Developmental Programming: Differential Effects of Prenatal Exposure to Bisphenol-A Or Methoxychlor on Reproductive Function

Home , Methoxychlor

Developmental Programming: Differential Effects of Prenatal Exposure to Bisphenol-A or Methoxychlor on Reproductive Function

Mozhgan Savabieasfahani, Kurunthachalam Kannan, Olga Astapova, Neil P. Evans, and Vasantha Padmanabhan Downloaded from https://academic.oup.com/endo/article/147/12/5956/2501261 by guest on 27 September 2021 Departments of Pediatrics and the Reproductive Sciences Program (M.S., O.A., V.P.), University of Michigan, Ann Arbor, Michigan 48109; Wadsworth Center (K.K.), New York State Department of Health, and Department of Environmental Health Sciences, State University of New York at Albany, Albany, New York 12201; and Division of Cell Sciences (N.P.E.), University of Glasgow Veterinary School, Glasgow G61 1QH, United Kingdom

Increased occurrence of reproductive disorders has raised maternal MXC concentrations in fat tissue and BPA in blood ,concerns regarding the impact of endocrine-disrupting chem- averaged approximately 200 ␮g/g fat and 37.4 ؎ 3.3 ng/ml icals on reproductive health, especially when such exposure respectively. Birth weights of BPA offspring were lower (P < occurs during fetal life. Prenatal testosterone (T) treatment 0.05) relative to C. There was no difference in the time of leads to growth retardation, postnatal hypergonadotropism, puberty between groups. BPA females were hypergonado- compromised estradiol-positive feedback, polycystic ovaries, tropic during early postnatal life and ended their breeding and infertility in the adult. Prenatal dihydrotestosterone season later, compared with C. Characterization of cyclic treatment failed to affect ovarian morphology or estradiol- changes after synchronization with prostaglandin F2␣ in five positive feedback, suggesting that effects of prenatal T may be C, six MXC, and six BPA females found that the onset of the LH facilitated via conversion of T to estradiol, thus raising con- surge was delayed in MXC (P < 0.05) and the LH surge mag- cerns regarding fetal exposure to estrogenic endocrine-dis- nitude severely dampened (P < 0.05) in BPA sheep. These rupting chemicals. This study tested whether fetal exposure findings suggest that prenatal BPA and MXC exposure have to methoxychlor (MXC) or bisphenol A (BPA) would disrupt long-term differential effects on a variety of reproductive en- -docrine parameters that could impact fertility. (Endocrinol ؍ cyclicity in the ewe. Suffolk ewes were administered MXC (n (mg/kg⅐d sc in cotton seed oil) or the vehicle ogy 147: 5956–5966, 2006 5) (10 ؍ BPA (n ,(10 ,from d 30 to 90 of gestation. On d 60 of treatment (16 ؍ C; n)

OCIETAL CONCERN HAS been mounting over the po- well established in human medicine that fetal exposure to the S tential deleterious effects on animal/human health of synthetic estrogen, diethylstilbesterol, has resulted in a wide environmental exposure to endocrine-disrupting com- variety of problems in the daughters of mothers prescribed pounds (EDCs). EDCs are hormonally active, synthetic, or diethylstilbesterol during pregnancy, including increased natural compounds that are present within our environment risk of cancers and infertility (9, 10). Exposure to estrogenic and food sources at concentrations that can interfere with the EDCs such as dicofol and DDT (dichloro-diphenyl-trichlor- normal activity of endocrine systems/tissues, most notably ethane), has also been reported to disrupt sexual differenti- the reproductive endocrine axis (1, 2). The controversy re- ation in other species including turtles (11) and birds (12). garding the deleterious effects of EDC exposure has mainly Exposure to methoxychlor (MXC) or bisphenol A (BPA), been fueled by studies that point to likely effects on human EDCs that do interact with estrogen receptors, was found to health, including the recent dramatic increases in the inci- masculinize the female brain (13), advance puberty (14), and dence of estrogen sensitive cancers (breast, prostate and tes- cause sex reversal (15). In addition to these specific effects of ticular) (3, 4), the decline in human sperm quality and quan- EDCs on the reproductive system, EDCs have the potential tity (5), a notable rise in endometriosis (2), an increase in to alter the endocrine status of the developing fetus and thus genital abnormality in boys (6), and early puberty (7) in girls. lead to adaptations that may predispose the fetus to obesity EDCs that can interact with estrogen receptors have re- and other metabolic/endocrine diseases in adulthood (16– ceived considerable attention because they can modulate 19). Based on current evidence, the U.S. Environmental Pro- signaling by native estrogen, a key regulator of several phys- tection Agency (EPA) has thus advocated that although iologic functions including reproduction (8). In addition, it is exposure to single estrogenic compounds, at current environmental levels, is insufficient to cause adverse effects in First Published Online August 31, 2006 adult humans, more information is needed to determine Abbreviations: BPA, Bisphenol A; C, control; CV, coefficient of vari- whether the same holds true for the human fetus and neo- ation; MXC, methoxychlor; P, progesterone; PG, prostaglandin; T, testosterone. nate, which lack some of the protective mechanisms found in the adult (20). Another deficiency of the current EDC Endocrinology is published monthly by The Endocrine Society (http:// www.endo-society.org), the foremost professional society serving the literature is the scarcity of information relative to levels of endocrine community. EDCs achieved in human. Low-dose studies carried out in

5956 Savabieasfahani et al. • Prenatal EDC Exposure and Reproductive Dysfunction Endocrinology, December 2006, 147(12):5956–5966 5957 recent years (21–23) target intake rather than levels achieved differentiation at d 30 (43, 44), development of hypophyseal portal in circulation or tissue load. Considering that animals may vasculature around d 50 (43), detection of LH and FSH in circulation metabolize, store, or respond to similar intake levels differ- around d 55 (45), and completion of primordial follicular differentiation by d 90 (44). EDC doses were selected on the basis of the lowest observed ently, it is essential to also have other reference points such effect level established in the EPA National Toxicology Program’s Re- as circulating and tissue concentrations of EDCs. port of the Endocrine Disruptors (20). Recent rodent studies have Whereas a link has been suspected between EDC exposure achieved effects with much lower intake levels (21–23). Controls (C; n ϭ and adverse effects on human and animal health, it is im- 16) received vehicle. portant to note that health risks have primarily been derived To determine concentrations of BPA achieved using this treatment regimen, blood samples were collected from both C and prenatal BPA- from epidemiological data and/or studies conducted in ro- treated animals (n ϭ 6/treatment group) at three different time points, dent models. Because the sensitivity to EDCs is likely to vary 50, 70, and 90 d of gestation (20, 40 and 60 d of BPA treatment). To avoid between species, studies in rodents need to be cross-vali- the stress associated with collection of biopsies from experimental ewes,

