



              EDSP: Weight of Evidence Analysis of Tier 1 Studies
                                       
                                       
                            Case Study: Chemical X
                                       
                                       
                         Office of pesticide programs
                   office of science coordination and policy
                   Office of Pollution Prevention and Toxics
                      U.S Environmental Protection Agency
                                       
                                       
                                       

1.	Case Study  -  Chemical X
A.	Introduction
Chemical X is presented as a specific case study to illustrate the process of applying a Weight of Evidence (WoE) analysis to determine a chemical's potential to interact with the endocrine system.  Unlike the EDSP List 1 case studies, Chemical X was not subject to EDSP Tier 1 screening and, therefore, only Other Scientifically Relevant Information (OSRI) is available for consideration in the WoE evaluation.  Therefore this case study will focus on how OSRI may in itself inform whether or not a chemical has the potential to interact with endocrine system. Chemical X is a non-pesticide that comes from a family of closely related, commercially important alkylphenols used in the production of ethoxylates which have a wide variety of industrial applications and are produced in large volumes.  The endocrine activity of Chemical X and its associated ethoxylates and carboxylates has been well studied in vitro and in vivo and, therefore, was chosen as a case study to illustrate the WoE process solely based on OSRI.

The OSRI chosen to represent the potential for Chemical X to interact with the endocrine system consists primarily of published, scientifically peer-reviewed studies from the public domain.  When possible, the selection of studies was based on those that closely resembled in principle and methodology of the EDSP Tier 1 screening assays.  It is emphasized that this case study, as with the EDSP List 1 case studies, is intended only to be illustrative of potentially challenging scenarios involved in the implementation of the WoE guidance.  The agency is intentionally keeping the identity of this chemical anonymous to avoid any chemical - specific recommendations and to ensure a broad interpretation of this case so that it may be generalized for other chemicals that might have a similar toxicity profile.  The presentation of Chemical X to the SAP will not have any impact on the current regulatory standing of Chemical X.  The OSRI selected and the WoE analyses presented for this case should not be interpreted as comprehensive, final or complete beyond this limited illustration.  The agency will fully evaluate Chemical X, taking into account SAP comments on the WoE process, to determine whether or not Chemical X has the potential to interact with the endocrine system when the final SAP report is available.  The SAP is asked to include Chemical X along with the other chemicals in their responses to respective charge questions.
B.	Data Available for Chemical X
Chemical X is a  phenolic compound with multiple isomers usually found as a pale yellow, highly viscous liquid, with an approximate molecular weight of 215.0 to 220.4 g/mole, specific gravity of 0.953 g/mL at 20°C (Budavari 1989), and a vapor pressure of 4.55 x 10[-3] (+-3.54 x 10[-3]) Pa (Roy F. Weston Inc. 1990). It has a dissociation constant (pKa) of 10.7+-1.0 and a log octanol/water partition coefficient (log Kow) of 3.80 to 4.77 (Roy F. Weston Inc. 1990).  The water solubility of Chemical X is pH-dependent: 4,600 ug/L at pH 5.0; 6,237 ug/L at pH 7.0; 11,897 ug/L at pH 9.0. The solubility of Chemical X in seawater is 3,630 ug/L and it is soluble in many organic solvents (Roy F. Weston Inc. 1990). Ahel and Giger (1993) measured the solubility of Chemical X at different temperatures in distilled water and demonstrated a nearly linear increase in solubility between 2°C (4,600 ug/L) and 25°C (6,350 ug/L).  The chemical has and does not volatize appreciably and is characterized as being moderately persistent, as it is resistant (stable) to abiotic routes of degradation (i.e., hydrolysis and photolysis) and to anaerobic biotic metabolism. The relatively high log10 octanol-water partition coefficient and high organic carbon partition coefficient (Koc~60,000 mL/L), when combined with the compound's low solubility, would limit the upper range of concentrations that could be tested. Although the compound is relatively lipophilic, it does not bioaccumulate appreciably in fish (BCF range 90  -  330x).

C.	Other Scientifically Relevant Information (OSRI) for Chemical X
Information that is submitted voluntarily and is applicable to support or clarify an EPA action is generally referred to as "other" scientifically relevant information.  Sources of relevant scientific and technical information may include toxicity study results from EPA or OECD equivalent test guidelines and information from published or publically available peer-reviewed studies.  Regardless of the source, the information is evaluated for quality and relevance, in accord with EPA's guidance on scientific integrity (EPA, 2012).
Chemical X has been widely investigated in the scientific community as having estrogenic activity; hence, the majority of published and publically available peer-reviewed studies available as OSRI involve the estrogen hormonal pathway.  OSRI that was selected to illustrate the WoE process applied to Chemical X consisted of in vitro and in vivo studies that are presented in a manner similar to the EDSP List 1 case studies for continuity.  The results from OSRI for the various modes or pathways of action involving E, A and T for receptor binding (agonist and antagonist) and activation/transcription, steriodogenesis (competitive induction or inhibition), and the hypothalamic-pituitary-gonadal and -thyroidal axes are summarized in the following sections in context with respective EDSP Tier 1 screening assays.
i.	ER Binding Assays
The results from several studies which are technically similar to the EDSP Tier 1 ER binding assay (OCSPP 890.1250) that have demonstrated the potential of Chemical X to bind with the estrogen receptor are presented.  In a study by Danzo (1997), ER binding was observed using rabbit uterine cytosol incubated with 7 nM [[3]H] 17β-estradiol alone overnight at 4[o]C or with 100-fold molar excess of radioinert 17β-estradiol to determine nonspecific binding, or 100 μM of Chemical X.  Results indicated Chemical X (red box) inhibited (75% reduction) binding of tritiated ([3]H)estradiol (E2) to the estrogen receptor as shown in Figure 1.

Figure 1.  The mean +- standard error is shown.  Percent inhibition of [3]H estradiol binding to the estrogen receptor is determined for a variety of chemicals.  On the far left, the specific binding of [3]H estradiol (*E) is shown (100%).  Second from the left, unlabeled estradiol (E) results in 100% inhibition (i.e. 0% [[3]H] 17β-estradiol bound to receptor).  Chemical X (100 μM) is indicated by the red box.  The numbers in parentheses above the bars indicate the number of experiments in which samples were assayed in triplicate. Figure adapted from Danzo 1997.
Blair et al. (2000) investigated 188 different chemicals for their relative binding affinity (RBA) to 17β-estradiol.  Chemical X obtained from five different sources competitively displaced radiolabled ligand from the ER with a mean IC50 of 2.4 to 4.7 x 10[-6] and a RBA of from 0.02 to 0.037 % as shown in Figure 2.

Figure 2.  Demonstrated affinity of Chemical X (open triangles) for the ER appears to be parallel to the linear portion of the curve of natural ligand estradiol (closed circles).  Figure adapted from Blair et al., 2000.
In an investigation by Satoh et al. (2001), human ERα binding was quantified using fluorescence-labeled 17β-estradiol, incubated with Chemical X for 2 hr at room temperature, Chemical X bound to ERα with an IC50 of 7.0 x 10[-7]M and an RBA of 2.22%.
Laws et al. (2000), in a large study including multiple in vivo experiments substantiated by in vitro ER binding and transcription activation, reported weak affinity of Chemical X for the rat uterine cytosolic ER as demonstrated by Ki of 0.67 μM compared to 0.4 nM for 17-β-estradiol.
Chemical X represents a variety of isomers with linear and branched side chains which affect the affinity of the chemical for the estrogen receptor.  For this reason, comparisons across studies may be difficult.  Despite these challenges, the results of estrogen receptor binding assays conducted in different cell backgrounds and following different methods indicate Chemical X binds to the estrogen receptor under a variety of assays conditions.
Thus, based on the combined results, Chemical X is considered a positive for ER binding in competitive receptor binding assays.


ii.	ER Transcriptional Activation Assays
Balaguer et al (1999) utilized four reporter gene assays using different cell lines transfected with various ER constructs to investigate the ability of chemical X to initiate ER mediated transcriptional activation.  For one assay, the authors used MCF-7 cells, with endogenous ERα expression, transfected with an ER-responsive luciferase reporter.  The authors also used a HeLa cell line containing endogenously expressed ER stably transfected with ERα, ERβ, or a chimeric ERα receptor along with a luciferase reporter gene.  After 3 days pretreatment with steroid-free calf serum to reduce steroid background of the culture medium, cells were incubated for 16 hr with appropriate reagents and luciferase activity was measured.  In all models, the Chemical X mixture tested was active.

