Final July 2015
A United Stales
Environmental Protection
m m.Agency
Chemical Safety EPA No. 740-C-15-002
and Pollution Prevention July 2015
(7101)
Endocrine Disruptor
Screening Program
Test Guidelines
OCSPP 890.2200:
Medaka Extended One
Generation Reproduction
Test (MEOGRT)
Page 1 of 124
-------
Final July 2015
NOTICE
This guideline is one of a series of test guidelines established by the United States Environmental
Protection Agency's Office of Chemical Safety and Pollution Prevention (OCSPP) for use in testing
pesticides and chemical substances to develop data for submission to the Agency under the Toxic
Substances Control Act (TSCA) (15 U.S.C. 2601, etseq.), the Federal Insecticide, Fungicide and
Rodenticide Act (FIFRA) (7 U.S.C. 136, el seq.), and section 408 of the Federal Food, Drug and
Cosmetic Act (FFDCA) (21 U.S.C. 346a). Prior to April 22, 2010, OCSPP was known as the Office
of Prevention, Pesticides and Toxic Substances (OPPTS). To distinguish these guidelines from
guidelines issued by other organizations, the numbering convention adopted in 1994 specifically
included OPPTS as part of the guideline's number. Any test guidelines developed after April 22,
2010 will use the new acronym (OCSPP) in their title.
The OCSPP test guidelines serve as a compendium of accepted scientific methodologies and
protocols that are intended to provide data to inform regulatory decisions under TSCA, FIFRA and/or
FFDCA. This document provides guidance for conducting the test, and is also used by EPA, the
public and the companies that are subject to data submission requirements under TSCA, FIFRA
and/or the FFDCA. As a guidance document, these guidelines are not binding on either EPA or any
outside parties, and the EPA may depart from the guidelines where circumstances warrant and
without prior notice. At places in this guidance, the Agency uses the word "should." In this
guidance, the use of "should" with regard to an action means that the action is recommended rather
than mandatory. The procedures contained in this guideline are strongly recommended for
generating the data that are the subject of the guideline, but EPA recognizes that departures may be
appropriate in specific situations. You may propose alternatives to the recommendations described in
these guidelines, and the Agency will assess them for appropriateness on a case-by-case basis.
For additional information about these test guidelines and to access these guidelines electronically,
please go to http://www.epa.gov/ocspp and select "Test Methods & Guidelines" on the left side
navigation menu. You may also access the guidelines in http://www. regulations, gov grouped by
Series under Docket ID #s: EPA-HQ-OPPT-2009-0150 through EPA-HQ-OPPT-2009-0159, and
EPA-HQ-OPPT-2009-0576.
Page 2 of 124
-------
Final July 2015
OCSPP 890.2200: Medaka Extended One Generation Reproduction Test
(a) Scope.
(1) Applicability. This guideline is intended to be used to help develop data to submit to
EPA under the Toxic Substances Control Act (TSCA) (15 U.S.C. 2601, et seq.), the Federal
Insecticide, Fungicide, and Rodenticide Act (FIFRA) (7 U.S.C. 136, et seq.), and the Federal
Food, Drug, and Cosmetic Act (FFDCA) (21 U.S.C. 346a).
(2) Background. The Endocrine Disruptor Screening Program (EDSP) reflects a two-
tiered approach to implement the statutory testing requirements of FFDCA section 408(p) (21
U.S.C. 346a). In general, EPA intends to use the data collected under the EDSP, along with other
information, to determine if a pesticide chemical, or other substances, may pose a risk to human
health or the environment due to disruption of the endocrine system.
Tier 2 testing is designed to identify any adverse apical effects caused by the substance which
may be attributable to possible endocrine interaction, and establish a quantitative relationship
between the dose and that adverse effect. The determination whether to require additional Tier 2
testing is made on a weight-of-evidence basis taking into account data from the Tier 1 assays and
other scientifically relevant information available. The fact that a substance may interact with a
hormone system, however, does not necessarily mean that when the substance is used, it will
cause adverse effects in humans or ecological systems. The EPA guidance document "Weight-
of-Evidence: Evaluating Results of EDSP Tier 1 Screening to Identify the Need for Tier 2
Testing" (Ref. 1) explains this process.
The Medaka Extended One Generation Reproduction Test (MEOGRT) is a two-generation test
used to characterize the likelihood, nature, and dose-response relationship of any estrogen-,
androgen-, and thyroid-related effects caused by a chemical on fish. The Japanese medaka
(Oryzias latipes) is the appropriate species for use in this test guideline, given its short life-cycle
and the possibility to determine its genetic sex (Ref. 2), a critical component of this test
guideline. The specific methods and observational endpoints detailed in this guideline are
applicable to Japanese medaka alone.
Chemicals that go through Tier 1 screening and are determined to be bioactive in the estrogen,
androgen, and/or thyroid hormone systems may require additional Tier 2 testing to characterize
adverse outcomes. The fact that a substance may interact with a hormone system, however, does
not necessarily mean that when the substance is used, it will cause adverse effects in humans or
ecological systems. The decision to require the MEOGRT is based on EPA's weight-of-
evidence determination of the Tier 1 screening data and other scientifically relevant information.
Tier 2 tests, such as the MEOGRT, are designed to identify any adverse apical effects which may
be caused by endocrine interaction of substance in the relevant taxonomic group, and to establish
a quantitative relationship between the dose and that adverse effect. Additional endpoints are
also included in the guidelines that may provide information regarding the general Adverse
Outcome Pathway(s) (AOPs) affected by the test substance. This test guideline is part of the
Tier 2 tests included in the OCSPP 890 series for the Endocrine Disruptor Screening Program.
Page 3 of 124
-------
Final July 2015
(3) Source. This OCSPP harmonized test guideline was developed through a process of
harmonization with Test Guideline 240 published by the Organization for Economic Cooperation
and Development (Ref. 3).
(b) Purpose. This guideline is intended for use in developing data, specifically
characterizing any adverse apical effect(s) which may be caused by endocrine interaction of a
chemical in fish and establish a quantitative relationship between the dose and that adverse
effect(s). The tests include exposure during the most sensitive life stages, and provide the
opportunity for identification of dose-response effects.
To fulfill this purpose, the MEOGRT is a longer-term study than the Tier 1 fish short-term
reproduction (FSTRA, OCSPP 890.1350; Ref. 4) assay and is designed to encompass critical life
stages and processes, cover a broad range of test concentrations, and employ a relevant route of
exposure. For the MEOGRT to be conclusive, a discernible cause-effect relationship needs to
exist between exposure to the test substance and an adverse effect.
(c) Introduction. This test protocol is intended to measure chemical effects on
reproduction and reproductive development in Japanese medaka (O. latipes). The test method
begins by exposing adult fish (F0 generation) to the test substance dissolved in water and
continues through development and successful reproduction in the F1 generation. Thus, the
MEOGRT is an extension of existing standard Fish Full Life-Cycle test protocols (Ref. 5; Ref.
6), OCSPP 850.1500 (Fish Life-Cycle Toxicity Test; Ref. 7), and OECD counterparts (Ref. 8-
15).
The MEOGRT provides data that can be used to simultaneously evaluate three general types of
AOPs ending in reproductive impairment: a) those primarily involving disruption of the
hypothalamus-pituitary-gonadal (HPG) endocrine axis; b) those that cause reductions in apical
effects on survival, growth, hatch, etc. through other endocrine axes (e.g., hypothalamus-
pituitary-thyroid); and c) those that cause reductions in apical effects on survival, growth, hatch,
etc. through non-endocrine mediated toxicity pathways. Some of the endocrine-relevant
endpoints, such as the presence of anal fin papillae in medaka males, are biomarkers only
minimally linked to adverse reproductive outcomes; whereas other apical endpoints such as
fecundity and fertility can be directly linked to adverse population outcomes through population
models. The endpoints (e.g. survival and growth) typically measured in the chronic toxicity
tests, the fish full life-cycle test (Ref. 7) and the early life-stage test (OCSPP 850.1400; Ref. 16)
are also included in the MEOGRT. These are used to evaluate the concentration-response of a
chemical working through any endocrine or non-endocrine mediated AOP. The question of
whether a test substance has endocrine-mediated effects at lower concentrations than non-
endocrine-mediated toxicities cannot be unambiguously addressed by the apical endpoints alone
and will rely on other diagnostic biochemical (e.g., vitellogenin induction), histopathological
(e.g., Ley dig cell hyperplasia) effects, or changes in secondary sexual characteristics (e.g., anal
fin papillae count). Definitions of key terms are provided in Appendix 1.
(d) General Experimental Design.
(1) Timeline. Table 1 summarizes the exposure and measurement endpoint timelines for
the MEOGRT. The MEOGRT protocol consists of continuously exposing parts of two
Page 4 of 124
-------
Final July 2015
generations of medaka to the test substance dissolved in treatment tank water. The FO generation
is started by exposing reproductively active fish (at least 12 wpf) for the first 21 days of the test
during which time the test substance and/or its metabolites are presumed to be distributed to the
gametes and tissues of these fish. At the beginning of the fourth week of exposure, eggs are
collected to start the F1 generation and growth measurements are taken of the FO fish. Fish in
the F1 generation are reared for a total of 15 weeks post fertilization (wpf). In addition, a subset
of the F1 fish is sampled at 9 wpf for evaluation of various endpoints. Reproduction is assessed
in adult F1 fish for 21 continuous days during the period of 12 tol4 wpf. Prior to the humane
killing of the F1 adults, eggs are collected and allowed to hatch ending the MEOGRT.
Table 1. Exposure and measurement endpoint timelines within the MEOGRT.
MEOGRT Exposure and Endpoint Timeline
Generation
Fo
1
2
3
4
Life-stage Key
Fi
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
Embryo
F2
1
2
Eleutheroembryo
Study Week
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
Juvenile
Endpoints
Subadult
Fecundity
Fo
Fi
Adult
Fertility
Fo
Fi
Hatching
Fi
F2
7 groups of replicates
• 5 for test substance treatments
• 2 for controls (4 if solvent is
Survival
Fi
Fi
Fi
Growth
Fo
Fi
Fi
used)
Within-group design
• 12 reps for reproduction,
Vitellogenin
Fi
Secondary sex
Fi
Fi
pathology, growth, SSC (wks
10-18)
• 6 reps for hatching, survival,
Histopathology
Fi
Study Week
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
vtg3 growth, (wks 1-9)
(2) Treatments and Replication. This guideline recommends five test substance
treatments and a negative (dilution water) control. The number of replicates per treatment does
not remain constant throughout the MEOGRT, and the number of replicates in the control
treatment is double of any single test substance treatment.
In the FO generation, each test substance treatment has six replicates while the negative control
treatment has 12 replicates. Solvents are highly discouraged, and if used, a justification for both
the use of a solvent and the choice of solvent should be included in the MEOGRT report. If a
solvent is used, two types of controls are necessary: a) a solvent control, and b) a negative
control. These two control groups should each consist of a full complement of replicates at all
points within the MEOGRT timeline (shown in Table 1).
During the first 9 weeks of the F1 generation, this replicate structure remains the same.
However, for the remainder of the F1 generation, the number of breeding pair replicates per
Page 5 of 124
-------
Final July 2015
treatment is optimally doubled to 12 replicate pairs and the number of replicates in the negative
control is doubled to 24 replicate pairs, and another 24 replicate pairs in the solvent control, if a
solvent is used. The determination of hatch from embryos spawned by the F1 pairs is done on
the same replicate structure as was done for the embryos spawned by the FO pairs, meaning
initially six replicates per test substance treatment and 12 replicates in the control group(s).
(3) Endpoints. In the FO generation, daily replicate fecundity (number of spawned
eggs), and daily replicate fertility (number of fertile eggs) are recorded for 21 continuous days.
Upon termination of the FO generation, growth parameters are measured. In addition, hatching
success is recorded for the embryos spawned by FO generation fish. At the adult life-stage of F1
generation (15 wpf), the primary data collected are related to reproduction: daily replicate
fecundity (number of spawned eggs) for 21 continuous days, daily replicate fertility (number of
fertile eggs) for 21 continuous days, and hatching success. After this reproductive assessment,
these F1 adults are sacrificed for growth, secondary sex characteristics, and histopathology
assessment. At the subadult (9 wpf) life-stage of Fl, fish are sampled for growth, secondary sex
characteristics, gonad phenotype, and liver vitellogenin (vitellogenin 1; vtgl) mRNA assessment
or liver vitellogenin protein assessment. Survival data are collected for the Fl generation
offspring at 4, 9, and 15 wpf. Table 1 (noted above) illustrates the timeline for assessing the
endpoints measured in the MEOGRT. The endpoints are summarized in Table 2.
Table 2. Endpoint overview of the MEOGRT*
Life-stage
Endpoint*
Generation
Embryo
(2 wpf)
Hatchability (% and time to hatch)
Fl, F2
Juvenile
(4 wpf)
Survival (hatch to 4 wpf)
Fl
Subadult
Survival (4 to 9 wpf)
Fl
(9 wpf)
Growth
(length and weight)
Vitellogenin
Secondary sex characteristics
(anal fin papillae)
External sex ratio
Time to 1st spawn
Adult
(12-14 wpf)
Reproduction
(fecundity and fertility for 21 days)
FO, Fl
Adult
Survival (from 9 to 15 wpf)
Fl
(15 wpf)
Growth
(length and weight)
Secondary sex characteristics
(anal fin papillae)
Vitellogenin (VTG protein)
Histopathology
(gonad, liver, kidney)
*These endpoints are to be statistically analyzed.
Page 6 of 124
-------
Final July 2015
(4) Genotypic Sex. All endpoints are analyzed in the context of the genetic sex of
the individual fish. Japanese medaka has a XX/XY sex determination system in which the only
functional gene identified on the Y chromosome is the dmy gene (Ref. 17). This presence of the
dmy gene indicates a XY individual regardless of phenotype (Ref. 18; Ref. 19), which may be
altered by exposure to a test chemical. The dmy gene determination for an individual fish
provides a description of the genetic predisposition of the gender phenotype for each fish.
Therefore, the dmy gene status of each fish is essential for proper analysis of biological data.
The dmy gene assessment is done on all fish just before the 9 wpf subadult sampling in order to
properly setup XX - XY breeding pairs. In addition, the DMY information on the individual fish
sampled at 9 wpf is retained to properly analyze the subadult data.
(e) MEOGRT Prologue
(1) Test Species. The test species is Japanese medaka, Oryzias latipes, because of its
short life cycle and the possibility to determine genetic sex. Japanese medaka have been used a
model fish for studying development and reproduction for over 60 years (e.g., Ref. 20). Many
published methods exist for its culture (Ref. 21; Ref. 22; Ref. 23), and data are available from
short-term lethality, early life-stage and full life-cycle tests (Ref. 5; Ref. 6; Ref. 24; Ref. 25;
Ref. 26). All fish are maintained on a 16 hour light: 8 hour dark photoperiod. The fish are fed
live brine shrimp, Artemia spp., nauplii which may be supplemented with a commercially
available flake food if necessary. Commercially available flake food should be regularly
analyzed for contaminants.
As long as appropriate husbandry practices are followed, no specific culturing protocol is
required. For example, medaka can be reared in 2 L tanks with 240 larval fish per tank until 4
wpf, then they can be reared in 2 L tanks with 10 fish per tank until 8 wpf, at which time, they
transition to breeding pairs in 2 L tanks.
The exposure phase should be started with sexually dimorphic adult fish from a laboratory
supply of reproductively mature animals cultured at 25 ± 2°C. The fish should be actively
spawning prior to the start of the test. The age and weight of the fish selected should be typically
more than 12 wpf, but not greater than 16 wpf and have recommended weights of > 300 mg for
females and > 250 mg for males. The range in individual weights, by sex, at the start of the test
should be kept within ± 20% of the arithmetic mean weight of the same sex. A subsample of fish
should be weighed before the test to estimate the mean weight.
(i) Selection of Test Fish. Test fish (F0 generation) should be selected from a single
laboratory stock which has been acclimated for at least two weeks prior to the test under
conditions of water quality and illumination similar to those used in the test. Note: this
acclimation period is not an in situ pre-exposure period. It is recommended that test fish be
obtained from an in-house culture, as shipping of adult fish is stressful and may interfere with
reliable spawning. Fish should be fed brine shrimp twice per day throughout the holding period
and during the exposure phase.
In addition, each breeding pair of F0 should be genetically verified to be XX - XY to avoid the
possible inclusion of spontaneous XX males. Small tissue samples of the tail fin are analyzed for
dmy using the same procedures described during the in-life portion of the MEOGRT. A
Page 7 of 124
-------
Final July 2015
minimum of 42 breeding pairs (54 breeding pairs if a solvent control is used) are considered
necessary to initiate the MEOGRT to ensure adequate replication. In addition, each breeding pair
of FO should be verified to be XX-XY (i.e. normal complement of sex chromosomes in each sex)
to avoid the possible inclusion of spontaneous XX males. It is strongly encouraged that several
more breeding pairs than the 42 pair minimum be available for inclusion in the test. In this way,
if pairs stop spawning prior to the start of the test they can be substituted with actively spawning
pairs. If this recommendation is followed, the MEOGRT should be initiated by selecting 42
breeding pairs from all available breeding pairs. All breeding pairs will have daily fecundity
values measured 4 to 7 days prior to the start of the test. The selected 42 breeding pairs are those
in the middle of the distribution of mean fecundities. The mean daily fecundity of the selected
breeding pairs should be greater than 20 eggs/pair/day. A randomized block design should be
used to ensure a balanced distribution of pairs to the treatments based on reproductive
performance.
(ii) Mortalities in Culture Fish. Mortalities in the culture fish should be recorded and
the following criteria applied following a 48 hour settling-down period:
• Mortalities of greater than 10% of the culture population in seven days preceding transfer
to the test system: reject the entire batch;
• Mortalities of between 5% and 10% of the population in the seven days preceding
transfer to the test system: acclimation for seven additional days to the 2 week
acclimation period; if more than 5% mortality during the second seven days, reject the
entire batch;
• Mortalities of less than 5% of the population in the seven days preceding transfer to the
test system: accept the batch.
Fish should not receive treatment for disease in the two-week acclimation period preceding the
test and during the exposure period, and disease treatment should be completely avoided if
possible. Moribund fish or fish with clinical signs of disease should not be used in the study. A
record of observations and any prophylactic and therapeutic disease treatments during the culture
period preceding the test should be maintained.
(2) Water. Any water in which the test species shows suitable long-term survival and
growth may be used as test water. It should be of constant quality during the period of the test.
In order to ensure that the dilution water will not unduly influence the test result (for example by
complexation of test chemical) or adversely affect the performance of the brood stock, samples
should be taken at intervals for analysis. Measurements of heavy metals (e.g. Cu, Pb, Zn, Hg,
Cd, Ni), major anions and cations (e.g. Ca, Mg, Na, K, CI, S04), pesticides, total organic carbon
and suspended solids should be made, for example, every six months where a dilution water is
known to be relatively constant in quality. The pH of the water should be within the range 6.5 to
8.5, but during a given test it should be within a range of ± 0.5 pH units. The dilution water
alkalinity, hardness, and total organic carbon should be reported. Some chemical characteristics
of acceptable dilution water are listed in Table 3.
Page 8 of 124
-------
Final July 2015
Table 3. Some Chemical Characteristics of Acceptable Dilution Water
Substance
Limit concentration
Particulate matter
5 mg/L
Total organic carbon
2 mg/L
Un-ionized ammonia
1 |ig/L
Residual chlorine
10 |ig/L
Total organophosphorous pesticides
50 ng/L
Total organochlorine pesticides plus polychlorinated biphenyls
50 ng/L
Total organic chlorine
25 ng/L
Aluminum
1 |ig/L
Arsenic
1 |ig/L
Chromium
1 |ig/L
Cobalt
1 |ig/L
Copper
1 |ig/L
Iron
1 |ig/L
Lead
1 |ig/L
Nickel
1 |ig/L
Zinc
1 |ig/L
Cadmium
100 ng/L
Mercury
100 ng/L
Silver
100 ng/L
(3) Exposure System. The design and materials used for the exposure system is not
prescribed, but certain test conditions are essential to the proper implementation of the
MEOGRT. A flow-through diluter system is recommended for this protocol. A flow-through
exposure system is highly recommended to achieve performance criteria. Static renewal systems
are strongly discouraged. The literature describes several different exposure systems used in
medaka testing (e.g., Ref. 6; Ref. 24, Ref. 25; Ref. 27; Ref. 28). For further information on
setting up flow-through exposure systems, please refer to the ASTM Standard Guide for
Conducting Acute Toxicity Tests on Test Materials with Fishes, Macroinvertebrates, and
Amphibians (Ref. 29). Examples of exposure systems are shown in Appendix 2. A summary of
test conditions is presented in Table 4.
(i) Components. Glass, stainless steel, or other chemically inert material should be used
for construction of the test system that has not been contaminated during previous tests. The
system components including aquaria, dilution cells, pumps, and delivery lines, should have
water-contact components of glass, stainless steel, Teflon®, or other inert material. However,
suitable plastics can be utilized if it has been shown that they will not compromise the test.
Exposure tanks should be glass aquaria with a recommended volume of approximately 1.8 L.
Page 9 of 124
-------
Final July 2015
(ii) Water Delivery. The system is required to support all chemical treatments and
treatment controls up to 84 aquaria during reproduction (up to 108 aquaria during reproduction if
a solvent is used). The flow rate to each aquarium should be sufficient to maintain both
biological conditions and chemical exposure. With 1.8 L aquaria, a flow rate of 20 mL/minute
usually is sufficient. Delivery of each treatment to the appropriate aquaria can be accomplished
with different systems including appropriately sized peristaltic pumps or metering pumps with
manifold systems (see Appendix 2 for example systems). A flexible system will have design
options that provide for a wide range of flows that are easily adjusted. This enables the flow to
be adjusted during a study, for instance, if biomass reduces test agent concentrations or lowers
water quality.
(iii) Lighting. Fluorescent lighting should be wide spectrum and should provide a 16
hour light:8 hour dark photoperiod at an intensity range of approximately 150 lux at the water
surface. If additional light is required during certain test tasks (i.e., collection of eggs during
reproduction), increased light intensity may be used, but should be reduced back to the specified
range as soon as possible.
(iv) Temperature. Temperature conditions should be maintained throughout the test
(see Table 3). One option uses low wattage silicon rubber heaters controlled by a digital
temperature controller. This dry system controls tank temperatures as well as water bath
systems, and has the advantage of allowing exposure aquaria to be removed from the system
without water dripping and allows leaking aquaria to be easily detected. Regular recording of
the temperature of the aquaria should be done. If a consistently high performing temperature
regulation system is in place, a minimum of two replicates in each treatment should be measured
daily. The replicates in each treatment should be rotated so that each replicate is measured
approximately the same number of times during the test. If temperature deviations occur, more
frequent measurements of more than two replicates per treatment should be done to adequately
record the temperature profile of the bioassay system.
(v) Curtains. Medaka can react adversely (e.g., reduced fecundity) by activity outside
their aquaria. Elevated light intensity may also lower fecundity. To maximize medaka
reproduction within the MEOGRT, it is recommended that the test aquaria are visually isolated
with curtains (or by other means) from the general laboratory environment.
(4) Test Chemical Dilution System. Due to variations in the physicochemical
properties of test substances, different approaches for preparation of exposure water will be
required. When possible, direct addition to water should be utilized, if the test substance has
sufficiently high water solubility. Test substances which are liquid at room temperature and
moderately soluble in water can be introduced using liquid:liquid saturator methods (Ref. 30).
Test substances which are solid at room temperature and are moderately soluble in water can be
introduced using glass wool column saturators (Ref. 30). Characteristics which indicate that the
test substance may be difficult to test in aquatic systems include: high log octanol-water
partitioning coefficients (log Kow), high volatility, susceptibility to hydrolysis, and susceptibility
to photolysis under ambient laboratory lighting conditions. Other factors may also be relevant to
determining testability and should be determined on a case-by-case basis. For guidance on
testing difficult substances and mixtures in flow-through systems refer to OECD Guidance
Document on Aquatic Toxicity Testing of Difficult Substances and Mixtures (Ref. 31) and for
Page 10 of 124
-------
Final July 2015
conducting aquatic toxicity testing in general, refer to USEPA Guidance Document on Special
Considerations for Conducting Aquatic Laboratory Studies.
All efforts should be made to avoid solvents or carriers. If solvent carriers are used, appropriate
solvent controls should be evaluated in addition to non-solvent (negative) controls (dilution
water only). In the event that use of a solvent is unavoidable, and microbial activity (bio-
filming) occurs, recommend recording/reporting of the bio-filming per tank (at least weekly)
throughout the test. Ideally, the solvent concentration should be kept constant in the solvent
control and all test treatments. If the concentration of solvent is not kept constant, the highest
concentration of solvent in the test treatment should be used in the solvent control. If a solvent is
used, the concentration of the solvent should be kept as low as possible, (e.g., <20 |iL/L) to avoid
potential effect of the solvent on endpoints measured (Ref. 32).
All aqueous stock solutions should be encased in a light proof container during agitation and
subsequent storage to prevent photo-oxidation or excessive microbial growth. The flow of the
test solution should be verified at least weekly and the flow between replicates should not be
greater than 10% from mean. Additionally, the system should be checked at least twice daily to
ensure proper delivery and function.
The precise preparation of test chemical solutions to be delivered to tanks is critical to the
success of a long-term bioassay. There are several design considerations for a test chemical
preparation system. The preparation of individual test chemical concentrations should be
independent from each other to allow the adjustment of individual treatment concentrations, and
the addition or elimination of treatments. The chosen system should be capable of producing a
large range of concentrations with a short residence time in the system allowing for rapid and
continuous dilution and delivery to the exposure aquaria. The system should be designed so that
system failure does not result in an increase in the test substance concentration delivered to the
exposure aquaria. Example dilution and delivery systems are shown in Appendix 2.
(5) Test Substance Concentration Selection. It is recommended to use five chemical
concentrations plus control(s). For the purposes of this test, results from the Tier 1 EDSP
studies, in particular, the fish short-term reproduction assay (Ref. 4; Ref. 33), should be used, as
well as any other pertinent information, to the extent possible, in determining the highest test
concentration so as to avoid concentrations that are overtly toxic. Prior to running the
MEOGRT, a range-finding experiment is recommended.