dated using sensitive animal models with established peri- a preliminary estimate of MXC load was obtained after the collection of Downloaded from https://academic.oup.com/endo/article/147/12/5956/2501261 by guest on 27 September 2021 ods of developmental susceptibility. Sheep provide a pow- fat biopsies from one pregnant ewe and her twin lambs (a female and erful model system to investigate reproductive consequences a male) after the animals were killed on d 90 of gestation (d 60 of MXC treatment). Fat biopsies were also obtained on d 30 and 60 of MXC of in utero estrogenic/androgenic EDC exposures because a treatment from two nonpregnant sheep. Fat biopsies were obtained from large body of literature already exists documenting critical three untreated C. Choice of measurement of BPA levels in maternal periods of fetal susceptibility to native steroids (24–35) in the blood and MXC in maternal fat was based on previous studies, which sheep. For instance, prenatal exposure of female sheep to found that BPA accumulates in liver, blood, or proteinaceous tissues testosterone (T), an estrogen precursor, has been reported to (46), and MXC is lipophilic and accumulates in fatty tissues (47). lead to growth retardation (25, 26, 32) and neuroendocrine (24, 27–29) and ovarian (31–33) defects that culminate in a Maternal care, neonatal measures, and postnatal care progressive loss of ovarian cyclicity (33–35). Six weeks before lambing, when maximal fetal growth was under- To assess the effects of EDC exposure, we selected two way, pregnant ewes were group fed an additional 0.5–1 kg alfalfa hay environmentally relevant endocrine disrupting compounds, and 250 mg Aureomycin crumbles (chlortetracycline) per ewe daily. All MXC and BPA. The organochlorine insecticide, MXC, has lambs, the majority twins, were born between March 15 and April 20, been used extensively to control pests in agricultural, dairy, 2002. Birth dates, the number of offspring, and offspring gender were and domestic settings and is known to be moderately per- recorded. At birth each lamb was given oral vitamin E and selenium and injections for Clostridium perfringens types C and D and tetanus. New- sistent in the environment (36, 37). Indeed, measurable con- born weight and body dimensions were recorded from all male and centrations of MXC can be found in the body fat of humans female lambs the day after birth to allow adequate time for maternal and rats (38, 39). BPA is a plasticizer, the production of which bonding. These measures included weight, height (determined with the is estimated to be about 1.7 billion kg/yr (40), and again lambs standing), chest circumference, and anourethral, anonavel, and studies have found measurable concentrations of BPA in the anoscrotal (males) distances. A blood sample was collected to measure maternal/fetal circulation, amniotic fluid, and placental tis- circulating levels of insulin and IGF-I. Each ewe and its lamb(s) were housed together for the first3dand sue of pregnant women (40–42). later group housed with other mothers and offspring, in a barn, under In this study, we aimed to address the threat EDCs may natural photoperiod. Lactating ewes were fed 1 kg shelled corn and 2–2.5 pose to reproductive well-being. The study was designed kg alfalfa hay while they were suckling lambs. A 60-W bulb in the lamb bearing in mind the shortcomings of some of the current EDC creep feed area was lit during the nights. Group-housed lambs had literature and as such used the following: 1) a robust sheep access to feed pellets (Shur-Gain, Elma, NY) containing 18% crude protein and alfalfa hay. At 8 wk of age, all lambs were weaned and the model with known critical developmental windows and de- females transferred to the Sheep Research Facility (Ann Arbor, MI) fined end points (24–35), 2) EDC exposure based on EPA- where they were maintained in pens, outside, under natural photope- published lowest observed effect level (21), 3) measures to riod. Once weaned all lambs were provided ad libitum access to com- quantify EDC burden in the circulation and tissues, and 4) mercial feed pellets (as above). When they reached a weight of approx- appropriate controls. We hypothesized that fetal exposure to imately 40 kg, to avoid fat deposition during the period of reduced growth, lambs were switched to a pelleted feed with 15% crude protein. MXC and BPA, at levels approaching human exposure levels, Trace mineralized salt with selenium and vitamins A, D, and E (Armada would disrupt reproductive cyclicity in the ewe. Grain Co., Armada, MI) was freely accessible throughout the study. Postnatal weight gain was monitored weekly for 8 months in all female Materials and Methods lambs. Breeding and prenatal EDC treatment Adult Suffolk ewes were maintained at a U.S. Department of Agri- Dietary control culture-inspected and University of Michigan Laboratory Animal Med- icine-approved farm for breeding. Starting 2–3 wk before breeding, ewes To ensure that there is no confounding effect from feed differences were group fed daily with 0.5 kg shelled corn and 1.0–1.5 kg alfalfa hay across treatments, food was purchased from a single supplier in bulk and per ewe to increase energy balance. Ewes were mated to fertility-proven stored for use. All ewes, before and after breeding as well as all lambs rams and the day of mating was noted. After breeding, all ewes were after weaning therefore received the same standardized food. Whereas housed under natural photoperiod in the same pasture and group fed there was a potential for interaction between experimental EDC expo- with a daily maintenance diet of 1.25 kg alfalfa/brome mix hay per ewe. sures with phytoestrogens from food, this feeding regimen ensured that Pregnant ewes (average weight 87.2 Ϯ 2.3 kg) received sc injections exposure of phytoestrogens was similar across all treatment groups and of MXC (purity 95ϩ%, catalog no. M-1501; Sigma, St. Louis, MO) (n ϭ thus allowed observation of effects of added experimental BPA and 10) or BPA (purity 99ϩ%, catalog no. 239658–250G; Aldrich Chemical MXC exposure. The feeding/EDC exposure paradigm used in this study Co., Milwaukee, WI) (n ϭ 10) at 5 mg/kg⅐d in cottonseed oil (Sigma) also had the added benefit that it reflects a true-life situation because from d 30 through 90 of gestation (term 147 d). Critical time points humans and animals are likely to be exposed to industrial pollutants and relative to timing of the EDC exposure include initiation of gonadal phytoestrogens in parallel. 5958 Endocrinology, December 2006, 147(12):5956–5966 Savabieasfahani et al. • Prenatal EDC Exposure and Reproductive Dysfunction