Figure 3.  Induction of luciferase activity in (A) a MCF-7 cell line (MELN) and (B) a HeLa cell line (ER α) based ER-mediated luciferase reporter gene assays.  Results are expressed as a percentage of luciferase activity measured per well (mean +- SEM of triplicate wells).  The 100% value represents the value obtained in presence of E2 10[-8]M (9500 arbitrary units/0.1 mg of protein).  Estradiol (E2) is shown in open triangles and Chemical X is indicated by open circles.  Adapted from Balaguer et al. 1999.

Bonefeld-Jorgensen et al. (2007) also demonstrated ER transcription activation in the MVLN cell line, which are MCF-7 cells stably transfected with a luciferase reporter. Chemical X was tested in three independent runs for agonist and antagonist activity at concentrations ranging from 10[-]8 to 10[-4]M.  Chemical X produced a clear estrogenic response with an EC50 of 8.9 μM.  No antagonist activity was observed when Chemical X was tested in the presence of 25 uM.

Figure 4.  Dose - response ER transactivation of E2 and Chemical X.  The graph on the left shows relative luminescence in MVLN cells exposed to E2 (0.05 - 500 pM, closed circles) and to 10-8 to 10[-4]M Chemical X (open diamonds) for 24 hr. The graph on the right shows agonistic/antagonistic ER activity of Chemical X tested alone or in co-exposure with 25 pM E2, which was set to 1. Mean values are shown (n >= 3). *Significantly different from the respective solvent controls (cells + 0.1% DMSO; 25 pM E2 + 0.1% DMSO).  Adapted from Bonefeld-Jorgenssen et al. 2007.
Using an ER-mediated, chemical-activated luciferase reporter gene (ER-CALUX) expressed in human T47D breast cancer cells, Legler et al. (1999) determined that Chemical X was among the most potent chemicals tested with an EC50 of 260 nM.
White et al. (1994) tested the effect of 4-Chemical X in rainbow trout (Oncorhynchus mykiss) hepatocytes and several human breast cancer cell lines.  Chemical X increased vitellogenin gene expression in trout hepatocytes and transcriptional activity in human breast cancer cells cotransfected with mouse estrogen receptor and a reporter gene (EREBLCAT) at concentrations in the range of 10[-6] to 10[-5] M.

Figure 5.  ER-mediated gene induction in A) rainbow trout hepatocytes and B) MCF-7 human breast cancer cells.  The graph on the left shows vitellogenin induction in rainbow trout hepatocytes and the graph on the right shows fold induction of a luciferase reporter gene linked to the ER stably transfected into MCF-7 cells in response to 10[-5], 10[-6], and 10[-7] M concentrations of Chemical X.  On both graphs, the response to 10[-8] E2 is indicated by the first bar on the graph.  Adapted from White et al. 1994.
Though different reporter gene constructs and cell backgrounds may have different sensitivities to native ligand and Chemical X, a variety of ER transcription activation assays comparable to the Tier 1 ERTA or quantifying similar ER-mediated gene activation support the ability of Chemical X to bind the ER and induce gene expression.
Thus, based on the combined results, Chemical X is considered positive for estrogenic activity in ER transcriptional activation assays.
iii.	AR Binding Assays 
The potential of Chemical X to bind the androgen receptor was tested in a variety of assay systems.  Satoh et al. (2001) incubated human androgen receptor for 1 hr incubation at 4[o]C with unlabeled testosterone, followed by the addition of an anti-testosterone antibody and peroxidase-labeled steroid for an additional hour.  After plates were washed, Chemical X and substrate were added and developed color was quantified. Chemical X did interact with the human AR used in the kits and the IC50 and relative binding affinity were determined to be 1.3 x 10[-5] M and 0.131%, respectively, though from the results, the authors could not determine if Chemical X acted as an AR agonist or antagonist.



Figure 6.  Competition by Chemical X for Testosterone Binding to Androgen Receptor.  Ligand binding experiments were carried out using a Ligand Screening System-Androgen Receptor kit (Toyobo, Japan). The SD was less than 2.6% (n = 5). Ordinates: % of inhibition = A  -  B/A (A and B were the fluorescence intensities in the absence and presence of competitor, respectively. Closed circles indicate the positive control, milbolerone.  Closed triangles indicate Chemical X. Adapted from Satoh et al. 2001.  

Using yeast transformed with full length mammalian AR and a luciferase reporter plasmid containing two AREs and a prostate specific antigen promoter, Lee et al. (2003) reported 5 nM Chemical X treatment inhibited [3H]5-dihydrotestosterone (DHT) binding to the AR by 30% in HeLa cells transfected with a luciferase-linked mouse AR, though the response was not dose-dependent.  Further, in transient transfection assays, Chemical X inhibited AR transcriptional activity (IC50 = 2.6 μM) and was a more potent inhibitor of androgen-induced gene expression than recognized antagonist cyproterone acetate.  The inhibitory effects of Chemical X on androgen receptor-mediated signaling is supported by the reported antiandrogenic effect of Chemical X in the range of 0.60 - 20 μM, on the induction of receptor activity by the androgen agonist R1881 with an IC50 of 14.1 μM, and a maximum inhibition (MI) of 56%.  Though high concentrations of Chemical X can cause cytotoxicity, the authors report yeast cell growth (measured by absorbance at 600 nm) was not significantly reduced in the concentrations of Chemical X tested in this study. 

Chemical X has been shown to significantly reduce AR gene transcription in Chinese hamster (Cricetulus griseus) ovary cells transfected with human AR and a luciferase reporter (Bonefeld-Jorgensen et al., 2007).  The potential AR antagonism of several compounds was tested in cells treated with increasing concentrations of Chemical X (0.15 to 40 x 10[-6]M) in the presence of an AR agonist (0.1 nM R1881).  AR binding was analyzed using a four parameter logistic curve fit and an IC50 was calculated. Chemical X elicited an antagonist response with an IC50 of 14.1 μM and a maximum inhibition of 56% at 10-6 M.  Chemical X was determined to be cytotoxic to CHO cells at concentrations > 40 x 10[-6]M. 



Figure 7. AR antagonism of Chemical X (open triangles) on R1881 action on CHO cells stably transfected with AR and a luciferase reporter.  Adapted from Bonefeld-Jorgensen et al. 2007.


These studies demonstrate that Chemical X interacts with the androgen receptor in in vitro assays, though it is important to note that steroidal estrogens and xenoestrogens will elicit a positive response in AR binding assays.
iv.	Steroidogenesis Assays
Effects of Chemical X on steroidogenesis were examined in assay systems substantially different than the in vitro EDSP Tier 1 Steroidogenesis assay (OCSPP 890.1550). Leydig cells from adult male Sprague-Dawley rats were isolated and treated with a range of low concentrations (0.0011, 0.0033, 0.0055, 0.011, 0.022 mg/L) and higher concentrations (0.11, 0.55, 1.1, 1.65, 2.2, 2.75, 3.3, 5.5 mg/L) of Chemical X disolved in 0.1% dimethylsulfoxide (DMSO).  Cell viability was tested by trypan blue exclusion. Testosterone production measured after 48 hr was significantly increased in the low concentrations and reduced at higher concentrations. Concentrations of Chemical X greater than 2.2 mg/L resulted in morphologic changes to Leydig cell cultures such as cell shrinkage and relaxation of cell junctions.  The decrease in testosterone production may be the result of an increased Leydig cell death at high concentrations.

Figure 8. Testosterone concentrations in media of Leydig cells treated with low concentrations (left) and high concentrations (right) of Chemical X.  * P < 0.05, ** P < 0.01, compared with the control group (n = 6).  Adapted from Gong and Ham 2005.

