In Utero Exposure to the Antiandrogen Di-(2-Ethylhexyl) Phthalate Decreases Adrenal Aldosterone Production in the Adult Rat¹

Abstract

We previously reported that in utero exposure of the male fetus to the plasticizer di-(2-ethylhexyl) phthalate (DEHP) resulted in decreased circulating levels of testosterone in the adult without affecting Leydig cell numbers, luteinizing hormone levels, or steroidogenic enzyme expression. Fetal exposure to DEHP resulted in reduced mineralocorticoid receptor (MR; NR3C2) expression in adult Leydig cells. In the present studies, treatment of pregnant Sprague-Dawley dams from Gestational Day 14 until birth with 20, 50, 100, 300, or 750 mg kg⁻¹ day⁻¹ of DEHP resulted in significant sex-specific decreases in serum aldosterone but not corticosterone levels at Postnatal Day 60 (PND60) but not at PND21. There was no effect on circulating levels of potassium, angiotensin II or adrenocorticotropin hormone (ACTH). However, there was reduced expression of AT receptor Agtr1a, Agtr1b, and Agtr2 mRNAs. The mRNA levels of proteins and enzymes implicated in aldosterone biosynthesis were not affected by in utero DEHP treatment except for Cyp11b2, which was decreased at high (≥500 mg kg⁻¹ day⁻¹) doses. The data presented herein, together with our previous observation that aldosterone stimulates testosterone production via an MR-mediated mechanism, suggest that in utero exposure to DEHP causes reduction in both adrenal aldosterone synthesis and MR expression in Leydig cells, leading to reduced testosterone production in the adult. Moreover, these results suggest the existence of a DEHP-sensitive adrenal-testis axis regulating androgen formation.

Introduction

Di-(2-ethylhexyl) phthalate (DEHP), a plasticizer commonly used by industry to add flexibility to polyvinyl chloride products, is highly prevalent in the environment and is present in a wide variety of consumer products, including medical devices such as intravenous tubing, catheters, and dialysis bags [1–4]. DEHP, which has been shown to have antiandrogenic properties, exerts its effects through its metabolite mono-2-ethylhexyl phthalate (MEHP), which is formed after cleavage of one of its side chains by lipases in the gut, and is 10 times more active than the parent compound [1, 5]. In humans, in utero exposure to DEHP has been linked to cryptorchidism [6], decreased anogenital distance [7], and reduced semen quality [8, 9] and is thought to contribute to decreased fertility by acting synergistically with other endocrine disruptors [10]. DEHP and its metabolites have been found in amniotic fluid [11], umbilical cord blood [12], and milk [13], suggesting prenatal and perinatal exposures. In utero exposure to DEHP in rodents has been shown to cause incomplete organogenesis of androgen-dependent tissues by disrupting androgen formation by fetal Leydig cells [14–17].

We recently reported that pregnant Sprague-Dawley rats gavaged with DEHP from Gestational Day (GD14) until birth resulted in male offspring with reduced circulating levels of testosterone in the adult [17, 18] without a reduction in Leydig cell numbers, steroidogenic protein and enzyme levels, or circulating luteinizing hormone (LH) concentrations. These results suggested that the long-term effects of DEHP may be independent of the classical steroidogenic pathway [17, 18] and identified the Leydig cell mineralocorticoid receptor (MR; NR3C2) as a gene target affected by DEHP exposure [18].

Aldosterone and corticosterone are produced by the adrenal gland and bind with similar affinities to MR. Selective activation of MR in tissues such as Leydig cells is achieved through expression of HSD11B1 and HSD11B2, enzymes that inactivate glucocorticoids [19]. Thus, in Leydig cells, aldosterone but not corticosterone activates MR, leading to stimulation of testosterone production [20]. Aldosterone is produced in response to the octapeptide angiotensin II (AT) and increased potassium levels and is known to act on the distal nephron to retain sodium and excrete potassium [21]. The renin-AT-aldosterone system (RAAS) regulates the production of AT II, and AT II binds to the AT receptors AGTR1A and AGTR1B located in the zona glomerulosa (ZG) of the adrenal gland. AT receptor activation triggers multiple signaling pathways and results in the stimulation of aldosterone synthesis and in adrenal cell growth and proliferation [22].

