Effects of mixtures of phthalates and other toxicants on sexual differentiation in rats: A risk assessment framework based upon disruption of common developing systems

USEPAUSEPA scientist grapplesgrapples withwith difficultdifficult environmentalenvironmental issuesissues

L Earl Gray Jr. This presentation does not necessarily reflect USEPA policy, but rather represents the author’s current view on the state of the science Reproductive Toxicology Branch, NHEERL, ORD, USEPA

Estrogens Methoxychlor Estradiol Ethinyl A Bisphenol Toxicants Cell Germ Fetal Busulfan Diazo dyes inhibitors Steroidogenesis Prochloraz Linuron Ketoconazole Fenarimol Androgens Testosterone Trenbolone Dioxins and PCBs Dioxins Dioxin PCB 169 congener Fetal Germ Cell Toxicants Cell Germ Fetal Busulfan Diazo dyes Estrogens Methoxychlor Estradiol Ethinyl Bisphenol A inhibitors Steroidogenesis Prochloraz Linuron Ketoconazole Fenarimol Androgens Testosterone Trenbolone Dioxins and PCBs and Dioxins Dioxin congener 169 PCB DEHP DCHP DPP, DBP, DiBP BBP, DHP,DHeP DiHP, DINP Linuron Prochloraz DEHP DCHP DPP, DBP, DiBP BBP, DHP,DHeP DiHP, DINP Linuron Prochloraz

Inhibitors of fetal fetal of Inhibitors synthesis androgen of synthesis the Prevent T and hormones natural DHT and can induce malformations in male and tract reproductive male rat in puberty delay Inhibitors of fetal fetal of Inhibitors synthesis androgen of Prevent synthesis the T hormones natural and DHT and induce can malformations in male and tract reproductive male rat in puberty delay Developmental Reproductive Toxicants Reproductive Vinclozolin Procymidone Linuron Prochloraz DDE and other p,p' DDT and p,p' o,p'- metabolites Vinclozolin Procymidone Linuron Prochloraz DDEother and p,p' DDT and p,p' o,p'- metabolites

AR Antagonists natural with Compete hormones T and DHT AR, for prevent AR-DNA in binding AR-dependent inhibit vitro, vivo, and in gene expression in may malformations induce and tract male reproductive male rat in puberty delay AR Antagonists Compete with natural hormones T andDHT for AR, prevent AR-DNA binding in AR-dependent inhibit vitro, vivo, and in gene expression in malformations induce may and tract male reproductive rat male in puberty delay •In Utero exposure and measure Hormone dependent endpoints male rat offspring as adults •Anogenital distance at birth •Nipple/ areolar numbers in infants •Reproductive Malformations •Undescended testes •Gubernacular abnormalities •Epididymal agenesis •Ventral prostate agenesis •Seminal vesicle agenesis •Vas deferens agenesis •Nipples •Hypospadias •Vaginal pouch •Reproductive Organ Weights •Glans penis •Ventral prostate •Seminal vesicle •Testes •Epididymides •Levator ani bulbocavernosus •Cowper’s glands •Testis and epididymal histopathology F1 male rat offspring: Reproductive endpoints sensitive to lower dosage levels of phthalates in utero. Critical “adverse effects”: % with Phthalate Syndrome and Fetal androgen production. DEHP DR Gray et al 2009 No Hypopadias 100 Testis weight Epididymal weight Seminal vesicle weight 80 Testis/EPI malfs NOT AGD 2 60 % no NIPS 13 Fetal t Production KLH % no PhSyn

Percent of control of Percent 40

0 0 0 50 00 5 50 0 1 1 200 2 3 DEHP mg/kg/d PE Dose Response reproductive effects on F1 Male rat offspring indicate that DEHP, DBP, DiBP, BBP and Dihexyl phthalate are equipotent. DPP is more potent and DINP is less potent than this group of PEs

TESTIS MALFS OR HISTOPATHOLOGY Testis weights are reduced by PE in utero treatments. Logistic regressin with shared slope 100 DEHP NTP 80 DBP WINE DEHP NTP BBP NAGAO 100 DBP WINE 60 BBP TYL BBP NAGAO BBP TYL DIBP SAILLENFAIT 80 40 DnHexyl P SAILLENFAIT DIBP SAILLENFAIT DnHexyl P SAILLENFAIT 20 BBP ASO DEHP GRAY 60 BBP ASO 0 DBP MYL 1999 DEHP GRAY Percent Affected

