

     Chemical Safety
     and Pollution Prevention
     (7101)
EPA No. 740-C-15-002
July 2015


Endocrine Disruptor Screening Program Test Guidelines



OCSPP 890.2200:

Medaka Extended One Generation Reproduction Test (MEOGRT)








                                       




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, et seq.), the Federal Insecticide, Fungicide and Rodenticide Act (FIFRA) (7 U.S.C. 136, et 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.


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.  
      (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., Leydig 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 generations of medaka to the test substance dissolved in treatment tank water.  The F0 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 F0 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 to14 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
                                       
                                      F0
                                       1
                                       2
                                       3
                                       4
                                       
                                       
                                       
                                       
                                       
                                       
                                       
                                       
                                       
                                       
                                       
                                       
                                       
                                       
                                       
                                       
                                Life-stage Key
                                       
                                      F1
                                       
                                       
                                       
                                       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
                                      F0
                                       
                                       
                                       
                                       
                                       
                                       
                                       
                                       
                                       
                                       
                                       
                                      F1
                                       
                                       
                                       
                                     Adult
                                       
                                   Fertility
                                      F0
                                       
                                       
                                       
                                       
                                       
                                       
                                       
                                       
                                       
                                       
                                       
                                      F1
                                       
                                       
                                       
                                       
                                       
                                   Hatching
                                       
                                       
                                       
                                       
                                      F1
                                       
                                       
                                       
                                       
                                       
                                       
                                       
                                       
                                       
                                       
                                       
                                       
                                       
                                      F2
                                       
7 groups of replicates
*  5 for test substance treatments
*  2 for controls (4 if solvent is used)
Within-group design
*  12 reps for reproduction, pathology, growth, SSC (wks 10-18)
*  6 reps for hatching, survival, vtg¸ growth, (wks 1-9)
                                       
                                   Survival
                                       
                                       
                                       
                                       
                                       
                                      F1
                                       
                                       
                                       
                                       
                                       
                                      F1
                                       
                                       
                                       
                                       
                                       
                                      F1
                                       
                                       
                                       
                                       
                                    Growth
                                       
                                       
                                       
                                      F0
                                       
                                       
                                       
                                       
                                       
                                       
                                       
                                      F1
                                       
                                       
                                       
                                       
                                       
                                      F1
                                       
                                       
                                       
                                       
                                 Vitellogenin
                                       
                                       
                                       
                                       
                                       
                                       
                                       
                                       
                                       
                                       
                                       
                                      F1
                                       
                                       
                                       
                                       
                                       
                                       
                                       
                                       
                                       
                                       
                                 Secondary sex
                                       
                                       
                                       
                                       
                                       
                                       
                                       
                                       
                                       
                                       
                                       
                                      F1
                                       
                                       
                                       
                                       
                                       
                                      F1
                                       
                                       
                                       
                                       
                                Histopathology
                                       
                                       
                                       
                                       
                                       
                                       
                                       
                                       
                                       
                                       
                                       
                                       
                                       
                                       
                                       
                                       
                                       
                                      F1
                                       
                                       
                                       
                                       
                                  Study Week
                                       1
                                       2
                                       3
                                       4
                                       5
                                       6
                                       7
                                       8
                                       9
                                      10
                                      11
                                      12
                                      13
                                      14
                                      15
                                      16
                                      17
                                      18
                                      19
                                       
                                       
                                       

      (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 F0 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
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 F0 pairs, meaning initially six replicates per test substance treatment and 12 replicates in the control group(s).  
      (3)  Endpoints.  In the F0 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 F0 generation, growth parameters are measured.  In addition, hatching success is recorded for the embryos spawned by F0 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 F1, fish are sampled for growth, secondary sex characteristics, gonad phenotype, and liver vitellogenin (vitellogenin 1; vtg1) mRNA assessment or liver vitellogenin protein assessment.  Survival data are collected for the F1 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)
F1, F2
Juvenile
(4 wpf)
Survival (hatch to 4 wpf)
F1 

Subadult
(9 wpf)
Survival (4 to 9  wpf)
F1


Growth
(length and weight)


Vitellogenin


Secondary sex characteristics
(anal fin papillae)


External sex ratio


Time to 1[st] spawn

Adult
(12-14 wpf)
Reproduction
(fecundity and fertility for 21 days)
F0, F1

Adult
(15 wpf)
Survival (from 9 to 15 wpf)
F1


Growth
(length and weight)


Secondary sex characteristics
(anal fin papillae)


Vitellogenin (VTG protein)


Histopathology
(gonad, liver, kidney)

*These endpoints are to be statistically analyzed.
 	(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 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 F0 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, Cl, SO4), 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.