Kortnet et al. (2009) reported that 1, 10, and 50 μM concentrations of Chemical X produced time- and concentration-specific effects on the expression of steriodogenic enzyme mRNA quantified by real-time PCR in in vitro Atlantic salmon (Salmo salar) ovarian tissues exposed for 3 and 7 days. Expression of steriodogenic acute regulatory protein (StAR), side-chain cleavage enzyme (P450scc), 3 β-hydroxysteroid dehydrogenase (3β-HSD), and 17α-hydroxylase/17, 20 lyase (P450C17) decreased with exposure to Chemical X.  Increases in aromatase (P450 CYP19) transcript following exposure to 10 μM concentrations of Chemical X paralleled elevated ovarian estradiol levels at the same exposure concentration. Tissue levels of ovarian testosterone and 11- ketotestosterone were significantly elevated following 7 days of exposure to all test concentrations of Chemical X. 

Results suggest Chemical X may interfer with in vitro steroidogenesis.

v.	Aromatase Assays 
Chemical X was tested in an EPA-funded study (RTI International, 2007) during the inter-laboratory validation studies for the Tier 1 Aromatase assay guideline (OSCPP 890.1200).  Briefly, Chemical X was incubated for 15 min with human recombinant aromatase and tritiated androstenedione (1-β [[3]H(N)]-androst-4-ene-3,17-dione ([[3]H]ASDN)) dissolved in ethanol and aromatase activity was determined by measuring the amount of tritiated water produced.  Four independent runs were conducted and each run included a full activity control, a background activity control, a positive control series (5, 15, and 25 μM) including a known inhibitor (aminoglutethimide), and a range of concentrations of the test chemical (5, 7.5, 10, 15, and 20 μM) with 2 replicates per concentration.  Chemical X inhibited aromatase activity with an IC50 of 21 μM.
In a second study, aromatase activity was measured in a human choriocarcinoma cell line was measured by tritiated water production from [3]H -androstenedione substrate.  Chemical X was added to cells and incubated for 18 hr after which cells were washed and provided with 0.2 μCi [1β-[3]H] androst-4-ene-3,17-dione and 10 nM unlabeled 4- 4-androstene-3,17-dione.  After 2 hr of incubation, the assay was terminated and cell media was extracted and analyzed for tritiated water production as a measure of aromatase activity. All concentrations of Chemical X were tested in triplicate and run in three independent assays, each of which included a positive control, the known aromatase inhibitor, 4-androsten-4-ol-3,17-dione.  Chemical X concentrations in the range of 10[-9] to 10[-5]M resulted in a maximum inhibition of 71% observed at 10[-5]M (Bonefeld-Jorgensen et al. 2007).
Based on the results of these studies, Chemical X appears to be an in vitro inhibitor of aromatase. 
vi.	Uterotrophic Assays
The results from several studies done in a uterotrophic assays similar in design and methodology to the EDSP Tier 1 screening assay (OCSPP 890.1600) are presented with the interpretation that a statistically significant (P<0.05) increase in uterine weight relative to controls is indicative of a positive response.  In a study by Laws et al. (2000), an experiment  was designed to evaluate the effects of Chemical X on uterine weight in sexually immature females rats (prepubertal) at 6 and 24 hours following the last of 3 daily doses (25, 50, 100, and 200 mg/kg) when administered by oral gavage for 3 days starting on PND 21-24.  Positive controls were 17 β-estradiol (E2) and ethynyl estradiol (EE).  Wet uterine weight (i.e., uterus and fluid in uterine lumen) expressed as percent of control weight was significantly increased in prepubertal animals at the 50 and 100 mg/kg doses at both 6 and 24 hr after the final dose as shown in Figure 9.

In another similarly designed experiment from the same study (Laws et al., 2000), the effects of Chemical X on uterine weight when administered by subcutaneous injection or oral gavage were examined.  Both dosing routes resulted in an increase in uterine weight, although results with oral gavage appeared to have a significant effect at a lower dose (> 50 mg/kg) than subcutaneous delivery (> 100 mg/kg) as shown in Figure 10.



Figure 9.  Three-day uterotrophic assay in prepubertal rats: comparison of effects 6 and 24 h after the last dose.  Wet uterine weight expressed as percent of control following exposure to 17--estradiol (E2) by sc injection, or to ethynyl estradiol (EE), or Chemical X (red box) by oral gavage. *Significant treatment effect by ANOVA or the Kruskal-Wallis Nonparametric Anova Test (p = 0.05) with comparison to the control by Dunnett's Multiple Comparison Test, Mann-Whitney test, or Dunn's Multiple Comparisons Test (p < 0.05).  Adapted from Laws et al. 2000.


Figure 10.  Three-day uterotrophic assay in prepubertal rats: comparison of effects sc injection or oral gavage dosing routes.  Wet uterine weight expressed as percent of control following exposure to 17-b-estradiol (far left) by sc injection, or to ethynyl estradiol (EE; second data set from left), or chemical X (red box) by oral gavage. *Significant treatment effect by ANOVA or the Kruskal-Wallis Nonparametric Anova Test (p = 0.05) with comparison to the control by Dunnett's Multiple Comparison Test, Mann-Whitney test, or Dunn's Multiple Comparisons Test (p < 0.05).  Adapted from Laws et al. 2000.

Again, in the same study by Laws et al. (2000), the effect of Chemical X on uterine weight in intact prepuberal and adult ovariectomized female rats was compared.  Uterine weights in adult ovariectomized females were significantly increased as shown in Figure 11, but at doses greater (100 mg/kg) than required to increase uterine weight in the prepubertal rats as shown in Figures 9 and 10.


Figure 11.  Three-day uterotrophic assay in adult ovariectomized rats. Wet uterine weight expressed as percent of control following exposure to 17-b-estradiol (far left) by sc injection, or to ethynyl estradiol (EE; second data set from left), or chemical X (red box) by oral gavage. *Significant treatment effect by ANOVA or the Kruskal-Wallis Nonparametric Anova Test (p = 0.05) with comparison to the control by Dunnett's Multiple Comparison Test, Mann-Whitney test, or Dunn's Multiple Comparisons Test (p < 0.05).  Adapted from Laws et al. 2000.

Chemical X was also used as a reference chemical in a multi-laboratory validation study during development of the EDSP Tier 1 Uterotrophic Assay.  The results are summarized in an integrated summary report (OECD, 2003). Briefly, Chemical X was administered to adult ovariectomized rats at 15, 75, 125, 250 and 350 mg/kg/d via oral gavage for 3 consecutive days.  A vehicle control group (corn oil) and positive controls (17α-Ethinyl Estradiol, EE) were included to evaluate estrogen agonist activity.  All animals survived to the end of treatment in the 15, 75 and 125 mg/kg dose-groups; however, extensive mortality in the 250 and 350 mg/kg dose groups (38 and 67%, respectively) indicated overt toxicity at these treatment levels.  At lower doses, uterine weight was statistically increased in the 75 and 125 mg/kg/d treatment groups in all 4 participating laboratories.

Thus, based on the combined results of the Uterotrophic Assays similar in design and methodology to the EDSP Tier 1 screening assay, Chemical X is positive for estrogenic activity as indicated by a significant increase in uterine weight.
vii.	Hershberger Assays
In a Hershberger assay that was designed similar to the EDSP Tier 1 screening assay (OCSPP 890.1400), Freyberger et al. (2007) castrated male Wistar rats on PND 45 and allowed them to acclimate for 1 week prior to test chemical administration in two distinct experiments.  A 160 mg/kg dose of Chemical X was administered in a corn oil suspension orally and subcutaneously for 10 days.  One day after the last treatment, animals were killed and ventral prostate (VP), seminal vesicles (SV), gland penis (GP), levator ani and bulbocavernosus muscles (LABC), and Cowper's glands (COWS) were removed and weighed.  The study design included a testosterone propionate treatment (TP, 0.4 mg/kg) as an androgenic stimulus for testing antiandrogenic potential and flutamide (FLU, 3 mg/kg) as an antiandrogenic positive control.  In accord with the OSCPP 890.1400 test guideline, a statistical increase (P<0.05) in the weights of two or more target tissues relative to controls is indicative of a positive response.  Chemical X exposure did not result in a significant androgenic or an antiandrogenic response in this study.
Chemical X was also used as a reference chemical in a multi-laboratory validation study during development of the EDSP Tier 1 Hershberger Assay.  The results are summarized in an integrated summary report (OECD, 2007). Briefly, Chemical X was administered at 160 mg/kg/d via oral gavage using the castrated rat model.  A vehicle control group as well as positive control groups to evaluate androgen agonist (TP) and antagonist (FLU) activities were included.

