This study demonstrates that DPeP reduces fetal testicular T production, StAR, Cyp11a, and insl3 gene expression levels and induces early postnatal reproductive alterations in male offspring. The potency of DPeP for inducing these changes is greater than that of previously examined antiandrogenic PEs. Potency results in the fetal male from the current study are consistent with results from an early study (Foster et al., 1980
), indicating that DPeP was the most potent phthalate examined in the pubertal rat model for inducing testicular injury after a 5-day dosing period. We also demonstrated testicular effects as early as 5 h following dosing, consistent with effects reported by Foster et al. (1982)
. For these reasons, we are using DPeP as a tool to study the earliest effects of PEs on fetal endocrinology and gene expression in more detail. DPeP is one of several phthalate compounds included in the U.S.EPA OCSPP’s action plan and is currently on the agenda for inclusion in a cumulative risk assessment on the phthalates (U.S.EPA, 2009
). Therefore, we generated dose-response data to fill a data gap for DPeP, which would be necessary for including this compound in an antiandrogen compound cumulative risk assessment. Additionally, data generated by this study further support the hypothesis that fetal testicular T production is a more sensitive endpoint for the antiandrogenic action of phthalate compounds than are genomic and early postnatal endpoints. T production may therefore be the most appropriate critical effect to consider in the risk assessment process.
In both humans and rodents, testicular androgen production begins during the sexual differentiation period and is partially responsible for masculinization of male reproductive features including differentiation of the Wolffian ducts into the epididymis, lengthening of the AGD, and formation of the vas deferens, seminal vesicles, and external genitalia. Several rodent studies have demonstrated that phthalate exposure throughout this critical period for male development inhibits testicular androgen production (Howdeshell et al., 2008
; Lehmann et al., 2004
; Parks et al., 2000
; Shultz et al., 2001
) and thereby interferes with organizational development of these features (Barlow and Foster, 2003
; Foster, 2006
). Blocking androgen-driven masculinization during this critical programming period (GD 14–18) has more impact on reproductive tract differentiation than does blocking androgen action on later morphological differentiation (after GD 18) (Welsh et al., 2008
) with antiandrogen treatments on GD 16–18 being more effective than earlier or later in gestation (Wolf et al., 2000
; Carruthers and Foster 2005
). However, the mechanism behind phthalate-induced disruption of T production within this window is unclear. Existing evidence of the mechanism driving phthalate-induced hormone decline relates to exposure-induced abnormal Leydig cell aggregation (Mahood et al., 2005
; Mylchreest et al., 2002
) and declines in gene expression of insl3
(Wilson et al., 2004
) and several steroid synthesis-related genes (Barlow et al., 2003
; Howdeshell et al., 2008
; Johnson et al., 2007
; Lehmann et al., 2004
; Liu et al., 2005
; Shultz et al., 2001
). Because most studies have examined the effects of PEs on androgen levels late in gestation (GD 18–21), information about the earliest pathways affected by phthalates is generally lacking. In the current study, we determined that the decline in T production can be acutely induced with 5-h exposure to a high dose of DPeP on GD 17. Ongoing efforts to pinpoint the earliest time point for phthalate disruption of T production within the critical programming window are important for further investigations related to the mechanism of action for the organizational toxicity of phthalates. A comparison of DPeP gene expression dose-response data for Cyp11a, StAR, and insl3 generated in this study to similar data for DEHP recently generated in our laboratory indicates that DPeP reduces expression of these genes with approximately three- to sixfold greater potency (Hannas, Howdeshell, Lambright, Furr, Gray, and Wilson, unpublished data). Therefore, DPeP is likely a good model phthalate for performing mechanistic studies at the molecular level that require robust responses. A comparison of the ED50s derived from the regression analyses for DPeP 5-day dosing and DEHP 10-day (GD 8–18) dosing data (Howdeshell et al., 2008
) demonstrates a nearly eightfold greater potency for the ability of DPeP to decrease fetal testicular T production on GD 18 as compared with DEHP (). Comparison of the ED50s for postnatal effects indicates that DPeP is approximately twofold more potent than DEHP for reducing AGD and 4.5-fold more potent for inducing male nipple retention (Gray et al., 2009
). It should be noted that the current study was conducted with Harlan SD rats, whereas the studies by Gray et al. (2009)
and Howdeshell et al. (2008)
used SD rats from Charles River Laboratory. We conducted a study in the Harlan SD rat in which DEHP treatment of pregnant dams resulted in comparable fetal testicular T production and postnatal AGD and nipple retention outcomes to the Charles River SD (unpublished data). Therefore, it is unlikely that the increased potency of DPeP detected in the Harlan SD is solely driven by different sensitivity of the Harlan and Charles River rats to phthalate exposure.
