Peroxisome Proliferator-Activated Receptors: Media
http://www.100md.com
《毒物学科学杂志》
ABSTRACT
Many phthalate ester plasticizers are classified as peroxisome proliferators (PP), a large group of industrial and pharmaceutical chemicals. Like PP, exposure to some phthalates increases hepatocyte peroxisome and cellular proliferation, as well as the incidence of hepatocellular adenomas in mice and rats. Most effects of PP are mediated by three nuclear receptors called peroxisome proliferator-activated receptors (PPAR,?,). An obligate role for PPAR in PP-induced events leading to liver cancer is well-established. Exposure of rats in utero or in the neonate to a subset of phthalate esters causes profound, sometimes irreversible malformations in the male reproductive tract. We review here the data that supports or discounts roles for PPARs in phthalate-induced testis toxicity including (1) toxic effects of phthalates on the male reproductive tract, (2) expression of PPARs in the testis, (3) activation of PPARs by phthalates, (4) role of PPAR in testis toxicity, (5) gene targets of phthalates involved in steroid biosynthesis and catabolism, and (6) interactions between PPARs and other nuclear receptors that play roles in testis development and homeostasis. Critical research needs are identified that will help determine the significance of PPARs in phthalate-induced effects in the rat male reproductive tract and the relevance of toxicity to humans.
Key Words: phthalate ester; PPAR; testis; peroxisome proliferator; male reproductive tract.
INTRODUCTION
Most animal cells contain peroxisomes, subcellular organelles that perform diverse metabolic functions including H2O2-derived respiration, ?-oxidation of fatty acids, and cholesterol metabolism (Lock et al., 1989). Peroxisome proliferators (PP) are a large group of structurally dissimilar industrial and pharmaceutical chemicals that were identified as inducers of both the size and number of peroxisomes after in vivo exposure. PP include a large number of phthalate ester plasticizers which are, through their widespread use in various industries, widely distributed in the environment (Staples et al., 1997). Unlike many PP, phthalates are metabolized to active species by esterases found in the gut and other tissues. These esterases cleave one of the two side chains from the parent diester phthalate producing an active monoester phthalate. Rodent exposure to PP, including many phthalate esters leads to predictable adaptations in the liver consisting of hepatocellular hypertrophy and hyperplasia and transcriptional induction of fatty acid metabolizing enzymes (Lock et al., 1989). Chronic exposure to some PP causes an increased incidence of liver tumors in male and female mice and rats (Klaunig et al., 2003). Many studies also show that exposure to some phthalates results in profound, irreversible changes in the development of the male reproductive tract (Foster et al., 2001; Sharpe, 2001).
Many of the adaptive consequences of PP exposure are mediated by a subset of nuclear receptor superfamily members called peroxisome proliferator-activated receptors (PPARs). The PPAR family includes three distinct subtypes, PPAR, PPAR? (also known as NUC1 or PPAR), and PPAR encoded by separate genes (Lemberger et al., 1996). PP activate PPARs leading to altered expression of genes with roles in metabolism, cell growth, and stress responses (Corton et al., 2000). Most, if not all responses to PP including some phthalates in the liver are dependant on PPAR (Klaunig et al., 2003). Given recent findings describing novel aspects of phthalate-PPAR interactions, we review here evidence supporting or discounting roles for PPARs in mediating the toxic effects of phthalates on the male reproductive tract.
Effects of Phthalates on the Developing Male Reproductive Tract in Utero
Exposure of male rats to some diester and monoester phthalates in utero results in profound changes in the testis and other androgen-dependent structures. The extent and severity of the changes depend in large part on the timing and duration of the exposure and how the compound is administered. Studies in which reproductive tract defects have been observed involve exposure of pregnant dams during the time of androgen-dependent sexual differentiation, estimated to begin in the rat gestation day 12 (GD 12). Studies in which di-n-butyl phthalate (DBP) or butylbenzyl phthalate (BBP) were dosed before GD 12 had no effect on reproductive tract development, as expected, but did produce other malformations (Ema et al., 2000; Ema and Miyawaki, 2002). Exposure regimens that resulted in the most consistent and robust effects involved administration of the phthalate by daily gavage throughout development of the reproductive tract until birth or shortly thereafter (Gray et al., 2000; Mylchreest et al., 1999). A number of studies showed common defects in the male rat reproductive tract after in utero exposure to DBP (Barlow and Foster, 2003; Barlow et al., 2004; Fisher et al., 2003; Mylchreest et al., 1999, 2000, 2002; Wine et al., 1997), di-(2-ethylhexyl) phthalate (DEHP) (Moore et al., 2001; Parks et al., 2000), and BBP (Nagao et al., 2000; Piersma et al., 2000; Tyl et al., 2004). These effects included those in the testis as well as other organs that depend on testosterone for proper development (Table 1). Some effects caused by DBP were found to be permanent including gross lesions in the testis, vasa deferentia, seminal vesicles, prostate, and penis, as well as decreases in anogenital distance and nipple retention (Barlow et al., 2004). In addition, DBP at 1% in the diet (Wine et al., 1997) and BBP at 750 mg/kg/day (Tyl et al., 2004) but not 500 mg/kg/day (Nagao et al., 2000) decreased total number of sperm and indices for mating, fertility, and pregnancy in F1 generation rats in multi-generation exposure studies. In contrast, di-isononyl phthalate (DINP) was a weak reproductive toxicant as in utero exposure led only to increased nipple retention (Gray et al., 2000). Dimethyl phthalate and diethyl phthalate had no adverse effects on the development of the male reproductive tract (Gray et al., 2000). Monobutyl phthalate (MBP), the principle metabolite of DBP, disrupted the descent of the fetal testis (Imajima et al., 2001; Shono and Suita, 2003).
Reproductive effects of DBP are not exclusively found in rats. Exposure of male tadpoles of the frog species Rana rugosa increased the incidence of undifferentiated gonads developing into those having complete or partial ovarian structure (Ohtani et al., 2000). Male rabbits exposed in utero exhibited decreased weights of testis and accessory sex glands, decreases in ejaculated sperm, and increases in abnormal sperm six weeks after exposure (Higuchi et al., 2003).
The effects of the active phthalates on the male rat reproductive tract share similarities with a condition in humans termed testicular dysgenesis syndrome (TDS). As proposed by Skakkebaek et al. (2001), TDS includes testicular germ cell cancers, poor semen quality, crytorchidism and hypospadias, all thought to arise from disruption of embryonic gonadal development. That TDS is increasing in developing countries and that environmental exposure to man-made male reproductive toxicants are responsible for TDS is unresolved, partly because of the lack of high quality data allowing comparison of studies (Sharpe, 2001). Only increases in testicular cancer over the last 50–90 years have been established (Bergstrom et al., 1996; Toppari et al., 1996). TDS, like exposure to phthalates is associated with abnormal function in both Leydig and Sertoli cells, resulting in downstream consequences in the testis and sex organs (Fig. 1). Exposure of rats to DBP in utero increased the proportion of immature Sertoli cells unable to support spermatogenesis and the number of multinucleated gonocytes, possibly through alterations in Sertoli cell-gonocyte interactions (Fisher et al., 2003). How DBP treatment interferes with these interactions and in particular, gonocyte cytokinesis is not understood.
Phthalates which produce changes in male sex organs may be operating through an anti-androgen-like mechanism. The effects of DBP and DEHP overlapped with, but were distinct from androgen receptor antagonists including linuron (Gray et al., 1999) and flutamide (Mylchreest et al., 1999). Unlike other anti-androgens, some active phthalates target the Leydig cell testosterone biosynthetic machinery (discussed in detail below). Decreases in fetal testicular testosterone levels or serum testosterone in young postnatal male rats (PND 35) were observed after in utero exposure to DBP (Fisher et al., 2003; Mylchreest et al., 2002; Shultz et al., 2001; Wilson et al., 2004), DEHP (Akingbemi et al., 2001; Parks et al., 2000; Wilson et al., 2004), and BBP (Nagao et al., 2000; Wilson et al., 2004). When exposed only in utero, testosterone levels returned to normal after PND 35 (Akingbemi et al., 2001; Fisher et al., 2003). Male rabbits exposed in utero to DBP also exhibited decreases in testosterone six weeks after exposure that were reversible at 12 weeks (Higuchi et al., 2003). In contrast, testicular descent is at least partially testosterone-independent and may be disrupted through an alternate mechanism involving decreases in insulin-like hormone 3 (INSL3) expression observed in fetal testis after exposure to DBP, DEHP, and BBP (Wilson et al., 2004). Overall, the data is consistent with some phthalates indirectly down-regulating the activity of the androgen receptor through decreases in testosterone levels, resulting in delayed or absent development of androgen-dependent male reproductive organs.
Alteration of Testis Structure and Function by Phthalates in Postnatal and Adult Rats
Exposure of postnatal, preadolescent, and adult rats to some phthalates produces a different set of sequelae in testis compared to that of male rats exposed in utero. The age of the rat at the time of exposure has a dramatic effect on the testis as younger animals are more sensitive to the effects of phthalates (Gray and Gangolli, 1986). Compared to rats, mice are relatively resistant at any age. Some phthalates induce testicular toxicity in guinea pigs (Gray et al., 1982) and ferrets (Lake et al., 1976). Resistance to phthalate toxicity in Syrian hamsters is likely due to inefficient metabolic activation of diesters to monoesters by gut esterases compared to that in rats (Gray et al., 1982). Exposure to phthalates in responsive species leads to decreases in seminiferous tubule diameter and testis weight. At least in rats, this occurs through increased germ cell apoptosis and necrosis preceding the sloughing of germ cells into the tubular lumen (Gangolli, 1982).