Pooled across multiple laboratories, there was a decrease (P<0.01) in mean body weight (approximately 4%) for Chemical X compared to the vehicle control, thus indicating that 160 mg/kg/d corresponded to a maximally tolerable dose (MTD). For the androgenic response, there was no statistically significant  (p<0.05) increase in any of the 5 targeted tissues for Chemical X compared to the control group among the 10 laboratories.  For the anti-androgenic response, 5 of the laboratories reported no significant decrease in any of the target tissue weights compared to the control.  For the other 5 laboratories, a statistically significant decrease was detected in one or more target tissues.  Notably, for those laboratories in which only one target tissue was affected, the tissue was different in each of the 4 laboratories.  In only 1 of the 5 laboratories was a significant decrease detected in two of the target tissues that would be indicative of a positive response.

Thus, based on the combined results of the Hershberger Assays similar in design and methodology to the EDSP Tier 1 screening assay, androgenic activity of Chemical X was negative.  

viii.	Pubertal Female Assay
The results for the pubertal female assays are represented by some studies in which the design and methodology was comparable to the EDSP Tier 1 screening assay (OCSPP 890.1450) and for others the results are from the offspring of generational and multigenerational studies.  In a study by Kim et al. (2002) prepubertal female Sprague-Dawley rats were dosed for 20 days beginning on PND 21 to 10, 50, and 100 mg/kg of Chemical X delivered by oral gavage.  Estrogenic activity of Chemical X was indicated by a significant advance in age at vaginal opening (VO).  Irregular estrous cyclicity was characterized by increases in time to diestrus following exposure to Chemical X at 50 and 100 mg/kg/d treatment.  Ovarian weight was significantly decreased among females exposed to the high dose (100 mg/kg/d).  Thyroid effects were indicated by a dose-dependent decrease in serum T4.
In a study by Nagao et al. (2000), Sprague-Dawley rats were dosed from PND 1 to 5 with 500 mg/kg/day of Chemical X delivered subcutaneously with animals examined through puberty.  All treated females experienced irregular estrous cycles with shortened diestrus.  Increased atretic follicles and decreased corpora lutea were observed in ovaries and increased incidence of uterine histopathology was observed among treated females.  Copulation and fertilization rates among male:female pairs were significantly reduced when compared to controls, as were the number of live embryos born to dams treated with Chemical X in the early postnatal period.
In a study by Laws et al. (2000), female Long-Evans rats were dosed from PND 21 to 35 with 25, 50, and 100 mg/kg of Chemical X delivered by oral gavage.  Age at vaginal opening was significantly advanced (i.e., the age of animals at VO decreased) following treatments of 50 mg/kg and greater.  The number of estrous cycles was significantly decreased following 100 mg/kg Chemical X dosing as shown in Table 1, although the same dose failed to alter vaginal cytology of ovariectomized females following 11 days of exposure and only affected estrus cyclicity of adult, intact females following longer exposures (25 days) of 100 mg/kg of Chemical X administrated orally.
Table 1.  Adapted from Laws et al. 2000.

In a multi-generation study, Nagao et al. (2001) treated adult female Sprague-Dawley rats beginning at 13 weeks of age with 2, 10, or 50 mg/kg of Chemical X delivered by oral gavage.  Females of the F0  (parent) generation were treated from two weeks prior to pairing until necropsy (21 days after delivery of the F1 generation). First generation (F1) females were dosed indirectly through lactation and directly from weaning (PND 21) until necropsy (21 days after delivery of the F2 generation) and the F2 generation was terminated at weaning (PND 21).  Chemical X dosing was continuous from the initiation of F0 generation through necropsy of the F2 animals.  Neonatal survival (i.e., number of live F1 and F2 pups per litter) was significantly reduced in the high treatment group (50 mg/kg), although postnatal growth was not affected.  Treatment did not significantly affect estrous cycles or result in histopathological lesions in any reproductive tissues examined.  Absolute and relative uterine weights were significantly increased among females exposed to 2 mg/kg and significantly decreased in the 50 mg/kg group.  Ovarian weight was significantly decreased and vaginal opening was significantly advanced in the high treatment group, although treatment did not have an effect on estrous cycles, mating, or fertility.  On postnatal day 22, serum concentrations of LH and FSH were decreased and T3 increased.

In a multi-generation study by Chapin et al. (1999), Sprague-Dawley rats were treated with 200, 650, and 2000 ppm of Chemical X via the diet.  Calculated dose rates for young rats were 9 to 35 mg/kg/d, 30 to 100 mg/kg/d, and 100 to 350 mg/kg/d, for each of the treatment groups respectively.  Adult and PND 21 animals were necropsied.  Vaginal opening was advanced (i.e., age decreased) in all generations and uterine weights were increased in the mid and high dose groups for only the F1 generation.
Thus, based on the combined results in female pubertal and multi-generational studies, Chemical X is positive for estrogenic activity as indicated by significant changes in various estrogen-dependent endpoints.  The results on thyroid hormones was inconsistent.
ix.	Pubertal Male Assay
The results for the pubertal male assays are represented by some studies in which the design and methodology was comparable to the EDSP Tier 1 screening assay (OCSPP 890.1500) and for others the results are from the offspring of generational and multigenerational studies.  

In a study by Benjamin et al. (2003), juvenile male Sprague-Dawley rats were exposed from PND 23 through 52/53 to 100 mg/kg/d.  Chemical X significantly delayed pubertal onset, increased incidence of histopathology observed in the testis, and affected spermatogenesis in treated males.

In a study by Nagao et al. (2000), Sprague-Dawley rats were treated from PND 1 to 5 with 500 mg/kg/d of Chemical X subcutaneously and examined animals through puberty.  No significant alterations to behavior or plasma testosterone were observed and the increased time to preputial separation was not significantly altered from control.  Chemical X-treated rat body weight was significantly reduced in males (though reductions were < 10% terminal body weight), however, relative testis weight was also significantly reduced.  Chemical X exposure may have altered male fertility as indicated by gonadal histopathological examination.  There was a decrease in germ cells in the seminiferous tubules and increased degenerate germ cells in the epididymides.

In a study by Nagao et al. (2001), Sprague-Dawley male rats were treated beginning at 6 weeks of age with 2, 10, or 50 mg/kg of Chemical X delivered by oral gavage.  Males were treated for 12 weeks prior to pairing.  No adverse effects of Chemical X were observed on reproduction or sperm characteristics, and no histopathological lesions were observed in reproductive tissues.  No significant differences were observed in relative reproductive organ weight, although the absolute weight of the epididymides of the 2 mg/kg treatment group were significantly greater than controls.  Relative thyroid and pituitary gland weights were elevated among the high treatment groups.  Preputial separation and anogenital distance were not affected.  No histopathological lesions were observed in reproductive tissues; however, there was a significant increase in testosterone.  Histologically, the thyroid gland appeared enlarged in the 50 mg/kg treatment group, and there was a dose dependent increase in TSH and a significant decrease in T3 and T4 in the 2 mg/kg group.

