
Office of Chemical Safety and Pollution Prevention (7101)
EPA Pub No. 740-C-15-001  
July 2015


Endocrine Disruptor Screening Program Test Guideline



OCSPP 890.2300:

Larval Amphibian Growth and Development Assay (LAGDA)




                                       
                                       
                                       
                                       

  
  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 890.2300:  Larval Amphibian Growth and Development Assay (LAGDA)

      (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 Larval Amphibian Growth and Development Assay (LAGDA) is a definitive Tier 2 test in the EDSP designed to characterize adverse apical effects on the development, growth and reproductive organ development in amphibians and to establish a quantitative relationship between the exposure concentration and that adverse effect.  Adverse effects identified by the LAGDA may be caused by endocrine interaction of a chemical in amphibians, especially those active within the hypothalamic-pituitary-thyroid (HPT) and hypothalamic-pituitary-gonadal (HPG) systems.  The method described in this guideline is derived from validation work on African clawed frog (Xenopus laevis) by the EPA with supporting work in Japan.  The LAGDA test guideline would follow other guidelines 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 241 published by the Organization for Economic Cooperation and Development (Ref. 2).
      
      (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 amphibians and establish a quantitative relationship between the dose and that adverse effect(s).  The LAGDA test guideline uses the African clawed frog (Xenpous laevis) as a surrogate to identify and characterize the adverse consequences of exposure to substances which interfere with the normal development and growth of amphibians from embryo-larval development, through metamorphosis and early juvenile development. The assay is intended to evaluate effects from exposure to pesticides and other chemicals, both through endocrine and non-endocrine mechanisms, during the larval and juvenile stages that may adversely affect populations.

(c)  Introduction.  
Much of our current understanding of amphibian biology has been obtained using the laboratory model species X. laevis.  This species can be routinely cultured in the laboratory, ovulation can be induced using human chorionic gonadotropin (hCG) and animal stocks are readily available from commercial breeders. 
Like all vertebrates, reproduction in amphibians is under the control of the hypothalamic pituitary gonadal (HPG) axis (Ref. 3), while development and growth is regulated by the hypothalamic pituitary thyroid (HPT) axis.  Estrogens and androgens are mediators of this endocrine system, directing the development and physiology of sexually-dimorphic tissues.  There are three distinct phases in the life cycle of amphibians when this axis is especially active: (1) gonadal differentiation during larval development, (2) development of secondary sex characteristics and gonadal maturation during the juvenile phase and (3) functional reproduction of adults.  Each of these three developmental windows are likely susceptible to endocrine perturbation by certain chemicals, ultimately affecting growth and reproductive fitness by the organisms. 
The gonads begin development at NF stage 43, when the bipotential genital ridge first develops.  Differentiation of the gonads begins at NF stage 52 when primordial germ cells either migrate to medullary tissue (males) or remain in the cortical region (females) of the developing gonads (Ref. 4).  This process of sexual differentiation of the gonads was first reported to be susceptible to chemical alteration in Xenopus in the 1950's (Ref. 5; Ref. 6).  Exposure of tadpoles to estradiol during this period of gonad differentiation results in sex reversal of males that when raised to adulthood are fully functional females (Ref. 7; Ref. 8).  Functional sex reversal of females into males is also possible and has been reported following implantation of testis tissue in tadpoles (Ref. 9). However, exposure to an aromatase inhibitor also causes functional sex reversal in X. tropicalis (Ref. 10), this has not yet been shown to occur in X. laevis.  Historically, toxicant effects on gonadal differentiation have been assessed by histological examination of the gonads at metamorphosis and sex reversal could only be determined by analysis of sex ratios.  Until recently, there had been no means to directly determine the genetic sex of Xenopus.  However, recent establishment of sex linked markers in X. laevis make it possible to determine genetic sex and allows for the direct identification of sex reversed animals (Ref. 11).
In males, juvenile development proceeds as blood levels of testosterone increase corresponding with the development of secondary sex characteristics as well as testis development. In females, estradiol is produced by the ovaries resulting in the appearance of vitellogenin (Vtg) in the plasma, vitellogenic oocytes in the ovary and the development of oviducts (Ref. 12).  Oviducts are female secondary sex characteristics that function in oocyte maturation during reproduction (Ref. 13).  Jelly coats are applied to the outside of oocytes as they pass through the oviduct and collect in the ovisac, ready for fertilization.  Oviduct development appears to be regulated by estrogens as development correlates with blood estradiol levels in X. laevis (Ref. 14) and X. tropicalis (Ref. 12).  The development of oviducts in males following exposure to polychlorinated biphenyl compounds (Ref. 15) and 4-tert-octylphenol (Ref. 16) has been reported.
Though capable of detecting perturbations of normal reproductive development through HPG axis interference, the LAGDA is also designed to detect substances that interfere with normal function of the HPT axis.  Amphibian metamorphosis from larval to adult life stages is a well-studied thyroid-dependent process (Ref. 17), which responds to substances that are active within the HPT axis (Ref. 3).  Metamorphosis occurs in all major chordate groups with the exception of amniotes (Ref. 18; Ref. 19) and is developmentally comparable to post-embryonic organogenesis in mammals (Ref. 20). Three primary physiological processes define amphibian metamorphosis: (1) reabsorption or regression of tissue/organs that are primarily functional during the larval life stage (e.g., gills), (2) remodeling larval organ systems to their adult forms (e.g., lungs, gut, liver), and (3) development of new tissues (e.g., limbs) for the adult life stage that are not present in the early larval life stage.  These changes allow adult amphibians to occupy an ecological niche distinct from that occupied by larvae.  The LAGDA is intended to characterize the adverse consequences of exposure to substances which interfere with these metamorphic and developmental processes that may affect populations.
(d)  General Experimental Design.  
The general experimental design entails exposing X. laevis embryos at Nieuwkoop and Faber (NF) stage 8-10 to a minimum of four different concentrations of test chemical and control(s) until 10 weeks after the median time to reach NF stage 62 in the control, with one interim sub-sample at NF stage 62  (See Ref. 4 for staging details).  There are four replicates in each test concentration with eight replicates in the control.  Endpoints evaluated during the course of the exposure include those indicative of generalized toxicity: mortality, abnormal behavior, abnormal morphological features, and growth determinations (length and weight), as well as endpoints designed to characterize potential apical consequences of endocrine toxicity acting through estrogen, androgen or thyroid-mediated physiological processes.
	(e)  Description of the Method.

	(1)  Test Species.  The representative amphibian test species is X. laevis, because this species is routinely cultured in laboratories worldwide, is easily obtainable through commercial suppliers, and is capable of having its genetic sex determined.  Reproduction can be induced in this species throughout the year using human chorionic gonadotropin (hCG) injections and the resultant embryos can be routinely reared in large numbers for use in this test protocol.  It is preferred that larvae used in the assay are derived from in-house cultures of adults. 
	
	(2)  Equipment and Supplies.  The following equipment and supplies are recommended for the LAGDA:

   * exposure system (described below);
   * glass or stainless steel aquaria (described below);
   * breeding tanks;
   * temperature controlling apparatus (e.g., heaters or coolers adjustable to 21+-1ºC);
   * thermometer; 
   * binocular dissection microscope and dissection tools; 
   * digital camera with at least 4 megapixel resolution and micro function; 
   * analytical balance capable of measuring to 0.001 mg or 1 ug; 
   * dissolved oxygen meter;
   * pH meter; 
   * light intensity meter capable of measuring in lux units; 
   * miscellaneous laboratory glassware and tools;
   * adjustable pipettes (10 to 5,000 μl) or assorted pipettes of equivalent sizes; 
   * the test chemical should be available in sufficient quantities, preferably from one lot, to conduct the study; and
   * analytical instrumentation appropriate for the chemical on test or contracted analytical services.
   
	(3)  Chemical Testability.  The LAGDA uses an aqueous exposure methodology whereby the test chemical is introduced into the test chambers via a flow-through system.  Flow-through methods however, introduce constraints on the types of chemicals that can be tested, as determined by the physicochemical properties of the compound.  Therefore, prior to using this protocol, baseline information about the chemical that is relevant to determining its testability should be obtained.  Both the Organization for Economic Cooperation and Development (OECD) Guidance Document on Aquatic Toxicity Testing of Difficult Substances and Mixtures (Ref. 21) and the USEPA Guidance Document on Special Considerations for Conducting Aquatic Laboratory Studies should be consulted (OCSPP 850.1000; Ref. 22).  Characteristics which indicate that the chemical may be difficult to test in aquatic systems include: high 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.  If it has been demonstrated through range-finding or other preliminary trials that quantitation of the test substance cannot be conducted because the test substance concentration cannot be kept constant, or because there are other analytical difficulties, a successful test for the chemical is unlikely using a flow-through test system.  Should this be the case, it is recommended that the chemical not be tested using this protocol.

	(4)  Exposure System.  This test was developed using a flow-through diluter system.  The system components should have water-contact components of glass, stainless steel, and/or chemically inert materials (e.g., Teflon[(R)]).  Plastics may be used if they have been demonstrated to be suitable for testing with amphibians.  Exposure tanks should be glass or stainless steel aquaria, equipped with standpipes that result in an approximate tank volume between 4.0 and 10.0 L and a minimum water depth of 10 cm.  The system should be capable of supporting all exposure concentrations, a control, and a solvent control, if necessary, with four replicates per treatment (eight in the control).  The flow rate to each tank should be constant in consideration of both the maintenance of biological conditions and chemical exposure.  It is recommended that flow rates accommodate at least 10 tank turnovers per day to avoid chemical concentration declines due to metabolism by both the test organisms and aquatic microorganisms present in the aquaria or abiotic routes of degradation (hydrolysis, photolysis) or dissipation (volatilization, sorption).  The treatment tanks should be randomly assigned to a position in the exposure system to reduce potential positional effects, including slight variations in environmental conditions (e.g., temperature and light intensity).  Specific experimental conditions under which the protocol should be conducted are discussed below (Section (e)(7)(ii)).  Further information on setting up flow-through exposure systems can be obtained from the ASTM Standard Guide for Conducting Acute Toxicity Tests on Test Materials with Fishes, Macroinvertebrates, and Amphibians (Ref. 23). 

	(5)  Water Quality.  Any dilution water that is locally available (e.g., spring water or charcoal-filtered tap water) and permits normal growth and development of X. laevis can be used; evidence of normal growth in this water should be available.  Because local water quality can differ substantially from one area to another, analysis of water quality should be undertaken, particularly if historical data on the utility of the water for raising amphibian larvae is not available.  Special attention should be given that the water is reasonably free of copper, chlorine, and chloramines, all of which can be toxic to amphibians.  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, fluoride, perchlorate, and chlorate (a by-product of drinking water disinfection) and suspended solids should be made before testing begins and/or, for example, every six months where a dilution water is known to be relatively constant in quality.  Some chemical characteristics of acceptable dilution water are listed in Table 1. 
Table 1. 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 
Aluminium  
                                   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 


	(6)  Iodide Concentration in Test Water.  In order for the thyroid gland to synthesize thyroid hormones to support normal metamorphosis, sufficient iodide should be available to the larvae through a combination of aqueous and dietary sources.  Currently, there are no empirically derived guidelines for minimum iodide concentrations in either food or water to ensure proper development.  However, iodide availability may affect the responsiveness of the thyroid system to thyroid active agents and is known to modulate the basal activity of the thyroid gland, which deserves attention when interpreting the results from thyroid histopathology.  Based on previous work, successful performance of the assay has been demonstrated when dilution water iodide (I[-]) concentrations ranged between 0.5 and 10 μg/L.  Ideally, the minimum iodide concentration in the dilution water throughout the test should be 0.5 μg/L.  If the test water is iodine-deficient or is reconstituted from deionized water, iodine should be added at a minimum concentration of 0.5 μg/L (added as the sodium or potassium salt).  The measured iodide concentrations from the dilution water and the supplementation of the test water with iodine or other salts (if used) should be reported.  Iodine content may also be measured in food(s) in addition to test water.  

	(7)  Animal Care.

	(i)  Adult Care and Breeding.   Appropriate care and breeding of X. laevis is conducted using a standardized guideline, Frog Embryo Teratogenesis Assay Xenopus (FETAX) in ASTM (Ref. 24).  Housing and care of X. laevis are also described by Read (Ref. 25). To induce breeding, three to five pairs of adult females and males are injected intraperitoneally with human chorionic gonadotropin (hCG).  Female and male specimens are injected with approximately 800-1000 IU and 500-800 IU, respectively, of hCG dissolved in 0.6-0.9% saline solution (or frog Ringer's solution, an isotonic saline for use with amphibians; www.hermes.mbl.edu/biologicalbulletin/compendium/comp-RGR.html ).  Injection volumes should be about 10 ul/g body weight (~1000 ul).  Afterwards, induced breeding pairs are held in large tanks, undisturbed and under static conditions to promote amplexus.  The bottom of each breeding tank should have a false bottom of plastic or stainless steel mesh (e.g., 1.25 cm openings) which permits the eggs to fall to the bottom of the tank.  Frogs injected with hCG in the late afternoon will usually deposit most of their eggs by mid-morning of the next day.  After a sufficient quantity of eggs are released and fertilized, adults should be removed from the breeding tanks. 

Eggs are then collected and jelly coats are removed by L-cysteine treatment (Ref. 24).  A 2% L-cysteine solution should be prepared and pH adjusted to 8.1 with 1M NaOH.  This 21°C solution is added to a 500 mL Erlenmeyer flask containing the eggs from a single spawn and swirled gently for one to two minutes and then rinsed thoroughly 6-8 times with 21°C culture water.  The eggs are then transferred to a crystallizing dish and determined to be >70% viable with minimal abnormalities in embryos exhibiting cell division. 
 
	(ii) Environmental Conditions.  A summary of exposure conditions and frequency of monitoring is presented in Table 2.  During the exposure period, temperature of test solutions should be measured daily and dissolved oxygen and pH of the test solutions measured twice per week.  Conductivity, alkalinity, and hardness are measured once per month.  For the water temperature of test solutions, the inter-replicate and inter-treatment differentials (within one day) should not exceed 1.0°C.  Also, for pH of test solutions, the inter-replicate and inter-treatment differentials should not exceed 0.5. 
