            	 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 μg/L
Residual chlorine 
10 μg/L
Total organophosphorous pesticides 
50 ng/L
Total organochlorine pesticides plus polychlorinated biphenyls 
50 ng/L
Total organic chlorine 
25 ng/L
Aluminum 
1 μg/L
Arsenic 
1 μg/L
Chromium 
1 μg/L
Cobalt 
1 μg/L
Copper 
1 μg/L
Iron 
1 μg/L
Lead 
1 μg/L
Nickel 
1 μg/L
Zinc 
1 μg/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[(R)], 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. 
       (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 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 uL/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.  
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); 
F1: 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
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 (F1) for fecundity assessment. See Figure 1.
Number of organisms per test
A minimum of 84 fish in F0 and 504 fish in F1 will be used.  If a solvent control is needed, then a minimum of 108 fish in F0 and 648 fish in F1 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)(1)(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  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 16[th] 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.   
The dmy gene information is used to establish XX  -  XY breeding pairs (in both F0 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 F0 generation and in Study Weeks 15 through 17 in the F1 generation.  F0 and F1 eggs are collected daily from each breeding pair for 21 consecutive days.  Eggs are gently 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 F0 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 F0 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 F0 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 F0 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.
	(3) Detailed Timeline.  A timeline for the MEOGRT was described previously in Table 1.  The MEOGRT includes 4 weeks of exposure to F0 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 F0 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 F1) and the F2 generation for two weeks until hatching. The exposure continues through development and reproduction in the F1 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(R) 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 (F0)
The F0 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 (F0 and F1)
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 F1 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 F0 to produce more eggs to reach the recommended 20 per replicate.
On Test Day 24, the F0 breeding pairs are humanely killed, and weight and length are recorded. If necessary F0 breeding pairs maybe kept for an additional 1-2 days in order to restart F1.
Study Weeks 5-6 (F1)
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 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. 
The number of eleutheroembryos are recorded and hatching success is calculated in each replicate.
Study Weeks 6-11 (F1)
  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.  (F1)
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. (F1)
The exposure continues as the subadult breeding pairs develop into adults. 
Study Weeks 15-17. (F1)
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 (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.
                                       
                                       
Figure 1.  Pooling strategy used during the MEOGRT.  The figure represents one treatment or (1/2) 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 Artemia 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 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.
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(R) 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 ug/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.

      
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 F1) should produce > 20 eggs/pair/day.
Fertility
Should be >80% for all eggs produced in F0 and F1 controls (i.e., F1 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.
                                 *       There should be sufficient reproduction in at least the third highest group and all lower groups of the F0 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 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 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.
                                       
Figure 2.  Flow chart for the recommended statistical procedures for MEOGRT data analysis. 
       (2)  Special Data Analysis Considerations.
      (i)  Use of Compromised Treatment Levels.  Several factors are considered when determining whether a replicate or entire treatment demonstrates overt toxicity and should be removed from analysis.  Overt toxicity is defined as >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 μl/L) to avoid potential effect of the solvent on endpoints measured (Ref. 32) and solvent concentrations should not exceed 100 μl/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 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.  
      (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-tert-pentylphenol on medaka (Oryzias latipes). Environ Toxicol Chem 22:1487-1496.
   6) Yokota H, Seki M, Maeda M, Oshima Y, Tadokoro H, Honjo T, Kobayashi K. 2001. Life-cycle toxicity of 4-nonylphenol to medaka (Oryzias latipes). Environ Toxicol Chem 20:2552-2560
   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., 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.
   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.
   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 54:330-338.
   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-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.
   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.