In a study by Gong and Han (2006), adult male Sprague-Dawley rats were treated from PND 35 to 37 days with 125 and 250 mg/kg/d of Chemical X administered by oral gavage for 50 days and measured serum testosterone and luteinizing hormone..  Terminal samples indicated testosterone levels had dramatically declined in the high dose group, while LH increased in the 125 and 250 mg/kg/d dose groups.
DeJager et al. (1999a) reported on the effects of Chemical X on male fertility after direct exposure to adult Sprague-Dawley rats in an OECD One-Generation Reproductive Toxicity Study (OECD Test Guideline 415, 1983).  Adult rats (20 animals/group) were treated daily via oral gavage with cotton seed oil (control) or Chemical X at 100, 250, and 400 mg/kg beginning at 12 weeks of age and ending at 22 weeks.  The 10-week treatment period was expected to extend beyond the duration of normal spermatogenesis and epididymal transit time to cover at least one complete spermatogenic cycle (i.e., approximately 52 days).  Endpoints included body mass, testicular and epididymal mass, cauda epididymal sperm count, and testicular histology. A dose-dependent decrease in survival was observed; 0 deaths in the control, 3 deaths in the low-, 15 deaths in the mid- and 18 deaths in the high-dose groups.  There was no significant difference in body mass change over the duration of the study between males treated with Chemical X and controls as shown in Table 2.  Testicular mass and testis mass:body mass ratio was significantly reduced (36 and 18%, respectively) in the high-dose group compared to the control group. Epididymal mass and ratio was significantly reduced in the mid- (12 and 10%, respectively) and high-dose (56 and 44%) groups compared to controls.  Total caudal epididymal sperm count was significantly reduced (79%) in the high-dose group compared to controls.  Testicular histology indicated normal spermatogenesis encompassing all 14 stages in all treatment groups; however, seminiferous tubule diameters were significantly reduced (10, 14, and 24%) in the low-, mid-, and high-dose groups, respectively, compared to the control group.  Additionally, although not indicated in Table 2, seminiferous lumen diameter and epithelium thickness were significantly reduced and thickened, respectively, in all treatment groups compared to the control group.  The extreme loss of life in the mid- and high-dose groups is indicative of overt toxicity; therefore, the low-dose group was the only group to allow a reliable assessment of Chemical X on male fertility.  Direct exposure of 100 mg/kg Chemical X to adult male rats resulted in a significant decreased in weight of the testis and epididymis and number of sperm.


Table 2.  Adapted from DeJager et al., 1999a.


In another study, DeJager et al. (1999b) reported on the effects of Chemical X on male fertility after gestational, lactational and direct exposure to neonatal and post-natal Sprague-Dawley rats in an OECD One-Generation Reproductive Toxicity Study (OECD Test Guideline 415, 1983). Dams (10 animals/group) were treated daily via oral gavage with cotton seed oil (control) or Chemical X at 100, 250, and 400 mg/kg beginning at Day 7 of gestation and ending at weaning (approximately 3 weeks after birth).  After weaning, male pups (20 animals/group) were directly exposed to Chemical X by oral gavage for 10 weeks to cover at least one complete spermatogenic cycle (i.e., approximately 52 days).  Endpoints included body mass, testicular and epididymal mass, cauda epididymal sperm count, and testicular histology.












Table 3.  Adapted from DeJager et al., 1999b.


No offspring were born from dams in the high-dose group (400 mg/kg), though all other groups had 100% survival of the first generation of male pups at the end of the treatment period.  Body mass, represented by the difference in body mass from weaning to end of experimental period, was significantly decreased relative to control males among the  low- (approximately 10%) and mid- (approximately 20%) dose groups.  Testicular mass was significantly reduced in the low- and mid-dose group compared to the control group.  Epididymal mass was significantly reduced in the mid-dose groups compared to controls.  Overall analysis of the epididymal ratio was not significantly different among groups.  Total caudal epididymal sperm count was significantly reduced in the mid-dose group compared to controls. Testicular histology indicated normal spermatogenesis encompassing all 14 stages in all treatment groups. Seminiferous tubule diameters were significantly reduced in the low- and mid-dose groups compared to the control group.  Additionally, although not indicated in Table 3, seminiferous lumen diameter and epithelium thickness were significantly reduced and thickened, respectively, in the low- and mid-dose groups compared to the control group.  Indirect and direct exposure to Chemical X during gestational, lactational and pubertal/peripubertal development impaired general growth that may be considered overtly toxic at 20% in the mid-dose group. Nonetheless, the negative impact on testicular and epididymal endpoints male is indicate of the effect of Chemical X on male fertility in rats.

Thus, Chemical X is considered positive either as estrogenic or anti-androgenic in the adult male rat.  Though many endpoints altered by Chemical X in the studies described in this section are not specific to either the E or A pathway, results strongly suggest the potential of chemical X to interact in the male rat endocrine system.
x.	Fish Endocrine Relevant Studies 
Harries et al. (2000) performed a short-term (6 week) reproductive performance test with 4-month-old fathead minnow (Pimephales promelas) breeding pairs exposed to nominal treatment concentrations of 1, 10 and 100 ug/L of chemical X (4 breeding pairs in two tanks, with 2 breeding pairs per tank separated by a perforated stainless steel partition) or fresh water (3 breeding pairs) and solvent (1 breeding pair exposed to 0.2 mL/L methanol) controls.  Replicate tanks were 30 L in volume with a flow-through and input of 0.5 L/min.  Reproductive performance of each breeding pair was measured for 3 weeks prior to exposure to the test chemical and for 3 weeks during exposure to the test chemical.  Mean total number of eggs spawned per pair was significantly reduced in the highest Chemical X concentration (100 ug/L) compared to controls (Figure 12A) for the exposure period, but there were no significant differences between breeding pairs in the pre-exposure period.  

The mean number of spawnings per breeding pair (frequency of spawning) also showed a significant reduction in response to 10 and 100 ug/L Chemical X as compared to the control (Figure 12B).   The highest dose of Chemical X also produced a significant reduction in mean egg batch size per breeding pair (Figure 12C).  

Figure 12. A) Number of eggs spawned; B) number of spawnings; and C) egg batch size by pairs of fathead minnow (Pimephales promelas) exposed to Chemical X.  Data are presented as the mean number of eggs spawned +- SEM for 3 week periods prior to (open bars) and during (solid bars) exposure.  Significant differences between the two periods are shown (* P < 0.05).  Dilution water controls (Control), n=3 breeding pairs; solvent control, n=1 breeding pair; Chemical X, n=4 breeding pairs for each treatment.  Adapted from Harries et al. 2000.  

                                                                               





Figure 13. Gonadosomatic index and plasma vitellogenin concentrations in fathead minnow (Pimephales promelas) after 3 weeks exposure to Chemical X.  Data are presented as means +- SEM.  Significant differences between treatments and dilution water (*) and solvent (Ø) controls are shown (* P < 0.05).  Dilution water controls (Control), n=3 breeding pairs; solvent control, n=1 breeding pair; Chemical X, n=4 breeding pairs for each treatment.  Adapted from Harries et al. 2000.  







Male fathead minnows showed a significant dose response increase to Chemical X as compared to dilution control water (control) in VTG levels but there were no significant differences in gonadosomatic index (Figure 13). 
                                       
To summarize, Harries et al. (2000) observed a clear dose response effect of Chemical X resulting in a decreased number of spawnings and an increase in male fathead minnow VTG plasma levels compared to controls.  Alterations in these two endpoints were not seen at 1 ug/L Chemical X, but were seen at 10 and 100 ug/L Chemical X.  Also the highest dose (100 ug/L) showed a significant reduction in number of eggs spawned and mean egg batch size.

Yokota et al. (2001) performed a reproduction study over two generations (F0 and F1) using medaka (Oryzias latipes) with exposure to control, solvent control (100 uL/L ethanol), and 1.85, 5.56, 16.7, 50, and 150 ug/L Chemical X treatments.  Exposure to the parental generation (F0) was from 24 hours post fertilization (hpf) to 104 days post hatch (dph) with a flow through system providing a 14 volume/day renewal rate to 1.8 L replicate test chambers.  There were 4 replicate test chambers from initiation (24 hpf) until 70 dph and each replicate began with 15 embryos.  At 60 dph, 5 larvae per replicate were sampled for secondary sex and histological analyses.  At 70 dph, 6 phenotypic male:female mating pairs of F0 fish were established and eggs from spawning pairs were collected daily from 71 to 103 dph.  Eggs spawned on 102 and 103 dph were collected and placed in 100 mL beakers containing 80 mL of test solution.  Test solution was replaced daily until all embryos hatched and then 15 larvae per replicate (4 replicate tanks per treatment (60 larvae per treatment) with the exception of the 17.7 ug/L treatment where fewer embryos from the three pairs of F0 fish only allowed 2 replicate tanks) were transferred to the same flow through system as used for the F0 generation.  F1 fish were exposed until 60 dph when all fish were sacrificed and sampled for secondary sex and histological analyses.  