Table 2.  Environmental conditions and frequency of monitoring.
Environmental condition
Targeted range
Frequency tested
Source water
Water temperature
21+-1°C
Daily
All tanks
Photoperiod
12:12  (hrs light:hrs dark)
n/a
n/a
Light intensity at water surface
600 to 2,000 lux (lumens/m[2])
Once
n/a
Test water dissolved oxygen               (% saturation)
>= 3.5 mg/L (>40%; however, >=60% preferred)
2x per week
All tanks by end of week
Test water pH
6.5-8.5
2x per week
All tanks by end of week
Test water conductivity (uS)
n/a
1x per month 
(rotate through replicates)
One replicate from each treatment and control
Test water hardness (mg/L of CaCO3)
10  -  250 mg CaCO3/L
1x per month
(rotate through replicates)
One replicate from each treatment and control
Test water alkalinity (mg/L of CaCO3)
10  -  250 mg CaCO3/L
1x per month
(rotate through replicates)
One replicate from each treatment and control
Relative humidity inside test system enclosure
>30%
1x per month post-metamorphosis
n/a

(iii) Husbandry.  The exposure tanks should be siphoned on a daily basis to remove uneaten food and waste products, being careful to avoid cross-contamination of tanks.  Care should be used to minimize stress and trauma to the animals, especially during movement of animals or aquaria, cleaning of aquaria, and manipulation of equipment adjacent to aquaria.  Stressful conditions/activities should be avoided such as loud and/or incessant noise, tapping on aquaria, vibrations in the aquaria, excessive activity in the laboratory, and rapid changes in environmental media (e.g., light availability, temperature, pH, dissolved oxygen, and water flow rates).  
		(iv) Feeding Regime.  Feed and feeding rates change during different life stages of X. laevis and are a very important aspect of the LAGDA.  Excessive feeding during the larval phase typically results in increased incidences and severity of scoliosis (bent tail) and should be avoided.  Conversely, inadequate feeding during the larval phase results in highly variable developmental rates among controls potentially compromising statistical power or confounding test results.  Therefore, use of the outlined feeding regime for larval and juvenile X. laevis in flow-through conditions is recommended (see Appendix 2).  Alternative feeding regimes are permissible providing the test organisms grow and develop satisfactorily according to the following performance criteria for control organisms (also outlined in Section (g)). 
               *       The median time to NF stage 62 is <=45 days.
               *       The mean weight at NF stage 62 is 1.0+-0.2 grams.
               *       The mean juvenile weight at test termination is 11.5+-3 grams.
   
 (A)  Larval feeding.  A recommended larval diet is provided in Appendix 2.  This diet consists of trout starter (e.g., Skretting USA, Tooele, UT), Spirulina algae discs (e.g., Wardley, Secaucus, NJ) and goldfish crisps (e.g., TetraFin[(R)] flakes from Tetra Sales, Blacksburg, VA) homogenized in culture (or dilution) water (see Appendix 2 for guidance).  This mixture is administered three times daily on weekdays and once daily on weekends.  Tadpoles are also fed 24-hour-old live brine shrimp nauplii (Artemia spp.) (e.g., Bio-Marine[(R)] Brand, Bio-Marine, Hawthorne, CA) twice daily on weekdays and once daily on the weekends starting on day 8 post-fertilization.  The larval feeding, which should be consistent in each test vessel, should allow appropriate growth and development of test animals in order to ensure reproducibility and transferability of the assay results: (1) the median time to NF stage 62 in controls should be <=45 days and (2) a mean weight of 1.0+-0.2 g at NF stage 62 in controls. 
 (B)  Post-NF stage 62 and juvenile feeding.  A recommended juvenile diet is provided in Appendix 2.  Once tadpoles have reached NF stage 62, the feeding regime consists of premium sinking frog food (e.g., Sinking Frog Food -3/32"; Xenopus Express, FL).  For larvae post-NF stage 62 and froglets (early juveniles), the pellets are briefly run in a coffee grinder, blender or crushed with a mortar and pestle in order to reduce their size.  Once juveniles are large enough to consume full pellets, grinding or crushing is no longer necessary.  The animals should be fed once per day.  The juvenile feeding should allow appropriate growth and development of the organisms to reach a mean weight of 11.5+-3 g in control juveniles at the termination of the assay.  
(8)  Analytical Chemistry.  Prior to initiating the study, the stability of the test compound (e.g., solubility, degradability, and volatility) and all analytical methods needed should be established using existing information or knowledge.  When dosing via the dilution water, it is recommended that test solutions from each replicate tank concentration be analyzed prior to test initiation to verify system performance.  During the exposure period, the concentrations of the test chemical are determined at appropriate intervals, preferably every week in all replicate tanks in each treatment group.  It is recommended that results be based on measured concentrations.  Also, the coefficient of variation (CV) of the measured test concentrations over the entire test period within a treatment should be maintained at 20% or less in each concentration.  When the measured concentrations do not remain within 80-120% of the nominal concentration (for example, when testing highly biodegradable or adsorptive chemicals), the effect concentrations should be determined and expressed relative to the arithmetic mean concentration for flow-through tests.  
The flow rates of dilution water and stock solution should be checked at appropriate intervals (e.g., three times a week) throughout the exposure duration.  In the case of chemicals which cannot be detected at some or all of the nominal concentrations (e.g., due to rapid degradation or adsorption in the test vessels, or by marked chemical accumulation in the bodies of exposed animals), it is recommended that the renewal rate of the test solution in each chamber be adapted to maintain test concentrations as constant as possible. 
(9)  Chemical Delivery: Preparation of Test Solutions.  To make test solutions in the exposure system, stock solution of the test chemical should be dosed into the exposure system by an appropriate pump or other apparatus.  The flow rate of the stock solution should be calibrated in accordance with analytical confirmation of the test solutions before the initiation of exposure, and checked volumetrically periodically during the test.  It is recommended that the test solution in each chamber be renewed at a minimum of 10 volume renewals/day. 
The method used to introduce the test chemical to the system can vary depending on its physicochemical properties.  Therefore, prior to the test, baseline information about the chemical that is relevant to determining its testability should be obtained.  Useful information about test chemical-specific properties include the structural formula, molecular weight, purity, stability in water and light, pKa and Log Kow, water solubility (preferably in the test medium) and vapor pressure as well as results of a test for ready biodegradability (OECD TG 301, Ref. 26; TG 310, Ref. 27). Solubility and vapor pressure can be used to calculate Henry's law constant, which will indicate whether losses due to evaporation of the test chemical may occur.  Conduct of this test without the information listed above should be carefully considered as the study design will be dependent on the physicochemical properties of the test chemical and, without these data, test results may be difficult to interpret or meaningless.  A reliable analytical method for the quantification of the test chemical in the test solutions with known and reported accuracy and limit of detection should be available. 
Water soluble test chemicals can be dissolved in aliquots of dilution water at a concentration which allows delivery at the target test concentration in a flow-through system.  Chemicals which are liquid or solid at room temperature and moderately soluble in water may require liquid:liquid or liquid:solid (e.g., glass wool column) saturators (Ref. 28).  While it may also be possible to dose very hydrophobic test chemicals via the feed, there has been no experience using that exposure route in this assay. 
Test solutions of the chosen concentrations are prepared by dilution of a stock solution.  The stock solution should preferably be prepared by simply mixing or agitating the test chemical in dilution water by mechanical means (e.g. stirring and/or ultrasonication).  Saturation columns/systems or passive dosing methods (Ref. 29) can be used for achieving a suitably concentrated stock solution.  The preference is to use a co-solvent-free test system; however, different test chemicals will possess varied physicochemical properties that will likely require different approaches for preparation of stock solutions.
If possible, solvents or carriers should be avoided because: (1) certain solvents themselves may result in toxicity and/or undesirable or unexpected responses, (2) testing chemicals above their water solubility (as can frequently occur through the use of solvents) can result in inaccurate determinations of effective concentrations, (3) the use of solvents in longer-term tests can result in a significant degree of "biofilming" associated with microbial activity which may impact environmental conditions as well as the ability to maintain exposure concentrations and (4) the absence of historical data that demonstrate that the solvent does not influence the outcome of the study, (5) use of solvents requires a solvent control treatment which has significant animal welfare implications, as additional animals are required to conduct the test.  For difficult to test chemicals, a solvent may be employed as a last resort, and the OECD Guidance Document on Aquatic Toxicity Testing of Difficult Substances and Mixtures should be consulted (Ref. 21) to determine the best method.  The choice of solvent will be determined by the chemical properties of the test chemical and the availability of historical control data on the solvent.  In the absence of historical data, the suitability of a solvent should be determined prior to conducting the definitive study.  In the event that use of a solvent is unavoidable, and microbial activity (biofilming) occurs, recording/reporting of the biofilming per tank (at least weekly) throughout the test is recommended.  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.  In cases where a solvent carrier is used, maximum solvent concentrations should not exceed 100 μl/L or 100 mg/L (Ref. 21), and it is recommended to keep the solvent concentration as low as possible (e.g, <20 μl/L) to avoid potential effects of the solvent on endpoints measured (Ref. 30).
      (10)  Exposure System Cleaning.  The exposure systems used for the LAGDA prepare and deliver treatment solutions over long periods of time.  A biofilm will invariably form on most wetted surfaces.  The microbiological community that establishes itself will often be able to metabolize the compound being assessed.  Detectable decreases in treatment concentration have been observed within three weeks of the start of the assay.  Routine cleaning and disinfection of all wetted components in the diluter/delivery system, including the aquaria themselves, is strongly recommended (approximately every three weeks).  As a precautionary note, the materials used in the construction of the diluter system 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 should be performed prior to use.  Analysis of chemical concentrations should be made before and after any cleaning of the exposure system.
An example protocol for cleaning would be to divert the exposure tank delivery lines to drain.  The exposure tanks 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 (e.g., Minncare[(R)] Cold Sterilant, a solution of 22% hydrogen peroxide and 4.5% peroxyacetic acid).  The cold sterilant solution should be mixed with water to a solution of 50X (2% concentration), and used to fill the toxicant delivery line circuit.  Allow the lines to soak without flow for a minimum of 20 minutes.  Rinse the delivery lines immediately with diluent water and check for residuals of the sterilant solution (e.g., with Minncare[(R)] Residual Test Strips).  Flush the lines until the cold sterilant concentration is below detection.  Reassemble the system and return the delivery lines to the exposure tanks. 
(11)  Randomization.  At all stages of the LAGDA and during daily tasks it is important to include randomization.  This includes, but is not limited to the following procedures:

    The order in which food is delivered to the aquaria should be randomized in some fashion to mitigate potential changes in food quantity or quality during the course of feeding.
    It is essential to randomize the selection of embryos when distributing them into the replicates at test initiation.
    When larvae at NF stage 62 or juveniles are sampled or netted, the order in which the replicate aquaria are sampled or manipulated should be random.
    Cleaning and siphoning of tanks should not be performed in the same order on a daily basis.  A separate set of tank cleaning supplies (e.g., siphons) per treatment and control to avoid cross-contamination of chemical and/or disease is recommended.
      (12)  Selection of Test Concentrations.  
      (i) Establishing the High Test Concentration.  For the purposes of this test, results from Tier I amphibian studies (i.e., Amphibian Metamorphosis Assay, Ref. 31; OECD 231, Ref. 32) should be used to the extent possible in determining the highest test concentration so as to avoid concentrations that are overtly toxic.  Other sources of information include: quantitative structure-activity relationships, read across and data from other existing amphibian studies such as the Frog Embryo Teratogenesis Assay - Xenopus (Ref. 24) and/or fish tests such as OECD TG229 (Ref. 33), TG234 (Ref. 34) and TG236 (Ref. 35) may contribute toward setting this concentration. 
Prior to running the LAGDA, it is recommended that a range finding experiment be conducted.  The range-finding exposure is initiated within 24 hours of fertilization and continued for 7-14 days (or more, if needed).  Note that if a solvent has to be used, then the suitability of the solvent (i.e., whether it may have an impact on the outcome of the study) could be determined as part of the range finding study.  The range finding test concentrations are set such that the intervals between test concentrations are no greater than a factor of 10.  The results of the range finding experiment should serve to set the highest test concentration in the LAGDA.
  
(ii) Test Concentration Range.  For the definitive test, a minimum of four concentrations of the test substance, plus a dilution water control are tested (and a solvent control, if used).  The results of the range finding experiment should serve to set the highest test concentration in the LAGDA.  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 test concentration differential between the highest and lowest should be approximately a factor of 10.  Determination of the spacing factor should be made based on concentration-response relationships characterized in previous studies (e.g., Tier 1 Amphibian Metamorphosis Assay, Ref. 31 or OECD 231, Ref. 32 and/or range-finding experiments), if available, in order to permit the highest likelihood of determining both a lowest observed effect concentration (LOEC) and no observed effect concentration (NOEC).  

(f)  Procedure.

(1)  Assay Overview.  The exposure is initiated with newly spawned embryos (NF stage 8-10) and continued until ten weeks after the median time to NF stage 62 (<=45 days from the assay initiation) in control group (Figure 1).  Generally, the duration of the LAGDA is 16 weeks (maximum 17 weeks).   Animals are examined daily for mortality and any sign of abnormal behavior.  As larvae reach NF stage 62, a larval sub-sample (5 animals per replicate) is collected and various endpoints are examined (Table 4).  Subsequently, a cull is performed to reduce tank density after all animals in the test have completed metamorphosis (i.e., reached NF stage 66)  or after 70 days from the assay initiation, whichever comes first.  The cull is carried out at random (but without sub-sampling) and the remaining animals (10 per tank) continue exposure until 10 weeks after the median time to NF stage 62 in the control.  At test termination (juvenile sampling), additional measurements are made (see Table 4). Illustrations of each stage of X. laevis development (from Ref. 4) are available on-line at http://www.xenbase.org/anatomy/alldev.do.

Figure 1.  Timeline for the LAGDA.  Images are adapted from Nieuwkoop and Faber (1994; Ref. 4).
a 41 days is the approximate median time to NF stage 62 in the controls; all control larvae will generally reach NF stage 62 over a 2 week period of time.     
[b] The cull takes place after all animals on test reach NF stage 66, or at 70 days, whichever comes first.


A minimum of four replicate tanks per test concentration and a minimum of eight replicates for the controls (and solvent control, if needed) should be used (i.e., the number of replicates in the control and any solvent control should be twice as large as the number of replicates of each treatment group to ensure appropriate statistical power).  Each replicate should contain the same number of animals; 20 animals per replicate is recommended, but replicates should contain no more than 30 animals.  While using the test method in this guideline, the minimum number of animals processed would be 15 (5 for NF stage 62 subsample and 10 juveniles).  However, additional animals are added to each replicate to factor in the possibility for unsuccessful hatch and/or mortality while maintaining the critical number of 15.

The test conditions for LAGDA are summarized in Table 3.


Table 3.  Test Conditions for LAGDA.