(l)  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 Vtg1 mRNA in Liver.......................................	61
         (11) Data Reporting Templates.......................................................	64
         (12) Histopathology Guidance.......................................................	66











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) 


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. 
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.

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.  
(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.

(c)   Isolation Chambers (Figure 4).  From Study Week 6 through Study Week 11 of the F1 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.



 


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.  

 
(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. 
      
(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 dilution 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 piston 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.
         
APPENDIX 3:  Example Daily Calendar. 

Medaka Extend One-Generation Reproduction Test  Daily Protocol
Enter Fertilization Date (F0 Generation):
4/9/2012
 
Chemical:

 
 
Enter Start Date (move F0 breeding pairs into system):
6/25/2012
 
Temp:
25o C
 
 
 
 
 
Event 
(1 row for each day of study)
Day
Date
Calendar Day
Test Wk
Test DayTest Day
F1 Wk
F1 dpf
F2 dpf
1st Tank to Feed
Brine Shrimp (dry wt.)/fish/day
Move breeding pairs into exposure system
Start exposure                                    
Mon
25-Jun-12
176
1
1
 
 
 
37
22.5 mg
Loading Phase
Tue
26-Jun-12
177
 
2
 
 
 
5
 
Loading Phase
Wed
27-Jun-12
178
 
3
 
 
 
17
 
Loading Phase
Thu
28-Jun-12
179
 
4
 
 
 
13
 
Loading Phase
Fri
29-Jun-12
180
 
5
 
 
 
7
 
Loading Phase
Sat
30-Jun-12
181
 
6
 
 
 
42
 
Loading Phase
Sun
1-Jul-12
182
 
7
 
 
 
16
 
Loading Phase
Mon
2-Jul-12
183
2
8
 
 
 
39
22.5 mg
Loading Phase
Tue
3-Jul-12
184
 
9
 
 
 
15
 
Loading Phase
Wed
4-Jul-12
185
 
10
 
 
 
24
 
Loading Phase
Thu
5-Jul-12
186
 
11
 
 
 
20
 
Loading Phase
Fri
6-Jul-12
187
 
12
 
 
 
22
 
Loading Phase
Sat
7-Jul-12
188
 
13
 
 
 
31
 
Loading Phase
Sun
8-Jul-12
189
 
14
 
 
 
38
 
Loading Phase
Mon
9-Jul-12
190
3
15
 
 
 
13
22.5 mg
Loading Phase
Tue
10-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
14-Jul-12
195
 
20
 
 
 
29
 
siphon tanks and remove eggs
Sun
15-Jul-12
196
 
21
 
 
 
8
 
Start F1 gen (Fill first set of incubators) 
Mon
16-Jul-12
197
4
22
1
0
 
1
22.5 mg
Second day of F1 egg collection if necessary ( Fill second set of incubators)
Tue
17-Jul-12
198
 
23
 
1
 
13
 
Terminate F0 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

Sat
28-Jul-12
209
 
34
 
12
 
26
0.7 mg
 
Sun
29-Jul-12
210
 
35
 
13
 
29
0.8 mg
Discard unhatched eggs from first spawn.
Mon
30-Jul-12
211
6
36
3
14
 
16
1.0 mg
Discard unhatched eggs from second spawn.
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
 
Sun
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
 
 
Sun
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
 
 
Sun
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
 
 
Sun
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
1-Sep-12
244
 
69
 
47
 
28
 
 
Sun
2-Sep-12
245
 
70
 
48
 
36
 

Mon
3-Sep-12
246
11
71
8
49
 
4
13.5 mg

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
 
Fin 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 fertility-fecundity 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
Fertility-Fecundity Assessment
Tue
9-Oct-12
282
 
107
 
85
 
 
 

Wed
10-Oct-12
283
 
108
 
86
 
 
 

Thu
11-Oct-12
284
 
109
 
87
 
 
 
 
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
Fertility-Fecundity 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
23-Oct-12
296
 
121
 
99
1
 
 
 
Wed
24-Oct-12
297
 
122
 
 
2
 
 
 
Thu
25-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 eggs from first spawn.
Mon
5-Nov-12
309
20
134
17
 
14
 
 
Discard unhatched eggs from second spawn.
Tue
6-Nov-12
310
 
135
 
 
15
 
 

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 μl proteinase K, and vortex.  Alternatively, Buffer ATL and proteinase K can be premixed immediately before sampling large numbers of fin clips and 200 μl 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 μl Buffer AL before vortexing another time.  Add 200 μl 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 μl Buffer AW1, centrifuge at greater than 6000 x g for 1 minute (discard flow-through/collection tube) then wash with 500 μl 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 μl 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.