The primary cellular target of toxic phthalates is the Sertoli cell which exhibits biochemical and morphological changes after exposure. FSH-stimulated cAMP accumulation in primary Sertoli cells was inhibited by Mon-(2-ethyl)-hexlphthalate (Heindel and Chapin, 1989; Lloyd and Foster, 1988), MBP and monopentyl phthalate, but not monomethyl or monoethyl phthalate (Heindel and Powell, 1992). Decreases in proliferation of Sertoli cells were observed after DEHP exposure in vivo (Li et al., 2000) or after exposure to MEHP in vitro (Li et al., 1998; Li and Kim, 2003), and may occur through a block in FSH-stimulated Sertoli cell proliferation (Li et al., 1998). Two genes which play roles in Sertoli cell proliferation and differentiation, Müllerian inhibiting substance and GATA-4, were decreased by MEHP in Sertoli cells within cultured testis from GD 18 and PND 3 rats (Li and Kim, 2003). Sertoli cell numbers recover somewhat during the course of exposure, as the decreases observed at one week were no longer evident after two to three weeks exposure to DEHP (Dostal et al., 1988). In rats exposed to MEHP, collapse in the Sertoli cell intermediate filament vimentin structure (indicative of changes in cell morphology) was observed as early as 3 h, followed by increases in perinuclear condensation of vimentin at 6–12 h. These changes preceded or were coincident with increased apoptosis in germ cells (Richburg and Boekelheide, 1996). Immunostaining for flamingo1/Celsr2, a G protein coupled receptor family member that may link cell-cell adhesion to G-protein dependent signaling was redistributed within 2 h of MEHP exposure and disappeared by 12 h, indicating that this protein is a proximate target of MEHP (Richburg et al., 2002).
Germ cells exhibit changes that may be secondary to effects on Sertoli cells. In the intact testis DEHP increased the generation of reactive oxygen species with concomitant decreases in the concentration of glutathione and ascorbic acid, and in primary cultures, oxidative stress was selectively induced in germ cells but not in Sertoli cells treated with MEHP (Kasahara et al., 2002). MEHP may specifically target the stages IX–XI of the mouse seminiferous cycle as apoptosis was increased in cells of these timed tubule segments treated in vitro (Suominen et al., 2003). Sertoli cells from mice treated with DEHP express the Fas ligand (FasL) while associated spermatocytes express Fas, the receptor for FasL, indicating that activation of the Fas system originating in Sertoli cells could lead to subsequent apoptosis in spermatocytes (Ichimura et al., 2003). Experiments in gld mice which carry a defective FasL gene revealed that germ cell apoptosis after MEHP exposure had both Fas-dependent (Richburg et al., 2000) and Fas-independent components (Giammona et al., 2002). Additionally, MEHP alters the expression and activity of death receptors 4,5,6 in the testis which may act in concert with the Fas-signaling pathway to initiate germ cell apoptosis (Giammona et al., 2002).
Leydig cells in postnatal and adult rats are also targeted by some active phthalates. DEHP (Jones et al., 1993; Kim et al., 2003) and di-n-octyl phthalate (Jones et al., 1993) decreased testosterone serum levels. Primary Leydig cells treated with MEHP or mono-octyl phthalate also exhibited decreases in testosterone secretion (Jones et al., 1993) although the concentrations used to achieve these effects were very high (10–3 M). Human chorionic gonadotropin-induced testosterone secretion was suppressed by prior exposure of Leydig cells in culture to PP (Biegel et al., 1995; Cook et al., 1992; Liu et al., 1996). No phthalates were tested in these studies, however. Decreases in testosterone levels may explain the decreased weights of accessory sex organs (seminal vesicle and prostate) in male rats exposed to DEHP that were partially reversed by co-administration of testosterone or gonadotropins (Gray and Gangolli, 1986). Decreased testosterone levels may also play a role in the genesis of Leydig cell tumors (Klaunig et al., 2003) through increases in compensatory Leydig cell proliferation observed with DEHP (Akingbemi et al., 2004) and the PP, WY-14,643 (Biegel et al., 1992). Two studies have shown increases in serum testosterone levels in male rats after exposure to DEHP at 200 mg/kg from PND 21–48 (Akingbemi et al., 2001) or relatively low doses of DEHP (10 or 100 mg/kg) for PND 21–90 (Akingbemi et al., 2004), which the authors attribute to an observed increase in Leydig cell numbers. Thus, short-term exposure to phthalates may lead to decreases in testosterone while longer exposures may stimulate Leydig cell proliferation resulting in consequent increases in testosterone. A more comprehensive time course would be useful to confirm these events for DEHP and other phthalates.
Expression of PPARs in the Testes
PPAR subtypes are expressed in the adult rat testes (Table 2). PPAR was expressed in Leydig and Sertoli cells of the adult rat (Braissant et al., 1996; Schultz et al., 1999). Conflicting reports exist as to the expression of PPAR in spermatocytes, with one report detecting expression (Schultz et al., 1999) and the other negative for expression (Braissant et al., 1996). PPAR? is expressed in Leydig and Sertoli cells but not in spermatocytes (Braissant et al., 1996). PPAR, in contrast, is either weakly expressed or not expressed at all in these cell types (Braissant et al., 1996). The spectrum of PPAR subtype expression in rat Leydig cells is similar to that in a Leydig mouse cell line (MA-10); PPAR and PPAR?, but not PPAR were expressed (Gazouli et al., 2002). Although there are no reports, that we know of, that examine PPAR target gene expression specifically in Sertoli cells or spermatocytes, there were increased levels of two PPAR gene targets involved in peroxisomal ?-oxidation (acyl-CoA oxidase, multifunctional protein-1) in Leydig cells after adult rats were fed the PP ciprofibrate in the diet for two weeks (Nemali et al., 1988) indicating that PPAR expressed in Leydig cells is responsive to PP. In contrast, neither WY or the PP ammonium perfluorooctanoate increased peroxisomal ?-oxidation in Leydig cells from chronically treated rats (Biegel et al., 2001).
The PPAR gene exhibited stage-specific expression during the spermatocyte differentiation cycle, peaking during stages II–VI and having the lowest expression at stages VII–VIII and IX–XII as assessed in isolated rat seminiferous tubules. Strong PPAR expression was observed in Sertoli cell nuclei during stages XIII–I (Schultz et al., 1999), a time when cells are most sensitive to FSH (Parvinen, 1982). Consistent with this, PPAR expression was increased in cultured seminiferous tubules after FSH treatment during all stages of the cycle (Schultz et al., 1999). These results indicate that PPAR expression is controlled in part by FSH, and that PPAR carries out a specific functional role during different stages of the differentiation cycle.
Only expression of PPAR has been examined in the early postnatal testis (Schultz et al., 1999). Expression of PPAR was high in rat seminiferous tubules at PND 1 followed by a steady decline until PND 30 and an increase at PND 60. Information about expression of PPAR subtypes in the rat testis in utero is lacking but PPAR-null (Lee et al., 1995) and PPAR?-null (Peters et al., 2000) mice remain viable and fertile indicating that these receptors are not essential for mouse testicular development and fertility. Compensatory mechanisms have not been ruled out (e.g., increased expression of another PPAR subtype) that would allow proper testicular development.
PPAR was expressed in human Leydig cells and spermatocytes, but not in Sertoli cells (Schultz et al., 1999). PPAR and PPAR? were expressed in human testis homogenates (Hase et al., 2002). PPAR may also be expressed although conflicting data exists (Elbrecht et al., 1996; Hase et al., 2002). Overall, these results indicate that PPAR subtypes are expressed in some types of adult cells in both rat and human testis. Further work is needed to determine if PPAR subtypes are expressed at both the transcript and protein levels in the testis throughout development from humans as well as a range of species susceptible to phthalate-induced testicular toxicity.
Activation of PPARs by Phthalate Esters
Despite their structural diversity compared to ligands that activate other nuclear receptors, PP do have similar structural requirements for activating PPARs in vitro and for initiating biological effects in animals. Most PP are amphipathic molecules containing a hydrophobic backbone (aliphatic or aromatic) linked to an acidic group or a moiety that can be metabolized to an acidic group. The acidic group is essential for ligand activity and typically consists of a carboxylic acid present in the parent compound. In the case of phthalates, esterases cleave a side chain from the diester resulting in a monoester containing a carboxylic acid. Some PP resemble endogenous lipid activators of PPARs such as long-chain saturated and unsaturated fatty acids (Xu et al., 1999).
Many phthalates activate PPARs in in vitro transactivation assays. In these assays either full length PPAR (Hurst and Waxman, 2003; Lampen et al., 2003; Lapinskas et al., 2004; Maloney and Waxman, 1999) or hybrid transcription factors consisting of the PPAR ligand binding domain cloned to the glucocorticoid receptor (Lampen et al., 2003) or GAL4 (Bility et al., 2004) DNA binding domains were co-transfected into a mammalian cell line with a reporter gene (e.g., luciferase) under control of either PPAR binding sites called peroxisome proliferator responsive elements (PPREs) (Hurst and Waxman, 2003; Lapinskas et al., 2004; Maloney and Waxman, 1999), glucocorticoid responsive elements (Lampen et al., 2003) or the binding site for GAL4, UASG (Bility et al., 2004). After time to express the receptor, the cells are treated with the phthalate esters. The reporter gene activity is then normalized to the activity of an additional reporter gene used as a transfection control and results are usually reported as a fold-change relative to a solvent control.
Mouse PPAR and PPAR are activated by a large number of phthalate monoesters (Table 3). Monoester phthalates with longer aliphatic or aromatic side chains tended to be more potent activators. Monoesters with short straight chains either do not activate at all or activate weakly. Monooctyl phthalate possessing a longer side chain activated all three subtypes (Bility et al., 2004), consistent with the observation that the longer the length of the fatty acid (up to 21 carbons), the better the fatty acid is at activating PPARs in vitro (Xu et al., 1999). Monooctyl and monoheptyl phthalates are unique in that they can activate PPARs in vitro but do not act as PP because the aliphatic side chains are rapidly cleaved in vivo (Albro and Moore, 1974). Steric restrictions appear to play a part in the ability of some monoesters to activate. Monoesters with the bulkiest side chains were less potent activators than MEHP or were completely inactive (e.g., mono-(1-methyl)-2-norbornyl phthalate, mono-(2,2-dimethyl-1-phenylpropyl) phthalate). The DEHP metabolite 2-ethylhexanoic acid weakly activated PPAR and compared to MEHP. In two studies the diesters DEHP and BBP were able to activate PPAR and PPAR, albeit weakly (Lampen et al., 2003; Lapinskas et al., 2004) despite the fact that the monoesters are considered to be the active toxic species. However, it is possible that the activation reflects a low level of esterase activity in the cell lines used.