Puberty and cycle characterization RIAs To establish the time of puberty (onset of progestogenic cycles), Circulating levels of LH were measured in twice-weekly samples biweekly blood samples were collected from all female lambs starting collected during the first month of life and all samples in the cycle at approximately 6 wk of age. Males were castrated by banding within characterization study (except the daily samples) using a well-validated first week of birth and not studied. At 40 wk of age, after achievement assay (50). Assay sensitivity, 50% displacement point, and intraassay of puberty, C (n ϭ 5) and prenatal EDC-treated females (n ϭ 6) received coefficient of variation (CV) at 80 and 20% displacement points of the LH two im injections of prostaglandin (PG) F2␣ (20 mg of 5 mg/ml Lutalyse; assay averaged 0.13 Ϯ 0.02 ng, 0.68 Ϯ 0.02 ng, 6.03 Ϯ 0.71%, and 3.70 Ϯ Pfizer Animal Health, Kalamazoo, MI) 11 d apart to synchronize their 0.42%, respectively (n ϭ 17 assays). Interassay CV based on three quality estrous cycles. To characterize cyclic changes in basal gonadotropin and control pools averaging 0.91 Ϯ 0.40, 13.29 Ϯ 0.23, and 22.40 Ϯ 0.55 ng/ml steroid concentrations during the follicular phase, blood samples were averaged 17.40, 8.40, and 9.56%, respectively. Circulating levels of FSH obtained from all C, prenatal MXC-, and prenatal BPA-exposed females were measured in two-hourly samples collected during the synchro- at 2-h intervals for 120 h, starting at the time of the second PGF2␣ nized estrus cycle using a validated RIA (51). Assay sensitivity, 50% injection. Only one female from any given dam was used in this study displacement point, and intraassay CV at 80 and 20% displacement to ensure dam is the experimental unit. To monitor temporal changes in points of the FSH assay averaged 0.08 Ϯ 0.01 ng, 0.58 Ϯ 0.01 ng, 10.59 Ϯ Downloaded from https://academic.oup.com/endo/article/147/12/5956/2501261 by guest on 27 September 2021 LH secretion, blood samples were collected frequently (12 min intervals 2.03%, and 5.30 Ϯ 1.02%, respectively (n ϭ 5 assays). Interassay CV based Ϯ Ϯ for 8 h) twice during the presumptive follicular phase (24–32 and 46–54 on two quality control pools averaging 3.50 0.25 and 30.13 1.63 h after administration of PGF2␣) and once during the luteal phase (d 8 ng/ml averaged 14.17 and 10.80%, respectively. after second PGF2␣ injection). To characterize the pattern of luteal pro- Plasma concentrations of P were measured using a commercial RIA gesterone (P) secretion, daily samples were collected for 2 wk starting kit (Coat-A-Count P; Diagnostic Products Corp., Los Angeles, CA) val- from completion of 2-h samples. All animal use procedures were ap- idated for use with sheep samples (52) in all twice-weekly samples to proved by the University of Michigan Committee for the Use and Care assess timing of puberty and the length of first breeding season and all of Animals. daily samples to assess luteal phase length. Assay sensitivity, 50% displacement point, and intraassay CV at 80 and 20% displacement points of the P assay averaged 0.03 Ϯ 0.004 ng, 0.17 Ϯ 0.001 ng, 11.43 Ϯ 0.62%, EDC measurements and 5.71 Ϯ 0.32%, respectively (n ϭ 7 assays). Interassay CV based on three quality control pools averaging 0.92 Ϯ 0.04, 1.82 Ϯ 0.08, and 13.07 Ϯ BPA levels in plasma samples were quantified by reverse phase 0.46 ng/ml averaged 11.69, 12.32, and 9.34%, respectively. Plasma in- HPLC with fluorescence detection, as described previously (48). Briefly, sulin concentrations were measured using ImmuChem-coated tube In- a 1-ml aliquot of plasma was transferred to a tube and 10 ␮l of 1-ppm sulin 125I RIA kit (ICN Pharmaceuticals, Costa Mesa, CA) in samples butylphenol added as an internal standard. Samples were extracted collected the day after birth. The sensitivity of the assay was 3.86 ␮U/ml twice with 5 ml ethyl ether. The solvent (2 ϫ 5 ml) was pooled and and intraassay CV was 5%. Circulating IGF-I levels were measured using evaporated to dryness under nitrogen. The sample extract was then a validated assay (53) after an acid-ethanol extraction solution using a reconstituted with 1 ml of acetonitrile and analyzed by HPLC. BPA recombinant human IGF-I (R & D Systems, Minneapolis, MN) as the recovery was determined by fortifying charcoal-stripped fetal bovine assay standard. The sensitivity of the assay was 2.9 ng/ml and intraassay serum (1 ml) with 50 and 100 ng of BPA and it averaged 77 Ϯ 12%. CV and recovery were 10 and 96%, respectively. For all assays, all values Standards of BPA were prepared in acetonitrile at concentrations of 0.1, below assay sensitivity were assigned the detection limit of the assay. 0.25, 0.5, and 1 ␮g/ml for calibration. Samples and standards were injected onto an analytical column (Prodigy ODS, 250 ϫ 4.6 mm column; Phenomenex, Torrance, CA), which was connected to a guard column Statistical analysis ϫ Prodigy ODS, 30 4.6 mm using a series 200 autosampler (PerkinElmer, In terms of newborn measures, the primary outcome measures were Norwalk, CT) and eluted with a flow of acetonitrile and water at a body weight, height, chest circumference, anogenital ratio (the ratio of gradient from 50% acetonitrile in water to 98% acetonitrile in water for anourethral to anonavel distance), anoscrotal to anonavel ratio (males), 20 min delivered by a PerkinElmer series 200 pump (flow rate was at 1 and circulating insulin and IGF-I levels. Each outcome measure was ml/min). Each sequence began with a blank and calibration standards. compared between groups using only the mean value from siblings from High-purity analytical grade solvents were used throughout. Detection multiple gestations by a general linear model that included treatment was accomplished using a Hewlett Packard 1046A fluorescence detector and number of males in the multiple gestations as a covariate. This is (Hewlett-Packard Co., Wilmington, DE) with an excitation wavelength similar to a repeated-measures ANOVA in which the siblings are re- of 229 nm and an emission wavelength of 310 nm. Florescence detector peated measures with gender and treatment as grouping variables (co- settings were: photomultiplier tube gain 12, lamp time -1, response time variates). To assess whether prenatal BPA exposure had an effect on 2 sec, stop time 27 min, and gate and delay at zero. Blanks were analyzed postnatal growth rate, a growth curve analysis was performed using a concurrently to check for interfering peaks. Detection limit of BPA was linear mixed model in which a separate regression line (random inter- 10 ng/ml. cept and random slope) was calculated for each ewe. The overall effect For measurement of MXC accumulation,3goffatwere weighed and of age, treatment, and the age by treatment interaction was examined. extracted with Soxhlet apparatus for 16 h using dichloromethane and Age at puberty was defined as the first increase of the first proges- hexane (400 ml). The extract was rotary evaporated to 6 ml. Two mil- togenic cycle. A progestogenic cycle was defined as a plasma P con- liliters of the extract was diluted in 2 ml dichloromethane and passed centration that remained 0.5 ng/ml or greater for at least two consecutive through a gel permeation chromatography column to remove fat, as twice-weekly time points. The duration of each progestogenic cycle was described elsewhere (49). One milliliter of the extract was used for lipid calculated from the day of P rise greater than 0.5 ng/ml to the day when determination. The extract from the gel permeation chromatography P concentration fell below this value. The onset of LH and FSH surges column was concentrated to 1 ml and injected into an Agilent 6890 N gas was determined based on modification of previously established criteria chromatograph equipped with an electron capture detector (Agilent (54). Briefly, the onset of an LH/FSH surge was defined as elevation of Technologies, Palo Alto, CA). A capillary column coated with DB-5 (5% circulating LH/FSH above baseline by 2ϫ assay sensitivity lasting at phenyl methylpolysiloxane, 30 m ϫ 0.25 mm inner diameter ϫ 0.25 ␮m least 8 h and peak concentration of LH/FSH exceeding at least twice the film thickness; Agilent Technologies) was used for the separation of average levels of LH/FSH during baseline periods. The end of the surge MXC. The column oven temperature was programmed to change from was defined as the time when LH/FSH levels fell below the established 120 C (1 min) to 180 C (2 min) at a rate of 10 C/min and then to 240 C criteria of surge onset. For secondary FSH surge, the sustained increase at 3 C/min, with a final hold time of 5 min. Inlet and detector temper- in FSH that followed termination of primary FSH surge was used and atures were held at 250 and 300 C, respectively. Concentrations were included same criteria as defined for primary surge. If secondary FSH calculated from the peak area of the sample to that of the corresponding levels did not fall back to baseline before end of the bleed, the duration external standard, MXC. Detection limit of MXC was 10 ng/g on a wet of surge was computed as interval between start of secondary FSH surge weight basis. Recovery of MXC through the analytical method was 82 Ϯ and time of termination of bleed. 7%. Data from first follicular bleed (24–32 h after PGF2␣) for control and Savabieasfahani et al. • Prenatal EDC Exposure and Reproductive Dysfunction Endocrinology, December 2006, 147(12):5956–5966 5959