The highest treatment (183 ug/L) resulted in 100% mortality of the F0 generation by 10 dph.  The gonadal sex ratio (based on histological examination) of F0 fish in the 51.5 ug/L Chemical X treatment was significantly different from control at 60 dph.  Induction of testis-ova in the gonads of the F0 fish at 60 dph was observed in both the 17.7 and the 51.5 ug/L Chemical X concentrations (Table 4).  The induction of testis-ova in the gonads of 17.7 and the 51.5 ug/L Chemical X F0 fish at 60 dph was concurrent with significant higher mortalities (20% and 35%, respectively) as compared to the control of (Figure 14).  

Table 4.  Total length and body weight of F0 fish at 60 d posthatch and their sex ratios as determined by gross examination of secondary sex characteristics and by gonadal histology.  Adapted from Yokota et al. 2001.  



Figure 14. Post-swim-up cumulative mortality from 20 to 60 d posthatch in each Chemical X treatment (4.2  -  51.5 ug/L) of the F0 generation.  Concentrations are expressed as mean measured concentrations (ug/L).  * and ** denote significant differences from the pooled controls at p = 0.031 and 0.002, respectively.  Adapted from Yokota et al 2001.  



Because there were no phenotypic males based on secondary sex characteristics, in the 51.5 ug/L treatment, no breeding pairs were established in the reproductive phase of the experiment (71 to 103 dph).  There were no significant effects on fecundity or fertility of F0 medaka exposed to 4.2, 8.2 or 17.7 ug/L concentration of Chemical X.  Female GSI (%) was significantly increased in the 8.2 and 17.7 ug/L Chemical X treatments as compared to the control at the end of the reproductive phase (Figure 15).  
                                       


Figure 15. Gonadosomatic index (GSI) in male (A) and female (B) of paired medaka (F0) at the end of the reproductive phase.  Data were expressed as mean +- standard deviation.  The sample size in each treatment was 6, except in the 17.7 ug/L treatment (n = 3).  * and ** denote significant differences from the pooled controls at p = 0.050 and 0.002, respectively.  Adapted from Yokota et al 2001.  












In the F1 generation, the only effect observed was a significantly different gonadal histology sex ratio in the 17.7 ug/L Chemical X treatment as compared to the control (Table 5).
Table 5.  Cumulative mortality, growth, and sex ratios as determined by gross examination of secondary sex characteristics and by gonadal histology at 60 d posthatch of F1 medaka.  Adapted from Yokota et al. 2001.

In summary, Yokota et al. (2001) found no effects of Chemical X in the F0 generation below the levels which produced significant mortality (17.7 ug/L Chemical X) with the exception of an increase in female GSI (%) at 8.2 ug/L Chemical X as compared to control.  In the F1 generation, there was no treatment induced mortality, but there was a skewed gonadal histological sex ratio in the 17.7 ug/L treatment as compared to control.  
Overall, the literature reviewed in this section indicates that Chemical X has endocrine responses in fish consistent with an estrogen agonist at doses where there is no overt toxicity.  These effects consist of increased VTG in males and decreased number of spawnings observed in fathead minnows and increased female GSI and skewed sex ratios in medaka.  
2.	DISCUSSION
D.	Effects on Hypothalamic-Pituitary-Gonadal (HPG) Axis
i.	Effects on Estrogen
Table 6 summarizes the results of the studies reviewed in this case study and endpoints relevant for determining the potential for Chemical X to interact with the estrogen pathway.  Chemical X displays clear estrogenic activity in the in vitro assays examined.  The chemical binds to the estrogen receptor and increases ER-linked gene activation.  Further, data from several types of in vitro ER binding and ERTA assays suggest Chemical X's interaction is both competitive and specific.  A number of published Uterotropic assays support the estrogenic activity of Chemical X based on increased uterine weight.  Chemical X also significantly advanced puberty (age at vaginal opening), altered estrous cyclicity, and reduced ovarian weight when administered to prepubertal female rats, further supporting hypothesized ER-mediated interaction of Chemical X with the estrogen pathway.  Chemical X also resulted in altered steroidogenesis and aromatase assay results.  Chemical X inhibited aromatase enzyme activity in a dose-dependent manner in several independent studies.   These data suggest Chemical X is capable of altering hormone production along the steroidogenic pathway, however, in vivo data support the estrogenic nature of Chemical X, and this effect is apparently strong enough to obscure any potential effects Chemical X may have on steroidogenesis.  In several mammalian studies, age at vaginal opening was significantly advanced and uterine weight was increased, though this effect appeared to be most pronounced when females were exposed to Chemical X before reproductive maturity. The length of phases of the estrous cycle and total number of estrous cycles were altered by chemical exposure, though again, this effect was less pronounced and observed at higher concentrations when ovariectomized females were exposed to Chemical X in adulthood.  Reduced gonadotropin levels and fecundity and fertility among mammals support hypothesized interaction of Chemical X with the hypothalamic-pituitary-ovarian axis.

Chemical X induced treatment-related effects on fish, though the fathead minnow and medaka studies reviewed in this case study were not identical to the Tier 1 EDSP Fish Short-Term Reproduction Assay (OCSPP 890.1350), and accordingly, more difficult to infer potential estrogen pathway interactions from these results.  Exposure to Chemical X did increase VTG in male fathead minnows and reduce spawnings and total number of eggs spawned .  The incidence of phenotypic and gonadal sex mismatch was significantly greater among treated medaka than controls and F1 female medaka showed a significantly increased GSI to untreated medaka.  
Based upon WoE evaluation of available data, Chemical X interacts with the estrogen signaling pathway and HPG axis.
Table 6.  Estrogenic/Anti-Estrogenic Pathway
 
 Lines of Evidence Indicating Potential Interaction with the Estrogenic/Anti-Estrogenic Pathway[a]
                                 Study Type /
                              Literature Citation
 ER Binding
 ER Activation
 Steroidogenesis
 Uterine Weight
 Ovarian Weight  or  Fish GSI
 Ovarian/Gonad Staging and Histopathology 
 Pituitary Weight
 Estrous Cyclicity[b]
 Age & Weight at VO
 Fertility/Fecundity
 Vitellogenin
 Estradiol
 Overt Toxicity Observed[C]
                                       
                                       
                                  EDSP Tier 1
ER Binding
                                       P
                                       
                                       
                                       
                                       
                                       
                                       
                                       
                                       
                                       
                                       
                                       
                                       
ERTA
                                       
                                       P
                                       
                                       
                                       
                                       
                                       
                                       
                                       
                                       
                                       
                                       
                                       
Aromatase
                                       
                                       
                                       P
                                       
                                       
                                       
                                       
                                       
                                       
                                       
                                       
                                       
                                       
Steroidogenesis
                                       
                                       
                                       P
                                       
                                       
                                       
                                       
                                       
                                       
                                       
                                       
                                       
                                       
Uterotrophic
                                       
                                       
                                       
                                       P
                                       
                                       
                                       
                                       
                                       
                                       
                                       
                                       
                                       
Female Rat
                                       
                                       
                                       
                                      ↑
                                      ↓
                                       N
                                       -
                                       P
                                      ↓
                                      ↓
                                       
                                       -
                                       
Fish
                                       
                                       
                                       
                                       
↑
                                      ↑
                                       
                                       
                                       
                                      ↓
                                     ↑M
                                       -
                                       
a	Key to responses:  Positive (P), negative (N) or equivocal (E) observation; arrows (↓ or ↑) indicate the direction of the response; A shaded cell  indicates that parameter was not evaluated or is not applicable.
b	Factors for estrous cyclicity include:  age of first estrous, length of estrous, and percent of animals cycling regularly.
c    An "X" in this column indicates that the effect(s) observed in the assay occurred in the presence of overt toxicity. 
ii.	Effects on Androgen Pathway
Table 7 below summarizes the results of the data reviewed in this case study and endpoints relevant for determining the potential for Chemical X to interact with the androgen pathway.  Various targets of the androgen pathway are delineated so as to facilitate determination of potential for androgenic, anti-androgenic, or HPG axis effects.  