The consistency in DPeP potency from fetal endpoints to postnatal effects supports the hypothesis that fetal declines in androgen production are causally linked to postnatal malformations in androgen-dependent tissues. We were able to accurately predict the relative magnitude of early postnatal effects (reduced AGD and nipple retention) based on the potency of DPeP for reducing fetal testicular T production relative to DEHP. Consequently, this study lends support to the notion of using fetal testicular T production decline as the critical effect for setting reference dose values in the risk assessment process.
The data collected in this study suggest that the reduction in T production would be a more sensitive critical endpoint in phthalate risk assessment than changes in gene expression data. In the U.S.EPA draft assessment of DBP, use of fetal T production as the critical effect endpoint (U.S.EPA, 2006
) fostered criticism claiming that gene endpoints should alternatively have been considered as the critical effects in the DBP phthalate risk assessment because they are disrupted at lower doses of DBP than T production and would therefore result in lower NOAEL or benchmark dose levels (Janssen, 2006
). In the current study, however, the resulting order of endpoint sensitivity to in utero
DPeP exposure was T production = StAR expression > insl3 expression > Cyp11a expression > retained nipples > AGD. A reduction in T production integrates the smaller individual reductions of the steroidogenic enzyme gene expression. In addition, T production typically is less variable than changes in fetal testis gene expression, and as a consequence, the NOAEL value for T production is usually below those for StAR, insl3, and Cyp11a mRNA. These results agree with the order of endpoint sensitivity identified in dose-response studies involving DBP (Howdeshell et al., 2010
in prep), BBP (Howdeshell et al., 2008
), and DEHP (Hannas, Howdeshell, Lambright, Furr, Gray, and Wilson, unpublished data). Currently, we are measuring gene expression levels for these genes, and several others, using a custom PCR array platform with 96 genes per plate to determine if this method provides more accurate and precise results than the methods used herein to determine gene expression levels.
AGD has become a widely accepted endpoint to identify the antiandrogenic activity of a compound administered during the sexual differentiation period (Foster et al., 2000
, Gray et al., 2000
, Nagao et al., 2000
, Parks et al., 2000
, Tyl et al., 2004
). Normally, androgens secreted by fetal testis during development lengthen the AGD in males compared with females. However, phthalate disruption of testicular androgen production during sexual differentiation in males results in decreased AGD on PND 1–2. As seen with the antiandrogenic compounds linuron, flutamide, and finasteride and the phthalates DEHP and DBP, this early decline in AGD length relative to untreated rats can be permanent through adulthood (Barlow and Foster 2003
; Bowman et al., 2003
; Hotchkiss et al., 2004
; McIntyre et al., 2001
) and males with shorter AGDs and retained female-like nipples have a higher probability of displaying severe reproductive tract malformations than do males with normal AGD lengths. Similarly, disruption of androgen action by phthalates during the male masculinization process results in retention of nipples through adulthood (Barlow and Foster 2003
; Gray et al., 1999
; Hotchkiss et al., 2004
). In the current study, a reduction in AGD on PND 2 and nipple retention on PND 13 were seen when fetal T production was reduced by approximately 80% on GD 18. It should be noted that use of T production data as a point of departure for predicting postnatal malformations would require further investigation with additional phthalate compounds, doses, and litters. However, results of the current study indicate that the fetal and postnatal effects of DPeP exposure evaluated in this study are causally linked and levels of reduced fetal T production are predictive of adverse postnatal outcomes. All male offspring from the DPeP GD 8–18 exposures performed in this study will be assessed further in the future for later postnatal reproductive malformations including delay in preputial separation, reduction in androgen-dependent organ weights, and histopathology of the testes and epididymides. These assessments will provide further information for making potency comparisons to other phthalates and correlations with T production levels.