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 et 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(R) assay is used to detect amplification of the dmy gene.  

This protocol utilizes a Taqman(R) 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(R) assay (5' DNA exo-nuclease assay) uses this 5' exo-nuclease activity in real time detection of PCR product.  A Taqman(R) 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(R) 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. 

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 et al. 2009 and Shinomiya et al. 2004.

   (b)  Major Materials and Reagents.

   * TaqMan(R) Universal PCR Master Mix (Applied Biosystems) 
   * 18S Reagent (Applied Biosystems))
   * 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(R) Universal PCR Master Mix (Master Mix)
12.5 μl
18S reagent 
0.25 μl
dmy Forward Primer (10 μM)
1.75 μl
dmy  Reverse (10 μM)
1.75 μl
dmy  Probe (1250 nM)
4.0 μl

Using a pre-chilled 96 well optical plate, load each well with 20.25 μl of Master Mix and add 4.75 μl 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.  
      
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). 
      
                                       

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). 


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).  

      (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. 

      


	
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 (Vtg I) 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).

      
      
      Figure 1.  Cuts made during necropsy of subadult samples.
      
      
      
      
      
      Figure 2.  Dissection of the liver at the subadult sampling.
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.


      Figure 1.  Anal fin papillae. 

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 (Vtg1) 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 Vtg1 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 Vtg1 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 RNAlater (Qiagen) prior to use.). 

   (2)  RNA Extraction using RNeasy Protect Mini Kit (Qiagen).  Follow the manufacturer's instructions to extract RNA.(RNeasy Mini Handbook 4[th] edition, September 2010) with the following exception.  Disruption of a RNAlater preserved liver is done in a 2 ml microcentrifuge tube with the liver, 600 μl 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 Vtg1 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 μl 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.

APPENDIX 9: Example Protocol for the Preparation of RNA for Standard Curve Used during Vtg1 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 (Vtg1; 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 VtgI 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 Vtg1 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 Vtg1 QPCR.  The intent is for this "standard RNA" to be replicated with the same efficiency as the target sequence (Vtg1 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 Vtg1 amplicon that would produce cDNA that is complimentary to the Vtg1 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 Vtg1 mRNA and its size in base pairs are used to calculate the number of copies of the Vtg1 target sequence present per unit volume.  The Vtg1 mRNA is serially diluted to a known copy number per unit volume (μl).  The dilution series is used to produce a standard curve, which can then be used determine the absolute copy number of Vtg1 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)


(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 /μl.

      (2)  Synthesizing cDNA from liver RNA.  Adjust the RNA concentration to 200 ng/μl 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 μl
RNasing(R) (Promega)
1 μl
Oligo d(T) (Promega)
1 μl
5x RM buffer (in Superscript(R) II Kit, Invitrogen)
4 μl
0.1M DDT (in Superscript(R) II kit, Invitrogen)
2 μl
dNTP (10mM, Promega U1511)
4 μl
Reverse transcriptase (Superscript(R) II, Invitrogen)
1 μl
      
Proceed with reverse transcription for 1 hour at 37°C.  Each sample consists of 16 μl of the above reaction mix + 4 μl RNA at 200 ng/μl from above.  The result is cDNA from the mRNA in the RNA samples.  Since Vtg1 mRNA is highly represented in the RNA samples, the cDNA produced from each of those samples will be well populated with Vtg1 cDNA.   This procedure is the reverse transcription step of RT-PCR. 

      (3)  Amplifying the Vtg1 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). 