Chemical X appears to interact with the androgen receptor in several in vitro assays, though interactions appear to vary somewhat with the receptor test system used.  Chemical X represent a number of isomers that likely vary in their affinity for hormone receptors.  The difference in isomers evaluated in different studies may explain disparate results.  Chemical X significantly reduced binding of physiological ligand to the rat AR.  Several AR transcription assays indicated an antagonistic effect of Chemical X on reporter gene expression. As discussed for the estrogen pathway, Chemical X may interfere with steroidogenesis, based on significant inhibition of aromatase enzyme activity, reduction in steriodogenic enzyme transcripts other than aromatase, and a biphasic alteration in testosterone production observed in in vitro assay described in pervious sections.  

The generally antagonistic interactions of Chemical X with the androgen pathway observed in in vitro assays were not uniformly supported by in vivo study data.  Though an EPA interlab validation study did support limited antagonistic action of Chemical X on androgen-responsive tissue weights in the Hershberger assay, none of data from the 10 labs involved in the interlab trial using Chemical X produced data that met the EDSP guideline criterion for androgen antagonism.  In other Hershberger studies, Chemical X is clearly negative for both agonist and antagonist activities, and in fact, Chemical X is sometimes used as a negative control in this assay. 

Effects of Chemical X treatment on the rat androgen signaling pathway appear to vary with age of exposure.  Animals exposed in the postnatal or pre pubertal stage had significant alterations to androgen levels and/or gonadal histology.  Animals exposed after puberty appeared to be less effected, though testosterone was significantly increased along with changes to the thyroid hormone pathway, suggesting these effects may be mediated through the HPG/HPT axis.
Early postnatal and prepubertal exposure resulted in reduced copulatory behaviors and germ cells, testis weight, and in some instances increased testosterone concentrations concomitant with elevated LH.  When exposed at 6 weeks of age, Chemical X resulted in reduced testis weight but no other changes to male reproductive parameters.  Additionally, there appeared to be no effect of Chemical X in a multigenerational rat study.  

It is important to note that this case study reviews OSRI rather than EDSP Tier 1 studies validated to screen for the potential of chemicals to interact in the EAT pathways.  While many studies reviewed are similar in underlying scientific principles, they are not identical in execution and thus, sensitivity. Many results summarized in this case study may be the result of alterations in the estrogen pathway or androgen pathway.  In validation studies, treatment with 17β estradiol or ethinyl estradiol results in similar perturbation.  Chemical X appears to interact with the AR and inhibit receptor-mediated transcription, however, the absence of significant results in the Hershberger assay, considered diagnostic for the A pathway, calls into question the ability of Chemical X to interact in the androgen pathway in in vivo test systems.  Based upon WoE evaluation of available data, Chemical X may alter androgen signaling pathway and the HPG axis.

Table 7.  Androgenic/Anti-Androgenic Pathway
 
 Lines of Evidence Indicating Potential Interaction with the Androgenic/Anti-Androgenic Pathway[a]
                                 Study Type /
                              Literature Citation
 AR Binding
 Steroidogenesis
 Testosterone
 Testes Weight or Male Fish GSI
 
 Gonad Staging and Histopathology
 Epididymides Weight
 Epididymides Histopathology 
 Pituitary Weight
 Accessory Sex Organ Weights/2° Sex Characteristics
 Age and Weight at PPS
 Fecundity/fertility
 Vitellogenin
 Overt Toxicity Observed[b]
                                       
                                       
                                  EDSP Tier 1
AR Binding
                                       P
                                       
                                       
                                       
                                       
                                       
                                       
                                       
                                       
                                       
                                       
                                       
                                       
Aromatase
                                       
                                       P
                                       
                                       
                                       
                                       
                                       
                                       
                                       
                                       
                                       
                                       
                                       
Steroidogenesis
                                       
                                       P
                                       
                                       
                                       
                                       
                                       
                                       
                                       
                                       
                                       
                                       
                                       
Hershberger
                                       
                                       
                                       
                                       
                                       
                                       
                                       
                                       
                                       -
                                       
                                       
                                       
                                       
Male Rat
                                       
                                       
                                      ↓
                                      ↓
                                      ↑
                                      ↓
                                      ↑
                                      ↑
                                       -
                                       -
                                       
                                       
                                       
a	Key to responses:  Positive (P), negative (N) or equivocal (E) observation; arrows (↓ or ↑) indicate the direction of the response; A shaded cell indicates that parameter was not evaluated or is not applicable; L, M, and H refer to low, middle and high test concentration, respectively; ♂ = male; ♀ = female
b   125 mg/kg bw/day dose in supplemental chronic/carcinogenicity study


E	Effects on Hypothalamic-Pituitary-Thyroidal (HPT) Axis
i.	Thyroid Hormones 
Table 8 below summarizes the results of the data reviewed in this case study and endpoints relevant for determining the potential for Chemical X to interact with thyroid regulation. The various targets of the thyroid pathway are delineated so as to facilitate determination of potential for thyroid or HPT axis effects.

The EDSP Tier 1 battery includes male and female pubertal rat studies with thyroid endpoints and an amphibian metamorphosis assay (OCSPP 890.1100) for screening chemicals with the potential to interact with the thyroid pathway.  While Chemical X does not have any Tier 1 screening submissions, several studies discussed in the case review included endpoints that may provide anecdotal information about the potential to interact with the thyroid pathway.  No equivalent EDSP Tier 1 assays designed to evaluate potential interaction with T were available for this case study.

Nagao et al. (2000) reported increased thyroid and pituitary weights and altered plasma T3, T4 and TSH. When treated over two generations, TSH was significantly increased, along with thyroid gland weight, in adults F0 males exposed to the highest dose (50 mg/kg/d) of Chemical X, and significantly decreased in F1 PND 22 males and females exposed to low (2 mg/kg/d) and high doses.  T3 and T4 were significantly decreased in F0 adults treated with low concentrations of Chemical X, and concentrations remained low in males exposed to high dose, though T3 of females exposed to the same concentration was elevated.  In the absence of a clear-dose response or consistent relationship, these results are difficult to interpret.  It is worth noting that Chemical X displays clear and somewhat consistently non-linear dose-response effects in the E and A pathways, so the thyroid effects summarized in this case study may be a result of exposure to the chemical.  

Interactions with the T pathway are not clearly delineated by the data summarized in this case study, however the studies reviewed not include assays specifically designed to detect potential chemical interaction with the thyroid.  Because of the limited data and ambiguous results, the potential of Chemical X to interact with the thyroid hormone pathway cannot be excluded.

Table 8.  Thyroid Pathway
Lines of Evidence Indicating Potential Interaction with the Thyroid Pathway[a]
                                 Study Type /
                              Literature Citation
 Hormones (T4 and TSH)
 Pituitary Weight
 Thyroid Weight
 Thyroid  Gross and Histopathology
 Frog Development Stage
 Hind Limb Length
 Snout to Vent Length
 Overt Toxicity Observed[b]
                                 EDSP Tier 1 
Male Rat
                                ↑ (Fo adult) 
                                 ↓ (F1 juv) 
                                      ↑
                                     ↑M
                                    altered
                                       
                                       
                                       
                                       
Female Rat
                                 ↑ (F1 juv)
                                      T3
                                       -
                                       -
                                       -
                                       
                                       
                                       
                                       

a	Key to responses:  Positive (P), negative (N); arrows (↓ or ↑) indicate the direction of the response; A shaded cell indicates that parameter was not evaluated or is not applicable.
b     An "X" in this column indicates that the potential endocrine effect(s) observed in the assay occurred in the presence of overt toxicity
c	While overt toxicity was observed in this assay in the two of the three treatment groups, at the lowest treatment group, where there was an no overt toxicity observed, there were no indications that the effects observed (delayed development and growth) were due the chemical's interaction with the thyroid pathway.