A range-finding test should be conducted under conditions (water quality, test system, animal
loading) similar to those used for the definitive test. As a suggestion, the range-finding test may
start with young larvae and continue with an exposure for 48 hours. Both the number of
replicates and the number of larvae per replicate can be kept to a minimum (e.g.., 2 replicates per
treatment and 10 larvae per replicate). The number of treatments will vary depending on the
quality of the information regarding the test substance, but minimally there should be three doses
of the test substance plus a control. While ancillary endpoints can be taken (i.e., growth), the
primary data to be collected are for survival. In addition, including some reproducing adult pairs
as another sensitive life stage with egg production as the primary endpoint is also recommended.
If use of a solvent is necessary (and no historical data are available), the range-finding test can be
used to identify suitability of the solvent.
Page 11 of 124
-------
Final July 2015
The results of the range-finding experiment should serve to set the highest test concentration in
the MEOGRT. The data from both these life-stages should be considered and the lower effective
concentration of the two studies used to determine a possible high concentration for the test.
Once the highest test concentration is established, the recommended concentration separation
(spacing factor) between adjacent test concentrations should generally be no less than 2-fold and
no greater than 3.2-fold (half-log). The lowest concentration should be a factor of 10- to 100-
times lower than the highest concentration.
(6) Analytical Measurement of Test Substance. After preparation of the test substance
stock solution, the concentration in the test system should be made using appropriate methods.
Concentrations in exposure tanks should be measured prior to adding fish to verify that target
concentrations are reached. Preceding the initiation of the test exposure it is important to verify
that the test substance dilution and delivery system can provide the desired test substance
concentrations to the exposure aquariums. The concentrations of the test substance should be
measured in the stock bottles (or saturator column elutes), mixing cells, and exposure aquaria.
While these measurements are necessary to set up the exposure system, they will also provide
insight regarding the aqueous stability of the test substance over the duration of the MEOGRT.
Parent chemicals and metabolites of concern should be measured using methods such as
spectrophotometry, gas chromatography-electron capture detection (GC-ECD), gas
chromatography-mass spectroscopy (GC-MS), HPLC (high performance liquid chromatography)
with an appropriate detector, or ion-chromatography with conductivity detection.
A summary of test conditions for the test is provided in Table 4.
Table 4. Summary of Test Conditions for the MEOGRT.
Parameter
Information
Fish species
Oryzias latipes (Japanese medaka); orange-red strain
Test type
Continuous flow-through
Water temperature
The nominal test temperature is 25 °C. The mean temperature
throughout the test in each tank is 24-26°C.
Illumination quality
Fluorescent bulbs (wide spectrum, 100-150 lux,-150 lumens/m2)
Photoperiod
16 h light: 8 h dark
Number of test organisms per
replicate
F0: 2 adults/replicate (aquarium);
Fl: initiated with maximum 20 embryos/replicate, 12 hatched
larvae (eleutheroembryos)/replicate at hatch, then 2 adults (XX-
XY breeding pair) at 9 wpf for reproductive phase.
Example aquarium size
18x9x15 cm
Example aquarium effective
volume
-1.8 liters
Volume exchanges of test
solutions
If using the size aquarium noted above, then use a minimum of 5
volume renewals/day; however, at least 16 volume
replacements/day (20 ml/min flow) is preferred.
Age of test organisms at
initiation
F0: >12 wpf recommended, not to exceed 16 wpf
Page 12 of 124
-------
Final July 2015
Parameter
Information
Number of treatments
Recommended 6 (5 test substance treatment plus 1 negative
control); 7 if a solvent is used
Number of. replicates per
treatment
Minimum 6 replicates/treatment for test substance treatment;
minimum of 12 replicates/control treatment*; replication
structure doubled within reproduction phase (Fl) for fecundity
assessment. See Figure 1.
Number of organisms per test
A minimum of 84 fish in F0 and 504 fish in Fl will be used. If a
solvent control is needed, then a minimum of 108 fish in F0 and
648 fish in Fl will be used. The unit counted is the post
eleutheroembryos.
Aeration
None, unless dissolved oxygen concentration approaches<60%
of air saturation value.
Dilution water
Well water, reconstituted water or dechlorinated tap water
previously demonstrated to be suitable for medaka
Exposure period
19 weeks total (from F0 to F2 hatching)
* If solvent is used, then 12 solvent controls as well as 12 negative controls recommended.
(f) Procedures
(1) Initiation of Test. Breeding pairs that have met the criteria detailed in Section
(e)(l)(i) above should be used to initiate exposure. It is recommended to begin with 12 replicate
breeding pairs in the control treatment. There should be six replicate breeding pairs in each of
the five test substance treatment levels for a total of 42 breeding pairs to start the MEOGRT (54,
if a solvent control is needed). Each breeding pair is randomly assigned a treatment (e.g., T1-T5
and control) and a replicate (e.g., A-L in controls and A-F in treatment), and then placed in the
exposure system with the appropriate flow to each replicate aquarium. While discouraged, if the
use of a solvent has been justified, 12 negative control replicates will receive dilution water only
and an additional solvent control group of 12 replicates will be added to the MEOGRT.
(2) Study Procedures.
(i) Egg Collection and Hatch. Eggs are collected during the first day (or first two days,
if needed) of Study Week 4 for initiating the F1 generation. After spawning, all eggs are
collected by carefully removing any attached eggs from the female and by siphoning any eggs
from the bottom of each aquarium. Filaments are removed from all eggs and the eggs from each
treatment are pooled. Fertile, healthy embryonated eggs are then randomly selected for
incubation. Using a good quality dissecting microscope, one can see hallmarks of early
fertilization/development such as raising of the fertilization membrane (chorion), ongoing cell
division, or formation of the blastula. If available, 120 embryos from each test substance
treatment and 240 from the controls (plus an additional 240 for the solvent control if one is used)
are placed in appropriate incubating devices containing 20 embryos per incubator, (see Appendix
2 for example apparatus) until hatch.
It is preferable that the embryos are collected on a single day; however, if there are not enough
embryos (120 per test substance treatment; 240 per control group), the embryos may be collected
over two days. If collected over two days (Study Days 22 and 23), all embryos within the
treatments that were collected on the first day are pooled with those collected on the second day
Page 13 of 124
-------
Final July 2015
then randomly redistributed to each of the treatment replicates. Each replicate incubator will
have 20 embryos per incubator (120 per test substance treatment; 240 per control group). If a
single treatment requires a second day of collection, all treatments (including controls) need to
follow this procedure. If after the second day of collection there are still inadequate numbers of
embryos within a treatment to load 20 embryos per incubator, reduce the number of embryos
loaded within that specific treatment to 15 embryos per incubator. If there are not enough
embryos to load 15 per incubator, reduce the number of replicate incubators until there are
enough embryos for 15 per incubator.
The incubators should continue to receive an adequate flow of the test substance for the duration
of the incubation period. These incubators can be located either in separate, dedicated aquaria or
in the same aquaria containing the breeding pairs. If the incubators are placed in the same
aquaria that contain the breeding pairs, care should be taken to ensure that the incubators do not
become fouled with food which may lead to embryo mortality.
To reduce mortality, embryos may be agitated within the incubator, for example by aeration,
delivery of the test solution or by moving the egg incubator vertically in the water column.
Incubators should be checked daily for mortalities. Mortalities are recorded and all dead
embryos are removed from the incubators. If agitation is used, the agitation is stopped on the
morning of the first expected day of hatch (e.g., at 25 C, hatching typically begins on the 8 dpf
and continues for about 2 more days).
For each treatment and control upon hatching, the hatched larvae (e.g., eleutheroembryos) are
counted daily, pooled, and distributed to each of the replicate aquaria (Figure 1). This can be
done by randomly selecting an eleutheroembryo from the treatment pool and then sequentially
adding it to a replicate aquarium. Each replicate aquarium should receive 12 eleutheroembryos.
If there are too few eleutheroembryos to fill all replicates, then the number of replicates that are
to be loaded is reduced until 12 eleutheroembryos per replicate can be achieved. Any additional
eleutheroembryos are humanely killed with anaesthetic.
Embryos that have not hatched after a prolonged incubation time are considered non-viable and
discarded. The day this occurs is defined as twice the median day of hatching in controls (e.g., at
a temperature of 25.5 °C, typically this is the 16th day post-fertilization).
(ii) Setup of Breeding Pairs. Determination of genotypic sex via fin clips is done at 9
wpf (i.e., Test Week 12 for F1 generation). All fish within a tank are anesthetized and a small
tissue sample is taken from either the dorsal or the ventral tip of the caudal fin of each fish to
determine the genotypic sex of the individual. See Appendices 3 and 4 for the detailed protocol
on extracting DNA from caudal fins and performing the dmy analysis. It is essential that the
integrity of the sample identification be maintained and contamination between samples be
prevented. The consequence of failure on either of these points is to have genetic sex to
phenotypic sex mismatches that are not related to test chemical exposure, making the
interpretation of the test data unreliable and potentially impacting the validity of the test. Once
the DNA is extracted, a polymerase chain reaction (PCR) method is used to amplify the dmy
gene, if present. As discussed in section (d), General Experimental Design, the presence of the
dmy gene is a definitive indication of an XY (male) individual, regardless of phenotype.
Page 14 of 124
-------
Final July 2015
The dmy gene information is used to establish XX - XY breeding pairs (in both FO and F1
generation adults) regardless of external phenotype which may be altered by exposure to an
endocrine-disrupting chemical. As a suggested procedure, the fish from a replicate can be
housed in small cages, if possible one per cage, in the replicate tank. Alternatively, two fish can
be held in each cage if they are distinguishable from each other. One method is to differentially
cut the caudal fin (e.g., dorsal vs ventral tip) when taking the tissue sample for dmy gene
analysis. Soon after the genotypic sex of each fish is determined, two XX fish and two XY fish
from each replicate are randomly selected for pooling (Figure 1). After the 12 total XX fish (24
for control(s)) and the 12 total XY fish (24 for control(s)) for each treatment are pooled, fish are
randomly selected to produce breeding pairs. If a replicate does not have either two XX or two
XY fish, appropriate fish should be obtained from other replicates within the treatment. The
priority is to have the recommended number of replicate breeding pairs in each treatment group
and in each control group.
Please note the following: a) pooling and re-distribution truncates the replicate lineage across
life-stages and generations preventing direct intra-replicate statistical comparisons, b) randomly
selected XX - XY breeding pairs are essential to the validity of the test, and c) fish with obvious
abnormalities (swim bladder problems, spinal deformities, extreme size variations, etc.) should
be precluded when establishing breeding pairs.
(iii) Sampling of Subadults (9 wpf). After the breeding pairs have been established, the
fish not selected as breeding pairs are humanely killed for measurement of subadult endpoints.
Unlike the fish selected to become breeding pairs, the replicate identity of each sampled subadult
fish is maintained. Also note that it is essential that the genotypic sex of each fish, that has
already been determined, is maintained since all endpoint data are analyzed in the context of the
genotypic sex of the specific fish.
Each fish is euthanized and measured for a variety of endpoints including: 9 wpf survival,
growth, liver vitellogenin (mRNA copy number or protein concentration), and anal fin papillae
number (secondary sexual characters (SSC)). The liver is dissected for quantification of
vitellogenin. If vitellogenin mRNA is quantified, the RNA extracted, and the copy number of
the vitellogenin I gene per ng of total mRNA is determined. The tail of the fish, including the
anal fin, is preserved in an appropriate fixative or photographed so that anal fin papillae may be
counted at a later date. It is strongly recommended that at this time another tissue sample from
each fish be collected and archived. This allows a post-hoc verification of the earlier dmy
analysis in the event that genotype and phenotypic endpoints are non-concordant. The following
appendices should be consulted for important guidance: Appendix 6 for the necropsy procedure,
Appendix 7 for counting anal fin papillae, and Appendix 8 and 9 for example protocols for the
extraction of RNA and the analysis of vitellogenin.
For the purposes of calculating a simple sex ratio, medaka with more than one anal fin papillary
processes are defined as a phenotypic males and those with no anal fin papillae are defined as
phenotypic females.
(iv) Assessment of Reproduction. Fecundity and fertility are assessed in Study Weeks
1 through 3 in the FO generation and in Study Weeks 15 through 17 in the F1 generation. FO and
F1 eggs are collected daily from each breeding pair for 21 consecutive days. Eggs are gently
Page 15 of 124
-------
Final July 2015
removed from netted females and siphoned from the bottom of the aquarium each morning.
Both fecundity and fertility are recorded daily for each replicate breeding pair. Fecundity is
defined as the number of eggs spawned, and fertility is functionally defined as the number of
fertile and viable eggs at the time of counting. Counting should be done as soon as possible after
egg collection.
For the most part, the process for the collection of eggs and hatch assessment is done in the same
manner for the FO and F1 generations; however, the replication per treatment and per control(s)
is different. In the F1 generation, eggs are collected from 12 replicate breeding pairs per
treatment (24 replicates in controls; 48 if a solvent control is needed); whereas, in the FO
generation, there are six replicate breeding pairs per treatment (12 replicates in controls; 24 if a
solvent control is needed). To assess embryo viability and hatch success, all viable embryos are
pooled and systematically distributed to six replicate incubators (12 incubators for controls; 24 if
a solvent control is needed) as was done in the FO generation. If needed, a second day of
collection can be used as long as the first and second collection days are pooled as in the FO
generation. Proceed as described previously in section (f)(2), and record the hatch per replicate
incubator so that the percent hatch can be calculated. Embryos that have not hatched by twice
the median control day of hatch are considered non-viable and should be discarded appropriately.
(v) Sampling of Adults. The adults are humanely killed and various endpoints are
assessed at 15 wpf (i.e., following Test Week 17). While the tail is removed and fixed or
photographed to assess the number of anal fin papillae (see guidance in Appendix 7), no other
tissue dissection is performed to maintain all tissues and organs in their in situ orientations. The
body cavity is opened to allow perfusion with appropriate fixative prior to submersing the entire
body in the fixative and each adult fish is evaluated histologically for pathology in the gonads,
kidneys and liver tissue as described in the Histopathology Guidance (Appendix 12).
Mechanistic endpoints evaluated in this assay (e.g., vitellogenin, SSCs and certain
histopathology effects) may be influenced by systemic or other toxicities. Consequently, liver
and kidney histopathology is assessed in detail to help better understand any responses in
mechanistic endpoints. "Reading down" from the highest treatment group (compared to the
control) to a treatment with no effect may be considered; however, it is recommended that user
consult the histopathology guidance in Appendix 12. The gonad phenotype is also derived from
this evaluation.
It is recommended that a tissue sample be taken to repeat the dmy analysis to verify the genetic
sex of specific fish when necessary. For instance, it is strongly recommended that if a breeding
pair fails to produce more than a minimal number of eggs, the genetic sex of both fish be re-
assessed to verify that the pair was indeed XX - XY.
(vi) Ongoing Observations. During the test, observations of behavior should be made at
least once daily, and any unusual behavior should be noted. In addition, any mortality should be
recorded and survival to 4 wpf (Test Day 44), from 4 wpf to 9 wpf (Test Day 80), and from 9
wpf to 15 wpf (Test Day 121) should be calculated (see Table 1 and Table 5). In addition,
during the time leading up to the selection of breeding pairs, each replicate should be monitored
for its first spawn. The study day that this occurs should be recorded, but statistical analysis of
this data is not performed.
Page 16 of 124
-------
Final July 2015
(3) Detailed Timeline. A timeline for the MEOGRT was described previously in
Table 1. The MEOGRT includes 4 weeks of exposure to FO adults and 15 weeks of exposure to
the F1 generation. In week 4 on approximately test day 24, the F1 generation is established and
the FO breeding pairs are humanely killed and weight and length are recorded (Table 1). This is
followed by exposure of the F1 generation for 14 more weeks (total of 15 weeks for Fl) and the
F2 generation for two weeks until hatching. The exposure continues through development and
reproduction in the Fl and hatching in the F2 generation. The total duration of the test is 19
weeks.
An example of a day-by-day schedule for conducting the test is provided as guidance in
Appendix 3 and as a Microsoft® Excel spreadsheet as a separate electronic file on the EPA
EDSP website: http://www.epa.gov/endo/. It is strongly recommended that this daily protocol
be continuously consulted before starting and during the implementation of the MEOGRT as it
details the activities performed during each day of the test.
A detailed week-by-week timeline for the test is presented below in Table 5.
Table 5. Detailed Weekly Timeline for the MEOGRT.
Study Weeks 1-3 (FO)
The FO generation spawning fish, that have met the selection criteria detailed above, are
exposed for three weeks to allow the developing gametes and gonadal tissues to be exposed to
the test substance. Each replicate aquarium houses a single XX - XY breeding pair of fish
There are a total of 42 breeding pairs (54 breeding pairs if a solvent control is needed).
Spawned eggs are collected, counted and assessed for fertility for 21 consecutive days starting
at Study Day 1.
Study Week 4 (FO and Fl)
On Study Day 22, eggs are collected from each aquarium and each female, pooled by
treatment, and systematically distributed to suitable incubation vessels (6 per treatment and 12
for controls) as detailed in (f)-(2)-(iv). Again, the vessels may be placed in separate
"incubator aquaria" set up for each treatment or in the replicate aquarium that upon hatch will
contain the eleutheroembryos.
Ideally, all embryos used to start Fl should be collected on the same day; however, a second
day of collection may be necessary if an insufficient number of viable embryos are collected
on the first day. If a second day of collection (Study Day 23) is needed, all embryos from both
days should be pooled and then systematically redistributed to each of the treatment replicates
at 20 embryos per incubator. Mortalities are recorded daily.
Note: If a single treatment requires a second day of collection, all treatments (including
controls) need to follow this procedure. If after the second day of collection there are
inadequate numbers of embryos within a treatment to load 20 embryos per incubator, then
reduce the number of embryos loaded within that specific treatment to 15 embryos per
incubator. If there are not enough embryos to load 15 per incubator, then reduce the number of
replicate incubators until there are enough embryos for 15 per incubator. Additionally, more
breeding pairs per treatment and controls could be added in FO to produce more eggs to reach
the recommended 20 per replicate.
Page 17 of 124
-------
Final July 2015
On Test Day 24, the FO breeding pairs are humanely killed, and weight and length are
recorded. If necessary FO breeding pairs maybe kept for an additional 1-2 days in order to
restart Fl.
Study Weeks 5-6 (Fl)
One day before the anticipated start of hatching, stop or reduce the agitation of the incubating
embryos to expedite hatching. As embryos hatch on each day, eleutheroembryos are pooled by
treatment and then systematically distributed to each replicate aquarium within a specific
treatment with no more than 12 eleutheroembryos placed into each aquarium. This is done by
randomly selecting eleutheroembryos and placing a single eleutheroembryo in successive
replicates in an indiscriminate draw, moving in order through the specific treatment replicates
until all replicates within the treatment have 12 eleutheroembryos. If there are not enough
eleutheroembryos to fill all replicates then ensure as many replicates as possible have 12
eleutheroembryos to start the Fl phase of the test. The appropriate data are collected to
calculate the hatching success of each replicate incubator. As detailed previously, embryos
that have not hatched by twice the median control day of hatch) are considered non-viable and
appropriately discarded.
The number of eleutheroembryos are recorded and hatching success is calculated in each
replicate.
Study Weeks 6-11 (Fl)
On Study Day 44, survival of the juvenile fish to this point is recorded as the number per
replicate out of the initial number of eleutheroembryos, nominally 12. The exposure continues
as the juvenile fish develop into subadults.
Study Week 12. (Fl)
On Study Day 78, a small tissue sample is taken from the caudal fin of each fish to determine
the genotypic sex of the individual by dmy analysis. This information is used to establish XX -
XY breeding pairs. The dmy genotpyic data for all the remaining subadult samples is retained
to ensure that all endpoint data can be related to the genetic sex of each individual fish.
On Study Days 80 and 81 (within three days), after the genotypic sex of each fish is
determined, 12 breeding pairs per treatment and 24 breeding pairs for controls are randomly
established as detailed in (f)-(2)-(ii).
The remaining fish (maximum 8 fish per replicate) are humanely killed and are sampled for
the various subadult endpoints as described in (f)-(2)-(vi). It is essential that the dmy gene
status (XX or XY) for all the subadult samples are retained to ensure that all endpoint data can
be related to the genetic sex See Appendices 3 and 4 for detailed protocols on DNA extraction
and dmy gene analysis.
Study Weeks 13-14. (Fl)
The exposure continues as the subadult breeding pairs develop into adults.
Study Weeks 15-17. (Fl)
On Study Day 98, eggs are removed from both the aquaria and the females to guarantee an
accurate enumeration of egg numbers spawned the next day. Spawned eggs are collected daily
for 21 consecutive days (Study Days 99-119) in each replicate, and assessed for fecundity and
fertility.
Study Week 18.
On Study Day 120, eggs are collected, pooled from each treatment, and systematically
distributed to suitable incubation vessels (6 per treatment and 12 for controls; plus an
additional 12 controls if a solvent control is used) with no more than 20 embryos per replicate
Page 18 of 124
-------
Final July 2015
(120 embryos for each test substance treatment and 240 embryos for the controls; plus an
additional 240 for the solvent control if one is used) allowed to incubate. If a second day of
embryo collection (Study Day 121) is needed, all embryos from the first day (Study Day 120)
should be pooled with the embryos from the second day, and then systematically redistributed
to each of the treatment replicates. This procedure is detailed in (f)-(2)-(iv).
After embryo collection, the F1 breeding pairs are terminated humanely killed and analyzed
for the various adult endpoints as described in (f)-(2)-(vi). If necessary F1 breeding pairs
maybe kept for an additional 1-2 days in order to restart F2.
Study Week 19. (F2)
One day before the anticipated start of hatching, stop or reduce the agitation of the incubating
eggs to expedite hatching. As embryos hatch on each day, eleutheroembryos are pooled by
treatment and then systematically distributed to each replicate aquarium within a specific
treatment with no more than 12 eleutheroembryos placed into each aquarium. This is done by
randomly selecting eleutheroembryos and placing a single eleutheroembryo in successive
replicates in an indiscriminate draw, moving in order through the specific treatment replicates
until all replicates within the treatment have 12 eleutheroembryos. If there are not enough
eleutheroembryos to fill all replicates then ensure as many replicates as possible have 12
eleutheroembryos to start the F1 phase of the test. The appropriate data are collected to
calculate the hatching success of each replicate incubator. As detailed previously, embryos
that have not hatched by twice the median control day of hatch) are considered non-viable and
appropriately discarded.
Page 19 of 124
-------
Final July 2015
Endpoints
hatch
survival
growth, VTG, SSC
reproduction, pathology, SSC
hatch
subadult
'samples
DMY
assessed
subadult
'samples
DMY
assessed
subadult
'samples
DMY
assessed
pool
eleuthero-
embryos
pool
viable e
pool
viable e;
subadult
'samples
distribute
distribute
DMY
assessed
distribute
subadult
•samples
DMY
assessed
subadult
•samples
DMY
assessed
fish
XX
XY
XX
XY
20
embryos
20
embryos
XX
XY
fish
20
embryos
20
embryos
XX
XY
embryos
fish
20
embryos
XX
XY
XX
XY
20
embryos
20
embryos
embryos
fish
fish
fish
embryos
20
embryos
XX
XY
XX
XY
XX
XY
XX
XY
XX
XY
imbryos
XX
XY
XX
XY
XX
XY
XX
XY
XX
XY
XX
XY
XX
XY
xx
XY
Figure 1. Pooling strategy used during the MEOGRT. The figure represents one treatment or V2 of a
control. Notice that at each pooling, the replicate lineage is re-established so that a specific replicate
identity is not continuous through time.
(3) Feeding Schedule. Fish can be fed brine shrimp Artemici spp., supplemented with a
commercially available flake food if necessary. Food should be regularly analysed for
contaminants such as organochlorine pesticides, polycyclic aromatic hydrocarbons (PAHs),
polychlorinated biphenyls (PCBs). Food with an elevated level of endocrine active substances
(i.e., phytoestrogens) that could compromise the response of the test should be avoided.The
following feeding schedule is recommended to ensure adequate growth and development to
support robust reproduction. Deviations from this feeding schedule may be acceptable, but they
should be tested to verify that acceptable growth and reproduction are observed. In order to
follow the suggested feeding schedule, the dry weight of brine shrimp per volume of brine
shrimp slurry needs to be determined prior to starting the test. This can be done by weighing a
defined volume of brine shrimp slurry that has been dried for 24 hours at 60°C on pre-weighed
pans. To account for the weight of the salts in the slurry, an identical volume of the same salt
solution used in the slurry should be dried, weighed, and subtracted from the dried brine shrimp
slurry weight. Alternatively, the brine shrimp can be filtered and rinsed with distilled water
before drying. Thereby eliminating the need to measure the weight of a "salt blank". This
information is used to convert the information in Table 6 from dry weight of brine shrimp to
Page 20 of 124
-------
Final July 2015
volume of brine shrimp slurry to be fed per fish. In addition, weekly aliquots of the brine shrimp
slurry should be weighed to verify the correct dry weight of brine shrimp is being fed.
Table 6. Example Feeding Schedule.
Time
(post-hatch)
Brine Shrimp
(mg dry weight/fish/day)
Time
(post-hatch)
Brine Shrimp
(mg dry weight/fish/day)
Day 1
0.5
Day 12
4.2
Day 2
0.5
Day 13
4.5
Day 3
0.6
Day 14
4.8
Day 4
0.7
Day 15
5.2
Day 5
0.8
Day 16-21
5.6
Day 6
1.0
Week 4
7.7
Day 7
1.3
Week 5
9.0
Day 8
1.7
Week 6
11.0
Day 9
2.2
Week 7
13.5
Day 10
2.8
Week 8-sacrifice
22.5
Day 11
3.5
—
—
(4) Analytical Chemistry Schedule. After the exposure is initiated, it is important to
monitor the concentration of the test substance in the exposure aquariums frequently enough to
allow accurate characterization of the treatment concentrations, and to determine whether the
dilution/exposure system is performing as desired. If it is not, then the appropriate adjustments
can be made to mitigate excursions of the test substance concentrations in each treatment as
necessary.
Weekly sampling of at least two replicate aquaria from each treatment is usually sufficient to
define the concentration of each treatment through time. One of the samples should be taken
from the same treatment replicate at each sampling time and the other sample should be selected
systematically from one the other possible treatment replicates. This sampling strategy will
provide estimates of test substance concentration in each treatment through time as well as
estimates of the within-treatment concentration variance.
It is also important to evaluate other possible sources of variance in the analytical chemistry
process by using appropriate measurement replication and chemical standards. For test
substances that appear to be unstable or are subject to biological degradation in the
dilution/exposure system, it may be necessary to measure treatments more than once per week.