Mouse PPAR? was activated by only a small number of phthalates and in general, higher concentrations of those phthalates were required to activate to the same level as PPAR and PPAR. PPAR? was activated by monoester phthalates with longer or bulkier side chains such as MEHP, mono-isohexyl phthalate, mono-(1-methyl)-heptyl phthalate, mono-isodecyl phthalate and monobenzyl phthalate but not monoesters with shorter side chains (Bility et al., 2004; Lampen et al., 2003).
Human PPARs were also activated by phthalates. All human PPAR subtypes were activated by MEHP, mono-(1-methyl)-heptyl phthalate and monobenzyl phthalate. Mono-sec-butyl phthalate, DL-mono-(1-methyl)-hexyl phthalate, mono-isoheptyl phthalate, mono-isodecyl phthalate, mono-isohexyl phthalate and mono-2-(methacryloyloxy)ethyl phthalate were more selective, activating only one or two of the subtypes. Importantly, the human PPARs required higher concentrations of the monoesters to be activated to the same levels as the corresponding mouse receptor (Bility et al., 2004; Hurst and Waxman, 2003; Maloney and Waxman, 1999).
Activation of PPARs comes about through two distinct mechanisms. Most PP are thought to bind directly to PPAR leading to activation. However, there is evidence that some PP may activate PPARs indirectly, through increasing the pool of endogenous activators, e.g., by displacing fatty acids from carrier proteins (Luebker et al., 2002). To determine whether phthalate esters interact directly with PPARs, the scintillation proximity assay (Nichols et al., 1998) was used to assess the ability of phthalate esters to bind to human PPAR and PPAR (Lapinskas et al., 2004). Some monoester phthalates interacted with both receptors, including, monohexyl phthalate, monooctyl phthalate, monobenzyl phthalate, but not shorter chain phthalates (monopropyl phthalate, monopentyl phthalate). Despite the short length of the side chain, monoethyl phthalate binds with low affinity to both PPAR and PPAR. BBP and DBP also bind weakly to both subtypes and this could explain why these diesters were able to weakly activate in transactivation assays. The data supports phthalate-induced changes in gene expression through binding to and activation of PPAR by monoester phthalates.
Activation of Other Nuclear Receptors by Phthalates
Phthalate esters have been tested for their ability to interact with sex hormone receptors. In transactivation assays, a wide range of diesters and monoesters did not act as androgen receptor agonists or antagonists in HepG2 cells (Gaido, personal communication). In addition, DEHP and its active metabolite, MEHP, do not bind to the human androgen receptor (Parks et al., 2000). Some phthalates behave like estrogen receptor ligands in that the phthalates inhibited binding of estrogen to isolated estrogen receptors and induced estrogen-responsive endpoints in vitro (summarized in Moore, 2000). The relevance of these findings is questioned as monoesters, considered to be the proximate metabolites, were inactive in these in vitro assays and the phthalate diesters that do alter estrogen receptor activity only do so at concentrations that approach solubility of the compound, which could lead to nonspecific effects. Importantly, the phthalates were uniformly negative in a number of in vivo tests for estrogenicity including the uterotrophic assay (Moore, 2000). Taken together, these studies indicate that phthalates do not mediate their effects through estrogen or androgen receptors, but rather, support PPAR subtypes as primary nuclear receptor targets for phthalate esters.
Correlation of PPAR Activation to Male Reproductive Tract Malformations
The strength of the monoester as a PPAR activator partially correlates with the ability of the parent phthalate to act as a male reproductive toxicant (Table 3). Monomethyl and monoethyl phthalate were either inactive or only weakly active as PPAR activators. The parent phthalates dimethyl and diethyl phthalate were inactive as in utero reproductive toxicants (Gray et al., 2000). In contrast, MEHP was an activator of all three mouse PPAR subtypes and DEHP exposure led to a spectrum of male reproductive tract changes in rats (Gray et al., 2000; Moore et al., 2001; Parks et al., 2000). BBP is hydrolyzed to either monobenzyl phthalate or monobutyl phthalate with the ratio of hydrolysis products in rats being approximately 3:5, respectively (Mikuriya et al., 1988). Monobenzyl phthalate activated all PPAR subtypes and BBP was positive as a reproductive toxicant (Gray et al., 2000; Nagao et al., 2000; Tyl et al., 2004).
There are two outliers in this comparison. DINP was a very weak reproductive toxicant while MINP was a moderate activator of mouse PPAR and PPAR (Bility et al., 2004). DBP is a reproductive toxicant in rats despite the fact that MBP was only weakly active in PPAR transactivation assays (Bility et al., 2004; Hurst and Waxman, 2003; Lapinskas et al., 2004). The weak activity of MBP in PPAR transactivation assays was surprising given that DBP, like DEHP and BBP, acts like a typical PP in the rat liver (Corton et al., 1996; Marsman, 1995) and effects in the mouse liver induced by DBP as well as DEHP and DINP are dependent on PPAR (Anderson et al., 1999; Lapinskas et al., 2004; Valles et al., 2003; Ward et al., 1998). Although treatment with MBP induced the same effects as DBP in the juvenile rat testis (Foster et al., 1981; Oishi and Hiraga, 1980) indicating that MBP is the proximate metabolite, no studies have been carried out in which MBP metabolites were comprehensively examined for toxic effects in the testis as well as abilty to activate PPARs. Candidate metabolites include MBP-glucuronide, a dominant MBP metabolite in adult rats (Foster et al., 1983; Tanaka et al., 1978) and omega and omega-1 products of MBP (Tanaka et al., 1978). It is also possible that a metabolite of DBP or DBP itself activates PPARs indirectly, i.e., by releasing a lipid activator of PPAR as discussed above a process which would not necessarily be detectable using transactivation assays. The fact that two out of six phthalates are outliers in this comparison weakens the hypothesis that PPARs are generally involved in phthalate-induced male reproductive tract defects.
Regulation of Sex Hormone Synthesis and Catabolism Genes by Phthalate Esters
As discussed above, BBP, DBP, and DEHP decrease testosterone levels in the fetus and in the young male rat. Consistent with this, genes involved in testosterone biosynthesis were uniformly down-regulated by DBP exposure in the fetal testis (Barlow et al., 2003; Shultz et al., 2001; Thompson et al., 2004) (Fig. 2A). The genes include Scavenger Receptor Class B, type 1 (SR-B1) and steroidogenic acute regulatory protein (StAR) (Barlow et al., 2003; Shultz et al., 2001). SR-B1 mediates the selective uptake of cholesterol esters from high-density lipoproteins, and StAR carries cholesterol from the outer to the inner mitochondrial membrane. The peripheral benzodiazepine receptor (PBR), which also carries cholesterol into the mitochondria was down-regulated by WY-14,643 and DEHP in adult mouse testes in a PPAR-dependent manner (Gazouli et al., 2002). In the fetal rat testis, PBR mRNA was up-regulated by DBP but by immunohistochemistry the protein was decreased in Leydig cells (Lehmann et al., 2004). Leydig cell mitochondria isolated from fetuses exposed to DBP in utero, exhibited decreased uptake of cholesterol supporting altered cholesterol handling for the decreases in testosterone synthesis (Thompson et al., 2004). Other genes involved in steroid biosynthesis that were down-regulated by DBP included P450 side chain cleavage enzyme (P450scc) thought to be the limiting enzymatic step in testosterone biosynthesis, 3?-hydroxysteroid dehydrogenase (3?-HSD) and CYP17 (Barlow et al., 2003; Shultz et al., 2001). Uniform decreases in the expression of these steroidogenic genes especially in a dose-dependent manner (Lehmann et al., 2004) is consistent with decreased levels of testicular testosterone.
The nuclear receptor steroidogenic factor 1 (SF-1) plays a prominant role in the development and differentiation of steroidogenic tissues and controls the expression of steroidogenic enzymes and cholesterol transporters required for steroidogenesis (Val et al., 2003). INSL3 (Emmen et al., 2000) and receptors for follicle-stimulating and leutinizing hormones (Val et al., 2003) are also regulated by SF-1. Genes altered by DBP exposure including SR-B1, StAR, P450scc, CYP17 and 3?-HSD, Type II (Barlow et al., 2003; Shultz et al., 2001) require SF-1 for basal promoter activity and for cyclic AMP induction. The hypothesis that phthalate ester exposure leads to suppression of SF-1 activity was tested in the Leydig cell line MA-10 in which expression of a reporter gene linked to the promoters of SR-B1, StAR, and CYP17 genes was measured after MBP exposure (Thompson et al., 2004). However, MBP had no effect on reporter gene expression. Although this finding indicates that phthalates inhibit steroidogenic gene expression independently of effects on SF-1 activity additional monoesters should be tested in this system before any definitive conclusion can be made of SF-1 involvement. The mechanism by which these genes are down-regulated by DBP and whether other phthalates target the same genes requires further study.
Phthalate esters along with other PP, also alter the expression of testosterone and estrogen metabolism genes (Figs. 2A and 2B). Among these, 5-reductase which converts testosterone to the more potent androgen, dihydrotestosterone was up-regulated in the prepubertal rat testis by DEHP (Kim et al., 2003). Multiple cytochrome P450 family members which hydroxylate testosterone were altered by PP exposure but in ways that do not clearly indicate a role in decreasing testosterone levels. CYP2C7 and CYP2C11 were down-regulated while CYP3A2 was up-regulated in the male rat liver (Corton et al., 1998; Fan et al., 2004-a,b). CYP2C11 and CYP3A were both up-regulated in the testes (Kim et al., 2003). WY-14,643 increased expression of fatty acyl-CoA enzyme A:testosterone acyl transferase in the liver (Xu et al., 2001). The down-regulation of CYP2C11 by a PP required PPAR; the site within the CYP2C11 promoter required for down-regulation by PP was identified and likely interacts with transcription factors other than PPAR implying that PPAR down-regulates CYP2C11 through an indirect mechanism (Ripp et al., 2003).