   Table 2. Amplification Reaction Mix.
Reagent
Per Sample Volume
RNase/DNase free water
25.36 μl
10x RM buffer (from AmpliTaq Gold, Applied Biosystems)
4 μl
MgCl2 (from AmpliTaq Gold)
2.24 μl
Vtg1 Standard Forward Primer 

(TAATACGACTCACTATAGGGAGATTCCTCGGATACGGCACAAT)
2 μl
Vtg1 Standard Reverse  Primer
(TAGACAGCTTTGCTGTAACGTAAGC)
2 μl
AmpliTaq Gold (Applied Biosystems,)
0.4 μl
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 Vtg1 cDNA to Vtg1 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 Vtg1 amplicon, synthesizes RNA that is nearly identical to the target sequence in the Vtg1 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).


Table 4. Reaction Mix to Convert cDNA to RNA.
Reagent
Per Sample Volume
cDNA (from above)
12 μl
10x buffer (from MEGAScript)
2 μl
10mM ATP (from MEGAScript)
1 μl
10mM CTP (from MEGAScript)
1 μl
10mM GTP (from MEGAScript)
1 μl
10mM UTP (from MEGAScript)
1 μl
RNA polymerase (Ambion MEGAScript, AM1333)
2 μl

Incubate the reaction mix for 1 hour at 37°C.  Add 1 μl 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 μg/μl.  Calculate the number of copies per μl 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)
      
      
      
      Example:  For CycA plasmid that has a concentration of 152 ng/μl = 1.52 X 10[ - 7] g/μl and a total size of 3,397 bp:
      
      	MW = 3,397 bp X 660 daltons/bp = 2.24 X 10[6] daltons
      	1 mol = 2.24 X 106 g
      	1 mol = 6 X 1023 molecules (= copies)
      
      
      
The quality of the RNA should be determined (Bioanalyzer 2100; RNA 6000 Nano Kit, Agilent Technologies or equivalent) following the manufacturer's instructions.  The Vtg1 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 μl to produce a series that is diluted a factor of 10 between each step, for instance 1010, 10[9], 10[8], etc.  To do this, since the copy numbers for the standards in this example are based upon loading 4 μl of standard  RNA into each well, the standard that has 1 x 10[10] copies per well (STD10) will have a concentration of 2.5 X 10[9] copies/ μl.  For instance, if the standard RNA has a concentration of 2.3 X 1011 copies/μl then this would be diluted to 2.5 x 10[9] copies/μl by adding 91 μl of RNase/DNase free water to every 1 μl of the standard RNA.  The other standards (typically at least 7 dilution steps) are serially diluted from STD10 with the following formula maintained: STDN = 2.5 X 10[N-1] copies/μl. 
APPENDIX 10:  Quantifying Vtg1 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 (Vtg1; 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 Vtg1.    

The following procedure is an example of precise quantification of Vtg1 mRNA in subadult liver samples using Taqman(R)-based QPCR.  Other suitable methods may be used. In this example, Vtg1 mRNA extracted from the liver of subadult Japanese medaka (sample procedure in Appendix 7), is amplified using QPCR and quantified using a Taqman(R) protocol. 

The following is a brief summary of the Taqman(R) 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(R) assay (5' nuclease assay) uses this 5' exo-nuclease activity in real time detection of PCR product.  A Taqman(R) 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(R) 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)


   (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/μl or copies/μl).
      (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)
μl/reaction
5x Taqman Buffer
6.0
Mn2+ acetate
3.6
dATP
0.9
dGTP
0.9
dCTP
0.9
dUTP
0.9
Vtg1 Forward (10 μM)
5'-AGGCAGTTTCTAAGGGCGAAC-3'
1.5
Vtg1 Reverse (10 μM)
5'-TGAATGGGCATAATCTTTGTGATT-3'
1.5
Vtg1 Probe (1250 nM)
5'-fam-TTTGGGAAATGCAAGACACCCTA-bhq-3'
4.8
rtTH polymerase
1.2
AmpErase(TM)
0.2
RNase/DNase-free water
3.6
Total
26



















Add 26 μl of the master mix to each well and then 4 μl 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
1
95°C
15 seconds
45
60°C
30 minutes

60°C
1 minute

95°C
5 minutes



Table 3. Example Plate Layout.