Test animal
Xenopus laevis
Initial larval stage
Nieuwkoop and Faber (NF) stage 8-10
Exposure period
Until 10 weeks after the median time to NF stage 62 in the clean water control
Test concentrations
Minimum of 4 different concentrations
Dilution water / laboratory control
Any water that permits normal growth and development of X. laevis (e.g., spring water or charcoal-filtered tap water)
Replication
4 replicate test vessels / test concentration
8 replicate test vessels / control
Exposure regime
Flow-through 
Test system flow-rate
Constant, in consideration of both the maintenance of biological conditions and chemical exposure (e.g., flow rate sufficient to provide 10 tank volume additions/day) 
Test solution / test vessel 
 4  -  10 L (10 cm minimum water depth) / glass or stainless steel
Initial larval density
~20 larvae / test vessel
Larval density after NF stage 66 cull
10 juveniles / test vessel (chosen randomly)
Feeding regime
See Appendix 2
Lighting
Photoperiod
12 hr light : 12 hr dark

Intensity
600 to 2000 lux (lumens/m[2]) at the water surface
Mean water temperature
21º +- 1°C (and the inter-replicate and -treatment differentials should be <1.0 ºC)
pH
6.5 to 8.5
Dissolved oxygen (DO) concentration
>=3.5 mg/L (>40% air saturation value, however, >=60% preferred)
Analytical chemistry schedule
initiation (Day 0), and weekly thereafter for each replicate tank at each concentration 
Control mortality
<= 20% in each replicate
Median time to NF stage 62 (controls)
<=45 days
Mean weight (controls)
NF stage 62 = 1.0+-0.2; At termination of assay = 11.5+-3 g
Concentrations of the test chemical in solution throughout the test
Maintain within +-20% of the mean measured values




As shown in Table 4, the endpoints evaluated during the course of the exposure are those indicative of toxicity including mortality, abnormal behavior such as clinical signs of disease and/or general toxicities, and growth determinations (length and weight), as well as pathology endpoints which may respond to both general toxicity and endocrine toxicity acting through estrogen-, androgen-, or thyroid-mediated pathways. 





Table 4.  Timing of endpoint measurements over the course of the LAGDA.

Endpoints*
Daily
Interim Sampling
(Larval Sampling at NF stage 62)
Test Termination
(Juvenile Sampling)
Mortality
X


Abnormal appearances - Clinical signs of disease and/or general toxicities
X


Time to NF stage 62

X

Histopathology  -  thyroid gland

X

Morphometrics (growth in weight and length)

X
X
Liver-somatic index (LSI)


X
Genetic/phenotypic sex ratios


X
Histopathology for gonads, reproductive ducts, kidney and liver


X
  * All endpoints are analyzed statistically. 

      (2)  Test Initiation.  Parent animals used for the initiation of the assay should have previously been shown to produce offspring that can be genetically sexed (Appendix 3).  After spawning of adults, embryos are collected, cysteine-treated to remove the jelly coat and screened for viability (Ref. 24).  Cysteine treatment allows the embryos to be handled during screening without sticking to surfaces.  Screening takes place under a dissecting microscope using an appropriately sized eye dropper to remove non-viable embryos.  It is preferred that a single spawn resulting in greater than 70% viability be used for the test.  Embryos at NF stage 8-10 are randomly distributed into exposure treatment tanks containing an appropriate volume of dilution water until each tank contains an equal number of embryos (20-30 embryos).  Embryos can be constrained to a similar location in the tank in a submerged beaker or glass petri dish that is later removed when larvae are free-swimming.  This method helps account for embryos and helps with observations of hatching success.  An appropriately sized container would be no less than 6 cm in diameter and would allow adequate test solution replacement, enough to avoid areas of stagnant test solution.  Embryos should be carefully handled during this transfer in order to minimize handling stress and to avoid any injury.  At 96 hours post fertilization, the tadpoles should have moved up the water column and begun clinging to the sides of the tank.
       
      (3)  Mortality and Morbidity.  All test tanks should be checked daily for dead animals and mortalities recorded for each tank.  Dead animals should be removed from the test tank as soon as observed.  The date, concentration, tank number and developmental stage of dead animals should be recorded as accurately as possible.  If the NF stage is not discernable, it should be categorized as either pre-NF stage 58 (pre-forelimb emergence), NF stage 58-NF stage 62, NF stage 63-NF stage 66 (between NF stage 62 and complete tail absorption), or post-NF stage 66 (post-larval).  Consistency in the NF stage at which death occurs may indicate a specific chemical mode of action and is important information for the study record.  Mortality rates exceeding 20% may indicate inappropriate test conditions or overtly toxic effects of the test chemical.  The animals tend to be most sensitive to non-chemical induced mortality events during the first few days of development after the spawning event and during metamorphic climax.  If an organism is euthanized due to morbidity (possibly determined by observing abnormal behaviors listed in the next section), this should be noted and reported.  Depending on when the individual is euthanized during the study, the carcass may need to be retained for possible histopathology analysis.  Procedures for euthanasia and fixation of tissues for histology are described below and should be followed according to the appropriate life-stage-specific procedures. 

      (4)  Additional Observations.  In addition, any observation of abnormal behavior, grossly visible malformations (e.g., scoliosis, also described as bent tail in tadpoles), or lesions should be recorded (include treatment and tank number).  Normal behavior for larval animals is characterized by suspension in the water column with tail elevated above the head, regular rhythmic tail fin beating, periodic surfacing, operculating, and being responsive to stimuli.  Abnormal behaviors would include, for example, floating on the surface, lying on the bottom of the tank, inverted or irregular swimming, lack of surfacing activity, and being nonresponsive to stimuli.  For post-metamorphic animals, in addition to the above abnormal behaviors, gross differences in food consumption between treatments should be recorded.  Gross malformations and lesions could include morphological abnormalities (e.g., limb deformities), hemorrhagic lesions, abdominal edema, and bacterial or fungal infections, to name a few.  The occurrences of lesions on the head of juveniles, just posterior to the nostrils, may be indications of insufficient humidity levels.  These determinations are qualitative and should be considered akin to clinical signs of disease/stress and made in comparison to control animals.  If the rate of occurrence is greater in exposed tanks than in the controls, then these should be considered as evidence for overt toxicity.
       
Observations of scoliosis should be counted (incidence) and graded with respect to severity (e.g., not remarkable  -  NR, minimal  -  1, moderate  -  2, severe  -  3).  Guidance for grading and minimizing scoliosis is found in Appendix 4.  Efforts should be made to ensure that the prevalence of moderate and severe scoliosis is limited (e.g., below 10% in controls) throughout the study, although greater prevalence of control abnormalities would not necessarily mean the test should be ended.    
Differences in food consumption may be determined by how quickly the tanks clear after feeding, how much uneaten food accumulates on the bottom of the tank, and the amount of feces that accumulates between daily tank cleaning.  

      (5)  Larval Sub-sample.  Those tadpoles that have reached NF stage 62 should be removed from the exposure tanks and either sampled or moved to the next part of the exposure in a new tank, or physically separated from the remaining tadpoles in the same tank with a divider. Tadpoles are checked daily and the study day on which an individual tadpole reaches NF stage 62 is recorded.  The defining characteristic for use in this assessment is the shape of the head shown in Figure 2a.  Once the head has become reduced in size such that it is visually approximately the same width as the trunk of the tadpole, the olfactory nerve length is shorter than the diameter of the olfactory bulb and the forelimbs are at the level of the middle of the heart, then that individual would be counted as having attained NF stage 62.  This determination can generally be made without netting or handling the individual by observing the dorsal characteristics of NF stage 62.  The goal is to sample a total of five NF stage 62 tadpoles per replicate tank.  This should be performed entirely at random, but decided a priori (explained below).
      
      
      
      
      
      
      
      
      
      
      
        

a. Larval sub-sampling (NF stage 62)		              b. Juvenile sampling
Ventral
Ventral





Snout-vent length
Snout-vent length


Dorsal
Dorsal

Snout-vent length 
Snout-vent length 
	  




Figure 2. Landmarks for measuring snout-vent length for the LAGDA in NF Stage 62 (a) and juvenile frogs (b).  The defining characteristics of NF stage 62 (a): the head is the same width as the trunk, the olfactory nerve length is shorter than the diameter of the olfactory bulb (dorsal view), and the forelimbs are at the level of the heart (ventral view).  Images adapted from Nieuwkoop and Faber (1994¸Ref. 4).
  
A hypothetical example of a replicate tank is provided in Figure 3.  Should there be 20 surviving tadpoles in a particular tank when the first individual reaches NF stage 62, five random numbers should be chosen from 1-20.  Tadpole #1 is the first individual to reach NF stage 62 and tadpole #20 is the last individual in a tank to reach NF stage 62.  Likewise, if there are 18 surviving larvae in a tank, five random numbers should be chosen from 1-18.  This should be performed for every replicate tank when the first individual on test reaches NF stage 62.  If there are mortalities during the NF stage 62 sampling, the remaining samples need to be re-randomized based on how many larvae are left less than NF stage 62 and how many more samples are needed to reach a total of five samples from that replicate.  
On the day a tadpole reaches NF stage 62, reference to the prepared sampling chart is made to determine whether that individual is sampled or physically separated from the remaining tadpoles for continued exposure.  In the example provided below (Figure 3), the first individual to reach NF stage 62 (i.e., box #1) is physically separated from the other larvae, continues exposure and the study day on which that individual reached NF stage 62 is recorded.  Subsequently, individuals #2 and #3 are treated the same way as #1 and then individual #4 is sampled for growth and thyroid histology (according to this example).  This procedure continues until the 20[th] individual either joins the rest of the post-NF stage 62 individuals or is sampled.  The random procedure used should give each organism on test equal probability of being selected.  This can be achieved by using any randomizing method, but also requires that each tadpole be netted at some point throughout the NF stage 62 sub-sampling period.








 

Time (duration ~2 weeks)

Single replicate tank  -  no mortalities
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
	 Individual larvae	 Sub-sample

Figure 3. Hypothetical example of NF stage 62 sampling regime for a single replicate tank. 
	
      (i)  Larval Sub-Sample Endpoints.  For the larval sub-sampling, the endpoints obtained are: 
       
   * time to NF stage 62 (i.e., number of days between fertilization and NF stage 62; n ≈ 20), 
   * external abnormalities (n ≈ 20), 
   * morphometrics (i.e., weight and snout-to-vent length; n = 5),  
   * thyroid histology (n = 5).  
   
      (ii)  Larval Sampling Procedure.  The following steps are in the larval sub-sampling procedure:

   * euthanize tadpole,
   * rinse and blot dry,
   * weigh tadpole (mg) and measure snout-vent length (mm),
   * note any gross morphologic abnormalities and/or clinical signs of toxicity,
   * remove and discard the lower torso posterior to the forelimbs (see Appendix 5 for guidance),
   * fix the upper torso for 48 hours in Davidson's solution, rinse, and store tissues in 10% neutral buffered formalin (NBF) (see Appendix 5 for guidance).

   	(A)  Tadpole Euthanasia. The sub-sample of NF stage 62 tadpoles (5 individuals per replicate) should be euthanized by immersion in a 0.03% (w/v) MS-222, tricaine methane sulfonate (CAS.886-86-2), solution (e.g., 500 ml).  The MS-222 solution should be buffered with sodium bicarbonate to a pH of approximately 7.0 because unbuffered MS-222 solution is acidic and irritating to frog skin resulting in poor absorption and unnecessary additional stress to the organisms.  The characteristics of the water used to make this solution may impact the pH and the amount of sodium bicarbonate needed to neutralize this solution.  In general, however, one part MS-222 to two parts of sodium bicarbonate should be used as a starting point.
   
Using a mesh dip net, a tadpole is removed from the experimental chamber and transported to the necropsy area in a suitable transport container.  The tadpole is then placed into the euthanasia solution.  The animal is properly euthanized and is ready for necropsy when it is unresponsive to external stimuli such as pinching the hind limb with a pair of forceps (approximately 5-8 minutes).

	(B)   Morphometrics (Weight and Length).  Measurements of wet weight (nearest mg) and snout-to-vent length (SVL) to the nearest 0.1 mm for each tadpole (see Figure 2a) should be made immediately after it becomes non-responsive by anesthesia.  These measurements are included in the test protocol to assess possible effects of a test chemical on the growth rate of organisms and are useful for detecting generalized toxicity.  Tadpoles should be blotted dry before weighing to remove excess adherent water.  Image analysis software may be used to measure SVL from a photograph.  After measurements of body size (SVL and weight) are made, any gross morphological abnormalities such as scoliosis (see Appendix 4) and/or clinical signs of toxicity, petechiae and hemorrhage should be recorded or noted, and digital documentation is recommended.  Note: Petechiae are small red or purple hemorrhages in skin capillaries. 

      (C)  Larval Sub-Sample Tissue Preparation for Histopathology.  Each larval sample (NF stage 62) is evaluated histologically for pathology in the thyroid glands.  The lower torso posterior to the forelimbs is removed and discarded (see Appendix 5, Figure 1 for trimming details and approximate gland locations).  The trimmed carcass is placed into individually labelled histological cassettes and fixed in Davidson's fixative (recipe below). 
      
Care should be taken not to compress tissues in cassettes as this will cause histological artifacts and could compromise pathology assessments.  The volume of fixative in the container should be at least 10 times the approximate volume of the tissues.  Appropriate circulation of the fixative should be achieved to adequately fix the tissues of interest.  All tissues remain in Davidson's fixative for at least 48 hours, but no longer than 96 hours, at which time they are rinsed in deionized water and stored in 10% neutral buffered formalin (recipe below).  

Histopathology guidance (Appendix 5) provides procedural information on trimming and processing tissues for histological examination as well as detailed information regarding pathologic effect evaluations and interpretation of diagnostic criteria. 

Davidson's Fixative
Glacial acetic acid...............115 ml
Formaldehyde (35-40%)........220 ml
95% Ethyl Alcohol...............330 ml
Distilled water....................335 ml

10% Neutral Buffered Formalin (NBF)
100 ml formalin, full strength (37-40% formaldehyde)
6.5 g sodium phosphate dibasic (anhydrous)
4.0 g sodium phosphate monobasic
900 ml distilled water

	(6)  End of Larval Exposure.  Given the initial number of tadpoles, it is expected that there will likely be a small percentage of individuals that do not develop normally and do not reach NF stage 66 in less than 70 days (a treatment-related delay in development could also cause individuals to not reach NF stage 66 in 70 days).  Because of this situation, larval exposure ends after all animals have reached NF stage 66 or after 70 days, whichever comes first.  Any tadpoles remaining at the end of larval exposure should be euthanized by immersion in 0.03% (w/v) buffered MS-222 (pH 7.0).  Their wet weight and SVL should be measured, the developmental stage recorded (Ref. 4), and any developmental abnormalities noted.  