Serum estrogen levels were increased in male rats after DEHP exposure (Akingbemi et al., 2004; Eagon et al., 1994) and after exposure to a number of PP (Biegel et al., 2001; Liu et al., 1996). Phthalates alter the expression of many estrogen metabolizing enzymes (Fig. 2B). Aromatase exhibits both up-and down-regulation by PP depending on the tissue and compound (Biegel et al., 2001; Kim et al., 2003; Lovekamp and Davis, 2001; Lovekamp-Swan et al., 2003; Mu et al., 2001; Rubin et al., 2000; Toda et al., 2003). Genes that decrease estrogen levels were increased after PP exposure including 17?-hydroxysteroid dehydrogenase, type IV (Corton et al., 1996) and fatty acyl-CoA enzyme A:estradiol acyl transferase (Xu et al., 2001). However, these changes may be offset by decreases in the estrogen hydroxylases CYP1B1 after exposure to DBP, DEHP, and BBP in the testis (Seo et al., 2004) or CYP2C11 (Corton et al., 1998) and estrogen sulfotransferase (Fan et al., 2004-b) after WY, GEM, or DBP exposure in rat livers. A handful of these genes involved in testosterone or estrogen metabolism are directly regulated by PPAR either in the liver or testis as determined by studies in PPAR-null mice. These include 17?-HSD, type IV (Corton et al., 1996) and the CYP3A2 ortholog Cyp3a11 (Fan et al., 2004-b). These results point to a role for PPAR in regulating expression of some sex steroid metabolism genes upon phthalate exposure in the adult animal. Although these studies help us appreciate the gene targets of phthalates involved in regulation of steroid metabolism, more work is needed to determine the mechanism by which phthalates alter the expression of these genes, whether these genes are also regulated in parallel at the protein level and the contribution of these collective changes to phthalate toxicity in the postnatal testis.
PPAR and Phthalate-Induced Testicular Toxicity
The role of PPAR in phthalate-induced developmental and testicular toxicity has been determined directly in wild-type and PPAR-null mice in two studies. In the first study (Peters et al., 1997), pregnant dams were dosed with DEHP at GD 8 and GD 9 (before development of the male reproductive tract) and malformations of the fetuses were evaluated on GD 10 and GD 18. DEHP exposure decreased crown-rump length and increased the incidence of open neural tubes (failure of the hind- and mid-brain to close) in both DEHP-treated wild-type and PPAR-null mice compared to controls, demonstrating that DEHP-induced malformations are PPAR-independent. However, this study cannot be used to determine the role for PPAR in phthalate-induced male reproductive tract defects from in utero exposure, because DEHP treatment did not occur during the critical window of development of the male reproductive tract and to our knowledge, the rat model of in utero male reproductive tract defects has not been successfully recapitulated in mice.
In the second study (Ward et al., 1998), male wild-type and PPAR-null mice were fed a diet containing 12,000 ppm DEHP and lesions in the liver and testis were examined. In the liver all of the expected effects of DEHP were dependent on PPAR. After four and eight weeks of exposure, testis from wild-type mice exhibited mild or moderate toxic effects including focal tubular degenerative lesions, decreased spermatogenesis, and giant cells within the epididymis. In sharp contrast, the testis from PPAR-null mice at these time points were predominantly normal except for a few tubules that lacked normal indicators of spermatogenesis. PPAR-null mice at 24 weeks exhibited moderate testicular effects. However, all of the wild-type mice were sacrificed at 12–16 weeks due to toxicity when there was severe testicular atrophy making it impossible to directly compare the effects in the two strains at the 24 week time point. The authors concluded that DEHP effects in the testis were the result of both PPAR-dependent and -independent mechanisms. This study indicates that PPAR determines the timing and severity of testicular toxicity by DEHP. To better characterize the role of PPAR in phthalate-induced testicular toxicity, additional studies using other phthalates should be carried out in wild-type and PPAR-null mice. Given that PPAR? is expressed in the testis, PPAR?-null mice could also be used to determine any role played by this receptor in phthalate-induced testicular toxicity. Determining a role for PPAR will require construction of testis-specific nullizygous mice as the PPAR-null mutation is embryonically lethal (Barak et al., 1999).
Cross-Talk between PPAR and Nuclear Receptor Signaling: Potential Impact on the Male Reproductive Tract
Several studies have demonstrated ‘cross-talk’ between PPAR and other nuclear receptors. Cross-talk occurs when signaling pathways other than the primary pathway are activated or repressed through interactions with components of the primary pathway. PPAR-induced interference of other nuclear receptor pathways occurs through competition between PPARs and other receptors for binding to (1) the same DNA response element or (2) a common heterodimerization partner (Fig. 3). In the Supplementary Materials we examine cross-talk between PPARs and nuclear receptors for estrogen, thyroid hormone, and retinoic acid that play roles in the development or homeostasis of the testis. Many of these studies supporting interactions between PPAR and other nuclear receptors were either performed with PP other than phthalates in in vitro cell models allowing only conjecture as to how PPAR-phthalate interactions may disrupt male reproductive tract function in vivo. Further work is needed before these proposed mechanisms should be considered to play a role in phthalate-induced effects in the male reproductive tract.
Conclusions
In this review we have assessed the involvement of PPAR subtypes in phthalate-induced effects on the male reproductive tract. Much of the evidence supporting a role for PPARs is correlative. A high level of PPAR activation and testicular toxicity requires metabolic conversion of the diester to the monoester. PPAR subtypes are activated by structurally diverse phthalate monoesters. Phthalate diesters activate weakly or not at all. PPAR and PPAR are usually more sensitive to monoester activation than PPAR?, and mouse PPARs are more sensitive than human PPARs. PPARs, principally the and ? subtypes are expressed in neonatal or adult Sertoli and Leydig cells, the target cells of phthalate ester effects. The testicular toxicity of DEHP occurred later and with diminished severity in PPAR-null mice, indicating PPAR acts as a modifier gene regulating the timing and severity of toxicity. There is a correlation between the ability of four of six diesters to induce reproductive tract changes and the corresponding monoesters to activate PPAR subtypes. Phthalates alter the expression of genes encoding sex steroid metabolizing enzymes and enzymes involved in testosterone biosynthesis, some in a PPAR-dependent manner. PPARs interact with and down-regulate the activities of other nuclear receptors that play roles in the developing testis including receptors for estrogen, thyroid hormone, and retinoic acid, leading to plausible mechanisms by which PPAR activation by phthalates could alter testicular function in vivo.
Other evidence exists to argue for little, if any participation of PPARs in phthalate-induced testicular effects. Although phthalates and PP can activate PPARs, only the phthalates are generally considered testicular toxicants. However, it is not unprecedented that structurally distinct classes of nuclear receptor ligands (e.g., ER ligands) have unique properties as agonists, partial agonists, or antagonists depending on receptor-ligand conformations and subsequent interactions with tissue-specific co-activators/co-repressors (McDonnell et al., 2002). With this in mind, it is interesting to note that many genes normally up-regulated by PP in the adult rat liver were down-regulated by DBP in the fetal testis and include those that are known targets of PPAR including fatty acid ?–oxidation genes (Shultz et al., 2001). Phthalate effects in the adult testis can occur rapidly (within 2 h), likely before any phenotypic effects of PPAR-mediated gene regulation would be observed. Two phthalates, DBP and DINP, do not neatly fit into a relationship of inducing male reproductive tract malformations while the corresponding monoester induces PPAR activation. DBP is a strong male reproductive toxicant but MBP only weakly activates PPARs; DINP is a weak toxicant but MINP is a moderately strong PPAR activator. The fact that phthalate-induced testicular toxicity occurs in species that do not respond to PP-induced liver effects has been used to invoke a PPAR-independent mechanism, but comprehensive information on PPAR expression in the testis of responsive vs. nonresponsive species is lacking.
One way to rationalize these seemingly disparate findings is to invoke a mechanism that at least in the neonatal testis, includes both PPAR-dependent and -independent events leading to testicular toxicity after phthalate exposure. Future investigations should be designed to directly determine the role of PPARs as potential mediators of phthalate effects. Resolving a role for PPARs will be expedited by a comprehensive determination of PPAR expression in the developing testis and testing the effects of multiple phthalates in established mouse models nullizygous for each PPAR gene during male reproductive tract development.
SUPPLEMENTARY DATA
Supplementary data are available online at www.toxsci.oupjournals.org.
ACKNOWLEDGMENTS
The authors thank Drs. Paul Foster and Kevin Gaido for a critical review of the manuscript and the Phthalate Esters Panel of the American Chemistry Council for funding to produce this review. We apologize to colleagues whose papers were not cited due to space limitations.
REFERENCES
Akingbemi, B. T., Youker, R. T., Sottas, C. M., Ge, R., Katz, E., Klinefelter, G. R., Zirkin, B. R., and Hardy, M. P. (2001). Modulation of rat Leydig cell steroidogenic funtion by di(2-ethylhexyl)phthalate. Biol. Reprod. 65, 1252–1259.
Biegel, L. B., Liu, R. C., Hurtt, M. E., and Cook, J. C. (1995). Effects of ammonium perfluorooctanoate on Leydig cell function: in vivo, in vitro, and ex vivo studies. Toxicol. Appl. Pharmacol. 134, 18–25.
Biegel, L. B., Hurtt, M. E., Frame, S. R., O'Connor, J. C., and Cook, J. C. (2001) Mechanisms of extrahepatic tumor induction by peroxisome proliferators in male CD rats. Toxicol. Sci. 60, 44–55.
Nichols, J. S., Parks, D. J., Consler, T. G., and Blanchard, S. G. (1998). Development of a scintillation proximity assay for peroxisome proliferator-activated receptor gamma ligand binding domain. Anal. Biochem. 257, 112–119.
Parvinen, M. (1982). Regulation of the seminiferous epithelium. Endocr. Rev. 3, 404–417.
Peters, J. M., Lee, S. S., Li, W., Ward, J. M., Gavrilova, O., Everett, C., Retiman, M. L., Hudson, L. D., and Gonzalez, F. J. (2000). Growth, adipose, brain, and skin alterations resulting from tageted disruption of the mouse peroxisome proliferator-activated reeceptor beta (delta). Mol. Cell Biol. 20, 5119–5128.