BPA (the second follicular bleed for most animals occurred during or compared with levels achieved after 30 d of treatment (Fig. after LH surge) and both the first and second follicular phase bleed 1, bottom right). The MXC concentrations in the pregnant ewe (46–54 h) for MXC animals and luteal period of all animals were sub- and her male and female fetus killed on d 90 of gestation (d jected to pulse analysis using the Cluster algorithm (55). The Cluster algorithm identifies pulses using criteria that define a pulse such that the 60 of MXC treatment) were comparable with levels achieved peak of the pulse differs significantly from both the preceding and in nonpregnant ewes that received the same MXC treatment following nadirs according to two-sample t tests. For analysis with for the same duration. Overall MXC levels achieved in Cluster, the minimum number of data points in a peak and nadir were treated ewes were markedly higher compared with controls. set at 2 and 2, respectively and the t statistic values used to identify a significant increase from preceding nadir and a decrease to following nadir were both at 2.0. Gestational length, birth measures, and growth rate Timing of puberty, duration and end of breeding season, number/ duration of progestogenic cycles and peak P concentrations achieved, Gestational length of C and prenatal MXC- and BPA- timing and attributes of gonadotropin surges, LH pulse frequency and treated females did not differ and averaged 148.3 Ϯ 0.1, amplitudes, and duration/peak concentration of luteal P increase 147.0 Ϯ 0.5, and 147.3 Ϯ 0.6 d, respectively. All animals were Downloaded from https://academic.oup.com/endo/article/147/12/5956/2501261 by guest on 27 September 2021 achieved in the synchronized cycle were analyzed by ANOVA after appropriate transformations to account for heterogeneity of variance. born between March 15 and April 19, 2002. Eleven female and 18 males were born in the C group. There were 11 Results females and six males in the prenatal BPA-treated group and BPA or MXC levels achieved nine females and nine males in the prenatal MXC-treated group. Prenatal BPA- but not MXC-treated female lambs Representative chromatographic profiles of BPA and MXC weighed less (P Ͻ 0.01) than the C at birth (Fig. 2). This from maternal plasma of a BPA-treated and abdominal fat of difference was associated with significant effects on height MXC-treated sheep are shown in Fig. 1 (top panels). Maternal (P Ͻ 0.01) and chest circumferences (P Ͻ 0.05). There were concentrations of BPA in blood were significantly higher in no differences in anogenital ratio among the three groups of BPA-treated females, compared with controls (below detec- female lambs (C: 0.28 Ϯ 0.09; prenatal MXC: 0.27 Ϯ 0.08; tion limit) at all three time points studied (50, 70, and 90 d prenatal BPA: 0.23 Ϯ 0.08). Circulating levels of IGF-I in of gestation corresponding to d 20, 40, and 60 of treatment) female lambs at1dofagewere similar across treatment (Fig. 1, bottom left). Thus, 20 d appeared sufficient to achieve groups and averaged 5.14 Ϯ 0.07, 4.85 Ϯ 0.26, and 5.23 Ϯ 0.14 a steady-state concentration of BPA. MXC was found in fat ng/ml in C and prenatal MXC- and BPA-treated females, samples collected from nonpregnant female sheep treated respectively. Circulating levels of insulin also did not differ with MXC with levels being higher 60 d after MXC treatment, significantly among these groups: 2.41 Ϯ 0.24, 2.00 Ϯ 0.24, 2.17 Ϯ 0.12 ng/ml in C and prenatal MXC- and BPA-treated females, respectively. In the males, prenatal MXC but not BPA treatment was associated with a significant (P Ͻ 0.05) increase in birth weight relative to the C (C: 5.0 Ϯ 0.18; prenatal MXC: 5.74 Ϯ 0.30; prenatal BPA: 4.96 Ϯ 0.59 kg). This increase was reflected in their chest circumference (C: 40.5 Ϯ 0.7; prenatal MXC: 42.4 Ϯ 0.9; prenatal BPA: 40.4 Ϯ 1.4 cm). Whereas there was no difference in the anogenital ratio between males of the three groups (C: 0.70 Ϯ 0.07; prenatal MXC: 0.75 Ϯ 0.09; prenatal BPA: 0.63 Ϯ 0.07 cm), the anoscrotal to anonavel ratio was increased in both the prenatal MXC- (P Ͻ 0.05) and BPA-treated (P Ͻ 0.01) males, compared with C (C: 0.38 Ϯ 0.01; prenatal MXC: 0.42 Ϯ 0.02; prenatal BPA: 0.46 Ϯ 0.004).

Postnatal growth As expected, there was a highly significant positive slope for age (P Ͻ 0.0001) across treatments. Growth rate tended to be higher in prenatal BPA- and not MXC-treated females (P ϭ 0.08), compared with C between 2 and 4 months of age.

Timing of puberty and reproductive cycles Age at puberty (start of first breeding season) was similar FIG.1.Top panels, Chromatographic profiles of BPA and MXC of BPA- and MXC-treated pregnant sheep. Bottom left panel, Levels of in C and prenatal MXC- and BPA-treated sheep (Fig. 2). All circulating BPA in C (open circles) and BPA-treated (closed circles) females irrespective of treatment had repetitive progesto- pregnant ewes on d 50, 70, and 90 of gestation (d 20, 40, and 60 of genic cycles during the first breeding season (Fig. 3). Whereas treatment). Bottom right panel, Levels of MXC found in fat obtained the date of onset of the breeding season was similar across from three control animals (open bar; C), two nonpregnant ewes on d Ͻ 30 and 60 of MXC treatment (hatched bars) and one pregnant ewe and treatment groups, the breeding season ended later (P 0.05) its male and female fetuses on d 60 of MXC treatment (d 90 of in the prenatal BPA-treated females (Fig. 3). Prenatal MXC- pregnancy; treatment began d 30 of pregnancy) (closed bars). (P Ͻ 0.05) and BPA-treated (P ϭ 0.054) animals had more 5960 Endocrinology, December 2006, 147(12):5956–5966 Savabieasfahani et al. • Prenatal EDC Exposure and Reproductive Dysfunction