	(7)  Cull After NF Stage 66.  After all juvenile animals have reached NF stage 66 or after 70 days (whichever occurs first), a cull should be conducted.  Post NF 66 animals that will continue the exposure should be selected at random.  Following the culling process, only 10 individuals (maximum) per tank should continue exposure until termination of the test. Animals that are not selected to continue exposure are euthanized by immersion in a 0.05% (w/v) MS-222 solution buffered with sodium bicarbonate to a pH of approximately 7.0.  Measurements of wet weight and snout-to-vent length (see Figure 2b), and a gross necropsy are conducted for each animal.  The phenotypic sex (based on gonad morphology) is noted as female, male, or indeterminate.
      (8)  Juvenile Sampling.   The remaining animals continue exposure until 10 weeks after the median time to NF stage 62 in the dilution water (and/or solvent control if relevant) control.  At the end of the exposure period, the remaining animals (maximum 10 frogs per replicate) are euthanized, and the various endpoints listed below are measured or evaluated, and recorded.  Final sample numbers will depend on genetic sex ratios.
      
      (i)  Juvenile Sample Endpoints.  For the juvenile sampling, the endpoints obtained are:
               *       morphometrics (weight and length);
               *       phenotypic/genotypic sex ratios;
               *       gonad histopathology;
               *       reproductive duct histopathology;
               *       liver histopathology;
               *       liver weight (Liver-Somatic Index [LSI]); and
               *       kidney histopathology.

      (ii)  Juvenile Sampling Procedure.  The juvenile sampling procedure has the following steps: 
*       euthanize frogs by immersion in buffered MS-222 solution (0.05% w/v) until subdued, followed by an intraperitoneal (IP) injection (e.g., 0.02 ml per g of frog) of buffered 2.5% (w/v) MS-222 in physiological buffer;
*       measure whole body weight (g) and snout-vent length (mm);
*       record any developmental abnormalities and obvious signs of disease (e.g., scoliosis, hemorrhage);
*       place the frog in dorsal recumbence;
*       open the abdominal cavity - excise, weigh, and fix the liver;
*       carefully remove the digestive organs (e.g., stomach, intestines) from the lower abdomen;
*       assess gross morphological abnormalities in gonads (see Appendix 5, Figures 2a & 2b);
*       remove a foot for a DNA sample and snap freeze in liquid nitrogen or dry ice; and,
*       trim and fix abdominal section according to Appendix 5, Figures 2a & 2b depending on the gonadal phenotype.
(A)  Juvenile Frog Euthanasia.  The juvenile frogs are euthanized by immersion in a neutral buffered 0.05% (w/v) MS-222 solution until subdued, followed by an IP injection of 2.5% (w/v) MS-222 in a physiological buffer solution (e.g., 0.1M phosphate buffered saline, pH 7.4).  Dosage for frogs is approximately 0.02 ml per gram of frog.  Injection provides efficient, fast euthanasia prior to sampling.  Using this procedure, frogs may be sampled after becoming unresponsive, usually around 2 minutes after injection, but can take longer.  Increasing the time of initial immersion and/or increasing the concentration of the immersion solution and/or increasing the volume of IP injection may be necessary depending on the effectiveness of the euthanasia procedure.  Care should be taken not to damage internal organs during the IP injection.
 (B)  Morphometrics (Weight and Length).  Measurements of wet weight and SVL (Figure 2b) are identical to those outlined for the larval sub-sampling. 

(C)   Juvenile Sample Tissue Preparation for Histopathology.   Each juvenile sample is evaluated histologically for pathology in the gonads, reproductive ducts, kidneys and liver tissue.  The abdominal cavity is opened and the entire liver is carefully dissected out and weighed.  Next, the digestive organs (e.g., stomach, intestines) and fat bodies are carefully removed from the lower abdomen to reveal the gonads, kidneys and reproductive ducts.  Any gross morphological abnormalities in the gonads should be noted.  Finally, the abdomen should be trimmed according to Appendix 5, Figures 2a & 2b depending on the phenotypic sex, and a hind limb (or appropriate tissue sample) should be retained (frozen) for genetic sex determination.  Collected livers and the carcass with the gonads left in situ should be immediately placed into separate labelled histological cassettes and fixed in Davidson's fixative (for recipe, see section (f)(5)(ii)(C)).  Care should be taken not to compress tissues in cassettes as this will cause histological artifacts and could compromise pathology assessments.  The volume of fixative in the container should be at least 10 times the approximated volume of the tissues.  The same fixatives used in the larval sub-sampling, as described previously, are used with adult samples.  All tissues should remain in Davidson's fixative for at least 48 hours, but no longer than 96 hours, at which time they are rinsed in de-ionized or tap water and stored in 10% NBF (for recipe, see section (f)(5)(ii)(C)).  Histopathology guidance is provided in Appendix 5 and contains procedural information on trimming and processing tissues for histological examinations.  It also contains detailed information regarding pathologic effect evaluations and interpretation of diagnostic criteria.

(D)   Genetic Sex Determination.  To determine genetic sex, DNA samples from frogs can be obtained from any tissues; however, one hind limb removed during dissection (or a portion of a hind limb) is easy to collect and store in a microfuge tube.  Tissue should be snap frozen in liquid nitrogen or dry ice and can be stored at -20°C until DNA isolation.  The isolation of DNA from tissues can be performed with commercially available kits and analysis for presence or absence of the marker is done by a polymerase chain reaction (PCR) method (see example protocol in Appendix 3).  Development of the markers used for genetic sex determination of X. laevis is described by Yoshimoto et al. (Ref. 11).  Generally, the concordance between juvenile histological sex (gonad phenotype) and genotype across control animals should be >95% (if the concordance is not 100% in control samples, there was likely a technical error).  It would be prudent to test the primers for genetic sexing on the adult breeders using toe clips prior to running the LAGDA with their offspring. 
      (g)  Performance Criteria and Test Acceptability/Validity.
(1)  Performance Criteria.  The performance criteria in Table 5 have been developed as guidance for determining the quality of the test performed and the general performance of the control organisms.  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. 




 
Table 5.  Performance criteria for the LAGDA.
Parameter 
Criteria
Test concentrations 
Maintain concentrations of the test chemical in solution within 20% of the mean measured value over the entire test period.
Viability of spawn
>=70% viability in the spawn chosen to start the study
Mortality in controls 
<= 20% mortality in each replicate in the controls.
Median time to NF stage 62 in controls.
The median time to NF stage 62 is <=45 days. 
Mean weight of test organisms in controls
The mean weight of test organisms at NF stage 62 and at the termination of the assay in controls and solvent controls (if used) should be 1.0+-0.2 and 11.5+-3 g, respectively. 
Dissolved Oxygen 
The dissolved oxygen concentration should be >=40% 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. 
Water temperature 
21 +- 1ºC - the inter-replicate/inter-treatment differentials should not exceed 1.0 ºC. 
Test concentrations without overt toxicity[a]
At least three treatment levels without obvious overt toxicity should be available for analysis. 

Excessive mortality, which compromises a treatment, is defined as >20% in 2 or more replicates that cannot be explained by technical error. 
Replicate performance 
1 replicate within a treatment can be compromised (see below for conditions for a compromised replicate).  An increase beyond this results in loss of that treatment level. 
Idiopathic scoliosis ("Bent tail") in tadpoles
Prevalence of moderate and severe scoliosis should be limited to <10% to improve confidence that the test can detect treatment-related developmental effects in otherwise healthy amphibian larvae. See Appendix 4 for guidance.

Special conditions for use of a solvent (use of a solvent is strongly discouraged)
If a carrier solvent is used, both a solvent control and clean water control should be used.

Statistically significant differences between solvent control and clean water control groups are treated specially. See below for more information.

Genotypic/phenotypic sex concordance in control juveniles
>95% concordance (if concordance is less than 100% in control juveniles, there was likely a technical error).
a Signs of overt toxicity may include, but are not limited to, floating on the surface, larvae lying on the bottom of the tank, inverted or irregular swimming, lack of surfacing activity, and being nonresponsive to stimuli, morphological abnormalities (e.g., limb deformities), hemorrhagic lesions, and abdominal edema.

(2)  Test Validity.  For a test to be considered valid, the following criteria should be met:
                                 *       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.
                                 *       At least three treatment levels without obvious overt toxicity are available for analysis. 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, lack of surfacing activity, and being nonresponsive to stimuli, morphological abnormalities (e.g., limb deformities), hemorrhagic lesions, and abdominal edema.
                                 *       A NOEC and LOEC should be determined.

(h)  Statistical Analyses.  

The LAGDA generates three forms of data to be statistically analyzed:
   * Quantitative continuous data including growth (weight, SVL, and LSI); 
   * Time-to-event data for developmental rates (i.e., days to NF stage 62 from assay initiation); 
   * Ordinal data in the form of severity scores or developmental stages from histopathology evaluations.  
   
Data are typically collected in spreadsheets and then transferred to and analyzed with a statistical software package.  Statistical analyses of the data should preferably follow procedures described in the document, Current Approaches in the Statistical Analysis of Ecotoxicity Data: A Guidance to Application (Ref. 36).  Figure 4 illustrates the recommended statistical analysis decision tree and guidance for the treatment of data from the LAGDA. 
                                       
Figure 4.  Statistical analysis decision tree for LAGDA data.
       (1)  Continuous Data.  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.  If the data are monotonic, a step-down Jonckheere-Terpstra trend test should be performed on replicate medians and no subsequent analyses should be applied.  If the data are non-monotonic (quadratic contrast is significant and linear is not significant), it should be analyzed using a mixed effects ANOVA model.  The data should then be assessed for normality (preferably using the Shapiro-Wilk or Anderson-Darling test) and variance homogeneity (preferably using Levene's test).  Both tests are performed on the residuals from the mixed effects ANOVA model.  Expert judgment can be used in lieu of these formal tests for normality and variance homogeneity, though formal tests are preferred.  If the data are normally distributed with homogeneous variance, then the assumptions of a mixed effect ANOVA are met and a significant treatment effect is determined from Dunnett's test.  Where non-normality or variance heterogeneity is found, then the assumptions of the mixed effect ANOVA model are violated and a significant treatment effect is determined with a Dunn's test.  Whenever possible, a one-tailed test should be performed as opposed to a two-tailed test, but it requires expert judgment to determine which is appropriate for a given endpoint.

      (2)  Mortality.  Mortality data should be analyzed for the time period encompassing the full test and should be expressed as percent that died in any particular tank. Tadpoles that do not complete metamorphosis in the given time frame, those tadpoles that are in the larval sub-sample cohort, those juvenile frogs that are culled, and any animal that dies due to experimenter error should be treated as censored data and not included in the denominator of the percent calculation.  Prior to any statistical analyses, mortality ratios should be arcsin-square root transformed and analyzed according to the "Other endpoints" branch of the decision tree (Figure 4).

      (3)  Time to NF stage 62.  Time to NF stage 62 data should be treated as time-to-event data. Mortalities or individuals not reaching NF stage 62 in 70 days are treated as right-censored data. Median time to NF stage 62 in clean water controls should be used to determine the test termination date.  This endpoint should be analyzed using a mixed-effects Cox proportional hazard model that can be implemented in SAS or the free open source statistical program "R".

      (4)  Growth:  Weight and Length.  Males and females are not sexually-dimorphic during metamorphosis so larval sub-sampling growth data should be analyzed independent of gender.  However, juvenile growth data should be analyzed separately based on genetic sex. These data should be analyzed according to the "Other endpoints" branch of the decision tree (Figure 4). Transformations are not recommended for these endpoints, unless deemed necessary by professional judgment.

      (5)  Liver-somatic-index (LSI).  Liver weights should be normalized as proportions of whole body weights (LSI) and analyzed separately based on genetic sex.  These data should be analyzed according to the "Other endpoints" branch of the decision tree (Figure 4). Transformations are not recommended for this endpoint, unless deemed necessary by professional judgment.

      (6)  Histopathology (severity scores and developmental stages).  Histopathology data are in the form of severity scores or developmental stages.  A test termed RSCABS (Rao-Scott Cochran-Armitage by Slices) uses a step-down Rao-Scott adjusted Cochran-Armitage trend test on each level of severity in a histopathology response (Ref. 37).  The Rao-Scott adjustment incorporates the replicate vessel experimental design into the test.  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 than controls (i.e., have more severe pathology than controls), but it also determines at which severity score the difference occurs thereby providing much needed context to the analysis.  In the case of developmental staging of gonads and reproductive ducts, an additional manipulation should be applied to the data since an assumption of RSCABS is that severity of effect increases with concentration.  The effect observed could be a delay OR acceleration of development.  Therefore, developmental staging data (gonad and reproductive ducts) should be analyzed as reported to detect acceleration in development and then manually inverted prior to a second analysis to detect a delay in development.

       (7)  Use of Compromised Replicates and Treatments.  Replicates and treatments may become compromised due to excess mortality from overt toxicity, disease, or technical error (see Section (g) for performance criteria).  If a treatment is compromised from disease or technical error, there should be at least three uncompromised treatments with three uncompromised replicates available for analysis.  In the case that a single replicate in a treatment is compromised due to excess mortality, that replicate should be removed from statistical analysis.  If overt toxicity occurs in the high treatment(s), it is preferable that at least three treatment levels with three uncompromised replicates are available for analysis (consistent with the Maximum Tolerated Concentration approach in Hutchinson et al. (Ref. 30).   In addition to mortality, signs of overt toxicity may include behavioral effects (e.g., floating on the surface, lying on the bottom of the tank, inverted or irregular swimming, lack of surfacing activity), morphological lesions (e.g., hemorrhagic lesions, abdominal edema) or inhibition of normal feeding responses when compared qualitatively to control animals.  A determination of test continuation should be made in consultation with the EPA.  