We report herein that in utero exposure to DEHP leads to a significant reduction of circulating aldosterone levels in the adult rat. In search of the mechanism underlying this long-term effect of DEHP, we examined RAAS components and found that adrenal AT receptor expression was reduced in a dose-dependent manner. These results suggest that in utero exposure to DEHP results in a decrease in the adult adrenal AT receptors and thus in reduced stimulation of aldosterone biosynthesis. In light of our previous study that showed that in utero exposure to DEHP decreased MR expression in adult Leydig cells [18], the present findings further suggest that in utero DEHP exposure results in a combination of decreased aldosterone production by the adrenal and MR expression in Leydig cells and that these lead to reduced androgen formation by the testis without an effect on the steroidogenic pathway.

Materials and Methods

DEHP Treatment and Animal Care

Timed pregnant Sprague-Dawley rats were purchased from Charles River Laboratories and gavaged daily with corn oil or with 20, 50, 100, 300, 500, 750 or 950 mg kg⁻¹ day⁻¹ DEHP (Sigma-Aldrich) from GD14 until parturition (Postnatal Day [PND] 0). The animals were weighed every 2 days, and the doses were adjusted accordingly. Male offspring were euthanized at PND60. Adrenals, liver, and kidney were collected and either snap frozen in liquid nitrogen or fixed in formaldehyde. Animals were handled according to protocols approved by the Georgetown and McGill University Animal Care and Use Committees.

Serum Measurements

Male offspring of pregnant dams treated with corn oil or with DEHP were euthanized at PND60 by CO₂. Blood was collected by percutaneous cardiac puncture. The serum was separated and shipped to Analytics Incorporated for measurements of electrolytes; low-density lipoprotein (LDL), high-density lipoprotein (HDL), and total cholesterol; triglycerides; AT II; ACTH; corticosterone; and aldosterone. Chemistries were performed on the Hitachi 717 Chemistry Analyzer (Roche Diagnostics) with reagents obtained from Randox. Corticosterone and estradiol were measured using double antibody radioimmunoassay (RIA) with reagents obtained from MP Biomedicals. Aldosterone was measured using a solid-phase coated-tube RIA with reagents obtained from Siemens Healthcare Diagnostics according to the manufacturer's instructions. Testosterone was measured as previously described [18].

Quantitative Real-Time PCR Analysis

Adrenals, liver, and kidney cortex were collected at PND60 and snap frozen in liquid nitrogen. Tissue extraction and quantitative PCR were performed as previously described [18] and carried out on the LightCycler 480 Real-Time PCR System (LC480; Roche Diagnostics). GAPDH was used as an endogenous control to normalize the gene targets obtained from one male offspring from each of three litters per treatment, with each sample processed in triplicate. The comparative Ct method was used to express the results relative to the reference gene. Supplemental Table S1 (all Supplemental Data are available online at www.biolreprod.org) contains a list of the TaqMan probes used.

Histology, Hematoxylin and Eosin, and Red Oil O Staining

Adrenal glands from the PND60 male offspring of pregnant dams treated with vehicle or DEHP (100–750 mg kg⁻¹ day⁻¹) were fixed in 4% paraformaldehyde and embedded in paraffin. Sections (4 μm) were stained with hematoxylin and eosin. For red oil O staining, PND60 adrenals were placed in cryomolds filled with Tissue-Tek CRYO-OCT and snap frozen in precooled (−80°C) isopentane. Samples were shipped to Cytochem (Montreal, QC) for staining.

Immunoblot Analysis

Adrenal glands were homogenized in RIPA buffer composed of 10 mM Tris, 150 mM NaCl, 1% triton X-100, 1% deoxycholate, 0.1% SDS, and 5 mM EDTA. Total protein was quantified using the bicinchoninic acid kit (Thermo Scientific). Protein (30 μg) was diluted into Laemmli buffer, loaded onto a 4–20% Tris-glycine gel, and transferred to a PVDF membrane. The membrane was incubated with antiserum against AT1 (AB15552-200UL; Millipore), AT2 (AB15554-200UL; Millipore), and GAPDH (2275-PC-100; Trevigen). The AT1 antibody targets the N terminus of the human AGTR1 receptor, and in the rat, both the AGTR1A and the AGTR1B receptors are recognized by this antibody [23]. Bound protein antibodies were detected with a secondary antibody coupled to horseradish peroxidase (goat anti-mouse, 554002, or goat anti-rabbit, 554021; DB Pharmingen) and ECL-enhanced chemiluminescence (GE Healthcare). Images were captured using the LAS-4000 gel documentation system (Fujifilm) and quantified using Multi Gauge v3.0 (Fujifilm).