1
2
3
4
5
6
7
8
9
10
11
12
A
NTC
STD3
STD3
STD3
Sample 3
Sample 3
Sample 11
Sample 11
Sample 19
Sample 19
Sample 27
Sample 27
B
NTC
STD4
STD4
STD4
Sample 4
Sample 4
Sample 12
Sample 12
Sample 20
Sample 20
Sample 28
Sample 28
C
known
STD5
STD5
STD5
Sample 5
Sample 5
Sample 13
Sample 13
Sample 21
Sample 21
Sample 29
Sample 29
D
known
STD6
STD6
STD6
Sample 6
Sample 6
Sample 14
Sample 14
Sample 22
Sample 22
Sample 30
Sample 30
E
Sample 1
STD7
STD7
STD7
Sample 7
Sample 7
Sample 15
Sample 15
Sample 23
Sample 23
Sample 31
Sample 31
F
Sample 1
STD8
STD8
STD8
Sample 8
Sample 8
Sample 16
Sample 16
Sample 24
Sample 24
Sample 32
Sample 32
G
Sample 2
STD9
STD9
STD9
Sample 9
Sample 9
Sample 17
Sample 17
Sample 25
Sample 25
Sample 33
Sample 33
H
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/μl; 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 Vtg1 mRNA per μl is recorded for each specific sample. For each sample, the derived copy number per μl and the concentration in ng/μl of the liver RNA is used to calculate the Vtg1 copy number per ng of total RNA. In addition, performers of this analysis should be familiar with the published MIQE guidelines (Bustin et al. 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. 
      
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


      
      (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/.  

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


      
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 NaHCO3/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 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).  
      


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.

        

      

   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.

      
        
            
   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 in 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.

      (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., F1, 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). 

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 al., 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 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 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).
      
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.
      
      (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 et al., 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.



Figure 8.  Endocrine pancreas, islet cell lesions.  A and B:  Large islet (Brockman body) from a control medaka.  It is not uncommon to observe large or bizarre looking cells in normal islets.  Bar = 25 um.  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 um (A and C), 25 um (B and D).  





Figure 9.  Gonads, germ cell neoplasms.  A: Dysgerminoma in the testis of an adult male.  The caudal pole of the testis is effaced by a mass (arrows) consisting of oogenic tissue.  B: 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 17β-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 um (A), 50 um (B).



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 um (A), 100 um (B), 25 um (C).



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 germ cell neoplasm (dysgerminoma).  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 um.




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 um. 



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.  A: 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).  B: 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 um (A), 50 um (B).



Figure 14.  Kidney, mineralization.  A: Renal tissue from an adult male.  At low magnification, extensive dilation of tubular lumina and Bowman's spaces is evident.  B: 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 um (A), 50 um (B).


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 with 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.  For EDC 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 um (A and C), 25 um (B and D).



Figure 16.  Kidney, tubular eosinophilia.  A: 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.


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 um.



Figure 18.  Liver, hepatocyte basophilia, increased.  A: Liver from an adult control male.  B: Liver from an adult male exposed to 100 ug/L 4-tert-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 17β-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 um.


Figure 19.  Liver, primary proliferative lesions.  A: 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 um (A and B), 250 um (C).



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.  A: Intravascular proteinaceous fluid (p) in the testis of an adult male exposed to 17u-estradiol at 100 ng/L for 4 weeks.  B: Intravascular proteinaceous fluid (p) in the ovary of an adult female exposed to 4-tert-octylphenol at 90 ug/L for eight weeks.  C: Heart from an untreated control fish.  D: Heart with intravascular proteinaceous fluid (p).  Bar = 25 um (A), 50 um (B, C, and D).




Figure 21.  Ovary, chorion.  Usually pale to dark eosinophilic and refractile, 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 um.




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 inhibitors such as prochloraz and fadrozole, and the non-aromatizable androgen trenbolone.  H&E, Bar = 500 um.




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.




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. 




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.  Histopathologically, 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 um.




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 um (Grades 2 through 4).



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 (EVS) 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 um.




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.  B: 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.  D:  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.  F: Mature spawning follicle (arrow).  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 um (A), 50 um (B through D), 100 um (E and F).
 