(5) System Cleaning Schedule. The chemical dilution and exposure systems used for
the MEOGRT are in contact with water and diluted chemical for a relatively long period.
Biofilms invariably form on most wetted surfaces. The microbiological community that
establishes itself will often be able to metabolize the compound being assessed. Often,
detectable decreases in test substance concentration can be observed within three weeks of the
start of the assay. Routine cleaning and disinfection of all wetted components in the
diluter/delivery system, and exposure aquaria is recommended approximately every three weeks.
Page 21 of 124
-------
Final July 2015
A suggested cleaning protocol would be to divert the exposure delivery lines from each replicate
aquarium. The replicate aquaria can usually be left static without flow for the necessary time.
The system components that can be removed should be cleaned with detergent and scrubbed with
an abrasive pad. The delivery lines should be sanitized using a cold sterilant product (e.g.,
Minncare® Cold Sterilant).
As a precautionary note, diluter systems are designed to the particular needs of each user. The
materials of construction may not be compatible with a specific sterilant. Before disinfecting
any dilution system, an evaluation as to whether the system is compatible with the sterilant
solution used needs to be done. Analysis of chemical concentrations should be performed both
before and after any cleaning of the exposure system.
(6) Environmental Conditions. Water quality characteristics should be measured
regularly across replicates and treatments to document adequate water quality to sustain healthy
medaka. Dissolved oxygen, and pH should be measured in two replicates in each treatment at
least once a week. Temperature should be monitored continuously and recorded daily. The
measured replicates should be rotated within each treatment so that all replicates are measured
roughly the same number of times over the test. Ammonia should be measured monthly in each
treatment. The frequency of measurement should increase if the specific water quality parameter
is approaching biologically significant levels: dissolved oxygen <5.0 mg/L, and un-ionized
ammonia >35 |ig/L. The dilution water alkalinity, hardness, and total organic carbon should be
documented at least once during the test.
(7) Randomization. At all stages and in most, if not all tasks, it is important to include
randomization. This includes, but is not necessarily limited to, the following procedures:
• The order in which food is delivered to the aquaria should be randomized to mitigate
potential changes in food quantity or quality during the course of feeding.
• It is essential to randomize the selection of fish when distributing them into the replicates
as eleutheroembryos, and designating them to either become part of breeding pairs or be
sampled as subadults.
• When fish are humanely killed either as adults or subadults, the order in which the
replicate aquaria are sampled should be random.
(g) Performance Criteria and Test Acceptability/Validity.
(1) Performance Criteria. Failure to meet a single performance criterion, while a
warning sign, would in general not be expected to compromise the performance of the entire test.
However, failure in several criteria or failure to meet the fecundity criterion could result in the
rejection of the test. When deviations from the test validity criteria are observed, the
consequences should be considered in relation to the reliability of the test results, and these
deviations and considerations should be included in the test report. The performance criteria are
generally expected for an acceptable test are summarized in Table 7.
Page 22 of 124
-------
Final July 2015
Table 7. Performance criteria for the MEOGRT.
Parameter
Criteria
Test concentrations
Maintain concentrations of the test chemical in solution
within ±20% of the mean measured values over the entire
test period.
Dissolved oxygen
concentration
> 60% of air saturation value throughout the test
pH
pH should be maintained between 6.5-8.5. The inter-
replicate/inter-treatment differentials should not exceed 0.5.
Mean water temperature
Between 24 and 26°C over the duration of study. Brief
excursions from the mean by individual aquaria should not
be more than 2°C.
Replicates within a treatment should not be statistically
different from each other.
Treatments within the test should not be statistically
different from each other.
Mortality in controls
< 20%) mortality in each replicate in the controls.
Hatchability of Eggs
> 80% in the controls (in each of the F1 and F2
generations).
Survival after hatching
Should be > 80%> (average) until 3 wpf in F1 controls.
Should be from > 90%> (average) from 3 wpf through
termination in F1 controls. Mortalities due to technical
errors (handling) should not be included in these analyses.
Fish weight at subadult
sampling
F1 control XX and XY fish mean weight should be >100
mg and mean length >20mm.
Growth of substantially less than these values suggests the
fish will not achieve sufficient size to spawn with
acceptable fecundity.
Occurrence of intersex fish
XX males should be <5% and XY intersex should be <2%
(determined by comparison of genotype and histological
sex) in F1 controls.
Fecundity
Should be >20 eggs/pair/day in F0 and F1 controls.
For adequate statistical power, 16 of 24 control pairs (in Fl)
should produce > 20 eggs/pair/day.
Fertility
Should be >80%> for all eggs produced in F0 and Fl controls
(i.e., Fl and F2 eggs) during the assessment.
(2) Test Validity. Achieving the following goals will likely deem a test to be considered
acceptable/valid:
• At least three treatment levels with three uncompromised replicates are available for
analysis. Excessive mortality, which compromises a treatment, is defined as >20%
mortality in 2 or more replicates that cannot be explained by technical error.
Page 23 of 124
-------
Final July 2015
• There should be sufficient reproduction in at least the third highest group and all lower
groups of the FO to fill the hatching incubators.
• There should be adequate survival in the third highest test exposure and lower exposure
groups in F1 to allow endpoint evaluation at the subadult sampling.
• Survival in the second highest exposure group of the F1 should be >20%.
• Note signs of overt toxicity. Signs of overt toxicity may include, but are not limited to,
floating on the surface, lying on the bottom of the tank, inverted or irregular swimming,
and being nonresponsive to stimuli.
• A NOEC and LOEC should be determined.
(h) Data Analysis. To expedite data reporting and statistical analysis, it is recommended
that common templates for data reporting be used. These templates take the form of Excel
spreadsheets that can easily be converted into the CSV file format which is required by the
Agency provided statistical tool. Example formats are provided in Appendix 11 and electronic
versions of all templates can be accessed on the EPA EDSP website, http://www.epa.gov/endo.
(1) Statistical Analysis. The types of biological data generated in the test are not unique
to it and except for pathology data, many appropriate statistical methodologies have been
developed to properly analyze similar data depending on the characteristics of the data including
normality, variance homogeneity, whether the study design lends itself to hypothesis testing or
regression analysis, parametric versus non-parametric tests, etc.
Data for continuous endpoints should first be checked for monotonicity by rank transforming the
data, fitting to an ANOVA model and comparing linear and quadratic contrasts. Additionally, the
issue of using a one-tailed statistical test versus a two-tailed statistical test should be considered.
Unless there is a biological reasoning that would make a one-tailed test inappropriate, it is
suggested that one-tailed tests be used. While certain statistical tests are recommended, if more
appropriate and/or powerful statistical methods are developed for application to the specific data
generated in the test, those statistical tests could be used to leverage those advantages.
The test data should be analysed separately for each genotypic sex. Failure to do this will greatly
reduce the statistical power of any analysis.
(i) Histopathology data. Histopathology data are reported as severity scores which are
evaluated using a newly developed statistical procedure, the Rao-Scott Cochran-Armitage by
Slices (RSCABS; Ref. 35). RSCABS uses a step-down Rao-Scott adjusted Cochran-Armitage
trend test on each level of severity in a histopathology response. The Rao-Scott adjustment
retains test-replication information; the by-slices procedure incorporates the biological
expectation that severity of effect tends to increase with increasing doses or concentrations,
while retaining the individual subject scores and revealing the severity of any effect found. The
RSCABS procedure not only determines which treatments are statistically different {i.e., have
higher prevalence of pathology than controls), but it also determines at which severity score the
difference occurs providing much needed context to the analysis.
"Reading down" from the highest treatment group (compared to the control) to a treatment with
no effect may be considered; however, it is recommended that user consult the histopathology
guidance in Appendix 12. Typically all samples are processed/sectioned after which are read by
Page 24 of 124
-------
Final July 2015
the pathologist. If using a "read-down" approach, it is noted that the Rao-Scott Cochrane-
Armitage by Slices (RSCABS) procedure uses the expectation that as dose levels increase the
biological impact (the pathology) will increase as well. Therefore, one will lose power if only
looking at a single high dose without any intermediate doses. If statistical analysis is not
necessary to determine that the high dose has no effect, then this approach may be acceptable.
(ii) Time to event data. Time to hatch and time to first spawn should be treated as time
to event data, with individual embryos not hatching in the defined period or replicates never
spawning treated as right-censored data. Time to hatch should be calculated from the median
day of hatch of each replicate. These endpoints should be analyzed using a mixed-effects Cox
proportional hazard model.
(iii) Fecundity data. The preferred analyses of fecundity examines the overall impact on
fecundity for the 21 day observation period. The raw data are recorded and presented in the
study report as the fecundity (number of eggs) per replicate for each day. The replicate mean of
the raw data should be calculated then a square root transformation applied. A one-way
ANOVA on the transformed replicate means should be calculated followed by Dunnett contrasts.
Alternatively, analyses for fecundity data consist of a step-down Jonckheere-Terpstra or
Williams' test to determine treatment effects, provided the data are consistent with a monotone
concentration-response. With a step-down test, all comparisons are done at the 0.05 significance
level and no adjustment for the number of comparisons made. The data are expected to be
consistent with a monotone concentration response, which can be verified by constructing linear
and quadratic contrasts of treatment means after a rank-order transform of the data. Unless the
quadratic contrast is significant and the linear contrast is not significant, the trend test is done.
Otherwise, Dunnett contrast is used to determine treatment effects if the data are normally
distributed with homogeneous variances. If those requirements are not met, then Dunn's test with
a Bonferonni-Holm adjustment is used. All indicated tests are done independently of any overall
F- or Kruskal-Wallis test. Further details are provided in OECD 2006.
If an understanding of time-by-treatment effects are desired to augment the primary fecundity
statistical results described above, the ANOVA model is given by Y= Treatment + Time +
Time*Treatment, with random effects of Replicate (Treatment), and Time*Replicate
(Treatment). Here Time refers to the frequency of egg counts (e.g., Day or Week). This is a
repeated measures analysis, with the correlations between observations on the same replicates
accounting for the repeated measures nature of the data.
Main effects of treatment are tested using the Dunnett (or Dunnett-Hsu) contrasts, which adjusts
for the number of comparisons. Adjustments for the main effect of time are not needed, for with
this factor there is no "control" level and every pair of levels is a comparison of possible interest.
For this main effect, if the F-test for the main effect is significant at the 0.05 level, then the
pairwise comparisons across levels of that factor can then be tested at the 0.05 level without
further adjustment.
The model includes two-factor interactions, so that a main effect for say time, may not be
significant even though time has a significant impact on results. Thus, if a two-factor interaction
Page 25 of 124
-------
Final July 2015
involving time is significant at the 0.05 level, then one can accept the comparisons of levels of
time at the 0.05 significance level without further adjustment.
Next are F-tests for significance of treatment within time, the so-called slices in the ANOVA
table. If, for example, the slice for treatment and time 12, is significant at the 0.05 level, then the
pairwise comparisons for treatment and time 12 can be accepted at the 0.05 level without further
adjustment.
Finally, for comparisons not falling under any of the above categories, comparisons should be
adjusted using the Bonferroni-Holm adjustment to p-values. Further information on analyses of
such models can be found in (Ref. 37) and (Ref. 38).
(iv) All other biological data. The first formal check of the data is to test whether it
violates the assumptions of monotonicity by using linear and quadratic contrasts. If the quadratic
contrast is significant and the linear contrast is not, the data are considered non-monotonic. If the
data are monotonic, a Jonckheere-Terpstra on replicate medians trend test (as advised in Ref. 36)
is recommended.
If the data are non-monotonic, in particular because of the reduced response of the highest one or
two treatments, consideration should be given to censoring the dataset so that the analysis is done
without those treatments. This decision will need to be made with professional judgment and all
available data, especially data that indicates overt toxicity at those treatment levels.
For weight and length, no transforms are recommended although they may occasionally be
necessary. However, a log transformation is recommended for the vitellogenin data; a square
root transformation is recommended for the SSC data (anal fin papillae); an arcsine-square root
transformation is recommended for the data on proportion hatching, percent survival, sex ratio,
and percent fertile eggs.
The biological data from adult samples has one measurement per replicate, that is, there are one
XX fish and one XY fish per replicate aquarium. Therefore, it is recommended that a one-way
ANOVA be done on the replicate means. If the assumptions of the ANOVA (normality and
variance homogeneity as assessed on the residuals of the ANOVA by Shapiro-Wilks test and
Levene's test, respectively) are met, Dunnett contrasts should be used to determine treatments
that were different from the control. On the other hand, if the assumptions of the ANOVA are
not met, then a Dunn's test should be done to determine which treatments were different from
control. A similar procedure is recommended for data that are in the form of percentages
(fertility, hatch, and survival).
The biological data from subadult samples has from 1 to 8 measurements per replicate, that is,
there can be variable numbers of individuals that contribute to the replicate mean for each
genotypic sex. Therefore, it is recommended that a mixed effects ANOVA model be used
followed by Dunnett contrasts, if the normality and variance homogeneity assumptions were met
(on the residuals of the mixed effects ANOVA). If they were not met, then a Dunn's test should
be done to determine which treatments were different than control.
Page 26 of 124
-------
Final July 2015
Type of Data
One-Way
ANOVA With
Apriori
Transforms
Dunn's
Test
Time to
event
Other
End points
Fecundity
Histology
Cox Mixed
Effect Model
Meets Normality
and Homogeneity
Assumptions
Are Time
Effects
Important?
—\ No 1—>
Pass Test for
Monotonicity
Age at
Sampling
-H No^l
RSCABS
y
Yes
Sub Adult
Yes
Yes
Square Root
Transform
Replicate Means
Across Time
Dunnett
Contrasts
Mixed Effect
Model With
Apriori
Transforms
Jonckheere-
Terpstra on
Replicate
Repeated
Measures Anova
Using Square
Root Transform
Pass Test for
Monotonicity
4 mortalities in any replicate between 3
wpf and 9 wpf that can only be explained by toxicity rather than technical error. Other signs of
overt toxicity include hemorrhage, abnormal behaviors, abnormal swimming patterns, anorexia,
and any clinical signs of disease. For sub-lethal signs of toxicity such as reduced growth and
pathology of non-gonadal tissue, qualitative evaluations (rather than statistical) may be
necessary, and should always be made in reference to the dilution water control group (clean
water only). If overt toxicity is evident in the highest treatment(s), it is recommended that those
treatments be censored from the analysis.
(ii) Solvent controls. A solvent should only be considered as a last resort when all other
chemical delivery options have been considered. If a solvent is used, then a dilution water
control should be run in concert and a justification for the use of a solvent and the choice of
solvent should be included in the test report.
It is recommended that solvent concentration are as low as possible (e.g, <20 |il/L) to avoid
potential effect of the solvent on endpoints measured (Ref. 32) and solvent concentrations should
not exceed 100 (al/L or 100 mg/L (Ref. 33). In addition, when solvents are used, tanks should be
checked for possible microbial activity (bio-filming) should be checked throughout the test.
At the termination of the test, the solvent control group should be statistically compared to the
dilution water control group for potential effects growth determinants (weight). If statistically
Page 27 of 124
-------
Final July 2015
significant differences are detected in these endpoints between the dilution water control and
solvent control groups, best professional judgment should be used to determine if the validity of
the test is compromised. If the two controls differ, the treatments exposed to the chemical should
be compared to the solvent control unless it is known that comparison to the dilution water control
is preferred. If there is no statistically significant difference between the two control groups it is
recommended that the treatments exposed to the test chemical are compared with the pooled
(solvent and dilution-water control groups), unless it is known that comparison to either the
dilution-water or solvent control group only is preferred.
(i) Data Reporting. The test report should, at minimum, include the following
information:
(1) Test Substance. Chemical source (and Lot number), chemical identification, such as
IUPAC or CAS name, CAS number, SMILES or InChI code, structural formula, purity (and
analytical method for quantification), chemical identity of impurities as appropriate and
practically feasible, etc. (including the organic carbon content, if appropriate),
For multi-constituent substance, (e.g., UVBCs and mixtures) characterized as far as possible by
chemical identity (see above), quantitative occurrence and relevant physicochemical properties
of the constituents.
(2) Test species. At a minimum, the supplier, available culture history including specific
strain information, and any pretreatment should be reported.
(3) Test conditions. Test procedure used (test type, loading rate, stocking density, etc.);
method of preparation of stock solutions and flow-rate; nominal test substance concentrations,
means of the measured values and standard deviations in test tanks and method by which these
were attained and evidence that the measurements refer to the concentrations of test substance in
true solution; dilution water characteristics (including pH, hardness, alkalinity, temperature,
dissolved oxygen concentration, residual chlorine levels, total organic carbon, suspended solids,
and any other measurements made); water quality within test tanks: pH, hardness, temperature,
and dissolved oxygen concentration; detailed information on feeding (e.g. brine shrimp source,
analyses for relevant contaminants if necessary, e.g., PCBs, PAHs and organochlorine pesticides,
any deviations from the feeding protocol and rationale for those deviations), source and
treatment of dilution water, average and ranges of water chemistry parameters, light intensity,
and tank dimensions. Test procedure used (test type, loading rate, stocking density, etc.).
(4) Results. Evidence that controls met the performance criteria; analytical techniques
used, statistics, treatment of data and justification of techniques used; tabulated data preferably
using the suggested data template for MEOGRT (Appendix 11); detailed report on
histopathology including tabulated data using suggested pathology spreadsheet (Appendix 11);
results of the statistical analysis preferably in tabular and graphical form; incidence of any
unusual reactions by the fish and any visible effects produced by the test substance; mean,
standard deviation, and range for each test endpoint; no observed effect concentration (NOEC)
for each response assessed; lowest observed effect concentration (LOEC) for each response
assessed (at p = 0.05); deviation from the test guidelines and deviations from the performance
criteria, and considerations of potential consequences on the outcome of the test.
Page 28 of 124
-------
Final July 2015
(j) Interpretation of Results. The MEOGRT as presented is intended to serve as a
definitive Tier 2 test within the EDSP. As such, the goal of the MEOGRT is to determine
whether adverse effects associated with putative endocrine-mediated pathways of a test
substance occur in medaka, and to quantitatively evaluate those apical effects. In addition, the
MEOGRT should include exposure during the most sensitive life stages, and provide the
opportunity for identification of dose-response effects. As a Tier 2 test, the MEOGRT should
complement the Tier 1 battery.
Because the statistical analysis methods recommended tend to favor detection of monotonic
responses, it is important to consider any significant finding an indication of an adverse response.
Also, the suite of endpoints included is deemed necessary to provide information both on
important apical endpoints that are relevant at the population level, and on endpoints that
characterize the AOPs that might be affected by exposure to the test agent.
The test should define a NOEC and have at least one concentration without a statistically
significant effect on any of the observed endpoints.
It is important to note, however, that if a given exposure level results in substantial mortality or
other overt signs of toxicity, responses in other endpoints may be due to general toxicity, and not
necessarily mediated via a primary interaction with the endocrine system. Any lower treatment
level(s) should be examined for effects outside of the range of general toxicity. If all test
substance concentrations exhibit mortality or effects on apical endpoint such as growth,
fecundity, sex reversal, and intersex, then the assay would need to be repeated with lower
concentrations before inferences about possible endocrine activity could be made.
(k) References.
1) US EPA. 2011. Endocrine Disruptor Screening Program. Weight-of-Evidence:
Evaluating Results of the EDSP Tier 1 Screening to Identify the Need for Tier 2 Testing.
US EPA, Washington, DC.
2) Padilla S, Cowden J, Hinton DE, Yuen B, Law S, Kullman SW, Johnson R, Hardman
RC, Flynn K, Au DWT, 2009. Use of medaka in toxicity testing. Current Protocols in
Toxicology 39: 1-36.
3) OECD. 2015. Medaka Extended One Generation Reproduction Test (MEOGRT).
Guidelines for the Testing of Chemicals No. 240. OECD, Paris.
4) US EPA. 2009. Endocrine Disruptor Screening Program. OCSPP 890.1350: Fish Short-
Term Reproduction Assay (FSTRA). US EPA, Washington, DC.
5) Seki M., Yokota H., Matsubara H., Maeda M., Tadokoro H., Kobayashi K. 2003. Fish
full life-cycle testing for the weak estrogen 4-/
-------
Final July 2015
7) US EPA. 1996. Ecological Effects Test Guidelines. OPPTS 850.1500: Fish life cycle
toxicity. US EPA, Washington, DC
8) OECD. 1998. Fish, Short-term Toxicity Test on Embryo and Sac-fry Stages. OECD
Guidelines for the Testing of Chemicals No. 212. OECD, Paris.
9) OECD. 2009. 21-day Fish Assay: A Short-Term Screening for Oestrogenic and
Androgenic Activity, and Aromatase Inhibition. OECD Guidelines for the Testing of
Chemicals No. 230. OECD, Paris.
10) OECD. 2010. Guidance document on the diagnosis of endocrine-related histopathology
in fish gonads. OECD Environment, Health and Safety Publications. Series on testing and
assessment No. 123. OECD, Paris.
11) OECD. 2011. Fish Sexual Development Test. OECD Guidelines for the Testing of
Chemicals No. 234. OECD, Paris.
12) OECD. 2012a. Fish Toxicity Testing Framework, OECD Environment, Health and Safety
Publications. Series on testing and assessment No. 171. OECD, Paris.
13) OECD. 2012b. Guidance Document on Standardised Test Guidelines for Evaluating
Endocrine Disrupters. OECD Environment, Health and Safety Publications. Series on
testing and assessment No. 150. OECD, Paris.
14) OECD. 2013a. Fish, Early-life Stage Toxicity Test. OECD Guidelines for the Testing of
Chemicals No. 210. OECD, Paris.
15) OECD. 2013b. Fish Embryo Acute Toxicity (FET) Test. OECD Guidelines for the
Testing of Chemicals No. 236. OECD, Paris.
16) US EPA. 1996. Ecological Effects Test Guidelines. OPPT 850.1400: Fish Early-Life
Stage Toxicity Test. US EPA, Washington, DC.
17)Matsuda M., Nagahama Y., Shinomiya A., Sato T., Matsuda C., Kobayashi T., Morrey
C., ShibataN., Asakawa S., ShimizuN., Hori H., Hamaguchi S., Sakaizumi M. 2002.
DMY is a y-specific dm-domain gene required for male development in the medaka fish.
Nature 417:559-563.
18)Nanda I, Hornung U, Kondo M, Schmid M, Schartl M. 2003. Common spontaneous sex-
reversed XX males of the medaka Oryzias latipes. Genetics 163: 245-251.
19) Shinomiya, A, Otake H. Togashi K. Hamaguchi S. Sakaizumi M. 2004, Field survey of
sex-reversals in the medaka, Oryzias latipes: genotypic sexing of wild populations,
Zoological Science 21: 613-619.
20)Matsui, K. 1949. Illustration of the normal course of development in the fish, Oryzias
latipes. Jpn. Exp. Morph. 5, 33-42.
Page 30 of 124
-------
Final July 2015
21)Denny J.S., Spehar R.L., Mead K.E., Yousuff S.C. 1991. Guidelines for culturing the
Japanese Medaka, Oryzias latipes. US EPA/600/3-91/064.
22)Koger CS, Teh SJ, Hinton DE. 1999. Variations of light and temperature regimes and
resulting effects on reproductive parameters in medaka (Oryzias latipes). Biology of
Reproduction 61: 1287-1293.
23)Kinoshita M, Murata K, Naruse K, Tanaka M. 2009. Medaka: Biology, Management, and
Experimental Protocols, Wiley- Blackwell.
24) Yokota H., Tsuruda Y., Maeda M., Oshima Y., Tadokoro H., Nakazono A., Honjo T.,
Kobayashi K. 2000. Effect of bisphenol A on the early life stage in Japanese medaka
{Oryzias latipes). Environ Toxicol Chem 19:1925-1930.
25) Seki M., Yokota H., Matsubara H., Tsuruda Y., Maeda M., Tadokoro H., Kobayashi K.
2002. Effect of ethinylestradiol on the reproduction and induction of vitellogenin and
testis-ova in medaka (Oryzias latipes). Environ Toxicol Chem 21:1692-1698
26)Gormley K., Teather K. 2003. Developmental, behavioral, and reproductive effects
experienced by Japanese medaka in response to short-term exposure to endosulfan.
Ecotox Environ Safety 54:33 0-3 3 8.
27)Kang I.J., Yokota H., Oshima Y., Tsuruda Y., Oe T., Imada N., Tadokoro H., Honjo T.
2002. Effects of Bisphenol A on the reproduction of Japanese Medaka (Oryzias latipes).
Environ Toxicol Chem 21:2394-2400.
28)Kang I.J., Yokota H., Oshima Y., Tsuruda Y., Yamaguchi T., Maeda M., Imada N.,
Tadokoro H., Honjo T. 2002. Effects of 17|3-estradiol on the reproduction of Japanese
medaka (Oryzias latipes). Chemosphere 47:71-80.
29) ASTM, 2002 ASTM Standard Guide for Conducting Acute Toxicity Tests on Test
Materials with Fishes, Macroinvertebrates, and Amphibians.
30)Kahl M.D., Russom C.L., DeFoe D.L., Hammermeister D.E. 1999. Saturation units for
use in aquatic bioassays. Chemosphere 39:539-551.
31) OECD. 2002. Guidance Document on Aquatic Toxicity Testing of Difficult Substances
and Mixtures. OECD Environment, Health and Safety Publications. Series on testing and
assessment No. 23. OECD, Paris.
32) Hutchinson TH., Shillabeer N., Winter MJ, Pickford DB. 2006. Acute and chronic effects
of carrier solvents in aquatic organisms: A critical review. Review. Aquatic Toxicology
76:.69-92.
33) OECD. 2012c. Fish Short Term Reproduction Assay. OECD Guidelines for the Testing
of Chemicals No. 229. OECD, Paris.
Page 31 of 124
-------
Final July 2015
34) Green J.W., Springer T.A., Saulnier A.N., Swintek J. 2014. Statistical analysis of
histopathology endpoints. Environ Toxicol Chem 33:1108-1116.
35) OECD. 2006. Current approaches in the statistical analysis of ecotoxicity data: a
guidance to application. Environmental Health and Safety Publications. Series on
Testing and Assessment, No. 54. Paris, France.
36) Hocking, RR. 1985. Analysis of linear models. Brooks/Cole Publishing Company. 400
Pg-
37)Hochberg, Y Tamhane, AC. 1987. Multiple Comparison Procedures, New York: Wiley
and Sons.