Bisulfite Sequencing of Agtr1a, Agtr1b, and Atrap Genes

CpG islands searcher software [24] was used to identify CpG-rich regions near promoter areas or first exons within target genes with a GC content of at least 55% and an observed CpG/expected CpG ratio of 0.65. Briefly, genomic DNA was extracted from adrenal glands using DNeasy kit (Qiagen) according to the manufacturer's instructions. DNA (2 μg) was bisulfite treated using the EpiTect Bisulfite kit (Qiagen) according to the manufacturer's instructions, followed by nested PCR amplification. The PCR mix consisted of 0.2 mM dNTP, 1.5 mM MgCl₂, 0.2 μM primers, 1 U PlatinumTaq polymerase (Invitrogen), and 2 μl of bisulfate-treated DNA in 20-μl total volume. The PCR conditions for first PCR round were preincubation at 94°C (5 min), 30 cycles of 30 sec 94°C denaturation, 30 sec annealing at temperatures shown in Supplemental Table S2, 2 min extension at 72°C, and 5 min incubation at 72°C. Then 2 μl of first PCR were nested, and second PCR was carried out using the following conditions: preincubation at 94°C (5 min), 35 cycles of 30 sec 94°C denaturation, 30 sec annealing at temperatures shown in Supplemental Table S2, 1 min extension at 72°C, and 5 min incubation at 72°C. After PCR, 20 μl of PCR reaction were mixed with loading dye and run on a 2% agarose gel containing ethidium bromide. Appropriate bands were cut, gel purified, and ligated in pGEM-T Easy Vector (Promega) for sequencing. Supplemental Table S2 shows a list of the primers used for nested PCR.

Statistical Analyses

Statistical analysis of the data using one-way analysis of variance followed by Dunnett's posttest was performed with Prism v4.02 (GraphPad, Inc.). For all experiments, the experimental unit was the pregnant dam, and the responses of male offspring from at least three dams (n = 3 pregnant dams) for each treatment were examined independently. For serum measurements, the average from two offspring was performed with the pregnant dam being the experimental unit (n = 1).

Results

In Utero Exposure to DEHP Decreases Aldosterone Serum Levels in the Adult

To establish the lowest observed adverse effect level of DEHP on aldosterone and testosterone production, pregnant Sprague-Dawley rats were gavaged from GD14 until birth with oil or with 20, 50, 100, 300, or 750 mg kg⁻¹ day⁻¹ of DEHP, and serum samples were collected at PND21 and PND60. Circulating testosterone and aldosterone concentrations decreased significantly to about 50% of control at each of the 100, 300 and 750 mg kg⁻¹ day⁻¹ doses (Fig. 1, A and B). Reduction of circulating aldosterone levels had no effect on sodium, potassium, chloride, or calcium levels (Fig. 2, B, D–F). Serum corticosterone did not change significantly at DEHP doses that affected aldosterone or testosterone levels (Fig. 1C). The reductions in circulating aldosterone levels seen in the PND60 adult offspring (Fig. 1B) were not seen at PND21 (Fig. 1D), and, as in the adult (Fig. 1C), there were no significant changes in corticosterone levels at PND21 (Fig. 1E). Circulating aldosterone levels were increased in the female offspring that were exposed in utero to 300 mg kg⁻¹ day⁻¹ (Fig. 3A), while estradiol levels were decreased to about 50% of controls (Fig. 3B).