      (8)  Solvent controls.   The use of 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 (negative) control should be run in concert with a solvent control. At the termination of the test, a statistical comparison of the solvent control group and the dilution water control group should be performed to determine any potential effects of the solvent.  The most relevant endpoints for consideration in this analysis are growth determinants (weight and length), as these can be affected through generalized toxicities.  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, then all test level comparisons should be made to the solvent control unless it is known that comparison to the dilution water control is preferred.  If there is additional evidence that there is no solvent interference then statistical comparisons with the dilution water control may be appropriate.  If there are no statistically significant differences 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 (see discussion in Ref. 36; Ref. 37; Ref. 38). 
      (i)  Data Reporting.	
      All data should be collected using electronic or manual systems which conform to good laboratory practices (GLP).  The test report should include the following information:
	(1)  Test Substance.  The report should include a complete characterization of the test substance.  
   * Physical nature and, where relevant, physicochemical properties. 
   * Mono-constituent substance: physical appearance, water solubility, and additional relevant physicochemical properties;  chemical identification, such as IUPAC (International Union of Pure and Applied Chemistry) or CAS name, CAS number, SMILES (Simplified Molecular Input Line Entry Specification) or InChI (IUPAC International Chemical Identifier code), structural formula, purity, chemical identity of impurities as appropriate and practically feasible, etc. (including the organic carbon content, if appropriate).
   * Multi-constituent substance, UVBCs (substances of unknown or variable composition, complex reaction products or biological materials and mixtures) characterized as far as possible by chemical identity (see above), quantitative occurrence and relevant physicochemical properties of the constituents. 
	(2)  Test Species.  Include: scientific name, strain if available, source and method of collection of the fertilized eggs and subsequent handling; incidence of scoliosis in historical controls for the stock culture used.   
	(3)  Test Conditions.  
    Photoperiod(s); 
    Test design (e.g., chamber size, material and water volume, number of test chambers and replicates, number of test organisms per replicate); 
    Method of preparation of stock solutions and frequency of renewal (the solubilizing agent and its concentration should be given, when used); 
    Method of dosing the test chemical (e.g., pumps, diluting systems);  
    The recovery efficiency of the method and the nominal test concentrations, the limit of quantification, the means of the measured values and their standard deviations in the test vessels and the method by which these were attained and evidence that the measurements refer to the concentrations of the test chemical in true solution; 
    Dilution water characteristics: pH, hardness, temperature, dissolved oxygen concentration, residual chlorine levels (if measured), total iodine, total organic carbon (if measured), suspended solids (if measured), salinity of the test medium (if measured) and any other measurements made; 
    Solvent (if other than water).  Justification of the choice of solvent, and characterization of solvent (nature, concentration used).
    The nominal test concentrations, the means of the measured values and their standard deviations; 
    Water quality within test vessels, pH, temperature (daily) and dissolved oxygen concentration;
    Detailed information on feeding (e.g., type of foods, source, amount given and frequency).
   
       (4)  Biological Observations and Data.  
         *       Daily observations of mortality, food consumption, abnormal swimming behavior, lethargy, loss of equilibrium, malformations (e.g., scoliosis), lesions, etc.  
         *       Observations and data collected at predetermined intervals include: growth (weight and SVL), clinical signs of disease or general toxicities, time to NF stage 62, and phenotypic as well as genetic sex.
      
      (5)  Histological Data.  The report should include narrative descriptions, as well as graded severity and incidence scores of specific observations.
      (6)  Statistical Analyses.  It is recommended that statistical analytical techniques outlined in this test guideline be used for analyses of LAGDA data.  If there are deviations from these techniques to determine statistically significant effects, justification of alternative techniques should be provided.  Results of the statistical analyses should be provided in tabular form.
      (7)  Ad hoc Observations.  The report should include narrative descriptions of observations that do not fit into the previously described categories.
      (8)  Results.
   * Evidence that controls met the performance criteria; 
   * Data for the control (plus solvent control when used) and the treatment groups as follows: mortality and abnormalities observed, time to NF stage 62, thyroid histology assessment (larval sample only), growth (weight and length), LSI (juvenile sample only), genetic/phenotypic sex ratios (juvenile sample only), and histopathology assessment results for gonads, reproductive ducts, kidney and liver (juvenile sample only). 
   * Approach for and results of the statistical analysis and treatment of data (statistical test or model used); 
   * No observed effect concentration (NOEC) for each response assessed; 
   * Lowest observed effect concentration (LOEC) for each response assessed (at α = 0.05); ECx for each response assessed, if applicable, and confidence intervals (e.g., 95%) and a graph of the fitted model used for its calculation, the slope of the concentration-response curve, the formula of the regression model, the estimated model parameters and their standard errors. 
   * Any deviation from the guideline and deviations from the acceptance criteria, and considerations of potential consequences on the outcome of the test. 
   * For the results of endpoint measurements, mean values and their standard deviations (on both replicate and concentration basis, if possible) should be presented.  Results of the statistical analyses should be provided in tabular form.
   * Median time to NF stage 62 in controls should be calculated and presented as the mean of replicate medians and their standard deviation.  Likewise, for treatments, a treatment median should be calculated and presented as the mean of replicate medians and their standard deviation. 

	(j)  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) OECD. 2015. Larvae Amphibian Growth and Development Assay (LAGDA). OECD Guidelines for the Testing of Chemicals No. 241; new TG, publication expected July 2015
   3) Kloas W, Lutz I. 2006. Amphibians as model to study endocrine disrupters. Journal of Chromatography A 1130: 16-27.
   4) Nieuwkoop, P.D. and Faber, J. (1994). Normal Table of Xenopus laevis (Daudin). Garland Publishing, Inc, New York, NY.
   5) Chang C, Witschi E. 1956. Genic control and hormonal reversal of sex differentiation in Xenopus. Journal of the Royal Society of Medicine 93: 140-144.
   6) Gallien L, 1953. Total inversion of sex in Xenopus laevis Daud, following treatment with estradiol benzoate administered during larval stage. Comptes Rendus Hebdomadaires des Séances de l'Académie des Sciences 237: 1565. 
   7) Villalpando I, Merchant-Larios H. 1990. Determination of the sensitive stages for gonadal sex reversal in Xenopus laevis tadpoles. International Journal of Developmental Biology 34: 281-285. 
   8) Miyata S, Koike S, Kubo T. 1999. Hormonal reversal and the genetic control of sex differentiation in Xenopus. Zoological Science 16: 335-340. 
   9) Mikamo K, Witschi E. 1963. Functional sex-reversal in genetic females of Xenopus laevis, induced by implanted testes. Genetics 48: 1411. 
   10) Olmstead AW, Kosian PA, Korte JJ, Holcombe GW, Woodis KK, Degitz SJ. 2009a. Sex reversal of the amphibian, Xenopus tropicalis, following larval exposure to an aromatase inhibitor. Aquatic Toxicology 91: 143-150. 
   11) Yoshimoto, S., Okada, E., Umemoto, H., Tamura, K., Uno, Y., Nishida-Umehara, C., Matsuda, Y., Takamatsu, N., Shiba, T., Ito, M.  (2008). A W-linked DM-domain gene, DM-W, participates in primary ovary development in Xenopus laevis.  Proc Natl Acad Sci USA 105(7):2469-2474.
   12) Olmstead AW, Korte JJ, Woodis KK, Bennett BA, Ostazeski S, Degitz SJ. 2009b, Reproductive maturation of the tropical clawed frog: Xenopus tropicalis. General and Comparative Endocrinology 160: 117-123.
   13) Wake, MH, Dickie, R. 1998.  Oviduct structure and function and reproductive modes in amphibians.  J. Exp. Zool., 282: 477-506.
   14) Tobias ML, Tomasson J, Kelley DB. 1998. Attaining and maintaining strong vocal synapses in female Xenopus laevis. Journal of Neurobiology 37: 441-448. 
   15) Qin ZF, Qin XF, Yang L, Li HT, Zhao XR, Xu XB. 2007. Feminizing/demasculinizing effects of polychlorinated biphenyls on the secondary sexual development of Xenopus laevis. Aquatic Toxicology 84: 321-327. 
   16) Porter KL, Olmstead AW, Kumsher DM, Dennis WE, Sprando RL, Holcombe GW, Korte JJ, Lindberg-Livingston A, Degitz SJ. 2011. Effects of 4-tert-octylphenol on Xenopus tropicalis in a long term exposure. Aquatic Toxicology 103: 159-169. 
   17) Tata, J. R. 2006. Amphibian metamorphosis as a model for the developmental actions of thyroid hormone. Molecular and cellular endocrinology, 246(1), 10-20.
   18) Dent, J. N. 1968. Survey of amphibian metamorphosis. In W. Etkin and L I. Gilbert (eds), Metamorphosis.  A Problem in Developmental Biology, New York: Appleton-Century Crofts, pg. 271-311.
   19) Just, J. J., Kraus-Just, J., & Check, D. A. 1981. Survey of chordate metamorphosis. In Metamorphosis (pp. 265-326). Springer US.
   20) Tata, J. R. 1993. Gene expression during metamorphosis: an ideal model for post‐embryonic development. Biossays, 15(4), 239-248.
   21) OECD. 2002. Guidance Document on Aquatic Toxicity Testing of Difficult Substances and Mixtures, OECD Series on Testing and Assessment, No. 23, OECD Publishing, Paris. DOI: 10.1787/9789264078406-en.  
   22) United States EPA. 1996.  OCSPP Harmonized Test Guidelines, Ecological Effects Test Guidelines, OCSPP 850.1000:  Special Considerations for Conducting Aquatic Laboratory Studies.  EPA 712-C-96-113. April 1996. Available on-line at 32. 
   23) American Society for Testing and Materials (ASTM). 2002. Standard Guide for Conducting Acute Toxicity Tests on Test Materials with Fishes, Macroinvertebrates, and Amphibians. American Society for Testing and Materials, ed., Philadelphia, PA.
   24) American Society for Testing and Materials (ASTM). 2004. Standard Guide for Conducting the Frog Embryo Teratogenesis Assay - Xenopus (FETAX). American Society for Testing and Materials, ed., Philadelphia, PA.
   25) Read BT. 2005. Guidance on the housing and care of the African clawed frog Xenopus laevis. Royal Society for the Prevention of Cruelty to Animals (RSPCA), Horsham, Sussex, U.K., 84 pp.
   26) OECD. 1992. Ready Biodegradability.  OECD Guidelines for Testing Chemicals.  No. 301.  OECD, Paris.
   27) OCED. 2006. Ready Biodegrabability  -  CO2 in sealed vessels (headspace test).  OECD Guidelines for Testing Chemicals. OECD, Paris.
   28) Kahl, M.D., Russom, C.L., DeFoe, D.L., and Hammermeister, D.E. (1999). Saturation units for use in aquatic bioassays. Chemosphere 39, 539-551.
   29) Adolfsson-Erici M, Åkerman G, Jahnke A, Mayer P, McLachlan M. 2012. A flow-through passive dosing system for continuously supplying aqueous solutions of hydrophobic chemicals to bioconcentration and aquatic toxicity tests. Chemosphere 86:  593-599.
   30) Hutchinson, T.H., Shillabeer, N., Winter, M.J., and Pickford, D.B. 2006. Acute and chronic effects of carrier solvents in aquatic organisms: a critical review. Aquat Toxicol 76, 69-92.
   31) US EPA. 2009. Endocrine Disruptor Screening Program Test Guidelines.  OCSPP 890:1100: Amphibian Metamorphosis (Frog).  US EPA, Washington, DC.
   32) OECD. 2009. The Amphibian Metamorphosis Assay. OECD Guidelines for the Testing of Chemicals No. 231. OECD, Paris. 
   33) OECD. 2012b. Fish Short Term Reproduction Assay. OECD Guidelines for the Testing of Chemicals No. 229. OECD, Paris. 
   34) OECD. 2011. Fish Sexual Development Test. OECD Guidelines for the Testing of Chemicals no. 234, OECD, Paris.  
   35) OECD. 2013b. Fish Embryo Acute Toxicity (FET) Test. OECD Guidelines for the Testing of Chemicals No. 236. OECD, Paris. 
   36) OECD. 2006. Guidance Document on Current Approaches in the Statistical Analysis of Ecotoxicity Data, OECD Series on Testing and Assessment, No. 54. ENV/JM/MONO(2006)18. OECD Publishing, Paris.
   37) Green J.W., Springer T.A., Saulnier A.N., Swintek J.  2014. Statistical analysis of histopathology endpoints. Environmental Toxicology and Chemistry. 33(5):1108-1116.
   38) Green, J.W. Wheeler, J. 2013.  The use of carrier solvents in regulatory aquatic toxicology testing:  Practical, statistical and regulatory considerations.  Aquatic Toxicology. 144-145C: 242-249.


(k)  List of LAGDA Appendices.								Page
(1)  Definitions.................................................................................		32	
(2)  Larval and Juvenile Diet Protocols....................................................			33
(3)  Genetic Sex Determination..............................................................		36
(4)  Considerations for Tracking and Minimizing the Occurrence of Scoliosis.......		39
(5)  Histopathology Guidance for LAGDA.................................................		41


                            APPENDIX 1: Definitions
Apical endpoint: Causing effect at population level.  
ELISA: Enzyme-Linked Immunosorbent Assay 
dpf: days post fertilization 
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 
HPT axis:  hypothalamic-pituitary-thyroid axis
InChI:  The IUPAC International Chemical Identifier is a textual identifier for chemical substances, designed to provide a standard and human-readable way to encode molecular information and to facilitate the search for such information in databases and on the web.
IUPAC: International Union of Pure and Applied Chemistry  
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. 
Liver-somatic index:  Liver-somatic index 
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. 
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 
SVL:  snout-to-vent length 
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. 
                APPENDIX 2:  Larval and Juvenile Diet Protocol

      The protocol described below is the recommended larval and juvenile diet protocol.  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.  Alternative diets are acceptable as long as the organisms meet the growth and development criteria.  
      
      (a) Larval Diet.
           
      This diet, consisting of Trout Starter, algae/TetraFin[(R)], and brine shrimp, allows control organisms to meet both of the following criteria:

Median time to NF stage 62 is <= 45 days, and
Mean weight at NF stage 62 is 1.0 +- 0.2 g.

      (1)  Larval Food Preparation.

      (i)  1:1 (v/v) Trout Starter: algae/TetraFin[(R)]:

   * Trout Starter: blend 50 grams of Trout Starter (fine granules or powder) and 300 mL of suitable filtered water on a high blender setting for 20 seconds.
   * Algae/TetraFin[(R)] mixture: blend 12 g spirulina algae disks and 500 ml filtered water on high for 40 seconds, blend 12 g Tetrafin[(R)] with 500 ml filtered water and then combine these to make up one L of 12 g/L spirulina algae and 12 g/L Tetrafin.[(R)]
   * Combine equal volumes of the blended Trout Starter and the algae/TetraFin[(R)] mixture.

      (ii)  Brine Shrimp. 15 ml brine shrimp eggs are hatched in 1 L of salt water (prepared by adding 20 mL of NaCl to 1 L deionized water).  After aerating 24 hours at room temperature under constant light, the brine shrimp are harvested.  Briefly, the brine shrimp are allowed to settle for 30 min by stopping aeration.  Cysts that float to the top of the canister are poured off and discarded, and the shrimp are poured through the appropriate filters and brought up to 30 ml with filtered water. 