Page 32 of 124
-------
Final July 2015
(1) List of MEOGRT Appendices.
Page
(1) Definitions 34
(2) Example Infrastructure 35
(3) Example Daily Calendar 42
(4) DNA Extraction from Fin Clip Tissue 46
(5) Example Real-Time PCR Protocol for dmy Gene Analysis in Japanese
Medaka 47
(6) Subadult Sampling - Necropsy 52
(7) Counting Anal Fin Papillae 54
(8) RNA Extraction from Medaka Liver 55
(9) Preparation of RNA for Standard Curve Used during VTG1 QPCR 56
(10) QPCR to Quantify Vtgl mRNA in Liver 61
(11) Data Reporting Templates 64
(12) Histopathology Guidance 66
Page 33 of 124
-------
Final July 2015
APPENDIX 1: Definitions.
ELISA: Enzyme-Linked Immunosorbent Assay
Fecundity = number of eggs;
Fertility = number of viable eggs/fecundity;
Fork length (FL): refers to the length from the tip of the snout to the end of the middle caudal
fin rays and is used in fishes in which it is difficult to tell where the vertebral column ends
www.fishbase.org
Hatchability = hatchlings/number of embryos loaded into an incubator
IACUC: Institutional Animal Care and Use Committee
Standard length (SL): refers to the length of a fish measured from the tip of the snout to the
posterior end of the last vertebra or to the posterior end of the midlateral portion of the hypural
plate. Simply put, this measurement excludes the length of the caudal fin. (www.fishbase.org)
Total length (TL): refers to the length from the tip of the snout to the tip of the longer lobe of
the caudal fin, usually measured with the lobes compressed along the midline. It is a straight-line
measure, not measured over the curve of the body (www.fishbase.org)
____________ standard length
i —.— Fork length 1 !
i i
| i
i i
1— Total length —. ¦
Figure 1: Description of the different lengths, used
ECx: (Effect concentration for x% effect) is the concentration that causes an x% of an effect on
test organisms within a given exposure period when compared with a control. For example, an
EC50 is a concentration estimated to cause an effect on a test end point in 50% of an exposed
population over a defined exposure period.
Page 34 of 124
-------
Final July 2015
Flow-through test: is a test with continued flow of test solutions through the test system during
the duration of exposure.
HPG axis: hypothalamic-pituitary-gonadal axis
IUPAC: International Union of Pure and Applied Chemistry.
Loading rate: the wet weight of fish per volume of water.
Lowest observed effect concentration (LOEC) is the lowest tested concentration of a test
chemical at which the chemical is observed to have a statistically significant effect (at p < 0.05)
when compared with the control. However, all test concentrations above the LOEC should have
a harmful effect equal to or greater than those observed at the LOEC. When these two conditions
cannot be satisfied, a full explanation should be given for how the LOEC (and hence the NOEC)
has been selected. Annexes 5 and 6 provide guidance.
Median Lethal Concentration (LC50): is the concentration of a test chemical that is estimated
to be lethal to 50% of the test organisms within the test duration.
No observed effect concentration (NOEC) is the test concentration immediately below the
LOEC, which when compared with the control, has no statistically significant effect (p < 0.05),
within a stated exposure period.
SMILES: Simplified Molecular Input Line Entry Specification.
Stocking density: is the number of fish per volume of water.
UVCB: substances of unknown or variable composition, complex reaction products or biological
materials.
VTG: vitellogenin is a phospholipoglycoprotein precursor to egg yolk protein that normally
occurs in sexually active females of all oviparous species.
WPF: weeks post fertilization.
Page 35 of 124
-------
Final July 2015
APPENDIX 2: Example Infrastructure.
(a) Background. This appendix illustrates examples of infrastructural
components that have been successfully used to conduct this test protocol.
(b) Egg Incubators. The smooth transition from the adults of a generation to the
larvae of the next generation is critical to the success of the assay. The embryos require
careful handling during this period. The design of an embryo incubator needs to address
both husbandry and exposure concerns. Careless embryo husbandry can result in
mortalities and lost embryos resulting in either insufficient eleutheroembryos to continue
to the next generation or artificially low numbers for hatching. Furthermore, embryo
development is an important exposure window in which the treatment chemical
concentration should remain stable. Therefore the incubator design should allow for both
continuous renewal of the test chemical and easy access to collect water samples from
inside the incubator in order to determine the chemical concentration within the
incubator. Easy and clean removal of the incubator from the exposure system is also a
requirement in order to observe the embryos under magnification, to remove non-viable
embryos and to remove eleutheroembryos. Elements critical to embryo hatching success
are: monitoring of the embryos and immediately removing ones that are non-viable;
precision handling during transfers between containers; and precise control of air and/or
water flow in the incubator.
The following are short descriptions and illustrations of incubators that have been used to
successfully hatch medaka embryos.
(1) Flow Though Incubator Tube.
(a) Pyrex No. 8240-100 tube (b) Tank
Figure 1. Flow Though Incubator Tube, (a) This type of incubator is constructed from
a Pyrex No. 8240-100 tube, cut to a length of 14 cm with slots cut for overflow, (b) A
stainless steel screen is placed over the slots, covering the slots and preventing the eggs
from washing out the slots while still allowing for overflow. The tank's influent delivery
line discharges directly into the incubator. A disposable pipette bubbles air to gently
agitate the embryos.
Page 36 of 124
-------
Final July 2015
(2) Threaded Cap Incubator.
//
Figure 2. Threaded Cap Incubator. This incubator consists of a transected
glass centrifuge tube, connected by a stainless steel sleeve and held in place by
the centrifuge screw top cap. A small glass or stainless steel tube projects through
the cap and is positioned near the rounded bottom, gently bubbling air to suspend
the eggs and reducing between-egg transmission of saprophytic fungal infections
while also facilitating chemical exchange between the incubator and the holding
tank.
(3) Syringe Style Incubator.
Figure 3. Syringe Style Incubator. This incubator consists of a fluid dispensing
syringe assembly containing stainless steel screens located at the top and bottom of the
syringe, ensuring that the eggs are contained within the syringe body. A series of valves
allow the flow to the incubator to be precisely regulated and/or diverted to the aquaria or
to drain as needed. The flow of fluid through the incubator should be sufficient to provide
agitation to the eggs.
Page 37 of 124
-------
Final July 2015
(c) Isolation Chambers (Figure 4). From Study Week 6 through Study Week
11 of the Fi generation, twelve fish are held in each replicate tank. The genotypic sex of
each fish is determined during Study Week 12. Each fish will have a small sample of
the caudal fin taken to determine its sex. The DMY analysis has a turnaround time of 24-
48 hours. The fish need to be held as identifiable individuals until their genetic sex has
been determined and they can be assigned to a breeding pair or sacrificed for the subadult
endpoints. Isolation chambers are used to hold the fish, with two fish contained in each
chamber. The caudal fin of each fish is differentially cut with one a dorsal and the other
receiving a ventral cut. Six chambers are set up in each aquarium. The placement of the
chamber in the replicate aquarium is pre-determined to maintain the identity of each fish.
Figure 4. Isolation Chamber Illustration. On the left, a single chamber. On the right, aquarium
with six chambers.
(d) Temperature Control and Monitoring. Fecundity and growth endpoints
require precise temperature control and monitoring. A dry system using low wattage
rubber silicon heaters controlled by a digital temperature controller is recommended
(Figure 5). The heating system should be sized to avoid large temperature swings if a
component fails. To further aid in temperature control, the aquaria should receive
treatment water that is temperature controlled at the nominal test temperature. The tank
heater system only inputs heat to compensate for ambient loss. It is also recommended
that the heating system is linked to a data collection system that provides real time
temperature monitoring and data logging for each aquaria.
Page 38 of 124
-------
Final July 2015
PANEL EX1 C/B17
Figure 5. Tank temperature and monitoring display.
(e) Delivery Systems. The flow of water to each tank should be uniform to
ensure that the water quality is comparable in all treatments. Requirements of such a
system are:
• The system should provide reliable and consistent flow. It should be easily
adjusted to a wide range of flow, allowing it to be adjusted in response to changes
in the biomass, water quality or chemical degradation.
• Ease of maintenance should be considered in the design of a water delivery
system.
Page 39 of 124
-------
Final July 2015
(1) Multi-channel Peristaltic Pump (Figure 6).
Figure 6. Multi-channel Peristaltic Pump. One system that meets these requirements is a
multi-channel peristaltic pump system which can deliver treatment water from the treatment
reservoir to each replicate exposure aquarium. These systems can reliably deliver flows up to 50
ml/min/channel. Delivery lines should be sterilized on a set maintenance schedule to control in-
line biological growth and the peristaltic tubes should be replaced on a regular basis since the
flow will diminish as the tubing wears.
(2) Positive Displacement Pump and Manifold (Figure 7).
Figure 7. Positive Displacement Pump and Manifold. Another type of system that meets the
above requirements is a positive displacement dilution pump pressurizing a manifold that has a
series of programmable solenoids directing flow to the replicate exposure aquaria. This system
eliminates the treatment reservoir and peristaltic pump. The treatment flow is controlled by the
sped of the pump which continuously prepares treatment solution. Treatment solution is
sequentially dispensed to the replicate tanks by the opening of a solenoid directing flow to a tank.
Page 40 of 124
-------
Final July 2015
(f) Chemical Dilution System. The precise preparation of toxicant solutions to
be delivered to exposure tanks is critical to the success of a long-term bioassay. The
design considerations for a toxicant preparation system are:
• The preparation of treatments should be independent. This enhances study design
flexibility by allowing the adjustment of individual treatment concentrations, the
addition or elimination of treatments, and the selection of concentration gradients
based on the experimental design and not on system limitations.
• The ability to produce a large range of treatment concentrations.
• The toxicant should have short residence time in the system. The toxicant should
undergo continuous dilution with rapid delivery to the exposure tanks.
• Excessive chemical should not be sent to waste.
• If the system fails the toxicant levels should drop.
One toxicant preparation system meeting these requirements incorporates dual headed
positive displacement pumps which have independent pump heads linked by a common
drive motor. Each treatment solution is prepared by a dedicated pump (Figure 8). The
pump links a flow of di lution water to a flow of toxicant which are combined at a T
fitting and directed through a static mixer into the treatment reservoir. The ratio of
dilution water to toxicant is set by adjusting the pi ston displacement volume of the pump
heads. The piston adjustment is independent for each pump head and flow ratios of 1:1
to 200:1 can be set up. This range can be expanded by swapping in higher or lower flow
pump heads. The flow to the treatment reservoir is controlled by the speed of the drive
motor. The pump speed can be manually adjusted or controlled with a circuit. Treatment
solutions have been diluted at rates from 10 to 260 ml / min.
Figure 8. Example of toxicant preparation system.
Page 41 of 124
-------
Final July 2015
APPENDIX 3: Example Daily Calendar.
Medaka Extend One-Generation Re
production Test Daily Protocol
Enter Fertilization Date (FO
Generation):
4/9/2012
Chemical:
Enter Start Date (move FO
breeding pairs into system):
6/25/2012
Temp:
25° C
1st
Event
Test
F1
F1
F2
Tan
Brine
(1 row for each day of
study)
Day
Date
Calenda
r Day
Test Wk
DayTes
t Day
W
k
dp
f
dp
f
kto
Feed
Shrimp (dry
wt.)/fish/day
Move breeding pairs into
exposure system
Start exposure
Mon
2 5-.Tun-12
176
1
1
37
22.5 mg
Loading Phase
Tue
26-.Tun-12
177
2
5
Loading Phase
Wed
27-.Tun-12
178
3
17
Loading Phase
Thu
28-.Tun-12
179
4
13
Loading Phase
Fri
29-.Tun-12
180
5
7
Loading Phase
Sat
30-.Tun-12
181
6
42
Loading Phase
Sun
1-Jul-12
182
7
16
Loading Phase
Mon
2-.Tul-12
183
2
8
39
22.5 mg
Loading Phase
Tue
3-.Tul-12
184
9
15
Loading Phase
Wed
4-.Tul-12
185
10
24
Loading Phase
Thu
5-.Tul-12
186
11
20
Loading Phase
Fri
6-.Tul-12
187
12
22
Loading Phase
Sat
7-.Tul-12
188
13
31
Loading Phase
Sun
8-.Tul-12
189
14
38
Loading Phase
Mon
9-.Tul-12
190
3
15
13
22.5 mg
Loading Phase
Tue
lO-Jul-12
191
16
38
Loading Phase
Wed
11-Jul-12
192
17
15
Loading Phase
Thu
12-Jul-12
193
18
9
Loading Phase
Fri
13-Jul-12
194
19
32
Loading Phase
Sat
M-Jul-12
195
20
29
siphon tanks and remove eggs
Sun
15-Jul-12
196
21
8
Start F1 sen (Fill first set of
incubators)
Mon
16-Jul-12
197
4
22
1
0
1
22.5 mg
Second dav of F1 egg
collection if necessary f Fill
second set of incubators)
Tue
17-Jul-12
198
23
1
13
Terminate FO Generation
Wed
18-Jul-12
199
24
2
Thu
19-Jul-12
200
25
3
Fri
20-Jul-12
201
26
4
Sat
21-Jul-12
202
27
5
Sun
22-Jul-12
203
28
6
Mon
23-Jul-12
204
5
29
2
7
Aeration turned off for first
spawn incubators.
Tue
24-Jul-12
205
30
8
Aeration turned off for
second spawn incubators.
Wed
25-Jul-12
206
31
9
35
0.5 mg
Thu
26-Jul-12
207
32
10
32
0.5 mg
Fri
27-Jul-12
208
33
11
27
0.6 mg
Page 42 of 124
-------
Final July 2015
Medaka Extend One-Generation Re
production Test Daily Protocol
Enter Fertilization Date (FO
Generation):
4/9/2012
Chemical:
Enter Start Date (move FO
breeding pairs into system):
6/25/2012
Temp:
25° C
Event
(1 row for each day of
study)
Day
Date
Calenda
r Day
Test Wk
Test
DayTes
t Day
F1
W
k
F1
dp
f
F2
dp
f
1st
Tan
kto
Feed
Brine
Shrimp (dry
wt.)/fish/day
Sat
28-.M-12
209
34
12
26
0.7 mg
Sim
29-.M-12
210
35
13
29
0.8 mg
Discard unhatched eaas from
first suawn.
Mon
30-Jul-12
211
6
36
3
14
16
1.0 mg
Discard unhatched eaas from
second suawn.
Tue
31-Jul-12
212
37
15
32
1.3 mg
Wed
1-Aug-12
213
38
16
36
1.7 mg
Thu
2-Aug-12
214
39
17
12
2.2 mg
Fri
3-Aug-12
215
40
18
40
2.8 mg
Sat
4-Aug-12
216
41
19
18
3.5 mg
Sim
5-Aug-12
217
42
20
6
4.2 mg
Mon
6-Aug-12
218
7
43
4
21
28
4.5 mg
Pool/Redistribute/Cull
Tue
7-Aug-12
219
44
22
23
4.8 mg
Wed
8-Aug-12
220
45
23
26
5.2 mg
Thu
9-Aug-12
221
46
24
39
5.6 mg
Fri
10-Aug-12
222
47
25
11
Sat
11-Aug-12
223
48
26
19
Sim
12-Aug-12
224
49
27
12
Mon
13-Aug-12
225
8
50
5
28
36
7.7 mg
Tue
14-Aug-12
226
51
29
31
Wed
15-Aug-12
227
52
30
20
Thu
16-Aug-12
228
53
31
26
Fri
17-Aug-12
229
54
32
17
Sat
18-Aug-12
230
55
33
8
Sim
19-Aug-12
231
56
34
40
Mon
20-Aug-12
232
9
57
6
35
22
9.0 mg
Tue
21-Aug-12
233
58
36
7
Wed
22-Aug-12
234
59
37
18
Thu
23-Aug-12
235
60
38
19
Fri
24-Aug-12
236
61
39
7
Sat
25-Aug-12
237
62
40
42
Sim
26-Aug-12
238
63
41
36
Mon
27-Aug-12
239
10
64
7
42
11
11.0 mg
Tue
28-Aug-12
240
65
43
29
Wed
29-Aug-12
241
66
44
18
Thu
30-Aug-12
242
67
45
10
Fri
31-Aug-12
243
68
46
12
Sat
l-Sep-12
244
69
47
28
Sim
2-Sep-12
245
70
48
36
Mon
3-Sep-12
246
11
71
8
49
4
13.5 mg
Page 43 of 124
-------
Final July 2015
Medaka Extend One-Generation Re
production Test Daily Protocol
Enter Fertilization Date (FO
Generation):
4/9/2012
Chemical:
Enter Start Date (move FO
breeding pairs into system):
6/25/2012
Temp:
25° C
Event
(1 row for each day of
study)
Day
Date
Calenda
r Day
Test Wk
Test
DayTes
t Day
F1
W
k
F1
dp
f
F2
dp
f
1st
Tan
kto
Feed
Brine
Shrimp (dry
wt.)/fish/day
Tue
4-Sep-12
247
72
50
22
Wed
5-Sep-12
248
73
51
10
Thu
6-Sep-12
249
74
52
2
Fri
7-Sep-12
250
75
53
31
Sat
8-Sep-12
251
76
54
42
Sun
9-Sep-12
252
77
55
19
Fill Clip.
Mon
10-Sep-12
253
12
78
9
56
35
22.5 mg
Tue
11-Sep-12
254
79
57
27
F1 Cull to pairs.
Wed
12-Sep-12
255
80
58
F1 Cull to pairs.
Thu
13-Sep-12
256
81
59
Fri
14-Sep-12
257
82
60
Sat
15-Sep-12
258
83
61
Sun
16-Sep-12
259
84
62
Mon
17-Sep-12
260
13
85
10
63
22.5 mg
Tue
18-Sep-12
261
86
64
Wed
19-Sep-12
262
87
65
Thu
20-Sep-12
263
88
66
Fri
21-Sep-12
264
89
67
Sat
22-Sep-12
265
90
68
Sun
23-Sep-12
266
91
69
Mon
24-Sep-12
267
14
92
11
70
22.5 mg
Tue
25-Sep-12
268
93
71
Wed
26-Sep-12
269
94
72
Thu
27-Sep-12
270
95
73
Fri
28-Sep-12
271
96
74
Sat
29-Sep-12
272
97
75
siphon tanks and remove eggs
Sun
30-Sep-12
273
98
76
Start F1 fertilitv-fecunditv
assessment.
Mon
1-Oct-12
274
15
99
12
77
22.5 mg
Tue
2-Oct-12
275
100
78
Wed
3-Oct-12
276
101
79
Thu
4-Oct-12
277
102
80
Fri
5-Oct-12
278
103
81
Sat
6-Oct-12
279
104
82
Sun
7-Oct-12
280
105
83
Mon
8-Oct-12
281
16
106
13
84
22.5 mg
F ertility-F ecundity
Assessment
Tue
9-Oct-12
282
107
85
Wed
10-Oct-12
283
108
86
Thu
11-Oct-12
284
109
87
Page 44 of 124
-------
Final July 2015
Medaka Extend One-Generation Re
production Test Daily Protocol
Enter Fertilization Date (FO
Generation):
4/9/2012
Chemical:
Enter Start Date (move FO
breeding pairs into system):
6/25/2012
Temp:
25° C
Event
(1 row for each day of
study)
Day
Date
Calenda
r Day
Test Wk
Test
DayTes
t Day
F1
W
k
F1
dp
f
F2
dp
f
1st
Tan
kto
Feed
Brine
Shrimp (dry
wt.)/fish/day
Fri
12-Oct-12
285
110
88
Sat
13-Oct-12
286
111
89
Sun
14-Oct-12
287
112
90
Mon
15-Oct-12
288
17
113
14
91
22.5 mg
F ertility-F ecundity
Assessment
Tue
16-Oct-12
289
114
92
Wed
17-Oct-12
290
115
93
Thu
18-Oct-12
291
116
94
Fri
19-Oct-12
292
117
95
Sat
20-Oct-12
293
118
96
Sun
21-Oct-12
294
119
97
Start F2 Generation, fill
incubators
Mon
22-Oct-12
295
18
120
15
98
0
22.5 mg
Terminate F1 Generation
Tue
2 3-Oct-12
296
121
99
1
Wed
24-Oct-12
297
122
2
Thu
2 5-Oct-12
298
123
3
Fri
26-Oct-12
299
124
4
Sat
27-Oct-12
300
125
5
Sun
28-Oct-12
301
126
6
Mon
29-Oct-12
302
19
127
16
7
Aeration turned off for first
spawn incubators.
Tue
30-Oct-12
303
128
8
Aeration turned off for
second spawn incubators.
Wed
31-Oct-12
304
129
9
Thu
1-Nov-12
305
130
10
Fri
2-Nov-12
306
131
11
Sat
3-Nov-12
307
132
12
Sun
4-Nov-12
308
133
13
Discard unhatched eaas from
first spawn.
Mon
5-Nov-12
309
20
134
17
14
Discard unhatched eaas from
second spawn.
Tue
6-Nov-12
310
135
15
Page 45 of 124
-------
Final July 2015
APPENDIX 4: Example Protocol for DNA Extraction from Fin Clip Tissues
(a) Background. In the MEOGRT, subadult medaka are assessed for the presence or
absence of the dmy gene in order to 1) setup up breeding pairs of one XX and one XY fish, and 2)
provide DMY data for each of the fish sacrificed at this time. The first step in this process is to
extract DNA from a tissue sample, typically a fin clip, of each fish. This can be done using
commercially available kits that are based upon selective binding of DNA by a silica-gel-based
membrane or other suitable methods of DNA extraction.
The example protocol presented is for DNA extraction from fin clips based on procedures used at
the U.S. Environmental Protection Agency laboratory in Duluth, MN. Specific products and/or
equipment listed can be substituted with comparable materials.
(b) Major Materials and Reagents. DNA may be extracted by a variety of methods. In
this example, a commercial DNA extract kit, DNeasy Blood and Tissue Kit (Qiagen), is used to
extract DNA from a fin clip tissue sample.
• DNA extraction kit ((DNeasy Blood and Tissue Kit, Qiagen)
• Microcentrifuge and tubes
(c) Protocol
(1) Tissue Collection. With the fish under anesthesia, remove a small piece (~1 mm
square) of tail fin, place in a 1.5 ml microcentrifuge tube and immediately quench freeze in liquid
nitrogen. Fins are stored at less than -20° C until further use. If tissue is going to be immediately
processed, freezing can be omitted and the tissue can be processed as indicated below.
(2) DNA Extraction. Extract DNA from fin clip samples using the DNeasy Blood and
Tissue Kit. Add 180 ul Buffer ATL to fin clip followed by 20 |il proteinase K, and vortex.
Alternatively, Buffer ATL and proteinase K can be premixed immediately before sampling large
numbers of fin clips and 200 (j.1 added to fin clip. Incubate microcentrifuge tubes at 56°C until
tissue is lysed (usually takes approximately 45 minutes but can be left overnight), vortex for 15
seconds, and add 200 jal Buffer AL before vortexing another time. Add 200 j_il of 100% ethanol,
and then vortex briefly. Pipet resulting mixture onto a DNeasy Mini spin column, centrifuge at
greater than 6000 x g for 1 minute, and discard the flow-through/collection tube. Wash bound
DNA with 500 j_il Buffer AW1, centrifuge at greater than 6000 x g for 1 minute (discard flow-
through/collection tube) then wash with 500 |il Buffer AW2, centrifuge at greater than 20000 x g
for 3 minutes to dry the column membrane (discard flow-through/collection tube). Place column
in a new 1.5 ml microcentrifuge tube, pipet 100 j_il Buffer AE onto the column, incubate at room
temperature for 1 minute, and centrifuge at 6000 x g for 1 minute to elute DNA. Store DNA at
less than -20°C until used.
Quantification of the extracted DNA is unnecessary because it will be used in a qualitative test to
detect the presence or absence of the dmy gene (see Appendix 5). In addition, there are sufficient
quality assurance procedures in place during the actual dmy assessment that the quality of the
extracted also does not need to be assessed. If experience with these types of DNA extraction kits
is lacking, the concentration and quality of the DNA can be assessed by standard methods.
Page 46 of 124
-------
Final July 2015
APPENDIX 5: Example Real-Time PCR Protocol to Detect dmy Gene in Japanese
medaka
(a) Background. In the MEOGRT, subadult medaka are assessed for the presence or
absence of the dmy gene (GenBank: AB071534.1) in order to 1) setup up breeding pairs of one
XX and one XY fish, and 2) provide DMY data for each of the fish sacrificed at this time. After
the DNA is extracted (see Appendix 3, the presence (or absence) of dmy, the medaka male
determining gene, is assessed in each sampleThe dmy gene is present on the Y chromosome of
every male somatic cell, but is not present in any of the female cells providing for a definitive
test for genotypic sex. A DNA sample that is positive for the dmy gene is from a genotypic male
while a DNA sample that is negative for the dmy gene is from a genotypic female (Matsuda el
al., 2002). These results are independent of phenotypic sex which may have been altered by
exposure to a test chemical that might disrupt endocrine hormone pathways.
The protocol discussed below is an example of real-time PCR assay to detect the dmy gene. The
example procedure is based on procedures used at the U.S. Environmental Protection Agency
laboratory in Duluth, MN. Specific products and/or equipment listed can be substituted with
comparable materials. In this example, a Taqman® assay is used to detect amplification of the
dmy gene.
This protocol utilizes a Taqman® assay, a variant of real-time PCR to detect amplification of the
DMY gene in each well of the 96-well plate. Starting at the designed primers, DNA polymerase
moves downstream synthesizing a new strand from the template strand and via its 5' exo-
nuclease activity, the polymerase removes any bases that would impede its progress down the
template strand. A Taqman® assay (5' DNA exo-nuclease assay) uses this 5' exo-nuclease
activity in real time detection of PCR product. A Taqman® probe, present in the PCR mix,
anneals to the template between the two primers in the path of the DNA polymerase as it
progresses on the template strand. The Taqman® probe is a short sequence of DNA (-20 bp)
with a reporter fluorophore on one end and a quencher on the other end. The close proximity of
these two moieties allows interaction prohibiting the fluorescence of the reporter and thus its
detection upon excitation within the real time instrument. However when the DNA polymerase
cleaves the probe, both reporter and quencher are released into solution, and the distance
between the two increases allowing the reporter to fluoresce upon excitation. Normally, the user
is interested in the quantity of fluorescence from the reporter during the exponential phase of
PCR as it is directly proportional to the amount of target sequence in the sample. However, in
this application, no quantification is needed because the purpose of this assay is determine the
presence or absence of the dmy gene. Any level of amplification above the threshold indicates
the presence of the dmy gene, while no amplification indicates the absence of the dmy gene (see
Figures 1 and 2).