Valles, E. G., Laughter, A. R., Dunn, C. S., Cannelle, S., Swanson, C. L., Cattley, R. C., and Corton, J. C. (2003). Role of the peroxisome proliferator-activated receptor alpha in responses to diisononyl phthalate. Toxicology 191, 211–225.(J. Christopher Corton*,1 )
Many phthalate ester plasticizers are classified as peroxisome proliferators (PP), a large group of industrial and pharmaceutical chemicals. Like PP, exposure to some phthalates increases hepatocyte peroxisome and cellular proliferation, as well as the incidence of hepatocellular adenomas in mice and rats. Most effects of PP are mediated by three nuclear receptors called peroxisome proliferator-activated receptors (PPAR,?,). An obligate role for PPAR in PP-induced events leading to liver cancer is well-established. Exposure of rats in utero or in the neonate to a subset of phthalate esters causes profound, sometimes irreversible malformations in the male reproductive tract. We review here the data that supports or discounts roles for PPARs in phthalate-induced testis toxicity including (1) toxic effects of phthalates on the male reproductive tract, (2) expression of PPARs in the testis, (3) activation of PPARs by phthalates, (4) role of PPAR in testis toxicity, (5) gene targets of phthalates involved in steroid biosynthesis and catabolism, and (6) interactions between PPARs and other nuclear receptors that play roles in testis development and homeostasis. Critical research needs are identified that will help determine the significance of PPARs in phthalate-induced effects in the rat male reproductive tract and the relevance of toxicity to humans.
Key Words: phthalate ester; PPAR; testis; peroxisome proliferator; male reproductive tract.
INTRODUCTION
Most animal cells contain peroxisomes, subcellular organelles that perform diverse metabolic functions including H2O2-derived respiration, ?-oxidation of fatty acids, and cholesterol metabolism (Lock et al., 1989). Peroxisome proliferators (PP) are a large group of structurally dissimilar industrial and pharmaceutical chemicals that were identified as inducers of both the size and number of peroxisomes after in vivo exposure. PP include a large number of phthalate ester plasticizers which are, through their widespread use in various industries, widely distributed in the environment (Staples et al., 1997). Unlike many PP, phthalates are metabolized to active species by esterases found in the gut and other tissues. These esterases cleave one of the two side chains from the parent diester phthalate producing an active monoester phthalate. Rodent exposure to PP, including many phthalate esters leads to predictable adaptations in the liver consisting of hepatocellular hypertrophy and hyperplasia and transcriptional induction of fatty acid metabolizing enzymes (Lock et al., 1989). Chronic exposure to some PP causes an increased incidence of liver tumors in male and female mice and rats (Klaunig et al., 2003). Many studies also show that exposure to some phthalates results in profound, irreversible changes in the development of the male reproductive tract (Foster et al., 2001; Sharpe, 2001).
Many of the adaptive consequences of PP exposure are mediated by a subset of nuclear receptor superfamily members called peroxisome proliferator-activated receptors (PPARs). The PPAR family includes three distinct subtypes, PPAR, PPAR? (also known as NUC1 or PPAR), and PPAR encoded by separate genes (Lemberger et al., 1996). PP activate PPARs leading to altered expression of genes with roles in metabolism, cell growth, and stress responses (Corton et al., 2000). Most, if not all responses to PP including some phthalates in the liver are dependant on PPAR (Klaunig et al., 2003). Given recent findings describing novel aspects of phthalate-PPAR interactions, we review here evidence supporting or discounting roles for PPARs in mediating the toxic effects of phthalates on the male reproductive tract.
Effects of Phthalates on the Developing Male Reproductive Tract in Utero
Exposure of male rats to some diester and monoester phthalates in utero results in profound changes in the testis and other androgen-dependent structures. The extent and severity of the changes depend in large part on the timing and duration of the exposure and how the compound is administered. Studies in which reproductive tract defects have been observed involve exposure of pregnant dams during the time of androgen-dependent sexual differentiation, estimated to begin in the rat gestation day 12 (GD 12). Studies in which di-n-butyl phthalate (DBP) or butylbenzyl phthalate (BBP) were dosed before GD 12 had no effect on reproductive tract development, as expected, but did produce other malformations (Ema et al., 2000; Ema and Miyawaki, 2002). Exposure regimens that resulted in the most consistent and robust effects involved administration of the phthalate by daily gavage throughout development of the reproductive tract until birth or shortly thereafter (Gray et al., 2000; Mylchreest et al., 1999). A number of studies showed common defects in the male rat reproductive tract after in utero exposure to DBP (Barlow and Foster, 2003; Barlow et al., 2004; Fisher et al., 2003; Mylchreest et al., 1999, 2000, 2002; Wine et al., 1997), di-(2-ethylhexyl) phthalate (DEHP) (Moore et al., 2001; Parks et al., 2000), and BBP (Nagao et al., 2000; Piersma et al., 2000; Tyl et al., 2004). These effects included those in the testis as well as other organs that depend on testosterone for proper development (Table 1). Some effects caused by DBP were found to be permanent including gross lesions in the testis, vasa deferentia, seminal vesicles, prostate, and penis, as well as decreases in anogenital distance and nipple retention (Barlow et al., 2004). In addition, DBP at 1% in the diet (Wine et al., 1997) and BBP at 750 mg/kg/day (Tyl et al., 2004) but not 500 mg/kg/day (Nagao et al., 2000) decreased total number of sperm and indices for mating, fertility, and pregnancy in F1 generation rats in multi-generation exposure studies. In contrast, di-isononyl phthalate (DINP) was a weak reproductive toxicant as in utero exposure led only to increased nipple retention (Gray et al., 2000). Dimethyl phthalate and diethyl phthalate had no adverse effects on the development of the male reproductive tract (Gray et al., 2000). Monobutyl phthalate (MBP), the principle metabolite of DBP, disrupted the descent of the fetal testis (Imajima et al., 2001; Shono and Suita, 2003).
Reproductive effects of DBP are not exclusively found in rats. Exposure of male tadpoles of the frog species Rana rugosa increased the incidence of undifferentiated gonads developing into those having complete or partial ovarian structure (Ohtani et al., 2000). Male rabbits exposed in utero exhibited decreased weights of testis and accessory sex glands, decreases in ejaculated sperm, and increases in abnormal sperm six weeks after exposure (Higuchi et al., 2003).
The effects of the active phthalates on the male rat reproductive tract share similarities with a condition in humans termed testicular dysgenesis syndrome (TDS). As proposed by Skakkebaek et al. (2001), TDS includes testicular germ cell cancers, poor semen quality, crytorchidism and hypospadias, all thought to arise from disruption of embryonic gonadal development. That TDS is increasing in developing countries and that environmental exposure to man-made male reproductive toxicants are responsible for TDS is unresolved, partly because of the lack of high quality data allowing comparison of studies (Sharpe, 2001). Only increases in testicular cancer over the last 50–90 years have been established (Bergstrom et al., 1996; Toppari et al., 1996). TDS, like exposure to phthalates is associated with abnormal function in both Leydig and Sertoli cells, resulting in downstream consequences in the testis and sex organs (Fig. 1). Exposure of rats to DBP in utero increased the proportion of immature Sertoli cells unable to support spermatogenesis and the number of multinucleated gonocytes, possibly through alterations in Sertoli cell-gonocyte interactions (Fisher et al., 2003). How DBP treatment interferes with these interactions and in particular, gonocyte cytokinesis is not understood.
Phthalates which produce changes in male sex organs may be operating through an anti-androgen-like mechanism. The effects of DBP and DEHP overlapped with, but were distinct from androgen receptor antagonists including linuron (Gray et al., 1999) and flutamide (Mylchreest et al., 1999). Unlike other anti-androgens, some active phthalates target the Leydig cell testosterone biosynthetic machinery (discussed in detail below). Decreases in fetal testicular testosterone levels or serum testosterone in young postnatal male rats (PND 35) were observed after in utero exposure to DBP (Fisher et al., 2003; Mylchreest et al., 2002; Shultz et al., 2001; Wilson et al., 2004), DEHP (Akingbemi et al., 2001; Parks et al., 2000; Wilson et al., 2004), and BBP (Nagao et al., 2000; Wilson et al., 2004). When exposed only in utero, testosterone levels returned to normal after PND 35 (Akingbemi et al., 2001; Fisher et al., 2003). Male rabbits exposed in utero to DBP also exhibited decreases in testosterone six weeks after exposure that were reversible at 12 weeks (Higuchi et al., 2003). In contrast, testicular descent is at least partially testosterone-independent and may be disrupted through an alternate mechanism involving decreases in insulin-like hormone 3 (INSL3) expression observed in fetal testis after exposure to DBP, DEHP, and BBP (Wilson et al., 2004). Overall, the data is consistent with some phthalates indirectly down-regulating the activity of the androgen receptor through decreases in testosterone levels, resulting in delayed or absent development of androgen-dependent male reproductive organs.
Alteration of Testis Structure and Function by Phthalates in Postnatal and Adult Rats
Exposure of postnatal, preadolescent, and adult rats to some phthalates produces a different set of sequelae in testis compared to that of male rats exposed in utero. The age of the rat at the time of exposure has a dramatic effect on the testis as younger animals are more sensitive to the effects of phthalates (Gray and Gangolli, 1986). Compared to rats, mice are relatively resistant at any age. Some phthalates induce testicular toxicity in guinea pigs (Gray et al., 1982) and ferrets (Lake et al., 1976). Resistance to phthalate toxicity in Syrian hamsters is likely due to inefficient metabolic activation of diesters to monoesters by gut esterases compared to that in rats (Gray et al., 1982). Exposure to phthalates in responsive species leads to decreases in seminiferous tubule diameter and testis weight. At least in rats, this occurs through increased germ cell apoptosis and necrosis preceding the sloughing of germ cells into the tubular lumen (Gangolli, 1982).