In Utero Exposure to DEHP Decreases Adult Adrenal AT Receptors

To address the possible mechanism by which in utero DEHP exposure results in reduction in serum aldosterone, we first determined the effect of in utero exposure on selected components of the RAAS and on known stimulants of aldosterone biosynthesis. No significant changes in serum levels of AT II, potassium, or ACTH were seen on PND60 after in utero exposure to 100–750 mg kg⁻¹ day⁻¹ DEHP (Fig. 2, A–C). No change in mRNA levels of the liver-made angiotensinogen (Agt) was seen (Fig. 2G). Similarly, the kidney-made renin (Ren1) mRNA levels did not change significantly (Fig. 2H). Histological analysis of the adrenals did not reveal obvious change in morphology between controls and DEHP-exposed adult tissues (Supplemental Fig. S1A). Adrenal weight was significantly reduced only following exposure to 750 mg DEHP kg⁻¹ day⁻¹ (Supplemental Fig. S1B). We then determined the effect of DEHP on the AT receptors in the adrenal (Fig. 4) and kidney (Supplemental Fig. S2). In the adrenal, Agtr1a mRNA levels significantly decreased at DEHP doses above 300 mg kg⁻¹ day⁻¹ (Fig. 4A), Agtr1b mRNA levels significantly decreased at all DEHP doses used (Fig. 4B), and Agtr2 mRNA was decreased significantly at the 750 and 950 mg DEHP kg⁻¹ day⁻¹ doses (Fig. 4C). Interestingly, the decreases in the AT receptors observed in adrenal tissue were not observed in the kidney (Supplemental Fig. S2); Agtr1a, Agtr1b, and Agtr2 mRNA levels were not significantly affected by DEHP exposures (Supplemental Fig. S2, A–C). Agtrap is a negative regulator of AGTR1 receptor signaling that modulates the signaling induced by AT II [25]. Agtrap mRNA levels were significantly decreased by all DEHP doses in both the adrenals and the kidney, with the effect on the adrenal more pronounced (Fig. 4D and Supplemental Fig. 2D). As detected by immunoblot analysis, angiotensin receptor protein levels, detected using an antibody that recognizes both AGTR1A and AGTR1B receptors, were significantly decreased at 300 and 750 mg kg⁻¹ day⁻¹ but not at 100 mg kg⁻¹ day⁻¹, while AT2 protein levels were significantly decreased at all doses tested (Fig. 4, E and F).

DNA methylation of AT receptor promoter regions induced by DEHP exposure could lead to the decreased mRNA levels observed. Bisulfate sequencing of CpG islands in promoter areas of the Agtr1a and Agtr1b genes showed no change in methylation patterns (Supplemental Figs. S3 and S4). Similarly, bisulfate sequencing of a CpG island in the Agtrap gene showed no change in methylation pattern (Supplemental Fig. S5).

In Utero Exposure to DEHP Reduces Adrenal Cyp11b2 Expression in the Adult

To determine mRNA levels of the proteins involved in the biosynthesis of aldosterone in the adrenal gland, adrenal gland mRNA was extracted from offspring of pregnant dams exposed in utero to 100–950 mg kg⁻¹ day⁻¹ DEHP. There were no significant changes in Star, Tspo, Cyp11a1, Hsd3b1, and Cyp11b1 mRNA levels (Fig. 5, A–F). The mRNA levels of Cyp11b2, the gene for aldosterone synthase, an enzyme involved in the last steps of aldosterone biosynthesis, decreased significantly at DEHP doses 500 mg kg⁻¹ day⁻¹ and higher (statistically significant at 500 and 950 mg kg⁻¹ day⁻¹; Fig. 5G).

In Utero Exposure to DEHP Up-Regulates Genes Involved in the Biosynthesis of Adrenal Cholesterol

We hypothesized that a decrease in aldosterone production could be due to reduced cholesterol bioavailability. Total cholesterol, LDL, HDL, and triglycerides were unaffected by in utero exposure to DEHP (Fig. 6, A–D). In adrenals, mRNA levels of the cholesterol uptake receptor Ldlr increased significantly at all DEHP doses used (Fig. 6E), but levels of Scarb1 (SR-BI), involved in mediating HDL cholesterol uptake, were not affected by any dose (Fig. 6G). In heart, Ldlr decreased significantly at 300 mg kg⁻¹ day⁻¹ and increased at 750 mg kg⁻¹ day⁻¹ (Fig. 6F), while Scarb1 was not significantly reduced (Fig. 6H).