FIG. 2. Newborn body weight, height, and chest circumference adjusted for litter size (mean Ϯ SEM) and age at puberty of female lambs born to ewes treated with cottonseed oil (C, open bars), sheep prenatally treated during d 30–90 of pregnancy with MXC (hatched bars)orBPA (closed bars). Asterisks, Significant differences from C (P Ͻ 0.05). Downloaded from https://academic.oup.com/endo/article/147/12/5956/2501261 by guest on 27 September 2021 progestogenic cycles than the C during the first breeding Effects of prenatal MXC and BPA treatment on estrous season (Fig. 3). Peak P levels and length of progestogenic cycle characteristics cycles did not differ among the three groups (Fig. 3). One C Patterns of circulating LH and FSH from three C, three and one prenatal MXC-treated female were excluded from prenatal MXC-, and three BPA-treated females beginning this analysis because of abnormal elevations in P that lasted from the time of second PGF2␣ injection are shown in Fig. 5. several weeks, precluding calculating the number of cycles. An LH surge was seen in all C and MXC-exposed females but Mean circulating concentrations of LH measured from bi- not all BPA-exposed females. Only 50% of BPA-treated an- weekly samples during the first month of life were higher in imals exhibited increases in LH that met the criteria defined prenatal BPA- but not the MXC-treated lambs (Fig. 4). for an LH surge (three of six animals) as opposed to 100% of controls (P ϭ 0.0637 by ␹2 analysis). Summary statistics of primary gonadotropin and secondary FSH surges are provided in Figs. 6 and 7 (excluding the three prenatal BPA- treated females that failed to show a definable LH surge). Relative to the second PGF2␣ injection, the time of onset of the LH surge and time to peak were both significantly delayed (P Ͻ 0.05) in prenatally MXC- but not BPA-treated sheep (Fig. 6, top two panels), compared with C. Timing of the primary FSH surge was similar to that of LH (not shown). The peak LH concentration and the total amount of LH released during the primary LH surge were significantly lower (P Ͻ 0.05) in prenatal BPA- but not MXC-treated sheep when compared with C (Fig. 6). There were no differences in the duration of the LH surge between C and the two EDC- treated groups. There were also no differences in the primary and secondary FSH surge magnitudes or total FSH secreted. Duration of the primary FSH surge in prenatal MXC-treated

FIG. 3. Patterns of biweekly P from three C (top left) and three prenatal MXC- (top middle) and three prenatal BPA-treated (top right) females during the first breeding season. Bottom panels, Mean Ϯ SEM FIG. 4. Mean Ϯ SEM of circulating LH during early postnatal period of end of first breeding season (bottom left) as well as the mean Ϯ SEM inC(open bar) and prenatal MXC-treated (hatched bar) and prenatal of peak level, number, and duration of the progestogenic cycles in C BPA-treated (closed bar) females. For analysis, serial values from and prenatal MXC- and BPA-treated females. Asterisks, Significant twice weekly samples taken during the first 2 months of postnatal life differences from C females. were averaged. Asterisk, Significant difference from C (P Ͻ 0.05). Savabieasfahani et al. • Prenatal EDC Exposure and Reproductive Dysfunction Endocrinology, December 2006, 147(12):5956–5966 5961

FIG. 5. Circulating patterns of LH (closed circles) and FSH (open circles) in three C (ewes 299, 244, and 268) and three prenatal MXC- (ewes 233, 273, and 274) and three BPA-treated (ewes 265, 234, and 262) females taken at 2-h intervals for 120 h after induction of luteolysis with two injections of PGF2␣ 11 d apart. Note the onset of LH surge is delayed in the three prenatal MXC- treated animals relative to C. The prenatal

BPA-treated females showed severely Downloaded from https://academic.oup.com/endo/article/147/12/5956/2501261 by guest on 27 September 2021 dampened LH surges (ewes 265 and 234). Ewe 262 showed a sustained but very small increase in LH, which fit the surge definition but not a clearly definable surge. The other three prenatal BPA- treated females failed to show LH increases that met the criteria established (see Materials and Methods).

females was, however, significantly (P Ͻ 0.05) longer than gan 9.2 Ϯ 1.6 h before the onset of LH surge. The second the other two groups (Fig. 7). follicular phase determination occurred during or after LH The LH pulse characteristics during the follicular and lu- surge in most of C and prenatal BPA-treated females. As teal phase of the cycle in C and prenatal MXC- and BPA- expected, LH pulse frequency was higher in the follicular treated females are summarized in Table 1. The first follicular than luteal phase across treatment groups (P Ͻ 0.05). There phase LH pulse determination began 12.4 Ϯ 2.8, 29.0 Ϯ 3.0, and 14.0 Ϯ 2.7 h before the onset of LH surge in C and prenatal MXC- and BPA-treated females, respectively. The second follicular phase determination for MXC animals be-

FIG. 6. Mean (Ϯ SEM) LH surge onset and peak relative to second PGF2␣ administration are shown in the top panels. Bottom panels FIG. 7. Characteristics of primary and secondary FSH surge (peak shows LH surge characteristics (LH surge peak, total LH, and du- height, total FSH, and duration) in C (open bar) and prenatal MXC- ration of the LH surge) from C (open bar) and prenatal MXC-treated (hatched bar) and prenatal BPA-treated (closed bars) females are (hatched bar) and prenatal BPA-treated (closed bar) females. Note shown in the top and bottom panels, respectively. Note measures in measures in prenatal BPA-treated females are from only three ani- prenatal BPA-treated females are from three animals. The other mals. The other three animals did not show clearly definable LH three animals did not show clearly definable LH surges and hence surge. Significant differences between treatment and control groups other parameters could not be calculated. Asterisk, Significant dif- are indicated by differing letters. ference from C (P Ͻ 0.05). 5962 Endocrinology, December 2006, 147(12):5956–5966 Savabieasfahani et al. • Prenatal EDC Exposure and Reproductive Dysfunction

TABLE 1. LH pulse dynamics during the follicular and luteal phase of the estrous cycle