In addition, an internal control is included in the Master Mix that amplifies the 18S ribosomal
subunit sequence (GenBank: X03205.1). Its presence is ubiquitous in eukaryotic cells, and thus
provides a clear indication of successful PCR in each well. If no amplification is present in the
control, either there is no DNA present in the tissue sample or there is inhibition of PCR by some
contamination.
Page 47 of 124
-------
Final July 2015
There is another variation of the described protocol that instead of a primer/probe combination
measuring fluorescence, only a primer set is used. This follows the typical protocol for
traditional PCR followed by separation of the amplicon(s) with agarose gel electrophoresis.
There are numerous references that can provide examples of protocols and possible primers,
(e.g., Padilla etal. 2009 and Shinomiya etal. 2004.
(b) Major Materials and Reagents.
• TaqMan® Universal PCR Master Mix (Applied Biosystems)
• 18S Reagent (Applied Biosy stems))
• dmy Forward Primer (5' TTC TGC CGC TGG AAA GAC 3')
• dmy Reverse Primer (5' TCT CTG GCG GAC CAT GAT 3')
• dmy Probe (5' FAM-CCA GTG CTT CAA ATG CGA GCA-BHQ 3')
• Real-time thermal cycler (e.g., ABI 7500)
• 96-well optical plates
• Centrifuge
(c) Protocol.
(1) Source DNA. DNA extracted using example protocol (or other suitable method)
discussed inAppendix 3.
(2) PCR. Although PCR is designed to amplify and quantify DNA, quantification is not
needed because the purpose of this assay is to simply detect the presence or absence of the dmy
gene in Japanese medaka. This protocol is a plus/minus assay. It maximizes throughput by
using 96-well plates and detecting amplification within each well, instead of running each
product, post-PCR amplification, through gel electrophoresis.
Prepare the following master mix keeping every reagent on ice shown in Table 1.
Table 1. Reagents
Reagents
Volume per Sample
TaqMan® Universal PCR Master
Mix (Master Mix)
12.5 [i\
18S reagent
0.25 [i\
dmy Forward Primer (10 (iM)
1.75 [i\
dmy Reverse (10 (iM)
1.75 [i\
dmy Probe (1250 nM)
4.0 [i\
Using a pre-chilled 96 well optical plate, load each well with 20.25 |il of Master Mix and add
4.75 [jl of the DNA sample. Seal, vortex, and centrifuge plate at 2000 rpm for 5 minutes to mix
the samples with the Master Mix and remove any air bubbles. Air bubbles in the well will alter
the fluorescent measurement.
Page 48 of 124
-------
Final July 2015
Run the following thermal cycler program on the ABI 7500 as shown in Table 2.
Table 2. Thermal Cycler Program.
Step #
Temperature
Time
1
50 C
2 min
2
95 C
5 min
3
95 C
15 sec
4
60 C
1 min
5
Go to step 3
30 times
(3) Interpretation of the Results. There are two detectors to consider when analyzing
the results. The first shows the amplification associated with the 18S reagents (internal positive
control). While the Ct value of the internal positive control will obviously be impacted by the
concentration of the DNA, it may also be impacted by the amount of fluorescence in the other
detector, the DMY detector, in particular if there is concurrent dmy gene amplification. In
addition, the level of fluorescence (delta Rn) of the detector may also be reduced if there is
concurrent dmy amplification. This is because the primers and probe in the internal positive
control are designed to have lower PCR efficiency than the dmy primers and probes to easily
detect amplified dmy DNA versus control DNA.
An XX sample will have virtually no amplification in the DMY detector through the 30 cycles of
PCR, but there will be amplification in the internal positive control detector in a XX sample. If
there is no amplification in either detector, there is a problem with the sample. The sample
should be tested to verify the presence of high quality DNA, and reprocessed through the DNA
extraction protocol, if appropriate. An XY sample will have amplification in the DMY detector,
and there will at least be some minimal evidence of amplification in the IPC detector if not
normal amplification. Below are example analyses of an XX fish (Figure 1) and an XY fish
(Figure 2).
Page 49 of 124
-------
Final July 2015
uena Kn vs uvcie
0.37
0.32
Delta Rn
y
-0.01
1
1
17 13 1
1 fi 1 fi 1
Cycle Number
1
a 19 ?
?
9
3
Figure 1. Example analysis of an XX fish. The curved plot (blue) represents the amplification of the
IPC detector and the plot at the bottom (black) represents the amplification of the dmy gene. In this
example, the dmy gene is not present in this sample (no amplification detected). The genotype of this fish
is XX (female).
Delta Rn vs Cvcle
0.40
0.37
1 ? 3 4 5 fi 7 fi 9 1(1 11 1? 13 14 15 1fi 17 1ft 19 ?fl 91 ?? 73 94 9fi 77 9fi 79 30
Cycle Number
Figure 2. Example of the analysis of an XY fish. The lower curved plot (blue) represents the
amplification of the IPC detector and the higher curved plot (black) represents the amplification of the
dmy gene. In this example, the dmy gene is present in this sample as shown by the elevated amplification
detected. The genotype of this fish is XY (male).
Page 50 of 124
-------
Final July 2015
(d) References.
1) Matsuda, M., Nagahama, Y., Shinomiya, A., Sato, T., Matsuda, C., Kobayashi, T.,
Morrey, C., Shibata, N., Asakawa, S., Shimizu, N., Hori, H., Hamaguchi, S., Sakaizumi,
M., 2002. DMY is a y-specific dm-domain gene required for male development in the
medaka fish. Nature 417: 559-563.
2) Padilla, S., Cowden, J., Hinton, D.E., Yuen, B., Law, S., Kullman, S.W., Johnson, R.,
Hardman, R.C., Flynn, K., Au, D.W.T., 2009. Use of medaka in toxicity testing. Curr
Protoc Toxicol. 39:1-36.
3) Shinomiya, A., Otake, H., Togashi, K., Hamaguchi, S., Sakaizumi, M., 2004. Field
survey of sex-reversals in the medaka, Oryzias latipes: genotypic sexing of wild
populations. Zool Sci 21:613-619.
Page 51 of 124
-------
Final July 2015
APPENDIX 6: Subadult Sampling - Necropsy
(a) Background. In the MEOGRT after the dmy gene assessment and breeding pairs
have been selected, the remaining subadult fish are humanely killed and used to assess growth,
secondary sex characteristics (SSC), gonad phenotyping, and liver vitellogenin mRNA
quantification. It is essential that the dmy status of these fish is maintained. If this information is
lost during the selection of the breeding pairs, fresh tissue should be taken to repeat the DNA
extraction and dmy analysis.
(b) Major Materials and Reagents.
• Dissecting microscope (with optional camera attached)
• Dissection tools including scissors
(c) Protocol
(1) Anesthesia. Place small numbers of fish in a solution of dilution water and buffered
tricane methanesulfonate (MS-222) at 100 mg/L.
(2) Growth. Measure and record length to the nearest 0.1 mm and weight to the nearest mg.
(3) Necropsy (Consult Figure 1). The first cut is a transverse cut done with high quality
dissecting scissors between the operculum and the connection site of the pectoral fins. This cut
is normally offset toward the pectoral fin. With experience, the cut can be made so that the heart
and the mesentery between it and the viscera stays with the head, while the liver is exposed at the
cranial opening of the carcass. The liver is separated from any connective tissue while it is
teased away from the rest of the viscera through the opening made by the first cut (Figure 2).
A second cut is made caudal to the vent. This isolates the anal fin from the carcass. The number
of anal fin papillae or SSC will be counted on the anal fin from each fish. At this time, an image
of appropriate quality can be taken through a dissecting microscope of the anal fin or the anal fin
can be preserved in an appropriate fixative like Davidson's (Note: Bouin's fixative is not suited
for this purpose). If fixation is done, care should be taken so that the fin remains flat to allow for
accurate counting. Regardless of the method, the number of anal fin papillae will be counted at a
later date.
A third cut is made on the tail fin to preserve some tissue for possible dmy gene analysis. The fin
tissue should be immediately snap frozen in liquid nitrogen and kept frozen until processing.
The result of the grossing procedure is a dissected liver for vitellogenin (Vtgl) mRNA analysis, a
head that can be discarded, a tail with the anal fin for counting of the anal fin papillae, and a
portion of the tail fin for possible DMY analysis (if needed at a later date).
Page 52 of 124
-------
Final July 2015
Figure 1. Cuts made during necropsy of subadult samples.
Figure 2. Dissection of the liver at the subadult sampling.
Page 53 of 124
-------
Final July 2015
APPENDIX 7: Counting Anal Fin Papillae
(a) Background. This method is used to count the number of anal fin papillae. Under
normal circumstances, only sexually mature male medaka have papillae, which develop on the
joint plates of certain anal fin rays as a secondary sexual characteristic, providing a potential
biomarker for exposure to an endocrine disrupting chemical.
(b) Major Materials and Reagents.
• Dissecting microscope (with optional camera attached)
• Fixative (if not counting from image)
(c) Protocol. After the necropsy procedure (described in Appendix 5), the anal fin
should be imaged to allow for convenient counting of anal fin papillae. While imaging is the
recommended method, the anal fin can be fixed with Davidson's fixative or other appropriate
fixative (Bouin's fixative is not recommended) for approximately 1 minute. It is important to
keep the anal fin flat during fixation to allow for easier counting of papillae. The carcass with
the anal fin can be stored in Davidson's fixative or other appropriate fixative until analyzed.
Count the number of joint plates with papillae (see Figure 1) which protrude from the posterior
margin of the joint plate.
Joint Plates
Figure 1. Anal fin papillae.
Page 54 of 124
-------
Final July 2015
APPENDIX 8: Example Protocol for RNA Extraction from Medaka Livers
(a) Background. In the MEOGRT, subadult medaka are humanely killed to measure
various endpoints. One of these endpoints is the number of copies of the vitellogenin 1 gene
(Vtgl) per ng of total RNA in the liver determined by quantitative PCR (QPCR). QPCR is
highly sensitive and has a large dynamic range (-7 orders of magnitude), making it an ideal tool
for efficiently assessing concentrations of a target gene, in this case, the Vtgl gene. RNA is
extracted from the liver of subadult medaka. The following protocol is intended to provide
guidance for the proper extraction and handling of hepatic RNA in support of the MEOGRT.
The protocol described is an example procedure used for Vtgl RNA extraction from subadult
Japanese medaka livers using real-time PCR. This protocol is based on procedures used at the
U.S. Environmental Protection Agency laboratory in Duluth, MN. Specific products and/or
equipment listed can be substituted with comparable materials.
(b) Major Materials and Reagents.
• RNA extraction kit (RNeasy Protect Mini Kit,Qiagen)
• Microcentrifuge
• Microcentrifuge tubes (2.0 or 1.5 ml)
• Spectrophotometer (NanoDrop ND-1000,Thermo Scientific)
(c) Protocol.
(1) Source Tissue. Liver tissue from subadult medaka (see procedure in Appendix 5) is the
source for the RNA extraction. The source tissue for the RNA extraction protocol has been
preserved in RNA/ater (Qiagen) prior to use.).
(2) RNA Extraction using RNeasy Protect Mini Kit (Qiagen). Follow the manufacturer's
instructions to extract RNA.(RNeasy Mini Handbook 4th edition, September 2010) with the
following exception. Disruption of a RNAIa/er preserved liver is done in a 2 ml microcentrifuge
tube with the liver, 600 |il of Buffer RLT, and a moderate number (-20 to 100) of 0.5 mm glass
beads (BioSpec Products). Disruption is completed by vortexing for 5 minutes on a Disruptor
Genie (Scientific Industries). This homogenate is centrifuged for 3 minutes at full speed and the
supernatant is pipetted into a new microcentrifuge tube. At this point, the manufacturer's
instructions are followed exactly.
(3) Assessment of the Quality and Concentration of the Extracted RNA. The
concentration (number of copies of the Vtgl gene) and quality of the extracted RNA should be
determined. These analyses are performed using a spectrophotometer (e.g., NanoDrop ND-1000,
Thermo Scientific). The spectrophotometer will provide a concentration in ng of RNA per (j.1 of
the sample and ratios of the absorbance at 260/280 nm and 260/230 nm. The 260/280ratio
should be -2.0 and the 260/230 ratio should be 2.0 - 2.2. If either ratio is appreciably lower it
may indicate the presence of contaminants. In this case, the sample should be re-purified using
the RNeasy Mini Kit (or similar product) or other purification protocol to remove the
contamination.
Page 55 of 124
-------
Final July 2015
APPENDIX 9: Example Protocol for the Preparation of RNA for Standard Curve Used
during Vtgl QPCR
(a) Background. In the MEOGRT, subadult Japanese medaka are humanely killed to
assess various endpoints. One of these endpoints is the quantification of vitellogenin in the liver.
One suggested method is to determine the number of copies of the vitellogenin 1 gene (Vtgl;
GenBank: AB064320.1) per ng of total RNA in the liver. The example procedure in this
Appendix using quantitative PCR (QPCR). QPCR is highly sensitive and has a large dynamic
range (~7 orders of magnitude), making it an ideal tool for efficiently assessing concentrations of
a target gene, in this case, the Vtgl gene.
The protocol described is an example procedure using quantitative PCR (QPCR). This protocol
is based on procedures used at the U.S. Environmental Protection Agency laboratory in Duluth,
MN. Specific products and/or equipment listed can be substituted with comparable materials.
For absolute quantification, RNA that contains the Vtgl sequence targeted by the primers and is
only minimally larger than this amplicon should by synthesized. Ideally this RNA would have
the identical sequence and number of base pairs as the amplicon produced by the primers used
for Vtgl QPCR. The intent is for this "standard RNA" to be replicated with the same efficiency
as the target sequence (Vtgl mRNA) extracted from livers during the test protocol. This ideal
RNA would be amplified during QPCR at an identical rate as the target sequence in actual
samples. In practice, primers were designed just outside the Vtgl amplicon that would produce
cDNA that is complimentary to the Vtgl target sequence. In addition, the forward primer
contained the T7 promoter sequence which allows for use of a RNA polymerase that recognizes
the T7 promoter sequence to synthesize RNA copies of this cDNA template.
The concentration of the synthesized Vtgl mRNA and its size in base pairs are used to calculate
the number of copies of the Vtgl target sequence present per unit volume. The Vtgl mRNA is
serially diluted to a known copy number per unit volume ([jl). The dilution series is used to
produce a standard curve, which can then be used determine the absolute copy number of Vtgl
mRNA in each subadult liver sample (Appendix 9).
In the following protocol, the use of specific reagents and equipment are detailed. While these
have been verified to work appropriately, reagents and equipment from other manufacturers can
be substituted as long as their appropriateness is verified.
(b) Major Materials and Reagents.
• RNase/DNase free water (Qiagen)
• NanoDrop ND-1000 (Thermo Scientific)
• Superscript II (Invitrogen)
• AmpliTaq Gold (Applied Biosystems)
• MEGAScript (Ambion AM1333)
• Bioanalyzer 2100 (Agilent Technologies)
Page 56 of 124
-------
Final July 2015
(c) Protocol.
(1) Source RNA. Extracted RNA from livers of adult female Japanese medaka (see
example procedure in Appendix 5) that are successfully spawning. Ensure that the extracted
RNA is of good quality and has a concentration greater than 200 ng /pi.
(2) Synthesizing cDNA from liver RNA. Adjust the RNA concentration to 200 ng/pl
with RNase/DNase free water. It is recommended that multiple samples of RNA be prepared in
case of problems occurring during the procedure. Prepare the following reaction mix shown in
Table 1
Table 1. Reaction Mix.
Reagent
Per Sample Volume |
RNase/DNase free water
3 ul
RNasing® (Promega)
1 pi j
Oligo d(T) (Promega)
1 (J |
5x RM buffer (in Superscript® II Kit, Invitrogen)
4 ul [
0.1M DDT (in Superscript® II kit, Invitrogen)
2 ul J
dNTP (1 OmM. Promega U 15 1 1)
4 ^ 1
Reverse transcriptase (Superscript® II, Invitrogen)
1 pi |
Proceed with reverse transcription for 1 hour at 37°C. Each sample consists of 16 j_il of the
above reaction mix + 4 jal RNA at 200 ng/pl from above. The result is cDNA from the mRNA in
the RNA samples. Since Vtgl mRNA is highly represented in the RNA samples, the cDNA
produced from each of those samples will be well populated with Vtgl cDNA. This procedure
is the reverse transcription step of RT-PCR.
(3) Amplifying the Vtgl cDNA. The cDNA produced above is a cDNA representation
of the various mRNA transcripts that were present in the original female liver RNA samples.
The next step is to selectively amplify the VTG1 cDNA, that is, to perform the amplification step
of RT-PCR with primers specific for VTG1. In addition, the T7 promoter sequence (TAA TAC
GAC TCA CTA TAG GGA GA) for T7 RNA polymerase is also included on the 5' end of the
forward primer so that in future steps the amplified VTG1 cDNA can be transcribed into RNA.
The amplification step is done with the following reaction mix (Table 2) and thermal cycler
profile (Table 3).
Page 57 of 124
-------
Final July 2015
Table 2. Amplification Reaction Mix.
J Reagent
Per Sample Volume
j RNase/DNase free water
25.36 (.il
lOx RM buffer (from AmpliTaq Gold, Applied Biosystems)
4 jxl
| MgCh (from AmpliTaq Gold)
2.24 ^1
Vtgl Standard Forward Primer
I (TA ATACG ACTC ACTATAGGG AG ATTCCTCGG ATACGGC AC A AT)
2 jxl
Vtgl Standard Reverse Primer
(TAGACAGCTTTGCTGTAACGTAAGC)
2 jxl
J AmpliTaq Gold (Applied Biosystems,)
0.4 ^1
Note: Underlined is the T7 promoter sequence.
Table 3. Thermal Cycler Profile.
Step
Temperature
Time
1
95°C
10 minutes
2
94°C
30 seconds
3
58°C
1 minute
4
72°C
1 minute
5
Go to step 2, run 34
cycles
(4) Convert Vtgl cDNA to Vtgl RNA. The amplified product from the RT-PCR is
used as template to synthesize RNA via RNA polymerase. Specifically, T7 RNA polymerase,
which recognizes the T7 promoter sequence that has been incorporated into the Vtgl amplicon,
synthesizes RNA that is nearly identical to the target sequence in the Vtgl mRNA. Therefore the
amplification efficiency of sample RNA from subadult medaka liver will be the same as the
amplification efficiency of the RNA used for the standard curve.
To convert the cDNA to RNA, first prepare the following reaction mix (Table 4).
Page 58 of 124
-------
Final July 2015
Table 4. Reaction Mix to Convert cDNA to RNA.
Reagent
i
| Per Sample Volume
cDNA (from above)
| 12 pi
lOx buffer (from MEGAScript)
| 2 jxl
lOmM ATP (from MEGAScript)
j 1 Ml
lOmM CTP (from MEGAScript)
1 1 Ml
lOmM GTP (from MEGAScript)
1 i (J
lOmM UTP (from MEGAScript)
11 Ml
RNA polymerase (Ambion MEGAScript, AM1333)
1 2 jxl
Incubate the reaction mix for 1 hour at 37°C. Add 1 (_il of DNase (Invitrogen) and incubate for
15 minutes at 37°C. Clean the RNA from free nucleotides, buffer components, and enzymes
with a MEGAclear Kit (Ambion) following the manufacturer's instructions.
(5) Determine the quantity and quality of the RNA. Quantify the RNA on a
NanoDrop ND-1000 (Thermo Scientific) following the manufacturer's instructions. This will
provide a concentration in jag/(_il. Calculate the number of copies per jj.1 using the following
equations (Roche Molecular Biochemicals Technical Note #LC 11/2000) shown in Table 5.
Table 5. Calculation of Number of dsDNA, ssDNA, or ssRNA Copies
For average molecular weight of:
Use this calculation
dsDNA
(number of base pairs) X (660 daltons/base pair)
ssDNA
(number of base pairs) X (330 daltons/base pair)
ssRNA
(number of base pairs) X (640 daltons/base pair)
(d) Example Calculation (For Demonstration Purposes Only):
General Formula:
MW = molecular weight [g/mol]
1 mol = 6 X 1023 molecules (= copies)
, , „?3 rcopiesl ^ .. fa ]
6 x lus-: —H x concentration revmrpvi
mot J L'U« ^ copies]
? p: i = amount —
Example: For CycA plasmid that has a concentration of 152 ng/pl = 1.52 X 10 7 g/pl
and a total size of 3,397 bp:
Page 59 of 124
-------
Final July 2015
MW = 3,397 bp X 660 daltons/bp = 2.24 X 106 daltons
1 mol = 2.24 X 106 g
1 mol = 6 X 1023 molecules (= copies)
23
6 x 10
"copies'
— 7
X 1.52 x 10 M
.14 i
m o!
4 X 101C
2.24 x 10
— Li
~ [mo
copies
ui
The quality of the RNA should be determined (Bioanalyzer 2100; RNA 6000 Nano Kit, Agilent
Technologies or equivalent) following the manufacturer's instructions. The Vtgl standard RNA
should be 612 base pairs. As a rule of thumb, the RNA should have a RNA Integrity Number
(e.g., calculated by the Bioanalyzer software) of ~7 or greater.
Dilute the RNA based upon its concentration in copies per j_il to produce a series that is diluted a
factor of 10 between each step, for instance 1010' 109, 108, etc. To do this, since the copy
numbers for the standards in this example are based upon loading 4 jj.1 of standard RNA into
each well, the standard that has 1 x 1010 copies per well (STDio) will have a concentration of 2.5
X 109 copies/ [jl. For instance, if the standard RNA has a concentration of 2.3 X 1011 copies/[jl
then this would be diluted to 2.5 x 109 copies/[ji by adding 91 |il of RNase/DNase free water to
every 1 j_il of the standard RNA. The other standards (typically at least 7 dilution steps) are
serially diluted from STDio with the following formula maintained: STDn= 2.5 X 10X"'
copies/[ji.
Page 60 of 124
-------
Final July 2015
APPENDIX 10: Quantifying Vtgl mRNA in Subadult Liver Samples using QPCR
(a) Background. In the MEOGRT, subadult Japanese medaka are sacrificed for the
assessment of various endpoints. One of these endpoints is the quantity of liver vitellogenin
which can be measured as the number of copies of the vitellogenin 1 gene (Vtgl; GenBank:
AB064320.1) per ng of total RNA in the liver determined by quantitative PCR (QPCR). QPCR
is highly sensitive and has a large dynamic range (~7 orders of magnitude), making it an ideal
tool for efficiently assessing concentrations of a target gene such as Vtgl.
The following procedure is an example of precise quantification of Vtgl mRNA in subadult liver
samples using Taqman®-based QPCR. Other suitable methods may be used. In this example,
Vtgl mRNA extracted from the liver of subadult Japanese medaka (sample procedure in
Appendix 7), is amplified using QPCR and quantified using a Taqman® protocol.
The following is a brief summary of the Taqman® protocol. Starting at the designed primers,
DNA polymerase moves downstream synthesizing a new strand from the template strand and via
its 5' exo-nuclease activity, the polymerase removes any bases that would impede its progress
down the template strand. A Taqman® assay (5' nuclease assay) uses this 5' exo-nuclease
activity in real time detection of PCR product. A Taqman® probe, present in the PCR mix,
anneals to the template between the two primers in the path of the DNA polymerase as it
progresses on the template strand. The Taqman® probe is a short sequence of DNA (-20 bp)
with a reporter fluorophore on one end and a quencher on the other end. The close proximity of
these two moieties allows interaction prohibiting the fluorescence of the reporter and thus its
detection upon excitation within the real time instrument. However when the DNA polymerase
cleaves the probe, both reporter and quencher are released into solution, and the distance
between the two increases allowing the reporter to fluoresce upon excitation. The quantity of
fluorescence from the reporter during the exponential phase of PCR is directly proportional to
the amount of target sequence in the sample. During PCR, amplicon production goes through
three phases: (1) the exponential phase where doubling of the amplicon is occurring every cycle
and the reaction is very specific and precise; (2) the linear phase where reagents are becoming
rate limiting, the reaction slows and amplicons may start to degrade; and (3) plateau phase where
the reaction has stopped with no additional product formed (Figure 1). Fluorescent
measurements are taken during the exponential phase during QPCR (real-time PCR), while with
traditional PCR, the reaction is often terminated in the plateau phase and amplicons are detected
using gel electrophoresis.
(b) Major Materials and Reagents.
• RNA from subadult liver samples
• Standard RNA
• RT PCR Reagents (e.g., Taqman EZ RT-PCR Reagents)
• Quantitative PCR thermocycler (e.g., ABI 7500,Applied Biosystems)
Page 61 of 124
-------
Final July 2015
(c) Protocol.
(1) Source RNA. There are two sources of RNA used in this example protocol: 1) RNA
extracted from subadult medaka (Appendix 7) and 2) RNA for the standard curve (Appendix 8).
In both cases, ensure that the RNA is of good quality as described in the specific protocol and
concentration is known (expressed as either ng/j_il or copies/[jl).
(2) QPCR. Prepare the following master mix keeping every reagent on ice as shown in
Table 1.
Table 1. QPCR Master Mix.
Taqman EZ RT-PCR Reagents
(Applied Biosystems)
(il/reaction
5x Taqman Buffer
6.0
Mn2+ acetate
3.6
dATP
0.9
dGTP
0.9
dCTP
0.9
dUTP
0.9
Vtgl Forward (10 (jM)
5'-AGGCAGTTTCTAAGGGCGAAC-3'
1.5
Vtgl Reverse (10 |iM)
5' -TGAATGGGC ATAATCTTTGTGATT-3'
1.5
Vtgl Probe (1250 nM)
5' -fam-TTTGGGAAATGCAAGACACCCTA-bhq-3'
4.8
rtTH polymerase
1.2
AmpErase™
0.2
RNase/DNase-free water
3.6
Total
26
Add 26 [jl of the master mix to each well and then 4 jal of the RNA sample from subadult liver or
standard RNA to each well. Unknown samples should be run minimally in duplicate while
standard RNA should be run in triplicate. In addition to samples and standards, appropriate
technical controls should be included. These should minimally include "no template controls"
and a liver mRNA sample that has previously been quantified. Other technical controls can be
included as warranted. The 96-well optical plate is sealed, vortexed, and centrifuged at -2000
rpm for ~ 5 minutes to remove air bubbles.