In the current study, the ED50 for decline in pup survival to PND 2 following in utero
DPeP exposure was similar to that for inducing AGD decline. Offspring mortality has not been detected previously at a comparable dose (which would induce postnatal malformations) with other PEs (Foster, 2006
; Gray et al., 2000
; Howdeshell et al., 2007b
). Howdeshell et al. (2008)
demonstrated that maternal oral doses of 300, 600, and 900 mg/kg/day DPeP administered for 10 days (GD 8–18) resulted in complete loss of litters. Clearly, DPeP is also acting simultaneously through other mechanisms beyond antiandrogen toxicity to induce overt offspring toxicity.
Although very little toxicity or exposure data are available for DPeP, this phthalate consistently ranks highest in potency of tested phthalates for inducing male reproductive toxicity (Benson, 2009
; Foster et al., 1980
; Howdeshell et al., 2008
). Although current levels of usage, production, and import of DPeP in the United States are unknown (U.S.EPA, 2009
), Silva et al. (2010)
of the Center for Disease Control recently reported that mono(4-hydroxypentyl) phthalate (a urinary biomarker of DPeP exposure) was detected above the LOD in 29% of the samples of human urine at concentrations ranging from < LOD to 8 ng/ml. The EPA action plan for phthalate risk assessment states that the agency may consider a requirement for manufacturers and processors of DPeP to notify the agency before taking action. Based on the results provided by the current study and previous studies characterizing DPeP as the most potent phthalate for inducing male reproductive effects, it would be prudent for the toxicity of this particular phthalate to be considered before increases in production and usage are approved to assure that usage would not result in exposure to potentially hazardous levels of DPeP. The toxicity threat posed by the increased potency of this phthalate in comparison to other antiandrogenic phthalates was further demonstrated in a cumulative phthalate toxicity study. Howdeshell et al. (2008)
determined that DPeP contributes to dose-additive phthalate mixture toxicity and also carried the largest burden within the mixture when compared with other phthalates within the mixture. Accordingly, the hazard associated with exposure to this compound is greater than for other phthalates that were examined.
Finally, this study was designed to characterize the dose-response relationships between DPeP exposure and fetal and neonatal androgen-dependent endpoints and to identify the ED50 for each endpoint to determine the relative potency of DPeP. In doing so, we determined that a smaller sample size (3–6 litters in most cases) was appropriate for obtaining a robust sigmoidal fit for the T production dose-response, with small variability between litters at each dose for most endpoints. A larger sample size or more precise methodology could improve the strength of the gene expression data, as these endpoints tend to vary more between litters than does T production. It should be noted that whereas the design of the current study produced data that would likely be useful for risk assessment, an alternative design that incorporates more litters per dose group would be better suited for determination of NOAELs. Alternatively, a benchmark dose analysis (BMD) would likely be more appropriate for the data we report in this study for risk assessment than NOAEL determinations because BMD analyses rely more heavily on the shape of the dose-response curve than statistical determination of the lowest dose that differs from the control group.
The results of the current study are noteworthy for three main reasons. First, we provide critical data for DPeP that are needed by the U.S.EPA and the Consumer Products Safety Commission to be able to appropriately assess the toxicity hazard associated with this compound. Second, we provide evidence supporting the use of T production as critical endpoint for risk assessment. And third, we have identified an antiandrogenic phthalate that can serve as a strong model for mechanistic investigations. The high potency and strong link between fetal and postnatal antiandrogenic endpoints associated with DPeP will facilitate future investigations related to the proximate mechanism of phthalate toxicity. Rodent studies focusing on these investigations would be applicable to humans because of a highly conserved androgen pathway, which is crucial to healthy reproductive development during sexual differentiation in both species. Overall, the results of the current study with DPeP have the potential to significantly advance our understanding of phthalate risk to human male reproductive health.