The primary cellular target of toxic phthalates is the Sertoli cell which exhibits biochemical and morphological changes after exposure. FSH-stimulated cAMP accumulation in primary Sertoli cells was inhibited by Mon-(2-ethyl)-hexlphthalate (Heindel and Chapin, 1989; Lloyd and Foster, 1988), MBP and monopentyl phthalate, but not monomethyl or monoethyl phthalate (Heindel and Powell, 1992). Decreases in proliferation of Sertoli cells were observed after DEHP exposure in vivo (Li et al., 2000) or after exposure to MEHP in vitro (Li et al., 1998; Li and Kim, 2003), and may occur through a block in FSH-stimulated Sertoli cell proliferation (Li et al., 1998). Two genes which play roles in Sertoli cell proliferation and differentiation, Müllerian inhibiting substance and GATA-4, were decreased by MEHP in Sertoli cells within cultured testis from GD 18 and PND 3 rats (Li and Kim, 2003). Sertoli cell numbers recover somewhat during the course of exposure, as the decreases observed at one week were no longer evident after two to three weeks exposure to DEHP (Dostal et al., 1988). In rats exposed to MEHP, collapse in the Sertoli cell intermediate filament vimentin structure (indicative of changes in cell morphology) was observed as early as 3 h, followed by increases in perinuclear condensation of vimentin at 6–12 h. These changes preceded or were coincident with increased apoptosis in germ cells (Richburg and Boekelheide, 1996). Immunostaining for flamingo1/Celsr2, a G protein coupled receptor family member that may link cell-cell adhesion to G-protein dependent signaling was redistributed within 2 h of MEHP exposure and disappeared by 12 h, indicating that this protein is a proximate target of MEHP (Richburg et al., 2002).
Germ cells exhibit changes that may be secondary to effects on Sertoli cells. In the intact testis DEHP increased the generation of reactive oxygen species with concomitant decreases in the concentration of glutathione and ascorbic acid, and in primary cultures, oxidative stress was selectively induced in germ cells but not in Sertoli cells treated with MEHP (Kasahara et al., 2002). MEHP may specifically target the stages IX–XI of the mouse seminiferous cycle as apoptosis was increased in cells of these timed tubule segments treated in vitro (Suominen et al., 2003). Sertoli cells from mice treated with DEHP express the Fas ligand (FasL) while associated spermatocytes express Fas, the receptor for FasL, indicating that activation of the Fas system originating in Sertoli cells could lead to subsequent apoptosis in spermatocytes (Ichimura et al., 2003). Experiments in gld mice which carry a defective FasL gene revealed that germ cell apoptosis after MEHP exposure had both Fas-dependent (Richburg et al., 2000) and Fas-independent components (Giammona et al., 2002). Additionally, MEHP alters the expression and activity of death receptors 4,5,6 in the testis which may act in concert with the Fas-signaling pathway to initiate germ cell apoptosis (Giammona et al., 2002).
Leydig cells in postnatal and adult rats are also targeted by some active phthalates. DEHP (Jones et al., 1993; Kim et al., 2003) and di-n-octyl phthalate (Jones et al., 1993) decreased testosterone serum levels. Primary Leydig cells treated with MEHP or mono-octyl phthalate also exhibited decreases in testosterone secretion (Jones et al., 1993) although the concentrations used to achieve these effects were very high (10–3 M). Human chorionic gonadotropin-induced testosterone secretion was suppressed by prior exposure of Leydig cells in culture to PP (Biegel et al., 1995; Cook et al., 1992; Liu et al., 1996). No phthalates were tested in these studies, however. Decreases in testosterone levels may explain the decreased weights of accessory sex organs (seminal vesicle and prostate) in male rats exposed to DEHP that were partially reversed by co-administration of testosterone or gonadotropins (Gray and Gangolli, 1986). Decreased testosterone levels may also play a role in the genesis of Leydig cell tumors (Klaunig et al., 2003) through increases in compensatory Leydig cell proliferation observed with DEHP (Akingbemi et al., 2004) and the PP, WY-14,643 (Biegel et al., 1992). Two studies have shown increases in serum testosterone levels in male rats after exposure to DEHP at 200 mg/kg from PND 21–48 (Akingbemi et al., 2001) or relatively low doses of DEHP (10 or 100 mg/kg) for PND 21–90 (Akingbemi et al., 2004), which the authors attribute to an observed increase in Leydig cell numbers. Thus, short-term exposure to phthalates may lead to decreases in testosterone while longer exposures may stimulate Leydig cell proliferation resulting in consequent increases in testosterone. A more comprehensive time course would be useful to confirm these events for DEHP and other phthalates.
Expression of PPARs in the Testes
PPAR subtypes are expressed in the adult rat testes (Table 2). PPAR was expressed in Leydig and Sertoli cells of the adult rat (Braissant et al., 1996; Schultz et al., 1999). Conflicting reports exist as to the expression of PPAR in spermatocytes, with one report detecting expression (Schultz et al., 1999) and the other negative for expression (Braissant et al., 1996). PPAR? is expressed in Leydig and Sertoli cells but not in spermatocytes (Braissant et al., 1996). PPAR, in contrast, is either weakly expressed or not expressed at all in these cell types (Braissant et al., 1996). The spectrum of PPAR subtype expression in rat Leydig cells is similar to that in a Leydig mouse cell line (MA-10); PPAR and PPAR?, but not PPAR were expressed (Gazouli et al., 2002). Although there are no reports, that we know of, that examine PPAR target gene expression specifically in Sertoli cells or spermatocytes, there were increased levels of two PPAR gene targets involved in peroxisomal ?-oxidation (acyl-CoA oxidase, multifunctional protein-1) in Leydig cells after adult rats were fed the PP ciprofibrate in the diet for two weeks (Nemali et al., 1988) indicating that PPAR expressed in Leydig cells is responsive to PP. In contrast, neither WY or the PP ammonium perfluorooctanoate increased peroxisomal ?-oxidation in Leydig cells from chronically treated rats (Biegel et al., 2001).
The PPAR gene exhibited stage-specific expression during the spermatocyte differentiation cycle, peaking during stages II–VI and having the lowest expression at stages VII–VIII and IX–XII as assessed in isolated rat seminiferous tubules. Strong PPAR expression was observed in Sertoli cell nuclei during stages XIII–I (Schultz et al., 1999), a time when cells are most sensitive to FSH (Parvinen, 1982). Consistent with this, PPAR expression was increased in cultured seminiferous tubules after FSH treatment during all stages of the cycle (Schultz et al., 1999). These results indicate that PPAR expression is controlled in part by FSH, and that PPAR carries out a specific functional role during different stages of the differentiation cycle.
Only expression of PPAR has been examined in the early postnatal testis (Schultz et al., 1999). Expression of PPAR was high in rat seminiferous tubules at PND 1 followed by a steady decline until PND 30 and an increase at PND 60. Information about expression of PPAR subtypes in the rat testis in utero is lacking but PPAR-null (Lee et al., 1995) and PPAR?-null (Peters et al., 2000) mice remain viable and fertile indicating that these receptors are not essential for mouse testicular development and fertility. Compensatory mechanisms have not been ruled out (e.g., increased expression of another PPAR subtype) that would allow proper testicular development.
PPAR was expressed in human Leydig cells and spermatocytes, but not in Sertoli cells (Schultz et al., 1999). PPAR and PPAR? were expressed in human testis homogenates (Hase et al., 2002). PPAR may also be expressed although conflicting data exists (Elbrecht et al., 1996; Hase et al., 2002). Overall, these results indicate that PPAR subtypes are expressed in some types of adult cells in both rat and human testis. Further work is needed to determine if PPAR subtypes are expressed at both the transcript and protein levels in the testis throughout development from humans as well as a range of species susceptible to phthalate-induced testicular toxicity.
Activation of PPARs by Phthalate Esters
Despite their structural diversity compared to ligands that activate other nuclear receptors, PP do have similar structural requirements for activating PPARs in vitro and for initiating biological effects in animals. Most PP are amphipathic molecules containing a hydrophobic backbone (aliphatic or aromatic) linked to an acidic group or a moiety that can be metabolized to an acidic group. The acidic group is essential for ligand activity and typically consists of a carboxylic acid present in the parent compound. In the case of phthalates, esterases cleave a side chain from the diester resulting in a monoester containing a carboxylic acid. Some PP resemble endogenous lipid activators of PPARs such as long-chain saturated and unsaturated fatty acids (Xu et al., 1999).
Many phthalates activate PPARs in in vitro transactivation assays. In these assays either full length PPAR (Hurst and Waxman, 2003; Lampen et al., 2003; Lapinskas et al., 2004; Maloney and Waxman, 1999) or hybrid transcription factors consisting of the PPAR ligand binding domain cloned to the glucocorticoid receptor (Lampen et al., 2003) or GAL4 (Bility et al., 2004) DNA binding domains were co-transfected into a mammalian cell line with a reporter gene (e.g., luciferase) under control of either PPAR binding sites called peroxisome proliferator responsive elements (PPREs) (Hurst and Waxman, 2003; Lapinskas et al., 2004; Maloney and Waxman, 1999), glucocorticoid responsive elements (Lampen et al., 2003) or the binding site for GAL4, UASG (Bility et al., 2004). After time to express the receptor, the cells are treated with the phthalate esters. The reporter gene activity is then normalized to the activity of an additional reporter gene used as a transfection control and results are usually reported as a fold-change relative to a solvent control.
Mouse PPAR and PPAR are activated by a large number of phthalate monoesters (Table 3). Monoester phthalates with longer aliphatic or aromatic side chains tended to be more potent activators. Monoesters with short straight chains either do not activate at all or activate weakly. Monooctyl phthalate possessing a longer side chain activated all three subtypes (Bility et al., 2004), consistent with the observation that the longer the length of the fatty acid (up to 21 carbons), the better the fatty acid is at activating PPARs in vitro (Xu et al., 1999). Monooctyl and monoheptyl phthalates are unique in that they can activate PPARs in vitro but do not act as PP because the aliphatic side chains are rapidly cleaved in vivo (Albro and Moore, 1974). Steric restrictions appear to play a part in the ability of some monoesters to activate. Monoesters with the bulkiest side chains were less potent activators than MEHP or were completely inactive (e.g., mono-(1-methyl)-2-norbornyl phthalate, mono-(2,2-dimethyl-1-phenylpropyl) phthalate). The DEHP metabolite 2-ethylhexanoic acid weakly activated PPAR and compared to MEHP. In two studies the diesters DEHP and BBP were able to activate PPAR and PPAR, albeit weakly (Lampen et al., 2003; Lapinskas et al., 2004) despite the fact that the monoesters are considered to be the active toxic species. However, it is possible that the activation reflects a low level of esterase activity in the cell lines used.