Folliculara Lutealb Groups Mean Frequency Amplitude Mean Frequency Amplitude C (first) 5.31 Ϯ 1.05 7.25 Ϯ 0.85 3.49 Ϯ 0.67 1.82 Ϯ 0.23 3.25 Ϯ 0.75 2.87 Ϯ 0.42 MXC (first) 3.03 Ϯ 0.44 6.50 Ϯ 0.85 2.51 Ϯ 0.50 MXC (second) 2.42 Ϯ 0.40 6.00 Ϯ 0.55 2.36 Ϯ 0.80 2.10 Ϯ 0.37 3.00 Ϯ 0.45 4.78 Ϯ 0.96 BPA (first)c 4.92 Ϯ 1.62 6.67 Ϯ 1.2 4.48 Ϯ 1.67 2.58 Ϯ 0.53 3.67 Ϯ 0.67 4.78 Ϯ 0.27d a Blood samples were collected frequently at 12-min intervals for 8 h twice during the presumptive early and late follicular phase (24–32 and 46–54 h after administration of PGF2␣). Because second follicular bleed occurred during or after LH surge in majority of the C and BPA-treated females, data from only the first follicular bleed is provided. For prenatal MXC animals, in which onset of LH surge was delayed, mean LH, LH pulse frequency, and amplitude are provided from both follicular bleeds. b Luteal phase blood samples were taken at 12-min intervals for8hond8after PGF2␣ injection. Downloaded from https://academic.oup.com/endo/article/147/12/5956/2501261 by guest on 27 September 2021 c Because three of the prenatal BPA-treated females failed to show definable LH surges, they were not included in these analyses. d P Ͻ 0.05. were no between treatment differences in frequency or am- ronmentally relevant EDCs, at environmentally relevant plitude of LH pulses during the follicular phase. LH pulse doses, can disrupt reproductive function in different ways. amplitude, but not frequency, was significantly increased Prenatal BPA exposure, at levels close to that seen in human (P Ͻ 0.05) in prenatal BPA- but not MXC-treated females maternal circulation, resulted in female offspring of low birth during the luteal phase. weight, which is a recognized risk factor for adult-onset All animals in C and prenatal MXC- and BPA-treated disease. Prenatal BPA exposure also caused early postnatal groups had luteal P increases. There was no difference in hypergonadotropism and prolonged the first breeding sea- peak or duration of luteal P increase between C and prenatal son, which is suggestive of a reduced sensitivity to estradiol- MXC- or BPA-treated groups (Fig. 8), despite the severely negative feedback. Prenatal MXC exposure, on the other disrupted LH surges in the prenatal BPA-treated females. hand, had no effect on birth weight or changes in gonadotropin secretion that are associated with activation of the Discussion reproductive axis but did affect hormone secretory dynamics during the estrous cycle such that the onset of primary go- It has been proposed that in utero exposure to environ- ␣ mental pollutants that can act as hormone agonists or an- nadotropin surges (after PGF2 ) was delayed. tagonists (8) could pose a considerable threat to mankind. Findings from this study using sheep as a model provide Robustness of the animal model used evidence that prenatal exposure to BPA or MXC, two envi- The majority of work testing EDC risk has relied on in vitro systems or rodent models. Because sensitivity to EDC may vary between species, comparative animal models using species other than rodents are also needed to establish safe exposure levels. The wealth of information available detail- ing the neuroendocrine and ovarian changes that regulate reproductive cyclicity in the sheep (56) and knowledge of the steroid sensitive developmental periods (24–35) makes the sheep a superb model to assess the deleterious effects of EDCs on reproductive function. For example, our previous studies and those of others found that prenatal exposure of sheep to T from d 30 to 90 of gestation can lead to low birth weight (25, 26) and severe estrous cycle defects, which included neuroendocrine (24, 27–30) and ovarian (31–33) defects. Some of these reproductive defects such as disruption of estradiol feedback and development of polycystic ovarian morphology in prenatal T-treated females appear to be programmed by aromatization of T to estradiol as an exposure to a similar pattern of dihydrotestosterone, a nonaromatiz- able androgen, failed to induce the same abnormalities (24, 31). These findings raise the possibility that environmental EDCs that interact with estrogen and androgen receptors may also disrupt reproductive cyclicity.

EDC exposure levels and relevance to human FIG. 8. Mean Ϯ SEM circulating patterns of daily P in C and prenatal MXC- and prenatal BPA-treated females after synchronized estrus It is highly debated whether EDC at current exposure are shown on the left and mean Ϯ SEM of peak height and duration of levels are detrimental to human health. As reviewed in recent luteal P increase are shown on the right. articles (20–23), to assess human risks, animal studies should Savabieasfahani et al. • Prenatal EDC Exposure and Reproductive Dysfunction Endocrinology, December 2006, 147(12):5956–5966 5963 include environmentally relevant doses as well as relevant its fetuses, although very preliminary, suggest MXC also and sensitive quantitative end points. Because uptake and crosses the placental barrier and reaches the fetus. metabolism are likely to vary from species to species and fetus to adult, it is also vitally important to establish a range Effects of EDC on offspring measures of appropriate reference points for exposure assessment. For In previous studies, we reported that prenatal treatment instance, species differences in the pharmacokinetics of a with T leads to intrauterine growth retardation (32) and compound may alter the effective exposure such that low consequently low birth weight (26). Our finding that prenatal dose administration may lead to high exposure if absorption exposure to BPA, an estrogen mimic, also results in low- is high and metabolism is reduced, whereas a high exposure birth-weight female offspring suggests that programming of may provide a low exposure, if absorption is low and/or the low birth weight in prenatal T-treated females may be fa- compound is degraded quickly. Because different animals cilitated by aromatization of T to estrogen. This is further may store or metabolize EDCs quite differently from hu- Downloaded from https://academic.oup.com/endo/article/147/12/5956/2501261 by guest on 27 September 2021 substantiated by the fact that offspring of dihydrotestoster- mans, blood/tissue levels need to be assessed in a variety of one-treated females are not of lower birth weight (Steckler, species in a variety of exposure paradigms to provide an T., and V. Padmanabhan, unpublished data). Low birth initial reference point to relate experimental studies of EDC weight is viewed as an early marker of several adult disor- exposure to human exposure levels and assess potential hu- ders including infertility (66–69). Prenatal MXC treatment, man risk. on the other hand, had no effect on birth weight of female but Because the aim of this study was to investigate the effects increased body weight and size of the male offspring. Con- of fetal exposure to EDCs, the model system targeted sidering that MXC can interact with both estrogen and an- achievement of maternal levels of EDCs that can be com- drogen receptors (70–72), the growth-promoting effects on pared with those reported in human. Subcutaneous injec- males may be facilitated by MXC acting as an estrogen an- tions were used to provide assurance that required levels are tagonist or androgen agonist rather than as an estrogen ag- achieved. Whereas the method of administration used (sc onist. The fact that both prenatal BPA and MXC treatments injection) differs from that which would be encountered increased the anoscrotal to anonavel ratio in the males sug- naturally, this does not detract from the result obtained using gests that this effect is likely programmed by the two EDCs this model system. acting as an estrogen agonist. Whereas the results are dis- Maternal levels of BPA achieved in the present study av- cussed in the context of estrogen/androgen signaling, any eraged 37.4 Ϯ 3.3 ng/ml on d 60 of BPA treatment and were number of mechanisms, many of which may have nothing to approximately 2-fold higher than the highest level reported do with their ability to interact with gonadal steroid hormone in human maternal blood (41). Levels of unconjugated BPA receptor, may have contributed toward this disruption. The in human maternal circulation ranged between 0.3 and 18.9 discussion is targeted to gonadal steroids due to the simi- ng/ml and in fetal blood between 0.2 and 9.2 ng/ml (41), larity of reproductive phenotype with T (androgenic ϩ es- suggesting that BPA crosses the placental barrier. High levels trogenic) and not dihydrotestosterone (androgenic) animals. of BPA have also been reported in human amniotic fluid and The subtle effects of prenatal MXC treatment on repro- placental tissue (40). Considering the widespread use of BPA ductive endocrine dynamics reflected as a delay in onset of in industrial and consumer products (57–61), our findings of LH surge, compared with prenatal BPA treatment, may re- reproductive disruptions after BPA exposure at concentra- late to the levels achieved, which may be beyond the optimal tions not far from those reported in human maternal blood response range. This is especially true because, as reviewed highlight the potential threat BPA poses to fetal and subse- by Vom Saal et al. (73), current assumptions of monotonic quent adult health. dose-response curve (lack of risk below a certain threshold Our experiments with MXC exposure began before the use and similar or increased responses with increasing dose) do of MXC was banned in the United States (62). Nonetheless, not appear to hold true. Alternatively, it is conceivable that studies focusing on disruptive effects of MXC are of rele- in sheep MXC gets quickly stored in fat before being me- vance because the risks of exposure to MXC are likely to tabolized to its active metabolite, 2,2-bis (p-hydroxyphenyl)- continue due to importation of many of the agricultural 1,1,1-trichloroethane (74). A third possibility relates to the products from countries in which the use of MXC is not mixed effects of MXC at the level of estrogen and androgen banned. Furthermore, adverse effects on developmental receptors (70–72). MXC exposure may not become evident for decades due to the long-term nature of the induced effects and the bioac- Effects of EDC on puberty cumulation of MXC in fat and its later release from people exposed to MXC in the recent past. Finally, there is also the Our findings that prenatal MXC and BPA had no effect on potential for transgenerational transfer of altered pheno- EDC differ from rodent studies. Rodent studies (14, 75, 76) types (63–65). The levels of MXC achieved with daily ad- found that both MXC and BPA advance onset of puberty. ministration of MXC for 60 d in this study are approximately Similar onset of progestogenic cycles in C and prenatal EDC- 1000-fold higher than the reported level of 156 ng/g lipid in treated sheep is not surprising, considering the strict pho- adipose fat of a human Spanish population (38). The increase toperiod requirement of female lambs for the synchronous in the concentration of MXC found in tissues of nonpregnant timing of puberty (77). Postnatal but not prenatal estrogen ewes after 60 d of exposure, compared with 30 d, substan- exposure has been shown to delay onset of puberty in female tiates the fact that MXC has the potential to bioaccumulate. lambs, possibly due to the profound postnatal hypogona- Similar levels of MXC achieved in the one pregnant ewe and dotropism it induces (78, 79). Whereas there were no differ- 5964 Endocrinology, December 2006, 147(12):5956–5966 Savabieasfahani et al. • Prenatal EDC Exposure and Reproductive Dysfunction ences in onset of progestogenic cycles, the early increase in plitude was increased in prenatal BPA-treated animals sug- postnatal LH seen in prenatal BPA-treated females is con- gestive of reduced sensitivity to P-negative feedback. The sistent with advancement of neuroendocrine puberty and an increased postnatal LH and delayed ending of first breeding early decline in estradiol-negative feedback, similar to that season in concert with reduced luteal P sensitivity in prenatal seen in prenatal T-treated females (24, 27, 28). Levels of LH BPA-treated females are consistent with estradiol being the are reported to be low in control females during the first 2 negative feedback regulator of LH during the prepubertal months of life (80). period and P assuming this role once cyclicity is established. The delayed/dampened LH surge in the prenatal MXC- Differential effects of BPA and MXC on reproductive and BPA-treated animals did not disrupt luteal function, endocrine parameters emphasizing the resiliency of the system to such cyclic per- turbations. The dominant follicle, once differentiated and