The mRNA levels of Hmgcr (HMG-CoA reductase) and Hmgcs1 (HMG-CoA synthase) were up-regulated by approximately 2-fold in the adrenal at all DEHP doses tested (Fig. 7, A and C). Insulin-induced gene 1 (Insig1), which plays an important role in cholesterol biosynthesis, was also up-regulated in the adrenal at all DEHP concentrations (Fig. 6E). In contrast to the adrenal, Hmgcr, Hmgcs1, and Insig1 mRNA levels in the heart were unaffected by in utero DEHP exposures (Fig. 7, B, D, and F).

Red oil O staining of adrenal glands showed dose-dependent differences in lipid droplet accumulation (Fig. 7G). The control group had a homogeneous distribution of lipids in the ZG, with few focal accumulations. Animals exposed to 100–300 mg kg⁻¹ day⁻¹ showed increased accumulations of lipid droplets confined to the ZG. At 750 mg kg⁻¹ day⁻¹, lipid droplet accumulation was seen both in the ZG and zona fasciculata (ZF).

Discussion

This study began from the observation that exposure of pregnant dams to DEHP from GD14 until birth resulted in a 50% reduction of circulating testosterone levels in the adult male offspring (PND60) [17, 18]. This window of exposure was chosen to correspond to the first peak of testosterone driving male sexual development. During the in utero treatment period, fetal Leydig cells are directly exposed to DEHP. However, DEHP and its metabolites clear by 3 days following their administration [26], and therefore the adult Leydig cell population, which differentiates long after DEHP clearance, is not exposed directly to DEHP and its metabolites. This suggests that in utero exposure to DEHP targets either the cells that give rise to the adult Leydig cells or the function of other cells or tissues that in some way negatively impact on testosterone production by the adult cells.

It is important to note that in utero exposure to another phthalate, the di(n-butyl) phthalate, was also shown to reduce testosterone levels in the adult, suggesting the possibility of a common endocrine-disrupting pathway involved in mediating the long-term effects of phthalates [27]. This raises public health concerns because phthalates could act additively and/or synergistically [28, 29], thus exacerbating the effects [30].

Although a direct effect of phthalates on fetal Leydig cell proteins and enzymes involved in steroidogenesis has been shown [31–33], the decrease in circulating testosterone seen in the adult rat following in utero to DEHP has been shown to occur despite near normal Leydig cell numbers, steroidogenic protein and enzyme gene expression, and circulating LH levels [17, 18]. These results suggested that in utero DEHP exposure impairs adult androgen formation by a mechanism that is independent of the classical steroid biosynthetic pathway. The rationale for the studies conducted herein was to search for such a mechanism.

We previously reported that in utero exposure to DEHP decreases MR expression in adult Leydig cells and thus affects aldosterone-regulated testosterone formation [18]. Based on these results, we evaluated whether DEHP exposure also affects adrenal aldosterone formation. During the in utero DEHP treatment, the fetal adrenal gland was exposed to concentrations that are close to or higher than human exposures observed with hemodialysis and total parenteral nutrition in infants, which can reach 10–20 mg kg⁻¹ day⁻¹ [4, 34]. We show herein that in utero exposure to 100 mg kg⁻¹ day⁻¹ DEHP reduced by 50% circulating aldosterone levels compared to control. These reductions were comparable in magnitude to the reductions in serum testosterone levels [17, 18]. Interestingly, the decrease in aldosterone was not sufficient to alter circulating levels of electrolytes at PND60. Similarly, AT II levels were not changed, suggesting that renal perfusion was not affected by the decreased aldosterone levels. DEHP had no effect on aldosterone levels at the 50 mg kg⁻¹ day⁻¹ dose and below, which correlated with our previous testosterone level measurements [17, 18]. The adrenal weights decreased, reaching statistical significance at 750 mg DEHP kg⁻¹ day⁻¹. Interestingly, adult body weight at PND60 was not affected by DEHP, suggesting that the development of the adrenal gland was affected specifically at higher doses [17]. We observed sex-specific long-term effects in steroid levels in response to DEHP exposure. In the female offspring, aldosterone levels were increased, although estradiol levels were reduced to half that in controls. To gain further understanding of the sex-specific effects of DEHP, a characterization of the aldosterone pathway similar to that performed herein must be carried out.

Aldosterone has been shown to induce MR activation in Leydig cells and thus to stimulate testosterone production by an MR-mediated mechanism [20]. Moreover, treatment with the MR antagonist spironolactone has been reported to decrease androgen formation [35]. Thus, decreased serum aldosterone levels, together with reduced MR expression in Leydig cells, both induced by in utero exposure to DEHP, could lead to the reduced testosterone production seen in the adult rat.