10 100 1000 control of Percent 40 DBP MYL 1999 mg/kg/d PE 10 100 1000 mg/kg/d PE

Epididymal Weight Reduction Seminal Vesicle Weight Reduction

100 DEHP NTP 110 DEHP NTP BBP NAGAO DBP WINE BBP TYL 100 BBP NAGAO 80 DIBP SAILLENFAIT 90 BBP TYL DnHexyl P SAILLENFAIT DIBP SAILLENFAIT 80 60 BBP ASO DnHexyl P SAILLENFAIT DEHP GRAY 70 BBP ASO DBP MYL 1999 60 DEHP GRAY Percent of Control 40 Affected Percent DBP MYL 1999 10 100 1000 10 100 1000 mg/kg/d PE mg/kg/d PE Testis/EPI malfs data pooled studies Adult male F1 epididymal weight data pooled from several rat studies 100

80 LogEC50 2.617 100 HillSlope 2.572 60 EC50 414.0 80 40 Bottom = 0.0 Top = 100.0 20 LogEC50 2.984 60 HillSlope -1.805 Percent Affected Percent 0 EC50 964.7

0.01 0.1 1 10 100 100010000 Percent of control 40 mg/kg/d PE 10 100 1000 mg/kg/d PE

Adult male F1 rat testis weight data Adult male F1 seminal weight data pooled from several rat studies pooled from several rat studies

100 100

80 80 Bottom = 0.0 Bottom = 0.0 Top = 100.0 Top = 100.0 60 LogEC50 2.971 60 LogEC50 3.068 HillSlope -2.438 HillSlope -1.740 EC50 935.1 EC50 1170

Percent control of 40 40 10 100 1000 10 100 1000 mg/kg/d PE mg/kg/d PE Fetal endocrine changes demonstrate the relative potency (RP) of individual PEs is consistent with their RP to induce reproductive tract alterations seen in postnatal life. Thus, changes in fetal androgen levels can be used to “screen” phthalates for their potential to induce malformations and possibly as a “critical effect” in individual and cumulative risk assessments.  Fetal Phthalate screen (FPS)  Dose dam with PE at a single high dosage group  Gestational exposure during critical period of sexual differentiation  On Gestational Day 18 necropsy dam  Measure testis fetal testis Testosterone (T) production  Testis gene expression  STAR  Insl3  Cyp11a  For “Positives”  Run dose response studies  Use potency factors to execute mixture studies  For “important” PEs with data gaps run a postnatal study FPSFPS ResultsResults toto datedate

 Positives  Negatives  DINP – both CAS #s *  Diethyl phthalate  Diheptyl phthalate  Dioctyl terephthalate  Diisoheptyl phthalate  Hexamoll DINCH  Dihexyl phthalate  Di-2-ethylhexyl  Dibutyl phthalate * tetrabromo phthalate  Diisobutyl phthalate *  Benzylbutyl phthalate  Future studies  Dicyclohexyl phthalate  DMP, Diproply P,  Dipentyl phthalate * DIDP, DPHP, others

* FPS DOSE RESPONSE STUDIES ONGOING Fetal T production is more sensitive to Phthlate ester (PE) disruption than are testis genes of postnatal effects (epididymal malformations) and T prod is less variable resulting in lower NOELs and BMDs. Below is a generic example.

PE IN UTERO EFFECTS ON FETAL AND POSTNATAL MALE RATS

120 100 Endpoint ED50 80 T Prod 231 T Production Extracted Testis T 275 60 Extracted Testis T STAR mRNA 280 40 STAR mRNA insl3 mRNA 372 Insl3 mRNA CYP 11a 431 20 CYP11a Epididymal agenesis 567 NORMAL EPI

Percent of control of Percent 0 10 100 1000 In utero mixture studies Summary

 Similar cellular and molecular mechanisms of action  Binary mixtures – pairs of chemicals  Mixture of 5 phthalates

 Diverse cellular and molecular mechanisms of action  Binary mixtures – pairs of chemicals  Mixture of 7 chemicals including pesticides and phthalates  Mixture of 10 chemicals Initial Step in a Cumulative Risk Assessment as outlined by the USEPA and other regulator agencies.