Whereas both prenatal MXC and BPA treatments resulted mature, appears to be capable of luteinizing and sustaining Downloaded from https://academic.oup.com/endo/article/147/12/5956/2501261 by guest on 27 September 2021 in altered reproductive endocrinological parameters in the normal corpus luteum function, even when the ovulatory female lambs after they had attained puberty, fertility out- trigger is delayed or dampened. Our finding that very small comes were not assessed. Prenatal treatment with MXC or but sustained increases in LH such as that seen in the prenatal BPA had differential effects on LH surge dynamics with BPA-treated female (ewe 262) was capable of eliciting a nor- MXC treatment delaying the onset and BPA treatment re- mal luteal P profile suggest a massive LH surge is not re- ducing the magnitude of the LH surge. Rodent studies have quired. These findings also parallel what is seen in the pre- found that BPA treatment disrupts estrous cyclicity (75, 81). natal T-treated sheep, in which repetitive P cycles are The cycle defects seen in rodents (75, 81) and sheep (this evident, despite severely compromised estradiol-positive study) are perhaps not surprising given the reports that BPA principally acts as an estrogen mimic (21–23), whereas MXC feedback (27, 29). If the delayed or dampened LH surge in can interact with both estrogen and androgen receptors (70– prenatal EDC-treated females has effects at other levels such 72). From a mechanistic perspective, the reduced or absent as oocyte quality (86) remains to be determined. LH surge magnitude such as that evidenced in prenatal BPA- treated females may be suggestive of impending loss of cy- Dietary considerations cles. Prenatal T-treated females, the model prenatal BPA- treated females mimic in terms of low birth weight (25, 26), It should be recognized that, whereas care was taken to early LH excess (27, 28), and severely dampened LH surge provide similar diet across all treatment groups, the effects (27), do show progressive loss of estrous cyclicity (33–35). of BPA or MXC described in this study might reflect syner- Whereas it is possible that severe dampening of LH surge gistic or additive interactions with phytoestrogens in the may be the result of reduced GnRH surge amplitude, earlier feed. Nonetheless, because the diet was constant across treat- studies in sheep have found minimal increases in GnRH is ments, any interaction with phytoestrogens does not reduce sufficient to elicit a full amplitude LH surge (82). A second the value of the findings because in the context of develop- explanation would be a pituitary effect whereby exposure to mental programming, this reflects a true-life situation with the EDC has negative effects on LH production or release humans being exposed to EDCs in concert with other from the pituitary. A third possibility for compromised LH phytoestrogens. surge dynamics in prenatal BPA-treated females may relate These studies are the first comprehensive study that re- to disrupted follicular function and a consequent reduction lates the impact of prenatal exposure to MXC or BPA on in the magnitude of preovulatory estradiol rise. These pos- programming of adult reproductive disruption to actual fetal sibilities remain to be investigated. exposure levels. Our findings draw attention to and validate The delayed onset of LH surge in the prenatal MXC- public concern over human fetal exposure to synthetic ex- treated sheep relative to the C cannot be explained by dif- ogenous steroids. ference in magnitude and duration of luteal P. Previous reports have linked the timing of the LH surge to the levels of P experienced during the previous luteal phase (83), the Acknowledgments LH surge being delayed in animals in which P levels in the The authors are thankful to Mr. Douglas Doop and Gary R. McCalla previous cycle were high. The delayed LH surge onset rel- for providing quality care and maintenance of the lambs used in this ative to PGF2␣ administration may be a function of com- study; Dr. Mohan Manikkam, Mr. James Lee, and Ms. Carol Herkimer promised follicular populations/development that induces a for their assistance with EDC treatment and blood sampling; and Dr. Teresa Steckler for assistance with the formatting of figures. consequent delay in of the preovulatory estradiol rise. In this regard, MXC treatment has been found to disrupt ovarian Received June 15, 2006. Accepted August 21, 2006. folliculogenesis in rodents (84). Alternatively, preovulatory Address all correspondence and requests for reprints to: Vasantha increase in estradiol may have occurred on time, but the Padmanabhan, Department of Pediatrics, 300 North Ingalls Building, required progression of changes in pulsatile GnRH secretion Room 1109 SW, Ann Arbor, Michigan 48109-0404. E-mail: might be compromised (85). However, follicular phase LH [email protected]. pulse frequency was not different between prenatal MXC- This work was supported by U.S. Public Health Service Grants HD 41098, National Institutes of Health Grants F32 ES012366 and T32 treated and C females. Whereas no difference in pulse dy- ES07062, and the Office of the Vice President for Research, University namics of LH was evident in the luteal phase of prenatal of Michigan. EDC-treated animals, compared with controls, LH pulse am- Disclosure summary: None of the authors has anything to declare. Savabieasfahani et al. • Prenatal EDC Exposure and Reproductive Dysfunction Endocrinology, December 2006, 147(12):5956–5966 5965