In an effort to identify the mechanism responsible for the reduction in aldosterone levels, we examined the effect of in utero exposure to DEHP on the primary stimulants of aldosterone biosynthesis, receptors and proteins involved in AT II signaling, the mineralocorticoid biosynthesis pathway, sources of extracellular cholesterol, and key enzymes in the de novo synthesis of cholesterol. The main stimulants for aldosterone secretion are AT II or potassium and, in extreme cases, such as profuse blood loss, ACTH. Unlike the sex hormones, the feedback loop regulating aldosterone secretion is not mediated primarily by its nuclear receptor, MR, but rather by the physiological effects of aldosterone. We found serum levels of AT II, potassium, and ACTH to be unaffected by DEHP, suggesting that aldosterone levels were sufficient to maintain the intravascular volume and, as mentioned above, potassium equilibrium. This, together with the observation that mRNA levels of the liver-made angiotensinogen and the kidney-made renin were normal, suggested that the cause for decreased aldosterone synthesis in response to DEHP treatment was most likely located in the adrenal gland. We measured mRNA levels of Agtr1a, Agtr1b, and Agtr2 and found them to be significantly decreased, to almost half of control levels, starting at 500 mg kg⁻¹ day⁻¹ DEHP dose. Agtr1b, however, was reduced significantly even at the lowest DEHP dose used. This effect of DEHP on Agtr1b paralleled its effect on circulating aldosterone and testosterone levels. In humans, there is only one AGTR1, which differs from rodents in which AGTR1A is in the ZG and medulla and AGTR1B is found exclusively in the ZG [36–38]. The AT receptor mRNA decrease correlated with reduced protein levels, suggesting that posttranslational regulation was unlikely. At the 100 mg kg⁻¹ day⁻¹ DEHP dose, there was no significant effect on AGTR1 protein levels, which could be due to the fact that the antibody used detects both AGTR1A and AGTR1B proteins. AT II-mediated stimulation of the AGTR1A and AGTR1B receptors in the ZG increases substrate availability for aldosterone formation [39], adrenal cell growth, and proliferation [22]. The decline in the expression of Agtr1b and thus its decreased signaling could be involved in the observed reduction in serum aldosterone.

Another possible explanation for the decrease in aldosterone serum levels in response to in utero DEHP exposure could be an effect on the mineralocorticoid biosynthesis pathway. We measured mRNA levels of the proteins and enzymes involved in aldosterone biosynthesis and found little effect on the mRNA levels of all but one enzyme, Cyp11b2, the final enzyme required for aldosterone biosynthesis [40]. CYP11B2 is selectively expressed in the adrenal ZG, and is up-regulated by AGTR1 receptor signaling [41]. The reduction in Cyp11b2 could explain the decrease in circulating aldosterone levels but only in males exposed in utero to high (500 mg kg⁻¹ day⁻¹) DEHP concentrations.

We then investigated the possibility that DEHP exposure might have reduced the bioavailability of circulating cholesterol, and this in turn could lead to reduced aldosterone formation. Thus, we determined the extracellular sources of cholesterol. In adrenal cells, the preferred cholesterol substrate for aldosterone biosynthesis comes from circulating HDLs, and its uptake into glomerulosa cells is mediated by the SR-BI receptors [42]. Serum measurements of total circulating cholesterol, LDL, and HDLs showed no significant changes. Taken together, these results suggest that there are enough extracellular sources of cholesterol available to support adrenal aldosterone formation.