 Identification of a group of chemicals to be included in a Common Mechanism Group  chemicals that induce a common toxic effect by a common mechanism of toxicity. Key Question:  Should chemicals that disrupt differentiation of the same reproductive tissue but by different molecular mechanisms of action in different tissues be included in a Common Mechanism Group?  The default answer has been:  NO because such a mixture would not be expected to produce adverse effects if each chemical is administered below the NOAEL Response addition (RA)  0 + 0……+ 0 = No Adverse Effects

 However if this assumption is incorrect Dose addition (DA)  0 + 0 +…….+ 0 = 100% Malformations ObjectivesObjectives ofof ourour researchresearch

 Determine how chemicals with similar and dissimilar mechanisms of toxicity interact during sexual differentiation

 To provide a framework for deciding what chemicals to include in a cumulative risk assessment

 Working Hypothesis: Chemicals that disrupt the development of a common reproductive tissue/system during sexual differentiation will produce dose additive responses, regardless of the molecular mechanism or the signaling pathway that is disrupted Cumulative effects:

Common mechanisms of toxicity AR antagonist mixture study

100 •Vinclozolin and Procymidone 95.8 •We90 nowHypospadias: know that – both 10+0=96% fungicides

80 Vaginal Pouch: 0+0=54% •Bind AR and antagonize androgen action in vitro 70 Hypospadias Vaginal Pouch 60•Inhibit androgen-stimulated tissue growth in vivo 54.2 50•Induce a common phenotype of male rat 40reproductive tract malformations

30 •Alter the same array of genes in the reproductive 20tract 10.3 10 0 0000

Percent of males aaffected Percent of males 0 Control Vinclozolin Procymidone Combination Cumulative effects of Phthalates A Common Mechanism of Toxicity: Altered fetal Leydig cell differentiation and reduced hormone synthesis DBP plus BBP mixture study. PhthalatesHypospadias: with a0+0=50% common Epididymalmetabolite agenesis: 9+23=100%

Hypospadias DBP plus DEHP mixture study 100 100 100 Epididymal Agenesis phthalates with no common Gubernacular agenesis 80 metabolite (Howdeshell et al 2007) 60 50 100 90 40 80 Hypospadias 75 23 Epididymal Agenesis 20 70 65 Gubernacular agenesis Percent of males aaffected Percent of 9 7.7 60 0 0000 00 50 Control BBP DBP Combination 40 30 26 27 25 20

Percent of males aaffected 10 3 3 4.7 Hypospadias: 3+0=25% 0 0 0 0 0 Epididymal agenesis: 26+5=65% Control DEHP DBP Combination Common mechanism of Toxicity Effects of a mixture of five phthalates on fetal testosterone production in the rat (Howdeshell et al, 2008)  Individual studies  DBP, DiBP, BBP, DEHP, DPP administered at several dosage levels on GD 8-18. Fetal testis collected on GD 18 and fetal T production measured.  Mixture study  Five phthalates were administered as a mixture and fetal T production measured on GD 18.  Mixture ratio designed so that each phthalate would contribute equally to reduction in fetal T production if the mixture behaved in a dose- additive manner. Comparison of the contribution of each of the five individual phthalates with the

observed effect of the mixture and the100 Observed effect predicted from dose addition80 DA RA modeling 60 40

100 of incidence % 20 0 80 malformations epididymal 100 1000 Total phthalate dose (mg/kg/d) 60 40 100 Observed

(% control) (% Mixture of 5 Phthalates: 80 DA Testosterone 20 Observed T reduction 60 RA Mixture of 5 phthalates: 0 Dose40 Addition Prediction

0 20 40 60 80 100 SV agenesis % incidence of incidence % 20

Percent of Top dose 0 100 1000 Total phthalate dose (mg/kg/d) Top dose=300 mg/kg/d four PEs except DPP at 100 mg/kg/d Unexpected high incidence of reproductive tract malformations in female rat offspring in the 5 phthalate mixture study

Uterine and vaginal agenesis

Shown here – uterus unicornis with agenesis of the lower vaginal canal and vaginal opening Cumulative effects:

Diverse mechanisms of toxicity that disrupt the same signaling pathway

Disruption of the Androgen signaling pathway in fetal tissues Binary mixtures studies with chemicals that act via different mechanisms

 The chemical pairs include:  1) a phthalate ester plus a herbicide that has dual modes of action (linuron - 75 mg/kg/d and BBP 500 mg/kg/d)(Hotchkiss et al. 2004).

