinon­
Lataillade,
G.
(
1995).
Lead
affects
steroidogenesis
in
rat
Leydig
cells
in
vivo
and
in
vitro.
Toxicology
103,
53­
62.

Tietz,
N.
W.,
(
1986).
In:
Textbook
of
Clinical
Chemistry,
W.
B.
Saunders,
Philadelphia,
691­
697.

Wilker,
C.
E.,
Welsh,
Jr.,
T.
H.,
Safe,
S.
H.,
Narasimhan,
T.
R.,
and
Johnson,
L.
(
1995).
Human
chorionic
gonadotropin
protects
Leydig
cell
function
against
2,
3,
7,
8­
tetrachlorodibenzo­
p­
dioxin
in
adult
rats:
role
of
Leydig
cell
cytoplasmic
volume.
Toxicology.
95:
93­
102.

Young,
D.
S.,
(
1990).
In:
Effects
of
Drugs
on
Clinical
Laboratory
Tests,
3rd
ed.,
AACC
Press,
Washington,
D.
C.,
3­
211
B
3­
224.