      (2)  Larval Feeding Protocol.  Table 1 summarizes the type and amount of feed used during the larval stages of the exposure. The animals are fed three times per day Monday through Friday and once per day on the weekends. 
      
      (b) Larval to Juvenile Diet Transition.  

      At NF stage 62, the tadpole opercula close and they no longer filter feed on suspended food.  However, it is unclear when exactly the organisms begin bottom feeding.  Therefore, pellet food is provided per post-NF stage 62 regardless of whether they consume it or not.  To aid the transition over to pellet food, crushed pellets are provided at a rate per post-NF stage 62 larvae.  Due to this diet transition, and to avoid overfeeding pre-NF stage 62 larvae (should all organisms be kept in the same divided tank), a decrease in larval feed concurrent with an increase in juvenile feed is recommended during this period of the test.  This can be accomplished by feeding post-NF stage 62 at the rate specified in Table 2 while decreasing the larval feeding volume proportionally when each group of five (5) individuals in a particular tank develops past NF stage 62. 

Table 1. Feeding regime for X. laevis larvae in flow-through conditions (Trout Starter, Algae/TetraFin[(R)] diet, and brine shrimp quantities for 20 tadpoles).

Trout Starter: algae/TetraFin(R)
Brine Shrimp
Time* 
Post-Fertilization (pf) 

Weekday
Feeding Volume
(3 times per day)
Weekend
Feeding Volume
(once per day)
Weekday
(twice per day)

Weekend
(once per day)
4-14 days (Weeks 0-1)
0.33 ml
1.2 ml
0.5 ml (from Day 8 to 15 pf)
0.5 ml (from Day 8 to 15 pf)
Week 2
0.67 ml
2.4 ml
1 ml (day 16)

1 ml (day 16)

Week 3
1.3 ml
4.0 ml
1 ml 
1 ml 
Week 4
1.5 ml
4.0 ml
1 ml 
1 ml 
Week 5
1.6 ml
4.4 ml
1 ml 
1 ml 
Week 6
1.6 ml
4.6 ml
1 ml 
1 ml 
Week 7
1.7 ml
4.6 ml
1 ml
1 ml
Weeks 8-10
1.7 ml
4.6 ml
1 ml 
1 ml 
* Day 0 is defined as the day hCG injection is done.

      (c) Post-NF stage 62 and Juvenile Diet.   

      After larvae reach NF stage 62, the feeding regime changes to 3/32 inch premium sinking frog food alone (Xenopus Express., FL) or equivalent.  
      
      (1)  Food Preparation for Larval to Juvenile Transition.  Sinking frog food pellets are briefly run in a coffee grinder, blender or mortar and pestle in order to reduce the size of the pellets by approximately 1/3.  Processing too long results in powder and is discouraged.
      
      (2)  Post-NF stage 62 and Juvenile Feeding Protocol.  Table 2 provides a reference regarding the type and amount of feed used during the post-NF stage 62 and juvenile life stages.  The animals should be fed once per day.  It should be noted that as animals metamorphose, they continue receiving a portion of the brine shrimp until >95% animals have completed metamorphosis (NF stage 66). The animals should not be fed on the day of test termination so feed does not confound weight measurements.  



Table 2.  Feeding regime for X. laevis juveniles in flow-through conditions.  It should be noted that un-metamorphosed animals, including those whose metamorphosis has been delayed by the chemical treatment, cannot eat uncrushed pellets.  
Weeks post-median NF stage 62 date in controls*
Crushed pellet volume (mg per organism)
Whole pellet volume (mg per organism)
As larvae reach NF stage 62 
25
0
Weeks 0-1
25
28
Weeks 2-3
0
110
Weeks 4-5
0
165
Weeks 6-9
0
220
*The first day of week 0 is the median NF stage 62 date in control animals.
                    APPENDIX 3:  Genetic Sex Determination.

      The example protocol for genetic sex determination using PCR and gel electrophoresis is based on procedures used at the U.S. Environmental Protection Agency laboratory in Duluth, MN.  Alternative methods, such as Q-RT-PCR, can be developed.  Additionally, alternative supplies such as specific products and/or equipment listed below can be substituted with comparable materials.

      (a) Xenopus laevis primers (see Ref.)

DM-W marker:  Gene ID: 100137717
Forward: 5'-CCACACCCAGCTCATGTAAAG-3'
Reverse: 5'-GGGCAGAGTCACATATACTG-3'
Positive Control (DMRT1): Gene ID:  735040
Forward: 5'-AACAGGAGCCCAATTCTGAG-3'
Reverse: 5'-AACTGCTTGACCTCTAATGC-3'

      (b) DNA purification.  

      Purify DNA from muscle or skin tissue using a commercially available kit according to kit instructions.  If using spin columns, DNA can be eluted using less buffer to yield more concentrated samples if deemed necessary for PCR. The final DNA concentration for purified samples should be ~4 ng/ul.  
      
      (c) PCR.  
      
      Sample protocol using JumpStart[TM] Taq (Sigma-Aldrich) is outlined below:
      
Master Mix
1x (uL)
[Final]
NFW
11
-
10X Buffer
2.0
-
MgCl2  (25mM)
2.0
2.5 mM
dNTP's (10mM ea)
0.4
200 uM
Marker for. primer (8 uM)
0.8
0.3 uM
Marker rev. primer (8 uM)
0.8
0.3 uM
Control for. primer (8 uM)
0.8
0.3 uM
Control rev. primer (8 uM)
0.8
0.3 uM
JumpStart[TM] Taq
0.4
0.05 units/ul
DNA template
1.0
~200 pg/ul

Note: When preparing Master Mixes, prepare extra to account for any loss that may occur while pipetting (e.g., 25x Master Mix will only yield approximately 23 reactions).  

Reaction: 		

Master Mix  	19.0 ul
Template	1.0 ul
Total		20.0 ul


Thermocycler Profile: 

Step 1. 94 ºC 1 min 
Step 2. 94 ºC 30 sec 
Step 3. 60 ºC 30 sec 
Step 4. 72 ºC 1 min 
Step 5. Go to step 2. 35 cycles 
Step 6. 72 ºC 1 min 
Step 7. 4 ºC hold 

PCR products can be run immediately in an agarose gel or stored at 4ºC. 

      (d) Agarose Gel Electrophoresis (3%)

50X TAE (Tris-acetate-EDTA)
24.2 g 	  Tris
5.71 ml  Glacial acetic acid
3.72 g 	  Na2(EDTA)·2H2O
Add water to 100 ml

1X TAE
392 ml	H2O
8 ml	50X TAE

3:1 Agarose
3 parts NuSieve(TM) GTG[(TM)] agarose
1 part Fisher agarose low electroendosmosis (EEO)

Electrophoresis Method
   * Prepare a 3% gel by adding, e.g., 1.2 g agarose mix to 43 ml 1X TAE.  Swirl to disassociate large clumps.
   * Microwave agarose mixture until completely dissolved (avoid boiling over).  Let cool slightly.
   * Add 1.0ul ethidium bromide (10 mg/mL).  Swirl flask.
   * Pour gel into mold with comb.  Cool completely.
   * Add gel to apparatus.  Cover gel with 1X TAE.
   * Add 1ul of 6x loading dye to each 10ul PCR product.
   * Pipette samples into wells.
   * Run at 160 constant volts for ~20 minutes. 


Figure 1.  Agarose gel image showing the band pattern indicative of a male (♂) individual (single band ~203 bp: DMRT1) and of a female (♀) individual (two bands at ~259 bp: DM-W  and 203 bp: DMRT1).  

      (e) Reference

Yoshimoto, S., Okada, E., Umemoto, H., Tamura, K., Uno, Y., Nishida-Umehara, C., Matsuda, Y., Takamatsu, N., Shiba, T., Ito, M.  (2008).  A W-linked DM-domain gene, DM-W, participates in primary ovary development in Xenopus laevis.  Proc Natl Acad Sci USA 105(7):2469-2474.

APPENDIX 4:  Considerations for Tracking and Minimizing the Occurrence of Scoliosis

      (a) Recognition and Grading the Severity of Idiopathic Scoliosis.  

      Idiopathic scoliosis, usually manifesting as "bent tail" in Xenopus laevis tadpoles, may complicate morphological and behavioral observations in test populations.  Efforts should be made to minimize or eliminate the incidence of scoliosis, both in stock and under test conditions.  In the definitive test, it is recommended that the prevalence of moderate and severe scoliosis be less than 10%, to improve confidence that the test can detect treatment-related developmental effects in otherwise healthy amphibian larvae.
       
Daily observations during the definitive test should record both the incidence (individual count) and severity of scoliosis, when present.  The nature of the abnormality should be described with respect to location (e.g., anterior or posterior to the vent) and direction of curvature (e.g., lateral or dorsal-to-ventral).  Severity may be graded as follows: 

   * (NR) Not remarkable: no curvature present 
   * Minimal: slight, lateral curvature posterior to the vent; apparent only at rest 
   * Moderate: lateral curvature posterior to the vent; visible at all times but does not inhibit movement 
   * Severe: lateral curvature anterior to the vent; OR any curvature that inhibits movement; OR any dorsal-to-ventral curvature 

      (b) Recommendations for Reducing Idiopathic Scoliosis.

      The EPA FIFRA Scientific Advisory Panel reviewed summary data for scoliosis in fifteen Amphibian Metamorphosis Assays (OCSPP 890.1100) with X. laevis (NF stage 51 through 60+) and provided general recommendations for reducing the prevalence of this abnormality in test populations (FIFRA SAP, 2013).  These recommendations are summarized below.
      
      (1)  Historical Spawning Performance.  Generally, high quality, healthy adults should be used as breeding pairs; eliminating breeding pairs that produce offspring with scoliosis may minimize its occurrence over time.  Specifically, minimizing the use of wild-caught breeding stock may be beneficial.  The LAGDA exposure period begins with NF stage 8-10 embryos, and it is not feasible to determine at the test outset whether given individuals will exhibit scoliosis.  Thus, in addition to tracking the incidence of scoliosis in animals that are placed on test, historical clutch performance (including the prevalence of scoliosis in any larvae allowed to develop) should be documented.  It may be useful to further monitor the portion of each clutch not used in a given study and to report these observations.

      (2)  Water Quality.  It is important to ensure adequate water quality, both in laboratory stock and during the test.  In addition to water quality criteria routinely evaluated for aquatic toxicity tests, it may be useful to monitor for and to correct any nutrient deficiencies (e.g., deficiency of vitamin C, calcium, phosphorus) or excess levels of selenium and copper, which are reported to cause scoliosis to varying degrees in laboratory-reared Rana sp. and Xenopus sp. (Marshall et al. 1980; Leibovitz et al. 1992; Martinez et al. 1992; as stated in the FIFRA SAP report (FIFRA SAP, 2013)).  The use of an appropriate dietary regimen (see Appendix 2), and regular tank cleaning, will generally improve water quality and health of the test specimens (FIFRA SAP, 2013).  
 
      (3)  Diet.  Specific recommendations for a dietary regimen, found to be successful in the LAGDA, are detailed in Appendix 2.  It is recommended that feed sources be screened for biological toxins, herbicides, and other pesticides which are known to cause scoliosis in X. laevis or other aquatic animals (FIFRA SAP, 2013).  For example, exposure to certain cholinesterase inhibitors has been associated with scoliosis in fish (Schultz et al. 1985) and frogs (Bacchetta et al. 2008).
         
      (c) References 
      
Bacchetta, R., P. Mantecca, M. Andrioletti, C. Vismara, and G. Vailati.  2008.  Axial-skeletal defects caused by carbaryl in Xenopus laevis embryos.  Science of the Total Environment 392: 110  -  118. 

FIFRA Scientific Advisory Panel (SAP).  2013.  Minutes of US EPA FIFRA SAP No. 2013-03. Endocrine Disruptor Screening Program (EDSP) Tier 1 Screening Assays and Battery Performance.  May 21-23, 2013.  Washington, DC. 

Schultz, T.W., J.N. Dumont, and R.G. Epler.  1985. The embryotoxic and osteolathyrogenic effects of semicarbazide. Toxicology 36: 185-198. 

Leibovitz, H.E., D.D. Culley, and J.P. Geaghan. 1982. Effects of vitamin C and sodium benzoate on survival, growth and skeletal deformities of intensively culture bullfrog larvae (Rana catesbeiana) reared at two pH levels. Journal of the World Aquaculture Society 13: 322-328. 

Marshall, G.A., R.L. Amborski, and D.D. Culley. 1980.  Calcium and pH requirements in the culture of  bullfrog (Rana catesbeiana) larvae. Journal of the World Aquaculture Society 11: 445-453. 

Martinez, I., R. Alvarez, I. Herraez, and P. Herraez. 1992. Skeletal malformations in hatchery reared Rana perezi tadpoles. Anatomical Records 233(2): 314-320. 
 

                APPENDIX 5:  Histopathology Guidance for LAGDA

      (a) Introduction.
  
      This Appendix describes the collection, histological preparation, and pathological evaluation of thyroid glands, gonads and gonadal ducts, kidney, and liver specimens from African clawed frogs (Xenopus laevis) in support of the Larval Amphibian Growth and Development Assay (LAGDA), a Tier 2 assay in the EDSP.  This guidance is provided to help ensure that histological procedures and pathological evaluations are performed accurately and consistently. The guidance provided 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 Appendix is divided into three 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 and Gross Trimming Procedures.  
      
      At study termination, each frog selected for sampling is euthanized via injection of a 10% solution (w/v) of MS-222 dissolved in a physiological buffer (e.g., 0.1M phosphate buffered saline pH 7.4) at a dose of 0.2 mL per 50 g body weight into the dorsal lymph sac.
Three different types of samples are collected for histopathology: 1) larval head and neck region for evaluation of the thyroid glands, 2) juvenile liver, and 3) juvenile dorsal trunk region for evaluation of the gonads, gonadal ducts, and kidneys.

      (1)  Larval Thyroid Glands (NF stage 62 sample).  A transverse cut is made at the posterior margin of the arms prior to fixation (solid line; Figure 1). The head / neck samples are placed in appropriately sized cassettes and fixed in Davidson's solution for 48 hours, and are then rinsed thoroughly in tap water, after which they may be maintained in 10% neutral buffered formalin prior to shipment or processing.  At any time following fixation, two transverse cuts are made through the head and neck region at the anterior margin of the eyes and anterior margin of the forelimbs, respectively (dashed lines; Figure 1).  Care should be taken to ensure that the cut made anterior to the forelimbs is as square to the axial midline as possible and includes as much of the neck region as possible. This ensures that the thyroid glands will be included in the head / neck sample.
      