In this example, QPCR is performed on an ABI 7500 (Applied Biosystems) using the following
thermal profile (Table 2). An example plate layout is provided in Table 3.
Table 2. QPCR Thermal Profile.
Temperature
Time
Cycles
Temperature
Time
Cycles
50°C
2 minutes
95°C
15 seconds
45
60°C
30 minutes
1
60°C
1 minute
95°C
5 minutes
Page 62 of 124
-------
Final July 2015
Table 3. Example Plate Layout.
i
:
3
4
5
(.
"
S
l>
III
1 1
i:
\
NTC
STD3
STD3
STD3
Sample
3
Sample
3
Sample 11
Sample
11
Sample
19
Sample
19
Sample
27
Sample
27
i:
NTC
STD4
STD4
STD4
Sample
4
Sample
4
Sample 12
Sample
12
Sample
20
Sample
20
Sample
28
Sample
28
(
known
STD5
STD5
STD5
Sample
5
Sample
5
Sample 13
Sample
13
Sample
21
Sample
21
Sample
29
Sample
29
i)
known
STD6
STD6
STD6
Sample
6
Sample
6
Sample 14
Sample
14
Sample
22
Sample
22
Sample
30
Sample
30
i:
Sample
1
STD7
STD7
STD7
Sample
7
Sample
7
Sample 15
Sample
15
Sample
23
Sample
23
Sample
31
Sample
31
i;
Sample
1
STD8
STD8
STD8
Sample
8
Sample
8
Sample 16
Sample
16
Sample
24
Sample
24
Sample
32
Sample
32
(i
Sample
2
STD9
STD9
STD9
Sample
9
Sample
9
Sample 17
Sample
17
Sample
25
Sample
25
Sample
33
Sample
33
ii
Sample
2
STD10
STD10
STD10
Sample
10
Sample
10
Sample 18
Sample
18
Sample
26
Sample
26
Sample
34
Sample
34
NTC = no template control; known = liver RNA sample with known copies/[jl; STD3-10 =
standard RNA
(3) Interpretation of QPCR. The guidance provided here is not meant to supersede or
replace the manufacturer's instructions on the analysis of QPCR data. The system software used
in this example (ABI 7500 System Software) is usually able to automatically set both the
baseline range and the critical threshold (Ct) value. If the software is unable to do this
automatically, the appropriate values for these parameters will be entered manually into the
software. Once these parameters are set, the software will generate a standard curve based upon
the inputed copy number for each standard RNA and the corresponding Ct value. The software
will also calculate the copy number, the mean copy number of the replicated sample and the
standard deviation of the replicated sample for each unknown sample based upon its specific Ct
value. It is recommended that the CV% of each replicated sample is less than 10%. In addition,
there should be no amplification in the "no template control" samples and the known sample
should fall within +/- 10% of its running mean from previous analyses.
(4) Data ReportingAll raw data should be maintained. Assuming the standard deviation
of any specific sample was acceptable, the mean copy number of Vtgl mRNA per |il is recorded
for each specific sample. For each sample, the derived copy number per jol and the concentration
in ngl\A of the liver RNA is used to calculate the Vtgl copy number per ng of total RNA. In
addition, performers of this analysis should be familiar with the published MIQE guidelines
(Bustin et ol. 2009; Shipley 2011) and all pertinent information required by the MIQE guidelines
should be recorded and provided in the test report.
(d) References.
(1) Bustin, S.A., Benes, V., Garson, J.A., Hellemans, J., Huggett, J., Kubista, M., Mueller,
R., Nolan, T., Pfaffl, M.W., Shipley, G.L., Vandesompele, J., and Wittwer, C.T., 2009.
The MIQE guidelines: minimum information for publication of quantitative real-time
PCR experiments. Clin Chem 55: 611-622.
(2) Shipley, G., 2011. The MIQE Guidelines Uncloaked. Polymerase Chain Reaction
Troubleshooting and Optimization: The Essential Guide, 151.
Page 63 of 124
-------
Final July 2015
APPENDIX 11: Example Data Reporting Templates for use with StatCHARRMS
(a) Background. To expedite data reporting and statistical analysis, especially if Agency
provided statistical tools are used, it is recommended that common templates for data reporting
be used. These templates take the form of Excel spreadsheets that can easily be converted into
the CSV file format which is required by the Agency provided statistical tool (StatCHARRMS).
Example formats are provided below and electronic versions of all templates can be accessed at
the EPA EDSP website: http://www.epa.gov/gov/endo/.
(b) Example Data Templates for StatCHARRMS
(1) Template for Recording Fecundity and Fertility. Table 1 shows a portion of the
electronic data sheet for recording both fecundity and fertility that represents the complete
dataset for one replicate. All fields except "Study" are required to properly analyze the data.
Other fields containing metadata can be added as needed. To be entered into StatCHARRMS,
the data should be in the CSV format which can easily be converted within Excel.
Table 1. Example Spreadsheet Template for Recording Fecundity and Fertility Data
Study
Date
Treatment
Replicate
Fecundity
Fertile Eggs
PTOP 03
10/1/2012
1
A
7
7
PTOP-03
10/2/2012
1
A
7
7
PTOP 03
10/3/2012
1
A
8
8
PTOP-03
10/4/2012
1
A
11
11
PTOP 03
10/5/2012
1
A
13
13
PTOP-03
10/6/2012
1
A
15
15
PTOP 03
10/7/2012
1
A
8
8
PTOP-03
10/8/2012
1
A
16
14
PTOP 03
10/9/2012
1
A
21
21
PTOP-03
10/10/2012
1
A
15
15
PTOP 03
10/11/2012
1
A
16
16
PTOP-03
10/12/2012
1
A
13
13
PTOP 03
10/13/2012
1
A
12
12
PTOP-03
10/14/2012
1
A
12
9
PTOP 03
10/15/2012
1
A
11
10
PTOP-03
10/16/2012
1
A
24
23
PTOP 03
10/17/2012
1
A
20
18
PTOP-03
10/1^/2012
1
A
20
20
PTOP 03
10/19/2012
1
A
16
16
PTOP-03
10/20/2012
1
A
17
16
PTOP 03
10/21/2012
1
A
8
5
(2) Example Template for Recording Endpoint Information. Table 2 illustrates a
portion of the electronic data sheet that can be used for recording endpoint information.
Additional fields for metadata can be added as necessary. An electronic version is available at
the EPA EDSP website: http://www.epa.gov/endo/.
Page 64 of 124
-------
Final July 2015
Table 2. Example Template for Recording Endpoint Information.
Study
ID
Treatment
Generation
Genotypic Sex
Vtg
ssc
Age
Weight
Length
PTOP
T1A1
1
F1
Male
1.50E+03
86
8 Week
183
22
PTOP
T1A3
1
F1
Female
6.60E+06
0
8 Week
182
23
PTOP
T1A4
1
F1
Female
1.20E+06
0
8 Week
196
24
PTOP
T1A5
1
F1
Male
2.70E+03
61
8 Week
199
21
PTOP
T1A6
1
F1
Male
8.50E+02
59
8 Week
192
23
PTOP
T1A7
1
F1
Male
1.10E+03
80
8 Week
181
22
PTOP
T1A8
1
F1
Male
1.20E+03
38
8 Week
115
25
(3) Example Template for Recording Pathology Information. Table 3 illustrates a
portion of the electronic data sheet that may be used for recording pathology information. Only
two fields of pathology diagnoses are shown below as examples. There are more potential
pathologies provided in the histopathology guidance provided in Appendix 11. Additional fields
for metadata can also be added as necessary.
Table 3. Example Template for Recording Pathology Information.
ID
Gen
Treatment
Rep
DMY
Age
Gon Phenotype
Gon Stage
T1A1
F0
1
A
Male
16 wk
1
3
T1A2
F0
1
A
Female
16 wk
5
3
T1B1
F0
1
B
Male
16 wk
1
3
T1B2
F0
1
B
Female
16 wk
5
3
T1C1
F0
1
C
Male
16 wk
1
2
T1C2
F0
1
C
Female
16 wk
5
2
T1D1
F0
1
D
Male
16 wk
1
2
T1D2
F0
1
D
Female
16 wk
5
3
T1E1
F0
1
E
Male
16 wk
1
1
T1E2
F0
1
E
Female
16 wk
5
3
T1F1
F0
1
F
Male
16 wk
1
2
T1F2
F0
1
F
Female
16 wk
5
3
Page 65 of 124
-------
Final July 2015
APPENDIX 12: Histopathology Guidance for the Medaka Extended One-Generation
Reproduction Test
(a) Introduction. The goal of this document is to serve as general guidance for the
collection, histological preparation, and pathological evaluation of gonads, kidney, and liver
specimens from Japanese medaka (Oryzias latipes) in support of the Medaka Extended 1-
Generation Reproduction Test (MEOGRT), a Tier 2 assay of the EPA's Endocrine Disruptor
Screening Program (EDSP). Tier 2 tests include endpoints that have been designed to provide
insight into the adverse outcome pathways (AOP) of test agents. The Agency is providing
histopathology guidance to help ensure that histological procedures and pathological evaluations
are performed accurately and consistently. This histopathology guidance is based on procedures
developed by the U.S. Environmental Protection Agency in consultation with other scientific
experts. Specific products and/or equipment listed can be substituted with comparable materials.
Laboratories may depart from some aspects of this guidance due to variability in some standard
laboratory practices (e.g., how much information is included on slide labels, etc.).
This document is divided into three general sections: Necropsy Procedures, Histology
Procedures, and Pathology Evaluation. The Pathology Evaluation section includes written
descriptions and illustrations of normal tissues and abnormal changes, with special emphasis on
findings that are likely related to endocrine disruption, and specific examples of lesion severity
grades as applicable. Additional guidance is provided on the topics of severity grading (in
general), data recording, statistical analysis, data interpretation, and report formatting.
(b) Necropsy Procedures. At the conclusion of the exposure, fish are anesthetized by
transfer to an oxygenated solution of MS-222 (100 mg/L buffered with 200 mg NaHC03/L) for
sampling. If potency of the solution is not adequate, additional MS-222 (<10 mg) may be added
to strengthen effectiveness.
If other observations and measurements are to be made at this time (e.g., length and body weight,
blood collection, and evaluation of secondary sex characteristics), these tasks should be
performed rapidly in order to avoid tissue autolysis which occurs rapidly in fish. To further
avoid autolysis caused by delays in data collection, the number of fish euthanized at any one
time should be kept to a minimum.
Prior to fixation, the tail is excised caudal to the abdominal cavity. Using a #11 blade or micro
scissors, a slit incision approximately 3 mm in length is made carefully through the ventral
abdominal wall to allow penetration of viscera by the fixative. Each fish is placed in an
individual tissue cassette along with its corresponding identification label, and the cassette is
placed in modified Davidson's solution for 24-48 hours. After the initial fixation period, the
tissues are rinsed thoroughly in 70% ethanol, after which they may be stored in 10% neutral
buffered formalin prior to histological processing.
(c) Histology Procedures.
(1) Decalcification. Decalcification of the specimens is usually not required due to the
presence of acetic acid in the modified Davidson's fixative. However, it is usually prudent to
Page 66 of 124
-------
Final July 2015
process and microtome a few control specimens ahead of the rest to ensure that decalcification is
complete. If further decalcification is necessary, specimens may be immersed in a commercial
formic acid/EDTA decalcifying solution for a short interval (e.g., several hours or overnight)
prior to processing.
(2) Processing and Embedding. Each whole fish specimen (i.e., minus the tail) is
processed in an automated tissue processor and infiltrated with paraffin according to routine
methods (Figure 1).
•JnWJ]
Figure 1. Fish are embedded in paraffin to allow sectioning in the parasagittal / sagittal plane, with the
left side being cut first. The cassette should include an appropriate label.
(3) Microtomy. Section thicknesses is set at 4-5 microns. Each fish is step-sectioned in
the parasagittal / sagittal plane at five distinct levels (Figure 2). Each of the five sections
acquired in this fashion will be placed on a single slide. A duplicate set of unstained sections is
obtained at each of the five levels; these will be placed on five additional slides. Specific
landmarks for each of the five levels are illustrated in Figures 3-7.
Figure 2. Each fish is step-sectioned in the parasagittal/sagittal plane at five distinct levels.
Eachsection is 4-5 microns thick. Each of the sections will be placed on a single slide.
Page 67 of 124
-------
Final July 2015
Figure 3. Level 1: The block is faced from the left toward the right side of the fish. Sections are
trimmed away until the left eye is revealed, and then trimming continues to the mid-portion of the
lens. The lens will be visible as a ring within the eye. The ring can be seen in the sections and the
block. The lens will be hard, the microtome blade will produce a scratch that can be heard and felt as
the blade cuts through. Sections acquired at this level should reveal the visceral cavity. A ribbon of
3-4 serial sections is obtained and mounted on a single slide. A second ribbon of 3-4 sections should
be obtained and mounted on a second slide. The first slide will be stained and the second slide will be
left unstained for possible future reference. This will be done for all sectioning levels.
Figure 4. Level 2: Trimming is continued until the left eye is no longer present in the sections, and
the dark brown pigment of the retinal epithelium has diminished. A ribbon of 3-4 serial sections is
obtained at that point and mounted on a single slide. A second ribbon of 3-4 sections should be
obtained and mounted on a second slide. The first slide will be stained and the second slide will be
left unstained for possible future reference.
Page 68 of 124
-------
Final July 2015
Figure 5. Level 3: The target organ for this slide is the pituitary gland, which is located at a level
that is midway between the eyes. Trimming is continued until the brain begins to elongate, leading
into the spinal cord (arrow). Four step sections are then obtained at 200 micron intervals, collecting
one good quality section at each step (sectioning halts before the right eye is reached). All four
sections are mounted on a single slide m the order they were obtained. A second ribbon of 3-4
sections should be obtained and mounted on a second slide. The first slide will be stained and the
second slide will be left unstained for possible future reference.
Figure 6. Level 4: Trimming is continued to the medial edge of the right eye, where the dark brown
retinal epithelial pigment is visible. A ribbon of 3-4 serial sections is obtained at that point and
mounted on a single slide. A second ribbon of 3-4 sections should be obtained and mounted on a
second slide. The first slide will be stained and the second slide will be left unstained for possible
future reference.
Figure 7. Level 5: Trimming is continued to the midpoint of the right lens, where light can be seen
through the lens. A ribbon of 3-4 serial sections is obtained at that point and mounted on a single
slide. A second ribbon of 3-4 sections should be obtained and mounted on a second slide. The first
slide will be stained and the second slide will be left unstained for possible future reference.
Following microtomy, each paraffin block is sealed with paraffin.
Page 69 of 124
-------
Final July 2015
(4) Staining and Coverslipping. Slides destined for staining are stained with
hematoxylin and eosin, and are covered with glass cover slips using an appropriate permanent
mounting medium.
(5) Labeling. Slides can be labeled with the following information:
• Study number
• Name of the test chemical
• Generation and age of the specimen (i.e., Fl, 16 wpf)
• Treatment (concentration) or Control
• Individual animal identification number
(d) Pathology Evaluation.
(1) General Approach to Pathologic Evaluations. Studies are to be read by
individuals experienced in reading toxicologic pathology studies, and who are familiar with
normal, small fish gonad histology, with gonadal physiology, and with general responses of the
gonads to toxicologic insult. Pathologists may be board certified (e.g. American College of
Veterinary Pathologists, The European Centre of Toxicologic Pathology, or other certifying
organizations); however, certification is not a requirement as long as the pathologist has obtained
sufficient experience with, and knowledge of, fish histology and toxicologic pathology.
Technicians should not be used to conduct readings due to the subtle nature of some changes and
the need for subjective judgments based on past experience.
It is recognized that there is a limited pool of pathologists with the necessary training and
experience that are available to read the gonadal histopathology for the MEOGRT assay. If an
individual has toxicological pathology experience and is familiar with gonadal histology in small
fish species, he/she may be trained to read the fish assay. If pathologists with little experience
are used to conduct the histopathological analysis, informal peer review may be necessary.
Pathologists generally read slides unblinded (i.e., with knowledge of the treatment group status
of individual fish) first and then they should be read blinded. This is because endocrinological
effects on histomorphology tend to be incremental, and subtle differences between exposed and
unexposed animals may not be recognizable unless tissue sections from high dose animals can be
knowingly compared to those from controls. Thus the aim of the initial evaluation is to ensure
that diagnoses are not missed (i.e., to avoid false-negative results). Re-evaluating the treatment-
finding by a pathologist in a blinded manner will prevent the reporting of false-positive results.
As general practice, .the control and high concentration samples are read first and then others are
read using a step-down procedure.
Pathologists should specifically evaluate the target tissues identified in the guidelines; however,
changes observed in other tissue types may also be recorded. This especially pertains to findings
suspected to be treatment-related, or findings that might otherwise impact the study results (e.g.,
systemic inflammation or neoplasia).
Page 70 of 124
-------
Final July 2015
It is suggested that the pathologist be provided with all available information related to the study
prior to conducting their evaluations. Information regarding gross morphologic abnormalities,
mortality rates, and general test population performance and health are useful for pathologists to
provide comprehensive reports and to aid in the interpretation of findings. For a more
comprehensive discussion of standard reading approaches for toxicologic pathology studies,
please refer to the Society of Toxicologic Pathology Best Practices for reading toxicologic
histopathology studies (Crissman et ol., 2004).
(2) Severity Grading. In toxicologic pathology, it is recognized that compounds may
exert subtle effects on tissues that are not adequately represented by simple binary (positive or
negative) responses. Severity grading involves a semi-quantitative estimation of the degree to
which a particular histomorphologic change is present in a tissue section (Shackelford et al.,
2002). The purpose of severity grading is to provide an efficient, semi-objective mechanism for
comparing changes (including potential compound-related effects) among animals, treatment
groups, and studies.
Severity grading should usually use the following system:
• 0 (not remarkable)
• Grade 1 (minimal)
• Grade 2 (mild)
• Grade 3 (moderate)
• Grade 4 (severe)
Findings that are not present are not graded and assigned a zero (0) to represent the tissue section
being not remarkable. This is not to mean "Grade 0." This practice provides continuity with
subsequent statistical analyses.
A grading system needs to be flexible enough to encompass a variety of different tissue changes.
In theory, there are three broad categories of changes based on the intuitive manner in which
people tend to quantify observations in tissue sections:
Discrete: these are changes that could be readily counted. Examples include atretic follicles,
oocytes in the testis, and clusters of apoptotic cells.
Spatial: these are changes that could be quantified by area measurements. Includes lesions that
are typically classified as focal, multifocal, coalescing, or diffuse. Specific examples include
granulomatous inflammation and tissue necrosis.
Global: these are generalized changes that would usually require more sophisticated
measurement techniques for quantification. Examples include increased hepatocyte basophilia,
thyroid follicular cell hypertrophy, or quantitative alterations in cell populations.
Listed below are general guidelines for the use of a severity grading system, with examples of
how the system could be applied to each of the above categories. Please understand that the
terms Discrete, Spatial, and Global are used for illustrative purposes only; it is not intended that
Page 71 of 124
-------
Final July 2015
these terms be incorporated into any diagnosis or grade. It should be stressed that the examples
below should be modified as needed for each particular type of change (diagnosis).
Grade 1:
• Discrete change example: 0 to 2 occurrences per microscopic field, or 1 to 2 occurrences
per tissue section.
• Spatial change example: the change occupies a miniscule area of either a specific tissue
type or the entire tissue section.
• Global change example: the least perceptible alteration relative to control animals or prior
experience.
Grade 2:
• Discrete change example: 3 to 5 occurrences per microscopic field or tissue section.
• Spatial change example: the change occupies a larger area than Grade 1, but still less than
or equal to 25% of either a specific tissue type or the tissue section.
• Global change example: the alteration is easily appreciated, but still not dramatic.
Grade 3:
• Discrete change example: 6 to 8 occurrences per microscopic field or tissue section.
• Spatial change example: the change occupies more than 25% but less than or equal to
50% of either a specific tissue type or the entire tissue section.
• Global change example: the alteration is dramatic, but a more pronounced alteration can
be envisioned.
Grade 4:
• Discrete change example: 9 or more occurrences per microscopic field or tissue section.
• Spatial change example: the change occupies more than 50% of either a specific tissue
type or the entire tissue section.
• Global change example: essentially, the most pronounced imaginable alteration.
At least some of the histomorphologic changes that have been associated with chemicals that
may cause endocrine disruption in fish are considered to be exacerbations of "normal",
physiologic findings (e.g., discussion of oocyte atresia in Nagahama, 1983 and Tyler and
Sumpter, 1996). Whenever possible, the severity of a given change should be scored relative to
the severity of the same change in concurrent control animals. For each important (i.e.,
treatment-associated) finding, the severity scoring criteria should be stated in the Materials and
Methods section of the pathology narrative report. By convention, it is recommended that
severity grading should not be influenced by the estimated physiologic importance of the change.
For example, the presence of two oocytes in the testis should not be graded as "severe", even if
the pathologist considers this finding to be highly significant in terms of endocrine modulation.
The reason is that estimating the physiologic importance adds a further layer of subjectivity to
the findings that complicates interlaboratory results comparisons.
(3) Data Recording. An example template for recording histopathology data for use
with StatsCHAARM is provided in Appendix 10. For each fish, the pathologist records the
Page 72 of 124
-------
Final July 2015
presence of a diagnosis by indicating the severity grade. In rare instances (e.g., tumor
diagnoses), severity grading may not be applicable. If there are no findings for a fish, this should
be recorded specifically. It is also important to record a notation if the target tissue is missing or
if the amount of tissue present is insufficient to make a diagnosis. Adding modifiers to a
diagnosis may help to further describe or categorize a finding in terms of chronicity, spatial
distribution, color, etc. In many instances, modifiers are superfluous or redundant (e.g., fibrosis
is always chronic); therefore, the use of modifiers should be kept to a minimum. An
occasionally important modifier for evaluating paired organs is unilateral; unless specified in this
manner, all diagnoses for paired organs are assumed to be bilateral. Other modifiers can be
created sparingly as needed by the pathologist.
(4) Statistical Analysis. Histopathology data are analyzed using the Rao-Scott Cochran-
Armitage by Slices, or RSCABS method (Green et al., 2013). Advantages of using RSCABS as
a statistical method for analyzing histopathology data include the ability to account for: 1)
experimental designs with multiple replicates, 2) lesion severity scores of individual animals in
addition to group-wise lesion prevalence, and 3) dose-response relationships. Additionally, the
RSCABS test is easy to perform and interpret.
(5) Data Interpretation. Once the microscopic examinations have been completed and
statistical analyses have been performed on the resulting data, the pathologist interprets the
histopathologic findings. The initial task is to determine which, if any, of the recorded findings
are related to administration of the test article, and which are not. The goal is to classify each
type of recorded finding (i.e., diagnosis) into one of three categories: 1) Treatment-related, 2)
Potentially treatment-related and 3) Non-treatment-related. Criteria for these determinations
are listed below.
(i) Determining Relationship to Treatment. A weight-of-evidence (WOE) approach is
used to determine if a particular finding should be considered treatment-related. Such evidence
may include any or all of the following as available:
• Differences between groups of control and treated animals in terms of lesion prevalence
and severity, utilizing statistical analytical results to test for significance as warranted.
• Ancillary data from the current study, involving information such as behavioral
observations, organ and body weights, secondary sex characteristics, genotypic sex,
reproductive performance data, and biochemical analyses (e.g., reproductive or thyroid
hormones, vitellogenin).
• Results from other submitted or pending agency studies.
• The at-large scientific literature, giving greater weight to studies in which the quality of
the research can be established and is considered superior.
• Overall biological, physiological, and toxicological plausibility.
Findings that are considered potentially treatment-related may be those that have borderline
statistical significance, or those in which the relationship to treatment is considered equivocal for
other reasons (e.g., lack of corroborating evidence from other sacrifices or other studies,
biological or toxicological implausibility, or commonality of the diagnosis as a background
finding).
Page 73 of 124
-------
Final July 2015
There are several points to be made regarding the determination of treatment-relatedness. First,
it is possible for a finding to be treatment-related but not be caused by the test article. This can
include situations in which group-wise differences may be associated with an uncontrolled (and
possibly unrecognized) variable involving conduct of the in-life assay, specimen preparation, or
some other non-systemic bias. Second, not all statistically significant differences are real, as a p-
value significance level of 0.05 allows for the probability that in 5% of cases the result occurred
by chance. Third, a finding may be statistically significant and not necessarily biologically or
toxicologically important. Fourth, in some instances, treatment-related findings may not be
statistically significant. For example, this can occur when at treatment induces a low frequency
of a lesion type that rarely occurs spontaneously.
(ii) Determining Relationship to Endocrine Disruption. A similar WOE approach can
be used to determine if a particular finding is likely to be endocrine-related; however, in this case
the WOE will more heavily depend on ancillary data, results of other assays, and the published
literature, including mechanistic studies where available.
(6) Report Format. The pathologist is responsible for deliverables that can include: 1)
Pathology Narrative Report, 2) Spreadsheet with recorded data (example provided in Appendix
10, and 3) digital image files of figures.
(i) Pathology Narrative. Each histopathology narrative report should contain at least
the first five of the following sections: Introduction, Materials and Methods, Results, Discussion,
Summary/Conclusions, References, Tables, and Figures. The Introduction section briefly
outlines the experimental design. The Materials and Methods section briefly describes
procedures used during the slide preparation and examination phases of the study. If specific
severity grading criteria were created for a particular finding, they should also be listed in this
section. The Results section should report findings that are: 1) treatment-related; 2) potentially
treatment-related; 3) non-treatment-related findings that are novel or unusual. Detailed
histomorphologic descriptions need only be included for findings that differ substantially from
diagnoses presented the Histopathology Atlas. It is intended that the Results section should be as
objective as possible {i.e., opinions and theories should be reserved for the Discussion section).
The Discussion section, which contains subjective information, should address relevant findings
that were reported in the Results section. Opinions and theories can be included in this section,
preferably backed by references from peer-reviewed sources, but unsupported speculation should
be avoided. The Summary/Conclusions section should encapsulate the most important
information from the Results and Discussion sections. The References section should include
only material that is cited specifically in the narrative report. A separate Tables section may not
be necessary if tables are embedded in the Results section. The Figures section should include
photomicrographic examples of treatment-related findings, plus unusual or noteworthy lesions.