3.	CONCLUSIONS
F.	Summary of Potential Endocrine Effects
Based on the studies evaluated in this WoE evaluation of Chemical X, there is strong support from in vitro and in vivo data that Chemical X has the potential to interact with the estrogen pathway.  It is important to recognize that "Chemical X" is in fact a collection of several isomers and composition of isomers is not always reported.  The degree of side chain branching may influence the affinity of Chemical X for the ER, and perhaps other potential endocrine interactions.  Chemical X appears to inhibit aromatase enzyme activity and may alter transcripts of steriodogenic enzymes upstream to estrogen synthesis.  In vitro AR binding assays and androgen receptor transactivation assays suggest Chemical X may be capable of interacting with the AR in a generally antagonistic manner.  These data were not corroborated by results of the Hershberger assay, though rat and fish in vivo assays suggest that Chemical X may act as an estrogen or antiandrogen in male animals and appears to be capable of interacting with the HPG axis.  Research summarized in this case study did not include experiments designed specifically to test for interaction in the thyroid endocrine pathway, though limited data suggest potential interaction with the mammalian thyroid system.  Though results discussed in this evaluation must be regarded carefully and do not indicate clear relationships, the potential of Chemical X with the thyroid hormone pathway cannot be discounted.

4.	REFERENCES
Ahel M. Giger W. 1993. Aqueous solubility of alkylphenols and alkylphenol polyethoxylates. Chemosphere 26:1461-1470.
Balaguer ES, P., Franois, F., Comunale, F., Fenet, H., Boussioux, A.M., Pons, M., Nicolas, J.C. Casallas, C. 1999. Reporter cell lines to study the estrogenic effects of xenoestrogens. Sci. Total Environ, 233: 47 - 56.

Blair RM, Fang H, Branham WS, Hass BS, Dial SL, Moland CL, Tong W, Shi L, Perkins R, Sheehan DM. 2000. The estrogen receptor relative binding affinities of 188 natural and xenochemicals: Structural diversity of ligands. Toxicol Sci. 54:138 -153.

Bonefeld-Jørgensen EC, Long M, Hofmeister MV, and Vinggaard AM.  2007. Endocrine-Disrupting Potential of Bisphenol A, Bisphenol A Dimethacrylate, 4-n-Nonylphenol, and 4-n-Octylphenol in Vitro: New Data and a Brief Review Environ Health Perspect. 115(S-1): 69 - 76.   
Budavari S (Ed.). 1989. The Merck Index: An encyclopedia of chemicals, drugs, and biologicals. 11th ed. Merck and Co., Inc. Rahway, NJ
Chapin RE, Delaney J, Wang Y, Lanning L, Davis B, Collins B, Minz N, Wolfe G. 1999. The effects of 4-nonylphenol in rats: A multi-generation reproduction study. Toxicol Sci. 52:80 - 91.

De Jager C, Borman MS, Van der Horst G. 1999a. The effect of p-nonylpnenol, environmental toxicant with oestrogenic properties on fertility parameters in male rats. Andrologia 31, 99 - 106.

De Jager, C, Bornman MS, Oosthuizen JM. 1999b. The effect of p-nonylphenol on the fertility potential of male rats after gestational, lactational and direct exposure. Andrologia , 31, 107 -113.

Danzo BJ. 1997. Environmental xenobiotics may disrupt normal endocrine function by interfering with the binding of physiological ligands to steroid receptors and binding proteins.  Environ Health Perspect. 105(3): 294 - 301.  

EPA. 2012. U.S. Environmental Protection Agency Scientific Integrity Policy http://www.epa.gov/osa/pdfs/epa_scientific_integrity_policy_20120115.pdf

Freyberger A, Ellinger-Ziegelbauer H, Krötlinger F. 2007. Evaluation of the rodent Hershberger bioassay: testing of coded chemicals and supplementary molecular-biological and biochemical investigations.  Toxicology. 24;239(1-2):77-88.

Gong Y, Han XD. 2006. Effect of nonylphenol on steroidogenesis of rat Leydig cells. J Environ Sci Health B. 41(5):705-15.

Harries E, Runnalls T, Hill E, Harris CA, Maddix S, Sumpter, J, Tyler CR. 2000.  Development of a reproductive performance test for endocrine disrupting chemicals using pair-breeding fathead minnows (Pimephales promelas).  Environ. Sci. Technol 34 (14), pp 3003 - 3011.

Hyun JL, Chattopadhyay S, Gong EY, Ahn RS, Lee K.  2003.  Antiandrogenic Effects of Bisphenol A and Nonylphenol on the Function of Androgen Receptor. Toxicol Sci 75(1): 40-46.

Kim HS, Shin JH, Moon HJ, Kang IH, Kim TS, Kim IY, Seok JH, Pyo MY, Han SY. 2002. Comparative estrogenic effects of p-nonylphenol by 3-day uterotrophic assay and female pubertal onset assay. Repro Tox.  16 (3): 259 - 268

Kortner TM, Vang SH, Arukwe A. 2009. Modulation of salmon ovarian steroidogenesis and growth factor responses by the xenoestrogen, 4-nonylphenol.  Chemosphere 77(7):989-98.

Laws  SC, Carey SA, Ferrell JM, Bodman GJ, Cooper RL. 2000.  Estrogenic Activity of Octylphenol, Nonylphenol, Bisphenol A and Methoxychlor in Rats .  Toxicol. Sci. 54(1): 154-167

Legler J, van den Brink CE, Brouwer A, Murk AJ, van der Saag PT, Vethaak AD, van der Burg B. 1999. Development of a stably transfected estrogen receptor-mediated luciferase reporter gene assay in the human T47D breast cancer cell line. Toxicol. Sci. 48: 55 - 66.

Nagao T, Saito Y, Usumi K, Nakagomi M, Yoshimura S, Ono H. 2000. Disruption of the reproductive system and reproductive performance by administration of nonylphenol
to newborn rats. Human Exp. Toxicol. 19: 284 - 296.

Nagao T., Wada K, Marumo H, Yoshimura S, Ono H. 2001. Reproductive effects of nonylphenol in rats after gavage administration: a two-generation study. Reprod. Toxicol.
15: 293-315.

NTP. 1997. Final Report on the reproductive toxicity of nonylphenol (CAS #84852-15-3) administered by gavage to Sprague-Dawley rats. R.O.W.Sciences 8989-30.

OECD. 2003. OECD Draft Report of the Validation of the Rat Uterotrophic Bioassay: Phase 2. Testing of Potent and Weak Oestrogen Agonists by Multiple Laboratories. ENV/JM/TG/EDTA(2003)1 http://www.epa.gov/scipoly/oscpendo/pubs/edmvac/uterotrophic_oecd_rodent_validation_p2_3_5_2003.pdf

OECD. 2007. Report of the Validation of the Rat Hershberger Assay: Phase 3: Coded Testing of Androgen Agonists, Androgen Antagonists and Negative Reference Chemicals by Multiple Laboratories. Surgical Castrate Model Protocol. Series on Testing and Assessment Number 73
http://search.oecd.org/officialdocuments/displaydocumentpdf/?cote=env/jm/mono(2007)20&doclanguage=en
Roy F. Weston Inc. 1990. Determination of the vapor pressure of 4-nonylphenol. Final Report Study No. 90-047. Roy F. Weston Inc., Environmental Fate and Effects Laboratory, 254 Welsh Pool Road, Lionville, PA. 
RTI International. 2007.  Characterization of the inhibition of aromatase activity by nonylphenol.  EPA Contract Number EP-W-06-026 EPA Task Order 3. 158pp.

Satoh K, Nagai F, Aoki N.  2001. Several environmental pollutants have binding affinities for both androgen receptorα and estrogen receptor β. J Health Sci. 47(5): 495 - 501 

Yokota H, Seki M, Maeda M, Oshima Y, Tadokoro H, Honjo T, Kobayashi K.  2001.  Life-cycle toxicity of 4-nonylphenol to medaka (Oryzias latipes).  Enviro Toxicol Chem. 20(11): 2552 - 2560. 

Ward TJ, Boeri RL. 1991. Chronic Toxicity of Nonylphenol to the Mysid, Mysidopsis bahia. Envirosystems Division, Resource Analysts. Final Report #8977-CMA to the Chemical Manufacturers Association, Washington, DC, USA

White R, Jobling S., Hoare S, Sumpter J, Parker M. 1994. Environmentally persistent alkylphenolic compounds are estrogenic.  Endo. 135(1): 175- 182.