      (2)  Juvenile Liver (Test termination).  The ventral abdominal wall is removed; the liver is carefully excised, weighed, transferred to an appropriately sized cassette that does not compress the tissue and fixed in Davidson's solution for 48 hours. The liver is then rinsed with tap water and maintained in 10% neutral buffered formalin prior to shipment or processing.  

 


Figure 1.  Dorsal view of  a larval (NF stage 62) X. laevis specimen to be evaluated for thyroid histopathology with solid red line indicating trimming plane prior to fixation, dashed red lines indicating trimming planes following fixation and red dots indicating approximate location of glands in relation to posterior trim plane.

      (3)  Juvenile Gonads, Gonadal Ducts, and Kidneys.  Following removal of the liver, the gastrointestinal tract is excised and discarded.  When manipulating the abdominal viscera, care should be taken not to disrupt the dorsally positioned gonads and fragile gonadal ducts.  One transverse cut is made through the trunk at the anterior margin of the hips (posterior margin of the kidneys) (Figure 2).  Another transverse cut is made through the middle of the spleen or several millimeters rostral from the anterior margin of the gonad.  The resulting trunk segment contains the intact gonads, gonadal ducts and kidneys.  The head, legs and other residual excised tissues are discarded.  The trunk is transferred to an appropriately sized cassette or container that does not compress the tissue and fixed in Davidson's solution for 48 hours.  The fixed tissues are then rinsed with tap water and maintained in 10% neutral buffered formalin prior to shipment or processing.  


Figure 2.  Ventral view of a:  A) male; and B) female; juvenile (10 weeks post-NF62) X. laevis specimens with lines approximately indicating the  tissue trimming to be performed before fixation, processing and evaluation for gonad and duct histopathology.
Prior to processing and embedment, the dorsal trunk segment is transected (bisected) at the axial midpoint of the gonads to create one anterior and one posterior segment.  The trunk is also transected at the posterior margin of the gonad which will be the plane microtomed for the posterior sections (dashed lines in Figure 3).  



Figure 3. Dorsal trunk segments of A) male; and B) female juvenile (10 week post-NF62) X. laevis specimens which have been trimmed for evaluation of gonad and gonadal duct histopathology. The anterior margins of the gonads are positioned toward the tops of the images. The dashed lines represent the cuts used to bisect the specimens into anterior and posterior segments and to provide a sectioning plane for the posterior margin of the gonad.

      (c)  Histologic Procedures.
      
      (1)  Larval Thyroid Glands
      
      (i)  Decalcification.  The acetic acid in modified Davidson's solution will soften bone and cartilage; therefore, decalcification prior to processing may or may not be required.  The study tissues can be tested by processing and microtoming a few control frogs prior to the other study animals.  If decalcification is found to be necessary, commercial preparations that contain a mixture of formic acid and EDTA are preferred, because they provide the optimum combination of gentleness and speed.  Decalcification times can often be 24 hours or less.	
      
      (ii)  Processing and embedding.  Each head / neck specimen is processed in an automated tissue processor and infiltrated with paraffin according to routine methods.  Samples are embedded in the paraffin block so that the cut surface of the posterior margin (the neck side) is sectioned first.

      (iii)  Microtomy.  Each block is faced (excess paraffin trimmed away) until at least one of the thyroid glands has been reached (approximately 500 microns into the tissue).  Step sections (4-5 microns thickness) are then taken at 50-micron intervals until the maximum diameter of at least one gland has been reached.  Two additional step sections are then cut at an interval of 50 microns, and these two sections are placed on a single glass slide.  Following microtomy, each paraffin block is sealed with paraffin.

      (iv)  Staining and coverslipping.  The thyroid gland sections are stained with hematoxylin and eosin (H&E) and are mounted with a glass coverslip using an appropriate permanent mounting medium.
      
      (v)  Labeling.  Slides are labeled with at least the following information:
      
   * Study number
   * Exposure Laboratory
   * Name of the test chemical
   * Individual animal identification number.

	(vi)  Quality control.  The bilateral thyroid glands in the two step sections should contain the maximum number of follicles.  If inadequate numbers of follicles are represented, and it is apparent that further thyroid tissue remains in the block, additional step sections are cut at 50 micron intervals until the appropriate sections are captured or until it is determined that additional recuts will not yield the required tissue.  The pathologist ultimately selects the sections to be evaluated. Following microtomy, each paraffin block is re-sealed with paraffin.

      (2)  Juvenile Liver.
      
	(i)  Processing and embedding.  Each liver specimen is processed in an automated tissue processor and infiltrated with paraffin according to routine methods.  Samples are embedded in the paraffin block so that the area of liver to be examined is maximized.

	(ii)  Microtomy.  Each block is faced (excess paraffin trimmed away) until the liver surface is reached.  Step sections (4-5 microns thickness) are then taken at 50-micron intervals until the maximum diameter of the liver samples is attained.  Two additional step sections are then microtomed at an interval of 50 microns, and these two sections are placed on a single glass slide.  Following microtomy, each paraffin block is sealed with paraffin.  

	(iii)  Staining and coverslipping.  The liver sections are stained with hematoxylin and eosin (H&E) and are mounted with a glass coverslip using an appropriate permanent mounting medium.

      (iv)  Labeling.  Slides are labeled with at least the following information:
      
   * Study number
   * Exposure Laboratory
   * Name of the test chemical
   * Individual animal identification number

      (3)  Juvenile Gonads, Gonadal Ducts, and Kidneys
      
      (i)  Decalcification.  The acetic acid in modified Davidson's solution will soften bone and cartilage; therefore, decalcification prior to processing may or may not be required.  The study tissues can be tested by processing and microtoming a few control frogs prior to the other study animals.  If decalcification is found to be necessary, commercial preparations that contain a mixture of formic acid and EDTA are preferred, because they provide the optimum combination of gentleness and speed.  Decalcification times can often be 24 hours or less.

      (ii)  Processing and embedding.  The trunk segments are processed in an automated tissue processor and infiltrated with paraffin according to routine methods.  The anterior and posterior trunk segments are embedded in separate blocks.  Blocks should be labeled as anterior (a) or posterior (p).  Each sample is embedded in the paraffin block so that the middle gonad surface of the anterior segment and the posterior gonad surface of the posterior segment are microtomed first.
      
	(iii)  Microtomy.  For blocks containing anterior segments, two serial sections (each 5 microns thick) are acquired as soon as the block is faced (excess paraffin trimmed away), and placed on a single slide; these will be the middle sections.  Each anterior segment block is then microtomed 500 microns for males, or 1000 microns for females, and then a second pair of serial sections is taken and placed on a single slide; these will be the anterior sections.  Finally each posterior section block is faced and then microtomed 1000 microns for males and females, and two serial sections are taken which are placed on a single slide; these will be the posterior sections.  If the gonads are much larger or smaller than anticipated, adjustments may be made in the depths at which anterior and posterior sections are acquired.  Such adjustments should be documented in the methods section of the pathology report.  Following microtomy, each paraffin block is re-sealed with paraffin.

      (iv) Staining and coverslipping.  The gonad/kidney/duct sections are stained with hematoxylin and eosin (H&E) and are mounted with a glass coverslip using an appropriate permanent mounting medium.
      
      (v)  Labeling.  Slides should be labeled with the following information:

   * Study number
   * Exposure Laboratory
   * Name of the test chemical
   * Individual animal identification number 
   * Lower case letters are used to indicate the anatomic locations of the sections as follows: a = anterior; m = middle; p = posterior to indicate the gonad region for the slide (e.g. 141a = slide 141 anterior gonad sections).  

      (d)  Pathologic Evaluation.

      (1)  General Approach to Pathologic Evaluations.  Slides are to be read by individuals experienced in reading toxicologic pathology slides, and who are familiar with normal, amphibian thyroid and gonad histology, physiology, and general responses of these organs 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, amphibian histology and toxicologic pathology.
        
It is recognized that there is a limited pool of pathologists with the necessary training and experience that are available to evaluate the histopathology endpoint for the LAGDA assay.  If an individual has toxicologic pathology experience and is familiar with thyroid, liver, and gonadal histology in amphibian species, he/she may be trained to evaluate the LAGDA 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 organisms) 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.  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 that the tissue section is not remarkable.  This is not to mean "Grade 0." This practice provides continuity with subsequent statistical analyses.  

It is recognized that this severity grading scheme differs from that of Grim et al., 2009, which advocated the following scale for the assessment of thyroid gland changes in X. laevis: Grade 1 = mild, Grade 2 = moderate, and Grade 3 = severe.  However, for consistency, all Tier 2 tests should use the Grade 1-Grade 4 system.

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 endocrine disrupting chemicals in wildlife are considered to be exacerbations of "normal", physiologic findings.  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 inter-laboratory results comparisons.
	
      (3)  Data Recording.  The pathologist records the results in an appropriate record-keeping manner in keeping with GLP.  Example spreadsheet templates for recording data are available on the EDSP website: http://www.epa.gov/endo/.  For each frog, 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 particular frog, 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 in the form of severity scores or developmental stages.  A test termed RSCABS (Rao-Scott Cochran-Armitage by Slices) uses a step-down Rao-Scott adjusted Cochran-Armitage trend test on each level of severity in a histopathology response (Greene et al., 2014). The Rao-Scott adjustment incorporates the replicate vessel experimental design into the test.  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 than controls (i.e., have more severe pathology than controls), but it also determines at which severity score the difference occurs thereby providing much needed context to the analysis.  In the case of developmental staging of gonads and reproductive ducts, an additional manipulation should be applied to the data since an assumption of RSCABS is that severity of effect increases with dose or concentration. The effect observed could be a delay OR acceleration of development. Therefore, developmental staging data should be analyzed as reported to detect acceleration in development and then manually inverted prior to a second analysis to detect a delay in development.
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 relatively 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, liver and body weights, genotypic sex, time-to-metamorphosis, and age at sacrifice.
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 chemical.  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 treatment induces a low frequency of a lesion type that rarely occurs spontaneously.
      
      (ii)  Determining Relationship to Endocrine Disruption.  A similar weight-of-evidence (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 include: 1) Pathology Narrative Report, 2) Spreadsheet with recorded data, and 3) TIFF 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; and 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 hypotheses 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 hypotheses 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, the completed spreadsheet should indicate the animals from which figure images were photographed, and the number of images obtained per photographed organism.  An example spreadsheet is provided on the EDSP website: http://www.epa.gov/endo/.
	
      (iii) Figures.  For record-keeping purposes, a complete set of unannotated photomicrographic figures should be submitted electronically on portable media as uncompressed TIFF files.
	
      (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 avian expert).  Examples of recommended procedures for conducting pathology peer reviews have been described (e.g., Morton et al., 2010; The Society of Toxicologic Pathologists, 1991; 1997).
      
	(8)  Summary of Diagnostic Criteria for Selected Histopathologic Findings. Table 1 lists diagnostic criteria for selected findings in LAGDA studies according to tissue type.
Table 1.  Diagnostic Criteria for Potential Histopathological Findings in LAGDA Studies

Tissue
Diagnostic Criteria
Thyroid
                                       Criteria for thyroid gland assessment are available in Grim et al. (2009).
Liver
Decreased (Glycogen) Vacuolation
1
Hepatocellular vacuolation is decreased slightly relative to controls, but at least 75% of the cytoplasm is vacuolated.


2
50-75% of the cytoplasm is vacuolated.


3
25-50% of the cytoplasm is vacuolated.


4
Less than 25% of the cytoplasm is vacuolated.
Gonad
Testis Stage
1
Undifferentiated gonad


2
Individual primary spermatogonia and undifferentiated somatic cells populate the medullary region


3
Seminiferous tubules with primary spermatogonia and cysts of secondary spermatogonia


4
Primary spermatocytes with rete testis formation; may have occasional spermatocysts that contain round or elongated spermatids


5
All stages of spermatogenesis evident

Tubule Development Score
1
Less than 50% of seminiferous tubules have visible lumens


2
50% or greater seminiferous tubules have visible lumens

Ovary Stage
1
Undifferentiated gonad


2
Gonad identifiable as an ovary based on the presence of a discontinuously open lumen lined with epithelial cells; germ cells within the cortex consist of primary oogonia, cysts of primary mitotic oogonia, secondary oogonia, and very early meiotic oocytes


3
First appearance of diplotene oocytes in cortex; the most prevalent germ cell types at this stage are cysts of secondary oogonia and cysts of leptotene-pachytene primary meiocytes


4
Pre-vitellogenic (Dumont Stage I) diplotene oocytes are the most prevalent germ cell type observed by area and absolute cell counts; the central lumen is proportionately smaller while the whole ovary grows greatly in size and volume due to the growth of the oocytes; cysts in earlier stages of oogenesis become fewer in number and are located along the periphery of ovary


5
Ovary consists almost entirely of vitellogenic oocytes (Dumont Stage IV); pre-vitellogenic diplotene oocytes can be found along the periphery of the ovary and germ patches of primary and secondary oogonia are difficult to locate

Gonad Phenotype Score
0
Gonad phenotype matches genotype


1
Gonad phenotype primarily matches genotype, with a relatively smaller portion of opposite sex gonad tissue


2
Gonad is approximately equal parts testis and ovary


3
Gonad phenotype primarily matches opposite genotype, with a relatively smaller portion of genotypic sex gonad tissue


4
Gonad phenotype matches opposite genotype (sex reversed)

Germ Cell Degeneration
1
A single degenerating germ cell to less than three clusters of degenerating germ cells per gonad


2
Three or more clusters of degenerating germ cells, but less than 25% of gonad affected


3
Gonad consists of greater than 25% but less than 50% of degenerating germ cells


4
Gonad consists of 50% or more of degenerating germ cells

Mononuclear Cell Infiltrates
1
Small, focal infiltrates in one gonad


2
Small focal infiltrates in both gonads, or large area of infiltrates in one gonad


3
Large areas of infiltrates in both gonads


4
Gonads contain greater than 50% of infiltrates by area
Oviduct
Oviduct Stage
1
The oviduct is either a fibrous connective tissue tag, or is essentially non-existent


2
The oviduct has an epithelial cell lining (1-3 cell layers), and is comparable in size to the Wolffian duct


3
The oviduct has convoluted, multilayered epithelial cell lining, and is substantially larger (1.5x or greater) than the size of the Wolffian duct


4
The oviduct is dwarfs the size of the Wolffian duct.  Epithelial cells have formed large, basophilic glandular structures
Wolffian Duct
Wolffian Duct Stage
1
Epithelial lining of Wolffian duct is completely or focally less than two cell layers thick


2
Epithelial lining of Wolffian duct is two cell layers thick or greater, but clear apical portion of cell (if present) is less than half the cell height


3
Epithelial lining of Wolffian duct is two cell layers thick or greater, and clear apical portion of the columnar cells is at least half the cell height but less than 2/3 of the cell height


4
The clear apical portion of the columnar cells represents 2/3 or greater of the total cell height
Kidney
Mineralization
1
Less than three small foci of mineralization per kidney


2
Three or more small foci of mineralization per kidney, or less than three large foci


3
Three or more large foci of mineralization per kidney


4
Mineralized deposits in 50% or more tubules

Tubule Dilation
1
Minimal to mild dilation of less than three tubule clusters


2
Mild to moderate dilation of three or more tubule clusters


3
Moderate dilation affecting 50% or more tubules


4
Massive dilation of one or more tubules
	
      (9)  Atlas of Histopathologic Findings (Figures 4-39).  The purpose of this section is to provide: 1) a common technical "language" for describing findings and 2) a reference atlas of both normal microanatomical structures and potential pathological findings. This atlas is organized first by organ type, followed by an alphabetical list of terms, then working definitions, and finally photomicrographic examples. 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 pathologists. 