 2) a phthalate ester plus an AR antagonist (DBP 500 mg/kg/d and procymidone 50 mg/kg/d)  Two studies

 SD rats were dosed on GD 14-18 with chemicals singly or in pairs at dosage levels equivalent to about one half of the effective dose which causes a 50% incidence (ED50) of hypospadias and/or epididymal agenesis. Binary mixture studies with a phthalate plus a pesticide

Linuron plus BBP mixture study. A pesticide AR Antagonist plus Phthalate mixture that is an AR antagonist and inhibits study (Hotchkiss et al, 2010) testosterone synthesis plus a phthalate (Hotchkiss et al. 2004) 60 Hypospadias Epididymal 100 97 50 Vaginal Pouch Agenesis 48.8 Hypospadias 80 Vaginal Pouch 40 Epididymal Agenesis 63 60 56 28.9 30 26.7 24.1 40 40 20

20 12 10 Percent of males aaffected Percent of males aaffected 0000000 1.5 0 0 0 000000 Control BBP Linuron Combination Control Procymidone DBP Combination

Hypospadias: 0+0=56% Hypospadias: 1.5+0=49% Disrupting the AR Pathway by Multiple mechanisms of toxicity. IN UTERO EXPOSURE TO THE FUNGICIDE PROCYMIDONE AND DIBUTYL PHTHALATE Hotchkiss et al, Reproductive Toxicology 2010.

 Timed-pregnant Sprague-Dawley dams (n=4- 10/dosage group) were gavaged daily from gestational day 14-18 with a mixture at 100, 83, 67, 50, 33, 17, 8, 4, or 0% of the top dose.

 The top dose of the mixture contained PRO at 150 mg/kg/d and DBP at 1125 mg/kg/d and was expected to induce 100% incidence of malformations.

 IN UTERO EXPOSURE TO THE FUNGICIDE PROCYMIDONE AND DIBUTYL PHTHALATE PRODUCE DOSE-ADDITIVE DISRUPTIONS OF MALE RAT SEXUAL DIFFERENTIATION Hypospadias  In addition, we observed a dose- related120 increase in testicular OBSERVED tumors100 with the mixture. DA 80 RA  Similar60 tumors have been seen with40 DBP at a lower incidence but Percent not20 with procymidone 0 0 20 40 60 80 100 % of top dose of the mixture “MegaMix1” Study: 7 antiandrogens Rider et al. 2008. Int J Androl

 Pregnant rats were dosed from GD 14-18 and male offspring were examined assessed for effects through adult life.

 The “high dose” group, termed the ED100, included the 7 chemicals, each at 1/7th their ED100 for malformations  vinclozolin 15 mg/kg/d, procymidone 15 mg/kg/d, prochloraz 35 mg/kg/d, linuron 20 mg/kg/d, and BBP, DBP and DEHP at 150 mg/kg/d per phthalate

 Doses: ED100 and 75%, 50% and 25% of the ED100 Hypospadias Epididymal Agenesis 100 80 Observed 90 Observed 70 80 DA 70 DA 60 60 IA 50 IA 50 40 40 30 RA 30 RA 20 20 10 10 Percent Affected Percent 0 Affected Percent 0 0 0 0 25 50 75 25 50 75 10 100 Undescended Testes Seminal vesicle weight 80 70 Observed Observed 70 60 60 DA 50 DA 50 IA 40 IA 40 Percent of Top Dose of the Mixture30 RA RA 30 20 20 10 10 0 Percent Affected 0 Reduced Percent 0 25 50 75 0 0 100 25 50 75 10 Values are no. affected/total no. Not litter means by Cynthia Disrupting the AR Pathway by Multiple mechanisms of toxicity. Ten chemical mixture study. Rider et al 2010

Ten “antiandrogenic” chemicals were administered orally to pregnant rats on gestational days 14-18 and the reproductive development of the male offspring was evaluated.

Data were analyzed to determine if the chemicals behaved in a response-, integrated or dose-additive fashion. “MegaMix 2” Study: 10 antiandrogens

 The “high dose” group, termed the ED100, included  vinclozolin (30 mg/kg/d)  procymidone (30)  prochloraz (60)  linuron (40)  Six phthalates: DPP (50) and BBP, DBP, DiBP, DiHP and DEHP (150 per phthalate)

 Doses: 100%, 80, 60,40, 20, 10 and 0% of the top dose Malformations in F1 male rats: Megamix 2 Hypospadias Obs vs predicted

100 OBS DA 80 IA 60 RA 40 20 0 Percent Affected 10 100 Percent of Top Dose

Epididymal Agenesis Undescended testes

100 OBS 100 OBS 80 DA 80 DA RA RA 60 IA 60 IA 40 40 20 20

0 affected Percent 0

Percent Affected 10 100 10 100 Percent of Top Dose Percent of Top dose Organ weights from F1 male rats: Megamix 2 data