Mouse PPAR? was activated by only a small number of phthalates and in general, higher concentrations of those phthalates were required to activate to the same level as PPAR and PPAR. PPAR? was activated by monoester phthalates with longer or bulkier side chains such as MEHP, mono-isohexyl phthalate, mono-(1-methyl)-heptyl phthalate, mono-isodecyl phthalate and monobenzyl phthalate but not monoesters with shorter side chains (Bility et al., 2004; Lampen et al., 2003).
Human PPARs were also activated by phthalates. All human PPAR subtypes were activated by MEHP, mono-(1-methyl)-heptyl phthalate and monobenzyl phthalate. Mono-sec-butyl phthalate, DL-mono-(1-methyl)-hexyl phthalate, mono-isoheptyl phthalate, mono-isodecyl phthalate, mono-isohexyl phthalate and mono-2-(methacryloyloxy)ethyl phthalate were more selective, activating only one or two of the subtypes. Importantly, the human PPARs required higher concentrations of the monoesters to be activated to the same levels as the corresponding mouse receptor (Bility et al., 2004; Hurst and Waxman, 2003; Maloney and Waxman, 1999).
Activation of PPARs comes about through two distinct mechanisms. Most PP are thought to bind directly to PPAR leading to activation. However, there is evidence that some PP may activate PPARs indirectly, through increasing the pool of endogenous activators, e.g., by displacing fatty acids from carrier proteins (Luebker et al., 2002). To determine whether phthalate esters interact directly with PPARs, the scintillation proximity assay (Nichols et al., 1998) was used to assess the ability of phthalate esters to bind to human PPAR and PPAR (Lapinskas et al., 2004). Some monoester phthalates interacted with both receptors, including, monohexyl phthalate, monooctyl phthalate, monobenzyl phthalate, but not shorter chain phthalates (monopropyl phthalate, monopentyl phthalate). Despite the short length of the side chain, monoethyl phthalate binds with low affinity to both PPAR and PPAR. BBP and DBP also bind weakly to both subtypes and this could explain why these diesters were able to weakly activate in transactivation assays. The data supports phthalate-induced changes in gene expression through binding to and activation of PPAR by monoester phthalates.
Activation of Other Nuclear Receptors by Phthalates
Phthalate esters have been tested for their ability to interact with sex hormone receptors. In transactivation assays, a wide range of diesters and monoesters did not act as androgen receptor agonists or antagonists in HepG2 cells (Gaido, personal communication). In addition, DEHP and its active metabolite, MEHP, do not bind to the human androgen receptor (Parks et al., 2000). Some phthalates behave like estrogen receptor ligands in that the phthalates inhibited binding of estrogen to isolated estrogen receptors and induced estrogen-responsive endpoints in vitro (summarized in Moore, 2000). The relevance of these findings is questioned as monoesters, considered to be the proximate metabolites, were inactive in these in vitro assays and the phthalate diesters that do alter estrogen receptor activity only do so at concentrations that approach solubility of the compound, which could lead to nonspecific effects. Importantly, the phthalates were uniformly negative in a number of in vivo tests for estrogenicity including the uterotrophic assay (Moore, 2000). Taken together, these studies indicate that phthalates do not mediate their effects through estrogen or androgen receptors, but rather, support PPAR subtypes as primary nuclear receptor targets for phthalate esters.
Correlation of PPAR Activation to Male Reproductive Tract Malformations
The strength of the monoester as a PPAR activator partially correlates with the ability of the parent phthalate to act as a male reproductive toxicant (Table 3). Monomethyl and monoethyl phthalate were either inactive or only weakly active as PPAR activators. The parent phthalates dimethyl and diethyl phthalate were inactive as in utero reproductive toxicants (Gray et al., 2000). In contrast, MEHP was an activator of all three mouse PPAR subtypes and DEHP exposure led to a spectrum of male reproductive tract changes in rats (Gray et al., 2000; Moore et al., 2001; Parks et al., 2000). BBP is hydrolyzed to either monobenzyl phthalate or monobutyl phthalate with the ratio of hydrolysis products in rats being approximately 3:5, respectively (Mikuriya et al., 1988). Monobenzyl phthalate activated all PPAR subtypes and BBP was positive as a reproductive toxicant (Gray et al., 2000; Nagao et al., 2000; Tyl et al., 2004).
There are two outliers in this comparison. DINP was a very weak reproductive toxicant while MINP was a moderate activator of mouse PPAR and PPAR (Bility et al., 2004). DBP is a reproductive toxicant in rats despite the fact that MBP was only weakly active in PPAR transactivation assays (Bility et al., 2004; Hurst and Waxman, 2003; Lapinskas et al., 2004). The weak activity of MBP in PPAR transactivation assays was surprising given that DBP, like DEHP and BBP, acts like a typical PP in the rat liver (Corton et al., 1996; Marsman, 1995) and effects in the mouse liver induced by DBP as well as DEHP and DINP are dependent on PPAR (Anderson et al., 1999; Lapinskas et al., 2004; Valles et al., 2003; Ward et al., 1998). Although treatment with MBP induced the same effects as DBP in the juvenile rat testis (Foster et al., 1981; Oishi and Hiraga, 1980) indicating that MBP is the proximate metabolite, no studies have been carried out in which MBP metabolites were comprehensively examined for toxic effects in the testis as well as abilty to activate PPARs. Candidate metabolites include MBP-glucuronide, a dominant MBP metabolite in adult rats (Foster et al., 1983; Tanaka et al., 1978) and omega and omega-1 products of MBP (Tanaka et al., 1978). It is also possible that a metabolite of DBP or DBP itself activates PPARs indirectly, i.e., by releasing a lipid activator of PPAR as discussed above a process which would not necessarily be detectable using transactivation assays. The fact that two out of six phthalates are outliers in this comparison weakens the hypothesis that PPARs are generally involved in phthalate-induced male reproductive tract defects.
Regulation of Sex Hormone Synthesis and Catabolism Genes by Phthalate Esters
As discussed above, BBP, DBP, and DEHP decrease testosterone levels in the fetus and in the young male rat. Consistent with this, genes involved in testosterone biosynthesis were uniformly down-regulated by DBP exposure in the fetal testis (Barlow et al., 2003; Shultz et al., 2001; Thompson et al., 2004) (Fig. 2A). The genes include Scavenger Receptor Class B, type 1 (SR-B1) and steroidogenic acute regulatory protein (StAR) (Barlow et al., 2003; Shultz et al., 2001). SR-B1 mediates the selective uptake of cholesterol esters from high-density lipoproteins, and StAR carries cholesterol from the outer to the inner mitochondrial membrane. The peripheral benzodiazepine receptor (PBR), which also carries cholesterol into the mitochondria was down-regulated by WY-14,643 and DEHP in adult mouse testes in a PPAR-dependent manner (Gazouli et al., 2002). In the fetal rat testis, PBR mRNA was up-regulated by DBP but by immunohistochemistry the protein was decreased in Leydig cells (Lehmann et al., 2004). Leydig cell mitochondria isolated from fetuses exposed to DBP in utero, exhibited decreased uptake of cholesterol supporting altered cholesterol handling for the decreases in testosterone synthesis (Thompson et al., 2004). Other genes involved in steroid biosynthesis that were down-regulated by DBP included P450 side chain cleavage enzyme (P450scc) thought to be the limiting enzymatic step in testosterone biosynthesis, 3?-hydroxysteroid dehydrogenase (3?-HSD) and CYP17 (Barlow et al., 2003; Shultz et al., 2001). Uniform decreases in the expression of these steroidogenic genes especially in a dose-dependent manner (Lehmann et al., 2004) is consistent with decreased levels of testicular testosterone.
The nuclear receptor steroidogenic factor 1 (SF-1) plays a prominant role in the development and differentiation of steroidogenic tissues and controls the expression of steroidogenic enzymes and cholesterol transporters required for steroidogenesis (Val et al., 2003). INSL3 (Emmen et al., 2000) and receptors for follicle-stimulating and leutinizing hormones (Val et al., 2003) are also regulated by SF-1. Genes altered by DBP exposure including SR-B1, StAR, P450scc, CYP17 and 3?-HSD, Type II (Barlow et al., 2003; Shultz et al., 2001) require SF-1 for basal promoter activity and for cyclic AMP induction. The hypothesis that phthalate ester exposure leads to suppression of SF-1 activity was tested in the Leydig cell line MA-10 in which expression of a reporter gene linked to the promoters of SR-B1, StAR, and CYP17 genes was measured after MBP exposure (Thompson et al., 2004). However, MBP had no effect on reporter gene expression. Although this finding indicates that phthalates inhibit steroidogenic gene expression independently of effects on SF-1 activity additional monoesters should be tested in this system before any definitive conclusion can be made of SF-1 involvement. The mechanism by which these genes are down-regulated by DBP and whether other phthalates target the same genes requires further study.
Phthalate esters along with other PP, also alter the expression of testosterone and estrogen metabolism genes (Figs. 2A and 2B). Among these, 5-reductase which converts testosterone to the more potent androgen, dihydrotestosterone was up-regulated in the prepubertal rat testis by DEHP (Kim et al., 2003). Multiple cytochrome P450 family members which hydroxylate testosterone were altered by PP exposure but in ways that do not clearly indicate a role in decreasing testosterone levels. CYP2C7 and CYP2C11 were down-regulated while CYP3A2 was up-regulated in the male rat liver (Corton et al., 1998; Fan et al., 2004-a,b). CYP2C11 and CYP3A were both up-regulated in the testes (Kim et al., 2003). WY-14,643 increased expression of fatty acyl-CoA enzyme A:testosterone acyl transferase in the liver (Xu et al., 2001). The down-regulation of CYP2C11 by a PP required PPAR; the site within the CYP2C11 promoter required for down-regulation by PP was identified and likely interacts with transcription factors other than PPAR implying that PPAR down-regulates CYP2C11 through an indirect mechanism (Ripp et al., 2003).