      (i)  Thyroid Glands



Figure 4. Normal versus Hypertrophic Thyroid Glands.  A. Transverse section through the neck region to demonstrate thyroid glands (arrows and inset) in control frog.  B. Severely hypertrophic thyroid glands.  Morphologic criteria for the histopathologic evaluation of the thyroid glands are described and illustrated in Grim et al., 2009.  Bar = 500 microns.
      (ii)  Liver



Figure 5. Hepatocellular Vacuolation, Increased.  Relative to the control liver, affected hepatocytes in the treated frog liver contain greater amounts of pale granular cytoplasm, the contents of which are morphologically consistent with glycogen.  Alterations in cell size can only be made by careful comparison of the treated frog livers to those of the concurrent controls.  Bar = 50 microns.




Figure 6. Hepatocellular Vacuolation, Decreased.  Relative to the livers of control frogs (see previous figure), affected hepatocytes contain lesser amounts of cytoplasmic glycogen.  This loss of glycogen vacuolation tends to occur preferentially in the basal (avascular) region of the cells, in the area adjacent to the biliary canals of Hering (arrow).  As vacuolation decreases, the liver progressively acquires a more basophilic overall appearance. Bar = 50 microns.
      (iii)  Gonads.



Figure 7. Normal Testes, Oviducts and Wolffian Ducts.  The testes (T) are attached to the peritoneal surface of the kidneys (K) by suspensory ligaments.  The black arrows indicate the oviduct (Müllerian duct) remnants (Stage 1 oviducts in this case), whereas the white arrows indicate the Wolffian ducts.  
Bar = 500 microns.




Figure 8. Normal Testis, Spermatogenic Stages.  Germ cell maturation occurs as follows: primary spermatogonia  secondary spermatogonia  primary spermatocytes  secondary spermatocytes  round spermatids  elongated spermatids  spermatozoa.  The majority of visible spermatocytes, which are primary spermatocytes, represent only prophase I of meiosis.  These include leptotene  zygotene  pachytene  diplotene phases.  Secondary spermatocytes, which are haploid representatives of meiosis II, are comparatively rare.  Sertoli cells are sparse, and those that are visible are usually surrounded by elongating spermatids.  Interstitial areas contain low numbers of Leydig cells (not indicated). 
Bar = 25 microns.



Figure 9. Meiotic Phases in the Vertebrate Testis.  Schematic diagram included for reference.



Figure 10. Staging of the Testis.  Stage numbers increase with increasing maturity of the testis.  Control males 10 weeks post NF stage 62 (LAGDA termination) typically have Stage 4 or 5 testes.



Figure 11. Stage 4 Versus 5 Testes, Higher Magnification.  Stage 4 testes can have elongating spermatids (arrow), but spermatocysts that contain spermatozoa are rare or not apparent.  Conversely, spermatozoa (S) are abundant in Stage 5 testes. Bar = 25 microns.



Figure 12. Germinal Epithelium Thinning and Dilation.  This finding is characterized by focal to diffuse attenuation of the testicular germinal epithelium (arrows), with expansion of the adjacent tubular lumen.  Germ cell degeneration is also evident.  In a previous report, similar changes were termed "dilated testis tubules" (Wolf et al., 2010). Bar = 100 microns. Grade 4 is left intentionally blank.




Figure 13. Testis, Germ Cell Degeneration.  This finding is characterized by the scattered presence of individual or clustered apoptotic germ cells (black arrows and inset), or germ cell syncytia (white arrow), within the germinal epithelium.  Affected testes may also have increased numbers of exfoliated cells and cellular debris in tubular lumina.  Bar = 25 microns (Grade 1), 50 microns (Grade 2).  Grades 3-4 are left intentionally blank. 




Figure 14. Spermatogonia, Increased.  This finding is characterized by a greater frequency of spermatogonia (arrows) in the testes of affected frogs as compared to those of the average control frog.  Bar = 50 microns. Grades 2-4 are left intentionally blank.




Figure 15. Mononuclear Cell Infiltrates.  Mononuclear cell infiltrates (arrows), which appear to consist primarily of lymphocytes, are relatively common in the testes.  Bar = 25 microns. Histogram provided for Grade 1 only.  Grades 2-4 are intentionally left blank.




Figure 16. Germ Cell Vacuolation.  This finding is characterized by increased amounts of clear cytoplasm in clusters of zygotene-phase primary spermatocytes.  Bar = 25 microns. Illustration for Grade 1 only.  Grades 2-4 are left intentionally blank.




Figure 17. Normal Ovaries, Oviducts and Wolffian Ducts.  The ovaries (O) are attached to the peritoneal surface of the kidneys (K) by suspensory ligaments.  The black arrows indicate the oviduct (Müllerian duct) remnants (Stage 2 oviducts in this case), whereas the white arrows indicate the Wolffian ducts.  Bar = 500 microns.




Figure 18. Staging of the Ovary.  Control females 10 weeks post NF stage 62 (LAGDA termination) typically have Stage 3 or 4 ovaries.



Figure 19. Mixed Sex Gonads.  The gonads contain both testicular (T) and ovarian (arrows) components.  Each of these gonads received a Phenotype Score of 2. Bar = 50 microns.


Figure 20. Complete Sex Reversal.  This normal appearing ovary was a treatment-related phenotypic finding in a genetic male frog.  This ovary would receive a Gonad Phenotype Score of 5.  Bar = 50 microns.




Figure 21. Hermaphroditism.  In addition to left and right testes (T), this animal had a small amount of ovarian tissue (arrow) within the abdominal mesentery.  K = kidney.  Bar = 250 microns.




Figure 22. Proteinaceous Fluid.  Proteinaceous fluid (arrows) appears as homogenous dark pink material within the central and subcapsular regions of the ovarian interstitium.  Bar = 100 microns (Grade 1), 50 microns (Grade 2). Grades 2-4 are left intentionally blank.



Figure 23. Oocyte atresia.  Spontaneous oocyte atresia appears to be relatively uncommon in the Stage 3 or 4 ovaries of most untreated control frogs.  Bar = 50 microns. Grades 2-4 are left intentionally blank.



Figure 24. Oocyte atresia, Additional Examples.  A. In this example, early oocyte degeneration is characterized by a loss of nuclear detail and increased nuclear density.  B. In this later stage of atresia, oocyte remnants are being scavenged by nests of macrophages (black arrows).  The white arrow indicates what appear to be hypertrophic granulosa cells from an atretic follicle.  Bar = 25 microns.


Figure 25. Ovary, Germ Cell Degeneration.  The term "germ cell degeneration" is used rather than "atresia" to characterize the in vivo deterioration of stem cells or oogonia, or the deterioration of phenotypically undifferentiated cells in early stage gonads.  Ovarian germ cell degeneration in the form of apoptotic-like cells (arrows) was observed as a treatment-related response in the Grade 2 image.  Bar = 25 microns.  Grade 1, 3, and 4 are left intentionally blank.
      (iv)  Gonadal Ducts.



Figure 26. Oviduct Staging.  As the oviduct (Müllerian duct) matures in the female, it becomes progressively larger, with an expanded lumen and a more convoluted mucosal lining.  Conversely, oviducts regress over time in male frogs.  Control male frogs typically have Stage 1 or 2 oviducts 10 weeks post NF stage 62 (LAGDA termination), and the average stage in males is usually less than the average stage in same-aged females of the same study.  In borderline cases, the size of the oviduct relative to the Wolffian duct is a useful criterion for differentiating between Stage 2 and Stage 3 oviducts.  Bar = 50 microns (Stages 1, 2, and 3), 100 microns (Stage 4).




Figure 27. Oviduct Staging, Additional Examples of Stage 1 Oviducts.  Stage 1 oviducts (arrows) range from small fibrous protuberances, to suspensory ligaments in which there is no apparent residual oviduct (e.g., Fig. D).  These oviducts do not have lumens or mucosal linings.  Bar = 50 microns.




Figure 28. Oviduct Staging, Additional Examples of Stage 2 Oviducts.  Stage 2 oviducts (arrows), which are comparable in size to the Wolffian ducts, all have a visible lumen and are lined by one to a few layers of mucosal epithelial cells, with or without slight mucosal folding.  Bar = 50 microns (Figures 28A-C), 25 microns (Figure 28D).




Figure 29. Oviduct Staging, Additional Examples of Stage 3 Oviducts.  Stage 3 oviducts (arrows), which greater than 1.5 times the size of Wolffian ducts, have intricate mucosal folding.  Bar = 500 microns (Figure 29A), 100 microns (Figures 29B-C), 25 microns (Figure 29D).




Figure 30. Oviduct Staging, Additional Examples of Stage 4 Oviducts.  Stage 4 oviducts (arrows) are markedly larger than Wolffian ducts, and the mucosal epithelium is forming, or has formed, large glandular structures.  Fig. D is enlarged relative to the other figures.  The oviduct in Fig. D represents the borderline between Stages 3 and 4.  Bar = 500 microns (Figure 30A- B), 100 microns (Figure 30C), 50 microns (Figure 30D).




Figure 31. Wolffian Duct Staging.  The Wolffian ducts, which are located within the lateral extremities of the right and left kidneys, also function as ureters in Xenopus spp.  With maturity, the Wolffian ducts become progressively larger, although not to the extent of oviducts in female frogs.  The mucosal lining of the ducts also becomes thicker and contains more mucous cells with age.  Generally, mean Wolffian duct scores are slightly higher in control males than in control females of the same study.  Bar = 25 microns (Stages 1-3), 50 microns (Stage 4).
      (v)  Kidneys.

FIGURE 32. Normal Kidney from a Control Frog.  Glomeruli (G) are located in the ventral third of the kidneys, whereas proximal tubules (T) occupy the dorsal portions.  Bar = 100 microns (Figure 32A), 50 microns (Figure 32B).
FIGURE 33. Proteinaceous Fluid.  Proteinaceous fluid (arrows) appears as homogenous dark pink material within the renal interstitium, blood vessels, tubules, and/or Bowman's spaces.  Bar = 25 microns (Grade 1), 100 microns (Grade 4). Grades 2-3 are left intentionally blank.
FIGURE 34. Proteinaceous Fluid, Grade 4, Additional Examples.  Tubules, glomeruli, and the renal interstitium are flooded by proteinaceous fluid in these examples.  BS = dilated Bowman's space.  Bar = 500 microns (Figure 34A), 100 microns (Figure 34B), 50 microns (Figures 34C-D).

















FIGURE 35. Fibrosis.  This finding, which is characterized by excessive amounts of fibrous connective tissue (F) within the renal interstitium, usually occurs as a sequel to chronic inflammation or other long standing parenchymal damage.  Bar = 100 microns (Grade 3). Grades 1, 2, and 4 are left intentionally blank.
FIGURE 36. Glomerulomegaly and Glomerular Hypercellularity.  This finding can be a consequence of low level chronic glomerular damage.  The arrow indicates a small deposit of proteinaceous material within a glomerular capillary.  Bar = 20 microns (Grade 3).  Grades 1, 2, and 4 are left intentionally blank.
FIGURE 37. Regenerative Blast Cell Hyperplasia.  This finding, which is another consequence of chronic renal damage and tubular loss, is characterized by streaming proliferations of cells with large, hyperchromatic (dark) nuclei (arrows).  Figure 37B is a higher magnification of Figure 37A.  This particular case was assigned a severity score of Grade 2 (mild).  Bar = 50 microns (Figure 37A), 25 microns (Figure 37B).


Figure 38. Mineralization with Tubular Dilation.  Mineralization appears to be a common background finding in laboratory reared X. laevis, and the severity of this finding appears to vary from facility to facility.  Husbandry factors (e.g., issues involving feed and/or water composition) are suspected causes.  The occurrence of mineralization is associated with focal to diffuse renal tubular dilation, presumably due to obstruction of urine flow.  Tubular dilation and mineralization are usually graded separately. In a given frog, the severity grade for tubular dilation is typically one grade less than that of mineralization (see example below).  Bar = 25 microns (Grade 1), 100 microns (Grades 2 and 3). Grade 4 is left intentionally blank.



Figure 39. Mineralization, Additional Examples.  Figure A represents another case of Grade 3 mineralization and Grade 2 tubular dilation.  Figure B illustrates a small amount of mineral (arrow) within a multinucleated giant cell macrophage.  Bar = 100 microns (Figure 39A), 25 microns (Figure 39B).
      (e)  References.

(1)  Green J.W., Springer T.A., Saulnier A.N., Swintek J. (2014). Statistical analysis of histopathology endpoints.  Environ Toxicol Chem. 33(5):1108-1116.

(2)  Grim K.C., Wolfe M., Braunbeck T., Iguchi T., Ohta Y., Tooi O., Touart L., Wolf D.C., Tiege J. (2009). Thyroid histopathology assessments for the amphibian metamorphosis assay to detect thyroid-active substances.  Toxicol Pathol. 37:415-424.

(3)  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.

(4)  Nieuwkoop P.D., Faber J.  (1994). Normal Table of Xenopus laevis (Daudin). Garland Publishing, Inc.  New York and London.

(5)  The Society of Toxicologic Pathologists. (1991). Peer review in toxicologic pathology: some recommendations.  Toxicol Pathol. 19:290-292.

(6)  The Society of Toxicologic Pathologists. (1997). Documentation of pathology peer review.  Position of the Society of Toxicologic Pathologists.  Toxicol Pathol. 25(6):655.

(7)  Wolf J.C., Lutz I., Kloas W., Springer T.A., Holden L.R., Krueger H.O., Hosmer A.J. (2010).  Effects of 17 beta-estradiol exposure on Xenopus laevis gonadal histopathology.  Environ Toxicol Chem. 29(5):1091-1105.