SeminalTenTen vesicle wt obs chemicalchemical vs predicted mixturemixtureEpididymal WTstudystudy obs vs predictions

100 OBS 100 OBS DA DA 75 IA 75 IA RA RA 50 50

25 25 Percent of Control of Percent Percent of Control of Percent 0 0 10 100 10 100 Percent of Top dose Percent of Top Dose

Ventral Prostate weight LABC wt obs vs predicted

OBS 100 OBS 100 DA 80 DA IA 75 IA 60 RA RA 50 40 20 25 Percent of Control 0 Percent of Control 0 1 10 100 10 100 Percent of Top dose Percent of Top Dose Summary of results on the AR Pathway

 DoseDose AdditionAddition isis thethe mostmost logicallogical modelmodel forfor thethe datadata  Response addition does not explain the results  DA is consistent with the biology of hormone action  Phthalates, vinclozolin, procymidone, linuron and prochloraz all act on the fetal tissues by disrupting a “Common Pathway”  What the tissues “see” is a reduction in AR bound to an androgen so in both cases androgen- dependent gene expression is attenuated  AR antagonists do this by preventing T from binding AR  T synthesis inhibitors do this by reducing T levels  The disrupted developing tissue does not distinguish among these two events. Disrupting Multiple Pathways in common tissues:

Altered sexual differentiation through disruption of AhR and AR signaling pathways

 Dibutyl phthalate (DBP) disrupts male development by decreasing testosterone production by the testes and decreasing expression of insl3 (a protein involved in testicular descent).

 2,3,7,8-Tetrachlorodibenzo-p-dioxin (TCDD) disrupts male reproductive tract development through an unknown mechanism of action that apparently does not involve the androgen signaling pathway. Disrupting Multiple Pathways in common tissues:

Altered sexual differentiation through disruption of AhR and AR signaling pathways

 Dibutyl phthalate (DBP) disrupts male development by decreasing testosterone production by the testes and decreasing expression of insl3 (a protein involved in testicular descent).

 2,3,7,8-Tetrachlorodibenzo-p-dioxin (TCDD) disrupts male reproductive tract development through an unknown mechanism of action that apparently does not involve the androgen signaling pathway. DBP+TCDD malformations and organ weight reductions that exceeded Response Addition predictions

Epididymal and Testicular Agenesis of the Malformations Rider et al 2010 Epididymal Weight Vas deferens 80 60 1500

60 1250 40 RA 40 1000 20 20 % Malformed % Malformed RA Weight (mg) 750 RA

0 500 0 l o ol P D h h BP r B D ow ig BP g ntr D x low high nt D l h ntrol D CDD o TCDD i o TC ix x o T C M ix C C Mix low M M Mi Mix hi

Paired Testis Weight Malformed External genitalia Epididymal Sperm Numbers 4.0 200 40 35 150 3.5 30 25 RA RA 100 3.0 20 15 (millions)

50 (mg) Weight 2.5 10 Sperm count Sperm 5 % Malformed 0 2.0 0 RA h l BP ow ol h o 2 00 20 00 l ig r ow ig tr D 5 3 5 D x h l n D P P P ontrol TCDD i x nt DBP x o C B B B C M i o TCDD i C T D D D M C 3 2 M Mix h 1. D DD CD TC T Cumulative risk assessment using a Framework based upon Disruption of a Common System/developing tissue

•Cumulative Risk assessments would be conducted on all the chemicals that disrupted common reproductive tissues

•Differentiation of androgen-dependent tissues depends upon critical interactions of dynamic interconnnected pathways.

•All the chemicals that affect the same tissue would be considered in a single cumulative risk assessment and the effects of a mixture would be predicted using the relative potencies on a tissue-by-tissue basis. The “Team” Dr. L. Earl Gray Jr. USEPA Dr. Vickie Wilson. USEPA Dr. Phillip Hartig. USEPA Johnathan Furr. USEPA Dr. Andrew Hotchkiss USEPA Christy Lambright USEPA Mary Cardon USEPA

Dr. Cynthia Rider Duke University

Dr. Jerry LeBlanc North Carolina State University

Dr Paul Foster NIEHS Dr. Chad Blystone NIEHS Dr. Kembra Howdeshell NIEHS

Supported in part by an NTP, NIEHS/EPA Interagency Cooperative Research Agreement

HHS Y1-ES-8014-01; EPA RW75922855-01-0. “AGD is forever”