The Figures section should include normal tissues for comparison, and digital images should be
taken at magnifications that clearly illustrate the salient features of the findings. Figures
embedded in the narrative should be in a universally readable compressed file format such as
JPEG.
Page 74 of 124
-------
Final July 2015
(ii) Spreadsheet. In addition to the recorded histopathology findings, a completed
spreadsheet (see guidance in Appendix 10) should indicate the animals from which figure images
were photographed, and the number of images obtained per photographed fish.
(iii) Figures. For record-keeping purposes, a complete set of unembedded and
unannotated photomicrographic figures should be included.
(7) Pathology Peer Review. Following the initial slide evaluation and creation of a
draft report by the study pathologist (SP), it is encouraged that at least a subset of the original
histologic sections be assessed by a second reviewing pathologist (RP). Known as pathology
peer review, the purpose of this exercise is to increase confidence in the histopathology data by
ensuring diagnostic accuracy and consistency. Commonly, this procedure involves the targeted
examination of one or more tissue types in which treatment-related findings were initially
detected (this helps to guard against false positive results), plus all tissues from a randomly
selected percentage (e.g., 10-20%) of animals from the control and high-dose groups (this helps
guard against false negatives). The RP is tasked with determining the accuracy and consistency
of diagnostic criteria, diagnostic terminology, severity grading, and the interpretation of findings.
The peer review can be performed in-house or (preferably) by an external pathologist, and
frequently the reviewing pathologist has at least equal or greater expertise than the SP.
Following the peer review, the SP and RP typically meet to resolve diagnostic differences. In
unusual cases in which such differences cannot be resolved, a panel of experts (Pathology
Working Group) may be convened to determine the final diagnoses. In addition to enhancing
confidence in the histopathology results, benefits of peer review may include decreased inter-
laboratory variability, and cross-training of pathologists (i.e., the initial study pathologist may
not always need to be an fish expert). Examples of recommended procedures for conducting
pathology peer reviews have been described elsewhere (e.g., Morton etal., 2010; The Society of
Toxicologic Pathologists, 1991; 1997).
(8) Atlas of Histopathologic Findings (Figures 8-49). The purpose of this section is to
provide: 1) to provide a common technical "language" for describing findings and 2) to create a
reference atlas of both normal microanatomical structures and potential pathological findings.
Listed alphabetically are a number of terms followed by working definitions or descriptions.
The information in this section is derived from a number of sources including scientific articles,
conference proceedings, related guidelines, toxicologic pathology textbooks, medical
dictionaries, and the personal experience of various fish pathologists.
Page 75 of 124
-------
Final July 2015
FigureS. Endocrine pancreas, islet cell lesions. A andB: Large islet (Brockman body) from a control medaka. It is not
uncommon to observe large or bizarre looking cells in normal islets. Bar = 25 urn. C and D: Islet cell carcinoma. Arrows
indicate the line of demarcation between the unaffected area of Brockman body (bb) and the islet cell carcinoma (icc). This was
an incidental finding in this study. Bar = 100 |im (A and C), 25 |im (B and D).
Page 76 of 124
-------
Final July 2015
Figure 9. Gonads, germ cell neoplasms. & Dysgemiinoma in the testis of an adult male. The caudal pole of the testis is
effaced by a mass (arrows) consisting of oogenic tissue. R A higher magnification of the tumor in A. The disorganization of the
oogenic tissue is apparent. Germ cell neoplasms such as seminomas and dysgerminomas are rare spontaneous findings in
medaka. There is currently little evidence to support the idea that such tumors are linked to EDC exposure, and control animals
seem to be affected as often as chemically-exposed individuals. Distinguishing features of germ cell neoplasms include
haphazard anatomic organization and progression of cell development, and a tendency to form mass-like lesions that distort the
gonad architecture. In early life stage studies in which fish are exposed to potent hermaphroditic chemicals such as 17|3-estradiol
or ethinylestradiol, it may be difficult to distinguish germ cell neoplasms from malformed intersex gonads. It is also important to
differentiate this neoplasm from other findings such as: 1) asynchronous development of the gonad in which different areas of the
gonad are in different stages of development that blend almost imperceptibly and do not form a mass; 2) testicular oocyte
formation, in which the scattered oocytes do not form a mass capable of distorting the gonad); and 3) possibly from
hermaphroditism, in which the anatomic arrangement and developmental progression of the aberrant tissue is orderly and
essentially resembles the normal gonad. H&E, bar = 250 urn (A), 50 pit (B).
Page 77 of 124
-------
Final July 2015
Figure 10. Gonads, hermaphroditism. A; Sagittal section of the abdomen of an adult medaka. Most of the abdomen is
occupied by a massively enlarged ovary that contains primarily atretic oocytes. Anterior to the ovary is a separate testis (arrow).
B and C: Higher magnifications of the testis (t) in A, and of the same testis (t) in another section. Hermaphroditism is a state in
which fully formed male and female gonad tissues are present in the same individual. The phrase "fully formed" indicates that: 1)
the male and female gonadal tissues are in discrete compartments; 2) the organizational architecture of the gonads is maintained;
and/or 3) there is visible evidence of supportive structures (e.g., tunica albuginea, ducts) in addition to germinal cells. Bar = 800
(Jtti (A), 100 Jim (B), 25 jim (C).
Page 78 of 124
-------
Final July 2015
Figure 11. Gonad, increased/decreased cells, [insert cell type], (testis or ovary). In this case, perinucleolar oocytes (po)
dominate this ovary, which also contains a few degenerating mature follicles (arrows). Another possible rule-out to consider in
this particular case would be a genii cell neoplasm (dysgeniiinoma). It is recognized that endocrine active compounds may alter
the proportional distribution of gametogenic and supportive cell types in the testis or ovary. Certain types of alterations (for
example, the proliferation or absence of single cell population) may not be adequately documented by gonadal staging. This
diagnostic term provides a mechanism for documenting such changes. For consistency, the pathologist should presume that these
semi-quantitative changes are: 1) relative to other cell types in the gonad; 2) relative to cell numbers in control animals; and 3)
estimates only, versus actual cell counts. Bar = 500 jiffl.
Page 79 of 124
-------
Final July 2015
Pheno 1
Pheno 2
Pheno 3
Pheno 4
Pheno 5
Figure 12. Gonads, phenotype scoring. For the purpose of the MEOGRT assay, gonads are scored for histologic phenotype
according to the following criteria: Phenotype 1 = entirely testicular tissue; Phenotype 2 = predominantly testicular tissue;
Phenotype 3 = approximately equal testicular and ovarian components; Phenotype 4 = predominantly ovarian tissue; Phenotype 5
= entirely ovarian tissue. Bar = 250 (im.
Page 80 of 124
-------
Final July 2015
Figure 13. Gonads, stromal tumors. These seem to be even rarer in medaka than germ cell neoplasms. Examples include
Sertoli cell tumors, granulosa cell tumors, and teratomas. Aj. Teratoma in the ovary of an adult female medaka. Various
embryonic tissue types are represented including cartilage (c), neural tissue (n), and gonad tissue (g). R Another area of the
tumor from A. The dominant feature in this section is a developing ocular mass in which the lens and retinal tissue are
recognizable. Bar =100 pit! (A), 50 (.im (B).
Page 81 of 124
-------
Final July 2015
Figure 14. Kidney, mineralization. Jfc Renal tissue from an adult male. At low magnification, extensive dilation of tubular
lumina and Bowman's spaces is evident. R Intratubular mineralization is obvious at higher magnification. The tubular dilation
is likely due to obstruction. It may be important in a study to differentiate this lesion from the nephropathy that can be induced
by exposure to estrogenic substances. Bar = 200 (tm (A), 50 pill (B).
Page 82 of 124
-------
Final July 2015
Figure 15. Kidney, nephropathy. A and B: Kidney from an adult female control, g = glomerulus. C and D: Kidney from an
adult female exposed to a compound will) estrogenic activity. Dilation of tubules and Bowman's space is evident at low
magnification (C). At higher magnification (D), changes include marked enlargement of glomeruli (g), eosinophilic deposits of
proteinaceous material in glomerular capillaries (black arrows), and vacuolation of the tubular epithelium (white arrows).
Degenerative renal disease has been observed in a variety of fishes that have been exposed to compounds with estrogenic activity
(Herman & Kincaid, 1988; Zillioux et al.. 2001; Palace et al., 2002). Renal impairment presumably occurs due to increased
production of vitellogenin that damages the kidney via protein overload. Such kidney changes are more likely to be observed in
males, presumably because there is no physiological outlet for the excess vitellogenin, but nephropathy can also be seen in
females exposed to high concentrations of estrogen-active substances. Microscopic lesions may include swelling of tubular
epithelial cells, tubular necrosis, dilation of Bowman's capsule, interstitial fibrosis, casts, and hyaline droplets in tubules or
glomeruli. ForEDC studies, the pathologist may elect to group these lesions under an umbrella diagnosis of "nephropathy", if
the data suggests that such changes are associated with estrogenic activity. Alternatively, the pathologist may choose to record
these types of changes as individual findings (e.g., kidney, tubular necrosis). Bar =100 jinx (A and C), 25 (.im (B and D).
Page 83 of 124
-------
Final July 2015
I '¦'<> 4&
^ jblZ*'- • . ut
» flfrltsffl a
Ml
. * ^ %
PwVl
MHffl
.£> ©o
Bp-V
'Cv o n
¦PH9
¦ " id. v5»
&Sji| «> -&%.,
¦* *5
\ O ©
?e. VI e
>, i\ ' • v "/
r-Tk - \ "V J Vi I
P ftrfl
ts- a
®o
W»
¦Si
¦ fat '-' f , * t (1, "5 L»0«>S -n~- I
»V siSHjL" I
as. — • •
1
*:
jE?
Figure 16. Kidney, tubular eosinophilia. Jfc Renal tissue from an adult female. Relative to the male (B), epithelial cells of the
proximal tubules (p) are smaller and have more basophilic cytoplasm. B; Renal tissue from an adult male. The plump epithelial
cells of the proximal tubules (p) have very fine granular eosinophilic material in their basal cytoplasm. Severity grading of
tubular eosinophilia is as follows: Grade 1 = essentially no eosinophilia; Grade 2 = small amount of eosinophilia; Grade 3 =
abundant eosinophilia. The kidney in B was scored Grade 3. Bar = 250 um.
Page 84 of 124
-------
Final July 2015
Figure 17. Liver, cystic degeneration. Liver from an adult female medaka. Cystic degeneration (cd) is characterized by various
numbers of single or multilocular, roughly spherical, fluid-filled spaces that are scattered throughout the hepatic parenchyma.
Individual lesions may or may not be associated with blood vessels, but the cysts themselves are not lined by endothelial cells.
Based on morphologic criteria, such lesions have also been termed hepatic cysts or spongiosis hepatis, although empirical
evidence suggests that these merely represent different stages in the progression of cystic degeneration. Cystic degeneration
tends to be relatively common in medaka, and especially older females. This particular liver also features bile duct concretions
(arrows), Bar= 100 pm.
Page 85 of 124
-------
Final July 2015
:
- *W.V
mMiwm
Figure 18. Liver, hepatocyte basophilia, increased. Aj. Liver from an adult control male. R Liver from an adult male exposed
to 100 jug/L 4-ferf-octylphenol, an estrogenic substance. There is a diffuse increase in hepatocyte basophilia, a loss of
cytoplasmic vacuolization, and hepatic blood vessels contain proteinaceous fluid. A generally diffuse increase in hepatocyte
cytoplasmic basophilia has been observed in male fish that have been exposed to compounds that are able to interact with hepatic
estrogen receptors, including E2 and 17|3-methyldihydrotestosterone (Wester et al., 2003). This increase in basophilia, which is
correlated with increased vitellogenin production, presumably mimics the heightened metabolic state (e.g., increased
endoplasmic reticulum) that is required for the production of vitellogenin in the reproductively-active female fish. Bar = 50 Jim,
Page 86 of 124
-------
Final July 2015
Figure 19. Liver, primary proliferative lesions. M Focus of hepatocellular alteration (altered focus). This is a non-
neoplastic, but likely pre-neoplastic, lesion that can be observed as a spontaneous or induced finding. Morphologic
characteristics include changes in hepatocyte size and color relative to the surrounding liver parenchyma, and blending with
unaffected hepatic tubules at the periphery of the lesion. B and C: Hepatocellular adenomas. Morphologic characteristics
include distinct margins, peripheral compression of unaffected hepatic tissue, little cytologic atypia (relative to carcinomas), and
generally larger size than foci. Hepatocellular carcinomas are less common but can occur also. Bar =100 Jim (A and B), 250
uni (C).
Page 87 of 124
-------
Final July 2015
s>. ,
: 'Jjy'i.-.'
'£'»*• ••
• • •*ry:y • { -
'•. "p":" ' •
¦ vi
• * ....'-J. ...*>» :.
T. . ¦* N ' .. . -
. ;¦ -
. V-V-ir-
Figure 20. Multiple tissues, proteinaceous fluid. This finding is characterized by the presence of homogeneous dark pink
translucent material, presumably vitellogenin, within the vascular and/or interstitial compartments of the testis, ovary, and other
tissues in fish that have been exposed to estrogenic substances. Aj. Intravascular proteinaceous fluid (p) in the testis of an adult
male exposed to 11 -estradiol at 100 ng/L for 4 weeks. B; Intravascular proteinaceous fluid (p) in the ovary of an adult female
exposed to 4-ferf-octylphenol at 90 (ig/L for eight weeks. C; Heart from an untreated control fish. Dk Heart with intravascular
proteinaceous fluid (p). Bar = 25 [ail (A), 50 tun (B, C, and D).
Page 88 of 124
-------
Final July 2015
« \L > '
%XL
jlLv
Figure 21. Ovary, chorion. Usually pale to dark eosinophilic and retractile, the chorion is the thick external layer of an oocyte
that surrounds the ooplasm. The terms zona radiata and vitelline envelope have been used synonymously. In mature, unspent
follicles, the chorion is noticeably surrounded by perifollicular cells (granulosa cells, theca cells, and surface epithelial cells). As
viewed by light microscope, the chorion is often minimally apparent or inapparent prior to the cortical alveolar phase of oocyte
development. Note the vast difference in thickness between the chorion of a cortical alveolar oocyte (small arrow) and the
chorion of a mature vitellogenic oocyte (large arrow). H&E, Bar = 25 imi.
Page 89 of 124
-------
Final July 2015
Grade 2
Grade 1
.
Grade 3
Xfi-A'C
Grade 4
Figure 22. Ovary, decreased yolk formation, grading. This finding is characterized by a progressive decrease in the quality
(i.e., the yolk becomes more watery) and amount of yolk in vitellogenic-sized follicles. In Grade 3 ovaries, follicles contain only
a scant amount of yolk (arrows), whereas in Grade 4, yolk is essentially not visible. Affected oocytes often have cortical alveoli
(yolk vesicles) that are fragmented or dissipated. Unlike oocyte atresia, the vitelline membrane (chorion) of affected oocytes is
often smooth and contiguous. However, decreased yolk formation is often accompanied by at least a low degree of oocyte atresia
(A). This type of change has been observed following exposure to aromatase inlribitors such as prochloraz and fadrozole, and the
non-aromatizable androgen trenbolone. H&E, Bar = 500 um.
Page 90 of 124
-------
Final July 2015
Figure 23. Ovary, edema. This ovary is markedly enlarged due to abundant ovarian edema (oe), which is represented by excess
clear space within the ovary. Numerous atretic follicles (af) are also present. The cause was undetermined in this case, sb =
swim bladder. Bar = 800 um.
Page 91 of 124
-------
Final July 2015
Zona radiata
Vitelline
envelope
Ooplasm
nucleus
Surface
epithelium
Theca
Granulosa
Oocyte
Follicle
Figure 24. Ovary, follicle. Diagram from Tyler and Sumpter, 1996. The functional unit of the ovary, this term generally refers
to an oocyte plus its surrounding sheath of perifollicular cells (granulosa cells, theca cells, and surface epithelium cells) (Tyler
and Sumpter, 1996). However, there are subtypes of follicles in which the oocyte is not present or may be difficult to appreciate;
these include post-ovulatory (spent), empty, and atretic follicles. A post-ovulatory follicle (the follicle has ruptured to release an
oocyte during spawning) is collapsed and often has enlarged (hypertrophic) granulosa and theca cells. Conversely, an empty
follicle (in which the oocyte has been dislodged from the histologic section as a post-mortem artifact) generally retains the shape
of the oocyte and may or may not have enlarged granulosa and theca cells. An atretic follicle should be distinguished from both
spent follicles and empty follicles; the presence of at least some ooplasmic material (often heterochromatic) within a follicle
indicates that it contains an atretic oocyte.
Page 92 of 124
-------
Final July 2015
Figure 25. Ovary, follicular atresia. Ovary from an adult female. The larger red arrow indicates a cortical alveolar oocyte that
is atretic, whereas the smaller black arrow denotes a large fragment of chorion that is partially surrounded by macrophages and
hypertrophic perifollicular cells. Essentially, degradation and resorption of an oocyte at any point in development, including
unspawned senescent oocytes, atresia can be the result of either physiological or pathological processes. For consistency, the
term atresia should generally be used in preference to the term "degeneration" et al. when referring to oocytes,
Histopatho logically,atresia is often characterized by clumping and perforation of the chorion, fragmentation of the nucleus,
disorganization of the ooplasm, and/or the uptake of yolk materials by perifollicular cells. Because even severe oocyte atresia
can be observed as an apparently spontaneous finding in one or more control females, it is important to compare populations
rather than individuals, and putative effects in studies with low animal numbers should be interpreted with caution. Although
increased oocyte atresia is a non-specific finding that is not limited to EDC exposure, it may contribute to an indication of
causality in a "weight-of-evidence" approach. The following is an example of a severity grading scheme for increased oocyte
atresia: Not remarkable = <3 atretic oocytes per ovary section; Grade 1 = 3 to 5 atretic oocytes per section; Grade 2 = 6 to 9
atretic oocytes per section; Grade 3 = greater than 9 atretic oocytes per section, but less than the vast majority; and Grade 4 = the
vast majority of oocytes in a section are atretic. Bar = 50 (jni.
Page 93 of 124
-------
Final July 2015
Grade 2
Grade 1
Grade 4
Figure 26. Ovary, follicular atresia grading. Severity grading for follicular atresia is based on the maximum number of atretic
follicles per ovary section as follows: Grade 1 = 3-5, Grade 2 = 6-8, Grade 3 = 9 or greater, but less than the vast majority, Grade
4 = the vast majority of follicles are atretic. Bar = 750 jun (Grades 2 through 4).
Page 94 of 124
-------
Final July 2015
Figure 27. Ovary, germinal epithelium Normal ovary from an adult female. Arrows indicate the germinal epithelium which,
at this magnification, is a membranous structure that separates the ovarian lumen (L) from the extravascular space tKVSi of the
ovarian stroma. The germinative parenchyma of the ovary, the membrane bound germinal epithelium constitutively contains
oogonia, pre-follicular and pre-thecal cells, epithelial cells, and occasionally small chromatin nucleolar (primary growth) oocytes
(Norberg et al., 1999; Parenti and Grier, 2003). The germinal epithelium separates the ovarian lumen from the stroma, the latter
of which often contains perinucleolar, cortical alveolar, and vitellogenic follicles within a variably-apparent extravascular space.
Bar =100 am.
Page 95 of 124
-------
Final July 2015
Figure 28. Ovary, oogenic cell types. A Oogonia (arrow). Unlike mammalian oogonia, which traditionally are considered to
be non-proliferative following the early post-natal period, piscine oogonia continue to divide in juvenile and adult fish. The
smallest of the oocytic cells, oogonia reside within the ovarian germinal epithelium, usually in comparatively low numbers.
Oogonia are characterized by a relatively large nucleus with small or inapparent nucleolus, and minimal amounts of cytoplasm.
13: Perinucleolar phase oocytes (p). Concomitant with oocyte growth, the nucleus (germinal vesicle) increases in size and
multiple nucleoli appear, generally at the periphery of the nucleus. The cytoplasm stains uniformly dark, although late
perinucleolar oocytes may have small clear or amphophilic vacuoles in the cytoplasm. These cells tend to be abundant in normal
adult ovaries. C Cortical alveolar oocytes (arrow). Generally larger than perinucleolar oocytes, this phase is characterized by
the appearance of cortical alveoli (yolk vesicles) within the ooplasm. The cortical alveoli are technically not yolk, as they do not
provide nourishment for the embryo (Selman and Wallace, 1989). The chorion becomes distinctly evident in this phase, the
nucleus becomes reduced, and the perifollicular cells are more easily visualized. JJ. Early vitellogenic oocytes (large arrow).
Larger than cortical alveolar oocytes, these cells are characterized by the centralized appearance of spherical, eosinophilic,
vitellogenic yolk granules / globules (small arrows). The nucleus has moved to the periphery of the cell and dissolved. E Late
vitellogenic oocytes (arrow). These cells are characterized by an increased accumulation of yolk material that fuses into a central
liquid mass which displaces the cortical alveolar material to the periphery of the cytoplasm. Jj. Mature spawning follicle
(aixow). In this phase of development, vitellogenesis has reached its peak, the cell has become larger and more hydrated, and
ooplasm consists almost entirely of yolk. Because of the transient nature of these cells in fractional spawning fish, mature /
spawning oocytes are uncommonly observed. Bar = 25 urn (A), 50 jtm (B through D), 100 pm(E and F).
Page 96 of 124
-------
Final July 2015
Grade 1
Grade 3
Grade 2
Figure 29. Ovary, perifollicular cell hyperplasia /hypertrophy, grading. Exposure to aromatase inhibitors (e.g., fadrozole,
proc-hloraz) has been associated with these perifollicular cell changes in medaka ovaries. A similar effect has also been linked
with exposure to the non-aromatizable androgen, trenbolone (unpublished data). This finding is characterized by an increase in
the height and number of granulosa cells, which gives this cell layer a "pseudostrafied" appearance in extreme cases. A common
coexisting change in affected medaka has been decreased yolk formation. Because perifollicular cells (i.e., granulosa cells) are
thought to be involved with aromatase production in fish (Nagahama, 1987; Devlin and Nagahama, 2002), it is possible that the
increased number and size of these cells is a compensatory mechanism aimed at restoring aromatase to levels required for
vitellogenesis. It is important to note that: 1) normal perifollicular cells may appear hypertrophic in tangentially-sectioned
oocytes, and 2) perifollicular cell changes are best identified by comparisons made with concurrent control fish. Bar = 25 (im
(all).
Page 97 of 124
-------
Final July 2015
Figure 30. Ovary, pigmented macrophage aggregate (histiocytic cells) in the ovary of an adult female. These aggregates are
present constitutively in the interstitium of the ovary, and rarely in the testis. Cells comprising pigmented macrophage
aggregates (PMA) have small condensed eccentric or peripheralized nuclei and various brown, yellow, red, or gold pigment
granules (lipofuscin, ceroid, hemosiderin, and/or melanin) that often impart a slightly crystalline appearance to their
comparatively abundant pale cytoplasm. In the normal ovary, these macrophage aggregates are likely involved in the processing
of breakdown products associated with atresia of unspawned oocytes. It has been demonstrated that macrophage aggregates may
become larger and/or more numerous following exposure to certain toxicants or infectious agents (Blazer et al., 1987).
Whenever possible, macrophage aggregates should be distinguished from granulomatous inflammation. Granulomatous
inflammation, which is a reaction to the presence of pathogens or foreign substances, is characterized by the presence of
epithelioid macrophages, with or without multinucleated giant cells, additional inflammatory cells, and necrosis. Distinguishing
PMA from inflammation is not always easy, as pigmented macrophage aggregates may become incorporated into areas of
granulomatous inflammation. Bar = 25 uni.
Page 98 of 124
-------
Final July 2015
Figure 31. Ovary, post-ovulatory follicles. A number of post-oviilatory follicles (POF), indicating recent spawning, are evident
in in this ovary from an adult female (arrows). Following release of an oocyte (i.e., spawning), the perifollicular sheath, which is
a membranous structure lined by granulosa cells, theca cells, and surface epithelium, collapses into a POF. Consequently, POFs
are most likely to be seen in Stage 2 and Stage 4 ovaries, and they are rarely present in Stage 3 ovaries. The granulosa cells of
POFs are much larger than those of intact follicles. Mammalian terms such as "corpus lutea" and "Graafian follicles", are
probably inappropriate, due to structural and functional differences between those entities and piscine POFs. POFs should be
differentiated from collapsed atretic follicles, the latter of which contain ooplasmic debris. Post-ovulatory follicles are graded
according to the maximum number per ovary section as follows: Grade 1 = 3-5 POF; Grade 2 = 6-8 POF, and Grade 3 = 9 or
greater POF. Bar = 250 pa.
Page 99 of 124
-------
Final July 2015
Figure 32. Ovary, post-ovulatory follicles, accelerated involution. A. Typical post-ovulatory follicle, in which only occasional
apoptotic-like cells (arrow) are present. B: In this ovary from a compound-treated fish, post ovulatory follicles contained myriad
apoptotic cells. Bar = 25 (im (A and B).
Page 100 of 124
-------
Final July 2015
Figure 33. Ovary, spermatogenesis. A andB: Ovary from an adult female control in which ovarian spermatogenesis (arrow)
was not a treatment-related finding. In B, spermatogenic cells of various phases are represented. This change is characterized by
the presence of non-neoplastic spermatogenic cells, usually immature, within the ovary. There is little or no evidence of lobular
or tubular testicular architecture. Care should be taken to distinguish ovarian spermatogenesis from mitotically dividing oogonia;
a key feature of ovarian spermatogenesis is the presence of multiple spermatogenic phases. Ovarian spermatogenesis should also
be distinguished from inadvertent carryover of spermatogenic tissue during the trimming or microtomy process. It should be
recognized that ovarian spermatogenesis may not always indicate masculinization. In some situations it may represent
incomplete conversion of a genotypic male to the female phenotype. Bar = 250 uin (A), Bar = 25 um (B).
Page 101 of 124
-------
Final July 2015
Figure 34. Swim bladder, gas gland adenoma. Gas gland adenomas (gga) of the swim bladder are uncommon, but not rare,
neoplasms in medaka. Thus far, this appears to be an incidental finding in toxicology studies. Anecdotal evidence suggests that
these lesions may be associated with congenital deformities of the spine and/or swim bladder, resulting in pneumatic duct
patency, swim bladder inflammation (pneumocystitis), and tumor formation. Related lesions include hyperplasia of the swim
bladder gas gland epithelium (increased amounts of epithelium without the formation of a distinct mass), and gas gland
adenocarcinomas (locally invasive tumors with cytologic pleomorphism). ak = anterior kidney, li = liver, gb = gallbladder. Bar
= 250 tun.