Figure 29.  Ovary, perifollicular cell hyperplasia / hypertrophy, grading.  Exposure to aromatase inhibitors (e.g., fadrozole, prochloraz) 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 um (all).  



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 um.



Figure 31.  Ovary, post-ovulatory follicles.  A number of post-ovulatory 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 um.  



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 um (A and B).




Figure 33.  Ovary, spermatogenesis.  A and B: 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 um (A), Bar = 25 um (B).
S
S
S
S



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 um. 


                                       

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 ug/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 um.  




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.  B: 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 um.



Figure 37.  Testis, germinal epithelium.  Normal testis from an adult male medaka.  The double arrow indicates 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.




Figure 38.  Testis, hypoplasia.  A and B: 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 um (A and C), 
25 um (B and D).




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 um.



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.  B: 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 um (A and B).  



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).




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 um (A and B).



Figure 43.  Testis, Sertoli cells.  Sertoli cells (arrows) tend to have sharply-defined elongated or triangular nuclei, variably evident nucleoli, and cytoplasm that is often indistinct.  The cytoplasmic arms of a Sertoli cell encircle a clonal group of spermatogenic cells, forming 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.




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 arms 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.




Figure 45.  Testis, spermatogenic cell types.  A: Spermatogonia.  The largest of the spermatogenic cells (~ 5-10 um), spermatogonia generally have pale vesicular nuclei, prominent nucleoli, variably distinct nuclear membranes, perinuclear cytoplasmic granules, and moderate amounts of granular cytoplasm (arrow).  Spermatogonia B are smaller than spermatogonia A, and spermatogonia B are usually present in larger clusters (e.g., >4 cells).  B: Spermatocytes.  Derived from spermatogonia, spermatocytes are of intermediate cell size (~ 4-6 um), and have comparatively dense nuclei and minimal to moderate amounts of indistinct cytoplasm.  Spermatocyte nuclei are usually evident in one of three meiotic phases: pachytene, leptotene, or zygotene.  Primary spermatocytes (p) are larger than secondary spermatocytes (s), and the latter are derived from primary spermatocytes following the first meiotic division.  Spermatocytes are usually one of the most abundant spermatogenic cells, and they tend to contribute to the largest spermatocysts.  C: Spermatids.  Derived from spermatocytes following the second meiotic division, these cells have dense nuclei and narrow rims of eosinophilic cytoplasm.  They are the smallest cells within the germinal epithelium (~2-3 um), and the cells lose their cytoplasmic attachments to one another during spermiogenesis.  D: Spermatozoa.  These cells have dark, round nuclei and minimal or no apparent cytoplasm.  Tails are generally not apparent in histologic sections.  Spermatozoa are the smallest spermatogenic cells (~ 2 um), and exist as scattered individual cells within tubular lumen.  Bar = 15 um (A), 8 um (B), 4 um (C and D).




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 no 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 um.



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 um (all).


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 and B: 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 ug/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 E), Bar = 25 um (B, D, and F).




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).

(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 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 3A 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 > (2/3) EWT
      *             Stage 2: EWG (2/3) to > (1/2) EWT
      *             Stage 3: EWG (1/2) to > (1/4) EWT
      *             Stage 4: EWG < (1/4) EWT
            


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 um (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 vitellogenic.
   * Stage 4  -  Late development/hydrated: majority are late vitellogenic and mature / spawning follicles; follicles are larger as compared to Stage 3.
   * Stage 5  -  Post-ovulatory: predominately spent follicles, remnants of theca externa and granulosa.
               
	Figure 51 illustrates some of these stages.
            


Figure 51. Staging the ovary.  Ovaries from four adult female medaka.  Bar = 100 um (Stages 0 and 1), 250 um (Stages 2 and 4). Stages 3 and 5 are not illustrated.

(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 17β-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., Denslow N.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 
17β-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.
        
         (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 (Pimephales promelas).  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 6[th] International Symposium on the Reproductive Physiology of Fish, pp. 233-236.
        
         (14)         Palace V.P., 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 Resour. 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 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 17α-ethynylestradiol. Toxicol. Chem. 20:1968-1978.
        
        
      
      

      