Serum estrogen levels were increased in male rats after DEHP exposure (Akingbemi et al., 2004; Eagon et al., 1994) and after exposure to a number of PP (Biegel et al., 2001; Liu et al., 1996). Phthalates alter the expression of many estrogen metabolizing enzymes (Fig. 2B). Aromatase exhibits both up-and down-regulation by PP depending on the tissue and compound (Biegel et al., 2001; Kim et al., 2003; Lovekamp and Davis, 2001; Lovekamp-Swan et al., 2003; Mu et al., 2001; Rubin et al., 2000; Toda et al., 2003). Genes that decrease estrogen levels were increased after PP exposure including 17?-hydroxysteroid dehydrogenase, type IV (Corton et al., 1996) and fatty acyl-CoA enzyme A:estradiol acyl transferase (Xu et al., 2001). However, these changes may be offset by decreases in the estrogen hydroxylases CYP1B1 after exposure to DBP, DEHP, and BBP in the testis (Seo et al., 2004) or CYP2C11 (Corton et al., 1998) and estrogen sulfotransferase (Fan et al., 2004-b) after WY, GEM, or DBP exposure in rat livers. A handful of these genes involved in testosterone or estrogen metabolism are directly regulated by PPAR either in the liver or testis as determined by studies in PPAR-null mice. These include 17?-HSD, type IV (Corton et al., 1996) and the CYP3A2 ortholog Cyp3a11 (Fan et al., 2004-b). These results point to a role for PPAR in regulating expression of some sex steroid metabolism genes upon phthalate exposure in the adult animal. Although these studies help us appreciate the gene targets of phthalates involved in regulation of steroid metabolism, more work is needed to determine the mechanism by which phthalates alter the expression of these genes, whether these genes are also regulated in parallel at the protein level and the contribution of these collective changes to phthalate toxicity in the postnatal testis.
PPAR and Phthalate-Induced Testicular Toxicity
The role of PPAR in phthalate-induced developmental and testicular toxicity has been determined directly in wild-type and PPAR-null mice in two studies. In the first study (Peters et al., 1997), pregnant dams were dosed with DEHP at GD 8 and GD 9 (before development of the male reproductive tract) and malformations of the fetuses were evaluated on GD 10 and GD 18. DEHP exposure decreased crown-rump length and increased the incidence of open neural tubes (failure of the hind- and mid-brain to close) in both DEHP-treated wild-type and PPAR-null mice compared to controls, demonstrating that DEHP-induced malformations are PPAR-independent. However, this study cannot be used to determine the role for PPAR in phthalate-induced male reproductive tract defects from in utero exposure, because DEHP treatment did not occur during the critical window of development of the male reproductive tract and to our knowledge, the rat model of in utero male reproductive tract defects has not been successfully recapitulated in mice.
In the second study (Ward et al., 1998), male wild-type and PPAR-null mice were fed a diet containing 12,000 ppm DEHP and lesions in the liver and testis were examined. In the liver all of the expected effects of DEHP were dependent on PPAR. After four and eight weeks of exposure, testis from wild-type mice exhibited mild or moderate toxic effects including focal tubular degenerative lesions, decreased spermatogenesis, and giant cells within the epididymis. In sharp contrast, the testis from PPAR-null mice at these time points were predominantly normal except for a few tubules that lacked normal indicators of spermatogenesis. PPAR-null mice at 24 weeks exhibited moderate testicular effects. However, all of the wild-type mice were sacrificed at 12–16 weeks due to toxicity when there was severe testicular atrophy making it impossible to directly compare the effects in the two strains at the 24 week time point. The authors concluded that DEHP effects in the testis were the result of both PPAR-dependent and -independent mechanisms. This study indicates that PPAR determines the timing and severity of testicular toxicity by DEHP. To better characterize the role of PPAR in phthalate-induced testicular toxicity, additional studies using other phthalates should be carried out in wild-type and PPAR-null mice. Given that PPAR? is expressed in the testis, PPAR?-null mice could also be used to determine any role played by this receptor in phthalate-induced testicular toxicity. Determining a role for PPAR will require construction of testis-specific nullizygous mice as the PPAR-null mutation is embryonically lethal (Barak et al., 1999).
Cross-Talk between PPAR and Nuclear Receptor Signaling: Potential Impact on the Male Reproductive Tract
Several studies have demonstrated ‘cross-talk’ between PPAR and other nuclear receptors. Cross-talk occurs when signaling pathways other than the primary pathway are activated or repressed through interactions with components of the primary pathway. PPAR-induced interference of other nuclear receptor pathways occurs through competition between PPARs and other receptors for binding to (1) the same DNA response element or (2) a common heterodimerization partner (Fig. 3). In the Supplementary Materials we examine cross-talk between PPARs and nuclear receptors for estrogen, thyroid hormone, and retinoic acid that play roles in the development or homeostasis of the testis. Many of these studies supporting interactions between PPAR and other nuclear receptors were either performed with PP other than phthalates in in vitro cell models allowing only conjecture as to how PPAR-phthalate interactions may disrupt male reproductive tract function in vivo. Further work is needed before these proposed mechanisms should be considered to play a role in phthalate-induced effects in the male reproductive tract.
Conclusions
In this review we have assessed the involvement of PPAR subtypes in phthalate-induced effects on the male reproductive tract. Much of the evidence supporting a role for PPARs is correlative. A high level of PPAR activation and testicular toxicity requires metabolic conversion of the diester to the monoester. PPAR subtypes are activated by structurally diverse phthalate monoesters. Phthalate diesters activate weakly or not at all. PPAR and PPAR are usually more sensitive to monoester activation than PPAR?, and mouse PPARs are more sensitive than human PPARs. PPARs, principally the and ? subtypes are expressed in neonatal or adult Sertoli and Leydig cells, the target cells of phthalate ester effects. The testicular toxicity of DEHP occurred later and with diminished severity in PPAR-null mice, indicating PPAR acts as a modifier gene regulating the timing and severity of toxicity. There is a correlation between the ability of four of six diesters to induce reproductive tract changes and the corresponding monoesters to activate PPAR subtypes. Phthalates alter the expression of genes encoding sex steroid metabolizing enzymes and enzymes involved in testosterone biosynthesis, some in a PPAR-dependent manner. PPARs interact with and down-regulate the activities of other nuclear receptors that play roles in the developing testis including receptors for estrogen, thyroid hormone, and retinoic acid, leading to plausible mechanisms by which PPAR activation by phthalates could alter testicular function in vivo.
Other evidence exists to argue for little, if any participation of PPARs in phthalate-induced testicular effects. Although phthalates and PP can activate PPARs, only the phthalates are generally considered testicular toxicants. However, it is not unprecedented that structurally distinct classes of nuclear receptor ligands (e.g., ER ligands) have unique properties as agonists, partial agonists, or antagonists depending on receptor-ligand conformations and subsequent interactions with tissue-specific co-activators/co-repressors (McDonnell et al., 2002). With this in mind, it is interesting to note that many genes normally up-regulated by PP in the adult rat liver were down-regulated by DBP in the fetal testis and include those that are known targets of PPAR including fatty acid ?–oxidation genes (Shultz et al., 2001). Phthalate effects in the adult testis can occur rapidly (within 2 h), likely before any phenotypic effects of PPAR-mediated gene regulation would be observed. Two phthalates, DBP and DINP, do not neatly fit into a relationship of inducing male reproductive tract malformations while the corresponding monoester induces PPAR activation. DBP is a strong male reproductive toxicant but MBP only weakly activates PPARs; DINP is a weak toxicant but MINP is a moderately strong PPAR activator. The fact that phthalate-induced testicular toxicity occurs in species that do not respond to PP-induced liver effects has been used to invoke a PPAR-independent mechanism, but comprehensive information on PPAR expression in the testis of responsive vs. nonresponsive species is lacking.
One way to rationalize these seemingly disparate findings is to invoke a mechanism that at least in the neonatal testis, includes both PPAR-dependent and -independent events leading to testicular toxicity after phthalate exposure. Future investigations should be designed to directly determine the role of PPARs as potential mediators of phthalate effects. Resolving a role for PPARs will be expedited by a comprehensive determination of PPAR expression in the developing testis and testing the effects of multiple phthalates in established mouse models nullizygous for each PPAR gene during male reproductive tract development.
SUPPLEMENTARY DATA
Supplementary data are available online at www.toxsci.oupjournals.org.
ACKNOWLEDGMENTS
The authors thank Drs. Paul Foster and Kevin Gaido for a critical review of the manuscript and the Phthalate Esters Panel of the American Chemistry Council for funding to produce this review. We apologize to colleagues whose papers were not cited due to space limitations.
REFERENCES
Akingbemi, B. T., Youker, R. T., Sottas, C. M., Ge, R., Katz, E., Klinefelter, G. R., Zirkin, B. R., and Hardy, M. P. (2001). Modulation of rat Leydig cell steroidogenic funtion by di(2-ethylhexyl)phthalate. Biol. Reprod. 65, 1252–1259.
Biegel, L. B., Liu, R. C., Hurtt, M. E., and Cook, J. C. (1995). Effects of ammonium perfluorooctanoate on Leydig cell function: in vivo, in vitro, and ex vivo studies. Toxicol. Appl. Pharmacol. 134, 18–25.
Biegel, L. B., Hurtt, M. E., Frame, S. R., O'Connor, J. C., and Cook, J. C. (2001) Mechanisms of extrahepatic tumor induction by peroxisome proliferators in male CD rats. Toxicol. Sci. 60, 44–55.
Nichols, J. S., Parks, D. J., Consler, T. G., and Blanchard, S. G. (1998). Development of a scintillation proximity assay for peroxisome proliferator-activated receptor gamma ligand binding domain. Anal. Biochem. 257, 112–119.
Parvinen, M. (1982). Regulation of the seminiferous epithelium. Endocr. Rev. 3, 404–417.
Peters, J. M., Lee, S. S., Li, W., Ward, J. M., Gavrilova, O., Everett, C., Retiman, M. L., Hudson, L. D., and Gonzalez, F. J. (2000). Growth, adipose, brain, and skin alterations resulting from tageted disruption of the mouse peroxisome proliferator-activated reeceptor beta (delta). Mol. Cell Biol. 20, 5119–5128.
Valles, E. G., Laughter, A. R., Dunn, C. S., Cannelle, S., Swanson, C. L., Cattley, R. C., and Corton, J. C. (2003). Role of the peroxisome proliferator-activated receptor alpha in responses to diisononyl phthalate. Toxicology 191, 211–225.(J. Christopher Corton*,1 )