Page 102 of 124
-------
Final July 2015
Figure 35. Testis, asynchronous development. This finding is characterized by the presence of distinctly different populations
(i.e., range of developmental stages) of gametogenic cells in different regions of a gonad, or the aberrant positioning of gonadal
cell populations. In this particular case, an 8-week old male had been exposed for approximately eight weeks to 27 jug/L 4-tert-
octylphenol. In addition to the presence of numerous testis-ova, the efferent duct system is abnormally irregular, and
spermatogonia-containing spermatocysts (arrows) are located in an atypical position adjacent to the ducts (asynchronous
development). Bar= 100 jutt.
Page 103 of 124
-------
Final July 2015
'k
B
mJSr*r* a®
. 5fsP
Figure 36. Testis, degeneration, increased. Examples of degenerative findings in the testis include: 1) individual or clustered
apoptotic germ cells; 2) vacuolated germ cells; 3) multinucleated (syncytial) cells in the germinal epithelium or testicular lumen.
Apoptotic germ cells are characterized by cell shrinkage, nuclear condensation, and fragmentation into spherical, membrane-
bound bodies, which are often phagocytized by neighboring cells. Typically, there is no associated inflammation associated with
these cells. Low numbers of degenerating germ cells are commonly found in the testes of control males. Extensive testicular
degeneration may lead to localized or generalized loss of the germinal epithelium. A Germ cell syncytium (arrow) in the testis
of a control male. R Moderate testicular degeneration characterized by the presence of numerous apoptotic cells within the
germinal epithelium (arrow). Moderate to severe testicular degeneration may also occur occasionally in untreated males.
Bar = 25 urn.
Page 104 of 124
-------
Final July 2015
Figure 37. Testis, germinal epithelium. Normal testis from an adult male medaka. The double arrow mdicates width of
germinal epithelium, which extends from the tunica albuginea to the efferent duct. Germ cell maturation occurs from the
periphery inward, sg = spermatogonia, sc = spermatocytes, st = spermatids, sz = spermatozoa. Bar =25 um
Page 105 of 124
-------
Final July 2015
Figure 38. Testis, hypoplasia. A aiidB: Normal testis in an adult male. C and D: Hypoplastic testis (arrows) from an
8-week-old male exposed to 450 mg/L 4-n-amylaniline for approximately 8 weeks. The hypoplastic testis is not only small, it is
poorly formed, consisting primarily of nests of spermatogonia with no clear efferent duct system. Indicating underdevelopment,
this condition may be associated with interstitial fibrosis and increased prominence of interstitial cells in affected areas of the
testis. Hypoplasia may be chemically induced, or it can occur spontaneously in rare instances. Bar = 250 jiin (A and C),
25 ).im (B and D).
Page 106 of 124
-------
Final July 2015
A**-'*
Figure 39. Testis, interstitial (Leydig) cells. Testis from a 16-week old control male. These androgen-producing cells have
dense, dark round or oval nuclei with little detail and moderate amounts of variably-evident, faintly vacuolated cytoplasm.
Compared to germinal cells, interstitial cells are usually present in low numbers, usually as single cells or small aggregates,
scattered irregularly throughout the interlobular interstitium. Although they may resemble spermatocytes, interstitial cells are
only present in intertubular areas. Bar = 25 pp..
Page 107 of 124
-------
Final July 2015
Figure 40. Testis, interstitial cell hyperplasia / hypertrophy. A; Testis from an adult male control. Scattered small clusters of
interstitial cells (arrows) are located between tubules. R Interstitial cell aggregates (arrows) are larger and more numerous in a
testis from an adult male exposed to fadrozole at 100 ppm. This finding is characterized by a relative increase in the number
and/or size of interstitial cells in the testis, as compared to the testes in the majority of control males. In moderate to severe
hyperplasia, the testicular interstitium may be expanded due to the proliferation of these cells. Hypertrophic interstitial cells
feature enlarged rounded nuclei with increased nuclear detail, and relatively abundant dense cytoplasm as compared to non-
hypertrophic interstitial cells. Bar = 25 inn (A and B).
Page 108 of 124
-------
Final July 2015
&§ii
No
Increase
iGrade 1 BM&SiS
vm
b 7^
Grade 3
Figure 41. Testis, interstitial cell hyperplasia / hypertrophy, grading. Testes of compound-treated males are scored relative to
the typical appearance of testes among concurrent controls. Bar = 25 um (all).
Page 109 of 124
-------
Final July 2015
. * • •• »
•v * ¦
u\
.'">0. . . 57^
feflwL. -• •<
T - »
¦*- -£/ s .
y* / * j r«» * * #»"« J
vtJ*\ •<•. •.;, .v.:¦."<* *
• liV • • j'* • »"» * ^ *
.«> U'.i' ;tV«
. . V ' I . . ••• *
• - v •• •• ? ,
I.V-^.V \ , ' I ••l/*.'
•-jt- V -'iV K> ' L V
'• vA-'Vr-'-W.v* *
.i .' ¦
;"V' ' . J*" ''. '' •!" .•'* .V V..* •,,H '•
•« ?. tv\ • -.. 'Jf.-v i i ». .r -
* ~ ••'lA.?!'
> -f/w H
$Sr*
yi& .
AC,
'¦jSa
«.£K?
tfW '/**
* t
V ' f
>„
\.. -ij
>• ¦ Xft
V?**
• j
3fiM
r."*?
r.V<- "
i'/4L\ JkT J?- 5-
n§§lw;-
• irf
^ .^4.* f I • i «* » « . < 11 1 »*
*
^-;£v » - v'
, ¦'••• C? . £r t «_- tt n «
vX' v&M ¦>¦• V ; *..v r *
V *WA '.i-.-'iV t VT*-*«
H&S- &<&:?&" fef
• '•£; *'<9$ 35K
ii. *a
btMRs k , ^ %
Wrap« 1' » * - -
o C JlJLAai \* . Mf •„
f £&"*<*• I • , #i*» V
Ji%« r»i \»
I %«<;
V>_8 *#
«•*
¦ \
&3i? .
•^fcvW *Y» • , *, *,s,.|. ;• «V
n?<5,.{'.'. • /a v v. *, j • •• • i
T. • ~ K.i ¦.. *k*f*A*
. ; <• v.*. •' 4 \ >\T -'v
^.n
fjfipd
. •».» J 'Kj,!-
fmSla
A';Vr^
* ' *• 1. * • A *A*4fc
v *.:r
r*r>;
Figure 42. Testis, interstitial fibrosis. Whether in the testis or ovary, fibrosis is characterized by the presence of increased
fibrous connective tissue (collagenous fibers and fibrocytes or fibroblasts) within the testicular or ovarian interstitium (stroma).
Due to a high degree of inter-animal variability among controls, it may be difficult to reliably distinguish subtle fibrosis in
treated fish. Bar =25 utn (A and B).
Page 110 of 124
-------
Final July 2015
V.
4* . 17/
• *• ff J
Jr.t * * *.
SRflKM
1**1* €
m m-
M
Figure 43. Testis, Sertoli ceUs. Sertoli cells (arrows) tend to have sharply-defined elongated or triangular nuclei, variably
evident nucleoli, and cytoplasm that is often indistinct. The cytoplasmic amis of a Sertoli cell encircle a clonal group of
spermatogenic cells, Conning a spermatocyst. Compared to germinal cells, Sertoli cells are usually present in low numbers,
usually as single cells located adjacent to lobular septa. Bar = 8 um.
Page 111 of 124
-------
Final July 2015
*
Figure 44. Testis, spermatocysts. The functional unit of the testis, this structure consists of a clonal group of spermatogenic
cells (spermatogonia, spermatocytes, or spermatids) that are surrounded by the cytoplasmic amis of (usually) one Sertoli cell.
Cells within spermatocysts exist as syncytia, maintained by intercellular attachments (cytoplasmic bridges), until final maturation
and release of spermatozoa occurs (spermiogenesis) (Grier, 1976), Each spermatocyst (packet of cells) represents a cohort of
germ cells in approximately the same developmental phase. Circled is a spermatocyst containing spermatogonia, and the arrow
indicates the Sertoli cell that appears to be associated with that particular spermatocyst. Bar =15 um.
Page 112 of 124
-------
Final July 2015
Figure 45. Testis, spermatogenic cell types. A: Spermatogonia. The largest of the spennatogenic cells (^5-10 iim).
spennatogonia generally have pale vesicular nuclei, prominent nucleoli, variably distinct nuclear membranes, perinuclear
cytoplasmic granules, and moderate amounts of granular cytoplasm (arrow). Spennatogonia B are smaller than spennatogonia
A, and spennatogonia B are usually present in larger clusters (e.g., >4 cells). B: Spermatocytes. Derived from spennatogonia,
spennatocytes are of intennediate cell size (~ 4-6 uin), and have comparatively dense nuclei and minimal to moderate amounts of
indistinct cytoplasm. Spennatocyte nuclei are usually evident in one of three meiotic phases: pachytene, leptotene, or zygotene.
Primary spermatocytes (p) are larger than secondary spennatocytes (s), and the latter are derived from primary spennatocytes
following the first meiotic division. Spennatocytes are usually one of the most abundant spennatogenic cells, and they tend to
contribute to the largest spennatocysts. C: Spermatids. Derived from spennatocytes following the second meiotic division,
these cells have dense nuclei and nanow rims of eosinophilic cytoplasm. They are the smallest cells w ithin the genninal
epithelium (~2-3 um i. and the cells lose their cytoplasmic attachments to one another during spenniogenesis. D Spermatozoa.
These cells have dark, round nuclei and minimal or no apparent cytoplasm. Tails are generally not apparent in histologic
sections. Spennatozoa are the smallest spennatogenic cells (~ 2 (im), and exist as scattered individual cells within tubular lumen.
Bar =15 jjm (A), 8 |im (B), 4 jim (C and D),
Page 113 of 124
-------
Final July 2015
v>v . • *. < v .-s>" Xm*m* ,,
r t! 4 ''ft" \V ' > i\ >*
y^v^-.¦¦..v • ¦<\
&$¥ " "¦"• :W
:• • • S&:;>;v - -V *rr ••;
> • v ' • . ' ,:1
'V/-' .?$£»&$ :v. *
rK' - ,>¦.& &&&| 301
: : 1_
Figure 46. Testis, testicular oocytes (testis-ova)._ This finding is characterized by the presence of one or more individualized or
clustered oogenic cells, usually immature, within the testis. There is little or 110 evidence of ovarian architecture. Testicular
oocytes may be chemically-induced or spontaneous; the incidence of spontaneous testicular oocytes in control fish may vary
according to test facility. This particular example is fairly unusual in that one of the oocytes (arrow) has progressed to the early
cortical alveolar phase. Bar = 50 11111.
Page 114 of 124
-------
Final July 2015
Figure 47. Testis, testicular oocytes, grading. Grade 1 is characterized by a single oocyte (arrow) per histologic testis section.
Arrows indicate multiple oocytes in the Grade 2 image. Note the progressive loss of testicular ductal architecture with increasing
grade score. Small remnants of spermatogenic tissue and the bi-lobed configuration of the Grade 4 testis provide evidence that
this is a testis rather than an ovary. This fish had been exposed to an estrogenic substance; otherwise, another potential rule-out
for this gonad might be dysgerminoma. Bar =100 pro (all).
Page 115 of 124
-------
Final July 2015
Figure 48. Thyroid glands. In medaka, the bilaterally symmetrical thyroid tissue is located caudal and somewhat lateral to the
branchial chamber, and similar to other fishes, the thyroid tissue is not a discrete encapsulated structure. Medaka differ from
other fishes in that the thyroid tissue of reproductively active adult males is often proliferative-looking. A andB: Thyroid tissue
from a control male. Not all males have thyroid tissue that appears this hyperplastic. C and D: Thyroid tissue from a control
female. Here the follicles relatively small and lined by flattened epithelium. E and F: Thyroid glands from a female exposed to
102 ng/L of 4-tert-octylphenol. The severity of follicular cell hypertrophy / hyperplasia in this fish was recorded as grade 1,
which is reduced compared to most control males. Bar =100 um (A, C, and 1'. i. Bar = 25 p,m (B, D, and E).
Page 116 of 124
-------
Final July 2015
No
Increasek
Grade 2
Grade 1
Grade 3
Figure 49. Thyroid glands, hypertrophy / hyperplasia, grading. Most control females will demonstrate little or no increase in
thyroid follicular cell size or number, whereas the thyroids of most reproductively active adult males will score as Grades 1 or 2.
Bar = 50 um (all).
Page 117 of 124
-------
Final July 2015
(e) Gondal Staging Criteria.
The goal of gonadal staging is to determine if the administration of a particular
endocrine-active substance affects the reproductive cycle status of adult male and female fish.
The purpose of this section is to describe a rapid, semi-quantitative method for assessing the
proportions of various gametogenic cell types (gonadal staging) based on the light microscopic
examination of hematoxylin and eosin-stained histologic sections.
Semi-quantitative gonadal staging has been proposed for, or employed in, studies involving
fathead minnows (Ankley et al., 2002; Jensen et al., 2001; Miles-Richardson et al., 1999; Nichols
et al., 2001; US EPA, 2002) and zebrafish (Van den Belt et al., 2002), among other fishes.
Although such studies generally included excellent descriptions of the different gametogenic
maturation stages (e.g., spermatogonium through spermatozoa for the testis), they did not
incorporate pre-defined categorical guidelines for evaluating and reporting the reproductive cycle
status of an individual fish. To maintain scientific integrity across the board in a program that
involves multiple studies, multiple laboratories, and large numbers of animals, it is essential that
observations are recorded on a fish-by-fish basis. The use of a categorization system can
improve the consistency and objectivity of reported observations within and among experiments;
consequently, comparisons of the results are more meaningful. Categorization systems also have
some drawbacks and limitations, the most significant of which are: 1) the potential loss of
discriminatory data when similar, but not identical, types of observations are combined (binned)
into a single class; 2) the questionable biological relevance of the classification criteria in some
cases; and 3) the inability of any single classification system to address every type of observation
(either predicted or unforeseen). To address this last limitation, gonadal staging is accompanied
by a complete histopathological evaluation of the gonads; in this manner, the loss or
overabundance of a specific gametogenic cell type, for example, can be documented. It should
be emphasized that gonadal staging results are virtually meaningless in terms of individual fish
(versus treatment groups). This is because considerable animal-to-animal variation in gonad cell
proportions is to be expected, even among fish of the control groups, as a consequence of
spawning cycle asynchrony. For example, the cellular composition of fathead minnow ovaries
was observed to vary substantially according to the day that each fish was sacrificed relative to
spawning (Jensen et al., 2001). Consequently, following the gonadal staging of individual fish,
each treatment group should be assessed as a whole and compared to the appropriate control
group to determine if a compound-related effect has occurred.
The semi-quantitative gonadal staging scheme proposed here is a modification of a system
adopted by the United States Department of the Interior, U.S. Geological Survey, Biological
Resources Division as part of the "U.S. Biomonitoring of Environmental Status and Trends
(BEST) Program" (McDonald et al., 2000). The authors of the BEST system credit previous
work by Treasurer and Holiday (1981), Nagahama (1983), Rodriquez et al. (1995), and
Goodbred et al. (1997). The foremost benefits of this system are speed and ease of use,
especially when compared to fully-quantitative staging. The basis of the BEST system is a
visual assessment of the density of gametogenic precursors as compared to mature gametocytes
in one or more gonad sections. Accordingly, the stage numbers (testis: Stages 0 to 4; ovary:
Stages 0 to 5) increase in direct relationship to the relative proportion of mature cells. Although
the BEST system was initially developed to assess reproductive function in seasonal spawners
Page 118 of 124
-------
Final July 2015
such as carp (Cyprinidae) and black basses (Centrarchidae), the same stage categories can be
applied to fractional spawners such as medaka.
A few modifications have been made to the BEST system to adapt it for use in small aquarium-
sized fishes. For example, there is currently no provision in the system for gonads that are
comprised entirely of spermatogonia or oogonia. Although many experiments will use
reproductively mature fish, it is possible that an occasional animal may not attain sexual maturity
by the time that the experiment is terminated, or that certain test compounds might cause
reversion of the gonads to a juvenile phenotype. Therefore, one modification of the BEST
system, a pre-staging category called "juvenile" has been added for both male and female fish.
Another modification to the system involves an apparent discrepancy between the BEST system
and Goodbred et al. concerning the thickness of the testicular germinal epithelium as a staging
criterion. As indicated by Goodbred et al., the germinal epithelium becomes thinner as the testis
stage increases, whereas, the reverse occurs according to the BEST system (as presented in
McDonald et al.). Although it is difficult to find corroborating statements in the scientific
literature, empirical evidence indicates that Goodbred et al. is correct on this point. A third
modification to the system is the option to subdivide a stage into two subordinate stages (e.g.,
Stages 3 A and 3B) if the pathologist believes that this tactic would reveal a subtle, compound-
related effect that might otherwise be missed. Other modifications to the system are relatively
minor and primarily involve rewording for clarification.
Granting that the cell distribution pattern is likely to vary throughout a given tissue section, the
gonad should be staged according to the predominant pattern in that section. This is
especially important for the medaka testis in which spermatogenesis progresses along axial
centrifugal and rostro-caudal gradients ("restricted spermatogonial" type testis). Gonads that
cannot be reasonably staged for various reasons (e.g., insufficient tissue, or extensive necrosis,
inflammation, or artifact) should be recorded as UTS (unable to stage).
(1) Criteria for Staging Testes. To derive each stage score, the estimated width of the
germinal epithelium (EWG) can be compared to the estimated width of the testis (EWT) as
shown in Figure 50.
• Stage 1: EWG > % EWT
• Stage 2: EWG % to > '/2 EWT
• Stage 3: EWG '/2 to > Vi EWT
• Stage 4: EWG < Vi EWT
Page 119 of 124
-------
Final July 2015
Figure 50. Staging the testis. Testes from four adult male medaka (transverse oblique sections). Black arrows represent the
EWT, and red arrows represent the EWG (measurements are illustrated unilaterally for simplicity). In order to obtain comparable
sections, it is imperative that each section contains a portion of the central duct (CD), preferably at its widest and longest extent
(paraffin, H&E). Bar= 100 (im (all).
(2) Criteria for Staging Ovaries. The following are morphologic criteria for staging
female fish:
• Juvenile: gonad consists of oogonia exclusively; it may be difficult or impossible to
confirm the sex of these individuals.
• Stage 0 - Underdeveloped: entirely immature phases (oogonia to perinucleolar oocytes);
no cortical alveoli.
• Stage 1 - Early development: vast majority (e.g., >90%) are pre-vitellogenic follicles,
predominantly perinucleolar through cortical alveolar.
• Stage 2 - Mid-development: at least half of observed follicles are early and mid-
vitellogenic.
• Stage 3 - Late development: majority of developing follicles are late vitellogenin
• Stage 4 - Late development/hvdrated: majority are late vitellogenic and mature /
spawning follicles; follicles are larger as compared to Stage 3.
• Stage 5 - Post-ovulatorv: predominately spent follicles, remnants of theca externa and
granulosa.
Figure 51 illustrates some of these stages.
Page 120 of 124
-------
Final July 2015
Figure 51. Staging the ovary. Ovaries from four adult female medaka. Bar =100 urn (Stages 0 and 1), 250 um (Stages 2 and
4). Stages 3 and 5 are not illustrated.
Page 121 of 124
-------
Final July 2015
(f) References.
(1) Crissman J.W., Goodman D.G., Hildebrandt P.K., Maronpot R.R., Prater D.A., Riley
J.H., Seaman W.J., and Thake D.C. (2004). Best Practices Guideline: Toxicologic
Histopathology. Toxicol Pathol. 32:126-131.
(2) Devlin R.H., Nagahama Y. (2002). Sex determination and sex differentiation in fish: an
overview of genetic, physiological, and environmental influences. Aquaculture 208:191-
364.
(3) Goodbred S.L., Gilliom R.J., Gross T.S., Denslow N.P., Bryant W.L., Schoeb T.R.
(1997). Reconnaissance of 17P-estradiol, 11-ketotestosterone, Vitellogenin, and Gonad
Histopathology in Common Carp of United States Streams: Potential for Contaminant-
induced Endocrine Disruption. U.S. Geological Survey, Open-File Report 96-627, pp.
47.
(4) Grier, H.J. (1976). Sperm Development in the Teleost Oryzias latipes. Cell. Tiss. Res.
168: 419-431.
(5) Herman R.L., Kincaid H.L. (1988). Pathological Effects of Orally Administered
Estradiol to Rainbow Trout. Aquaculture 72:165-172.
(6) Jensen K.M., Korte J.J., Kahl M.D., Pasha M.S., and Ankley G.T. (2001). Aspects of
Basic Reproductive Biology and Endocrinology in the Fathead Minnow (Pimephales
promelas). Comp Biochem Physiol C Toxicol Pharmacol. 128:127-141.
(7) McDonald K.K., GrossT.S., DenslowN.D., Blazer V.S. (2000). Reproductive
Indicators". In: Biomonitoring Of Environmental Status And Trends (BEST) Program:
Selected Methods For Monitoring Chemical Contaminants And Their Effects In Aquatic
Ecosystems. (C. J Schmitt and G. M. Dethloff, eds.) U.S. Geological Survey, Biological
Resources Division, Columbia, MO: Information and Technology Report USGS/BRD-
2000-0005, pp. 30-42.
(8) Miles-Richardson S.R., Kramer V.J., Fitzgerald S.D., Render J.A., Yamini B., Barbee
S.J., and Giesy J.P. (1999). Effects of Waterborne Exposure of
17P-estradiol on Secondary Sex Characteristics and Gonads of Fathead Minnows
(Pimephales promelas)." Aquat Toxicol 47:129-145.
(9) Morton D., Sellers R.S., Barale-Thomas E., Bolon B., George C., Hardisty J.F., Irizarry
A., McKay J.S., Odin M., Teranishi M. (2010) Recommendations for Pathology Peer
Review. Toxicol Pathol. 38(7): 1118-1127.
(10) Nagahama Y. (1983). The Functional Morphology of Teleost Gonads. In: Fish
Physiology. (W. S. Hoar, D. J. Randall, E. M. Donaldson, eds.), San Diego, California,
USA, Academic Press, pp. 223-275.
Page 122 of 124
-------
Final July 2015
(11) Nagahama,Y. (1987) Review: Gonadotropin action on gametogenesis and
steroidogenesis in teleost gonads. Zool. Sci. 4:209-222.
(12) Nichols K.M., Miles-Richardson S.R., Snyder E.M., and Giesy J.P. (2001). Effects of
Exposure to Municipal Wastewater in Situ on the Reproductive Physiology of the
Fathead Minnow (Pimephalespromelas). Environ Toxicol and Chem. 18:2001-2012.
(13) Norberg B., Kjesbu O.S., Taranger G.L., Andersson E., Stefansson S.O. (1999). Rhe
Teleost Germinal Epithelium: A Unifying Concept. Proceedings of the 6th International
Symposium on the Reproductive Physiology of Fish, pp. 233-236.
(14) Palace VP., Evans R.E., Wautier K., Baron C., Vandenbyllardt L., Vandersteen W., Kidd
K. (2002). Induction of Vitellogenin and Histological Effects in Wild Fathead Minnows
from a Lake Experimentally Treated with the Synthetic Estrogen, Ethynylestradiol. Water
Qual. Res. J. Canada. 37(3): 637-650.
(15) Parenti L.R., Grier H.J. (2003). Patterns and Processes in the Evolution of Fishes.
Annual Meeting of the Society of Integrative and Comparative Biology, Toronto,
Canada, p.8.
(16) Rodriguez J.N., Oteme Z. J., Hem S. (1995). Comparative study of vitellogenesis of two
African catfish species Chrysichthys nigrodigitatus (Claroteidae) and Heterobraanchus
longifilis (Clariidae). Aquat. Living Re sour. 8:291-296.
(17) Selman K., Wallace R. A. (1989). Cellular aspects of oocyte growth in teleosts. Zool.
Sci. 6:211-231.
(18) Shackelford C.C., Long G., Wolf J., Okerberg C., and Herbert R. (2002). Qualitative and
Quantitative Analysis of Nonneoplastic Lesions in Toxicology Studies. Toxicol Pathol.
30:93-96.
(19) The Society of Toxicologic Pathologists (1991). Peer Review in Toxicologic Pathology:
Some Recommendations. Toxicol Pathol. 19:290-292.
(20) The Society of Toxicologic Pathologists. (1997). Documentation pf Pathology Peer
Review. Position of the Society of Toxicologic Pathologists. Toxicol Pathol. 25(6):655.
(21) Treasurer J.W. and Holliday F.G.T. (1981). Some Aspects of the Reproductive Biology
of Perch Perca flaviatilis L.: A Histological Description of the Reproductive Cycle. J.
Fish Biol. 18:359-76.
(22) Tyler C.R. and Sumpter J.P. (1996). Oocyte Growth and Development in Teleosts. Rev
Fish Biol Fisheries 6:287-318.
(23) U.S. Environmental Protection Agency. (2002). A Short-term Test Method for
Assessing the Reproductive Toxicity of Endocrine-Disrupting Chemicals Using the
Page 123 of 124
-------
Final July 2015
Fathead Minnow (Pimephales promelas). EPA 600/R-01-067, pp. 141, 144.
(24) Van den Belt K.P., Wester W., van der Ven L.T.M., Verheyen R., and Witters H. (2002).
Effects of Ethynylestradiol on the Reproductive Physiology in Zebrafish (Danio rerio):
Time Dependency and Reversibility. Environ Toxicol and Chem 21:767-775.
(25) Wester P.W., van der Ven L.T.M., van den Brandhof E.J., Vos J. H. (2003).
Identification of Endocrine Disruptive Effects in the Aquatic Environment: A Partial Life
Cycle Assay in Zebrafish. RIVM Report 6409200001/2003, pp.31-39, 40-49.
(26) Zillioux E.J., Johnson I.C., Kiparissis Y., Metcalfe C.D., Wheat J.V., Ward S.G., Liu H.
(2001). The Sheepshead Minnow as an In Vivo Model for Endocrine Disruption in
Marine Teleosts: A Partial Life Cycle Test with 17a-ethynylestradiol. Toxicol. Chem.
20:1968-1978.
Page 124 of 124
------- |