We next assessed the levels of genes involved in cholesterol uptake and de novo cholesterol biosynthesis pathways. Ldlr mRNA levels were increased, suggesting the import of cholesterol by the cells. In contrast, SR-BI, the preferred pathway of cholesterol import into the glomerulosa cells [43], was unchanged, suggesting that the chronic effects of DEHP are driven by an unknown target involved in the Ldlr gene transcription regulation. An increase in AT II signaling is known to enhance specific LDL binding and uptake [44, 45], and in chronic exposures it up-regulates SR-BI and LDL receptor pathways by increasing gene transcription [46–49]. Similarly, the increase in genes involved in cholesterol uptake after incubation with AT II was also observed in the adrenocortical cell line LRP2PB [50]. Paradoxically, in our study, circulating levels of AT II were unaffected by DEHP exposure, whereas AT receptor mRNA levels decreased. Activation of the AT receptors leads to increased capacity of the adrenals to produce aldosterone by up-regulating de novo cholesterol synthesis and cholesterol uptake [22]. Hmgcr (HMG-CoA), which mediates the rate-limiting step of the de novo cholesterol synthesis, was up-regulated at all DEHP doses used. The up-regulation of cholesterol uptake and de novo cholesterol synthesis pathways would be expected to be associated with increased, not reduced, aldosterone output. Perhaps in utero DEHP exposure might, in some unknown way, result in a reduced cholesterol delivery to the mitochondria, thus triggering a feedback loop that results in the up-regulation of extracellular cholesterol import and de novo synthesis. This hypothesis is supported by the dose-dependent increase in lipid droplet accumulation observed in the ZG and, at the high dose of 750 mg DEHP kg⁻¹ day⁻¹, in the ZF as well. Additionally, expression of AGTR1 has been shown to occur through self-regulation and in response to changes in sodium diet [51]. It is possible that AGTR1 receptors are down-regulated as part of a feedback loop triggered by DEHP. It remains to be elucidated if the decrease in AT receptor mRNA levels is the cause or a consequence of the changes observed in the cholesterol pathways. Interestingly, the mRNA levels of the negative regulator of AGTR1 signaling, Agtrap, were also decreased [52]. An absence of negative modulation of the AGTR1 signaling pathway could lead to AGTR1 down-regulation and thus to reduced aldosterone.

The tissue specificity of the effects of in utero exposure to DEHP on the adrenal is supported by the findings that the liver and kidney mRNA measurements of RAAS pathway components and heart mRNA levels of Hmgcr, Hmgcs1, Insig1, and Ldlr were not changed after in utero exposure to DEHP. This is in agreement with our previous report in which MR and MR-driven genes were decreased in the testis but not in kidney and suggests that DEHP targets the testes and adrenals during the GD14–19 window of fetal exposure [18]. Leydig cells develop in two waves: the fetal and adult populations. In the experimental design used herein, fetal Leydig cells are directly exposed to DEHP. However, these cells disappear soon after birth and are replaced by cells of the adult Leydig cell lineage, long after DEHP has been cleared from circulation. Despite the direct detrimental effects of DEHP on fetal Leydig cell steroidogenic enzyme expression and function [17, 32], the adult Leydig cells could emerge unaffected from the fetal DEHP exposure, and the testis regains its ability to produce androgens although at reduced levels [17, 18]. The adrenal gland is present throughout development. Therefore, disruption of gene expression, possibly epigenetically mediated, could lead to long-term effects on adrenal function. It is likely that the effects initiated by the in utero exposure to DEHP emerge around puberty because reduction in aldosterone formation was not seen at PND21. It is also likely that the reduced mineralocorticoid levels may affect the differentiation process of the adult Leydig cell population. It remains to be seen if aldosterone or testosterone replacement therapy during the gestational stages or in the adult could rescue the effects induced by DEHP.

As with our previous observation of MR levels in Leydig cells exposed to DEHP [18], there appears to be an on/off mechanism for the adrenal-specific decreases of aldosterone and Agtr1b mRNA levels. Such a mechanism is also likely to be the cause of the increased expression of Ldlr, Hmgcr, Hmgcs1, and Insig1. Whether a combination of phthalates or other environmental factors can decrease this threshold remains to be investigated.

In summary, the data presented herein and previously indicate that in utero exposure to DEHP leads to reduced aldosterone formation by the adult adrenal and reduced MR expression in adult Leydig cells [17, 18], resulting in reduced testosterone formation. A schematic representation of the results presented herein in relation to Leydig cell and adrenal development is presented in Figure 8. These data bring forward the possibility that there exists an adrenal-testis axis regulating testosterone formation and perhaps adult Leydig cell differentiation. This axis could react to the environment, stress, exercise, or other stimulants, thereby playing an important role in the regulation of steroid hormone biosynthesis.

Acknowledgments

We thank Mr. Charles Essagian and Mrs. Annie Boisvert for their help with the animal handling and tissue collection.

References

Author notes