References tive feedback actions of estradiol but does not affect timing of puberty in female sheep. Biol Reprod 66:924–933 1. International Programme on Chemical Safety 2002 Global assessment of the 28. Sarma HN, Manikkam M, Herkimer C, Dell’Orco J, Welch KB, Foster DL, state-of-the-science of endocrine disruptors. Geneva: World Health Organi- Padmanabhan V 2005 Fetal programming: excess prenatal testosterone re- zation duces postnatal LH, but not FSH responsiveness to estradiol negative feedback 2. Crisp TM, Clegg ED, Cooper RL, Wood WP, Anderson DG, Baetcke KP, in the female. Endocrinology 146:4281–4291 Hoffmann JL, Morrow MS, Rodier DJ, Schaeffer JE, Touart LW, Zeeman 29. Unsworth WP, Taylor JA, Robinson JE 2005 Prenatal programming of re- MG, Patel YM 1998 Environmental endocrine disruption: an effects assess- productive neuroendocrine function: the effect of prenatal androgens on the ment and analysis. Environ Health Perspect 106:11–56 development of estrogen positive feedback and ovarian cycles in the ewe. Biol 3. Kelsey JL, Berstein L 1966 Epidemiology and prevention of breast cancer. Ann Reprod 72:619–627 Rev Public Health 17:47–67 30. Robinson JE, Forsdike RA, Taylor JA 1999 In utero exposure of female lambs 4. Bergstrom R, Adami HO, Mohner W, Zatonski W, Storm H, Ekbom A, Tretli to testosterone reduces the sensitivity of the gonadotropin-releasing hormone S, Teppo A, Akre A, Hakulinen T 1996 Increase in testicular cancer incidence neuronal network to inhibition by progesterone. Endocrinology 140:5797–5805 in six European countries: a birth cohort phenomenon. J Natl Cancer Inst 31. West C, Foster DL, Evans NP, Robinson J, Padmanabhan V 2001 Intrafol- 88:727–733 licular activin availability is altered in prenatally androgenized lambs. Mol Cell 5. Toppari J, Larsen JC, Christiansen P, Giwercman A, Grandjean P, Guillette Endocrinol 185:51–59 Downloaded from https://academic.oup.com/endo/article/147/12/5956/2501261 by guest on 27 September 2021 Jr LJ, Jegou B, Jensen TK, Jouannet P, Keiding N, Leffers H, McLachlan JA, 32. Steckler T, Wang J, Bartol F, Roy SK, Padmanabhan V 2005 Fetal program- Meyer O, Muller J, Rajpert-De Meyts E, Scheike T, Sharpe R, Sumpter J, ming: prenatal testosterone treatment causes intrauterine growth retardation, Skakkebaek NE 1996 Male reproductive health and environmental xenoestro- reduces ovarian reserve and increases ovarian follicular recruitment. Endo- gens. Environ Health Perspect 104:741–803 crinology 146:3185–3193 6. Paulozzi LJ, Erickson JD, Jackson RJ 1997 Hypospadias trends in two US 33. Manikkam M, Steckler TS, Welch KB, Inskeep EK, Padmanabhan V 2006 surveillance systems. Pediatrics 100:831–834 Fetal programming: prenatal testosterone treatment leads to follicular persis- 7. Herman-Giddens ME, Bourdony C, Slora E, Wasserman R 2001 Early pu- tence / luteal defects. Partial restoration of ovarian function by cyclic pro- berty: a cautionary tale. Pediatrics 107:609–610 gesterone treatment. Endocrinology 147:1997–2007 8. Danzo B 1997 Environmental xenobiotics may disrupt normal endocrine func- 34. Clarke IJ, Scaramuzzi RJ, Short RV 1977 Ovulation in prenatally androg- tion by interfering with the binding of physiological ligands to steroid recep- enized ewes. J Endocrinol 73:385–389 tors and binding proteins. Environ Health Perspect 105:294–301 35. Birch RA, Padmanabhan V, Foster DL, Robinson JE 2003 Prenatal program- 9. Senekjian EK, Potkul RK, Frey K, Herbst AL 1988 Infertility among daughters ming of reproductive neuroendocrine function: fetal androgen exposure pro- either exposed or not exposed to diethylstilbestrol. Am J Obstet Gynecol duces progressive disruption of reproductive cycles in sheep. Endocrinology 158:493–498 144:1426–1434 10. Newbold RR 2004 Lessons learned from perinatal exposure to diethylstilbes- 36. NRC 1999 Hormonally active agents in the environment. Washington DC: trol. Toxicol Appl Pharmacol 199:142–150 National Academy Press; 430 11. Crews D, Bergeron JM, McLachlan JA 1995 The role of estrogen in turtle sex 37. Reynolds PJ, Lindahl IL, Cecil HC, Bitman J 1976 A comparison of DDT and determination and the effect of PCBs. Environ Health Perspect 103(Suppl methoxychlor accumulation and depletion in sheep. Bull Environ Contam 7):73–77 Toxicol 16:240–247 12. Fry DM, Toone TK 1981 DDT-induced feminization of gull embryos. Science 38. Botella B, Crespo J, Rivas A, Cerrillo I, Olea-Serrano MF, Olea N 2004 213:922–924 Exposure of women to organochlorine pesticides in southern Spain. Environ 13. Gray Jr LE, Ferrell JM, Ostby JS 1985 Alteration of behavioral sex differen- Res 96:34–40 tiation by exposure to estrogenic compounds during a critical neonatal period: 39. Kunze FM, Laug EP, Prickett CS 1950 The storage of methoxychlor in the fat effects of zearalenone, methoxychlor, and estradiol in hamsters. Toxicol Appl of the rat. Proc Soc Exp Biol Med 75:415–416 Pharmacol 80:127–136 40. Ikezuki Y, Tsutsumi O, Takai Y, Kamei Y, Taketani Y 2002 Determination of 14. Howdeshell KL, Hotchkiss AK, Thayer KA, Vendenbergh JG, vom Saal FS bisphenol A concentrations in human biological fluids reveals significant early 1999 Exposure to bisphenol A advances puberty. Nature 401:763–764 prenatal exposure. Hum Reprod 17:2839–2841 15. Stoker C, Rey F, Rodriguez H, Ramos JG, Sirosky P, Larriera A, Luque EH, 41. Schonfelder G, Wittfoht W, Hopp H, Talsness CE, Paul M, Chahoud I 2002 Munoz-de-Toro M 2003 Sex reversal effects on Caiman latirostris exposed to Parent bisphenol A accumulation in the human maternal-fetal-placental unit. environmentally relevant doses of the xenoestrogen bisphenol A. Gen Comp Environ Health Perspect 110:A703–A707 Endocrinol 133:287–296 42. Yamada H, Furuta I, Kato EH, Kato EH, Kataoka S, Usuki Y, Kobashi G, Sata 16. Tchernitchin AN, Tchernitchin N 1992 Imprinting of paths of heterodiffer- F, Kishi R, Fujimoto S 2002 Maternal serum and amniotic fluid bisphenol A entiation by prenatal or neonatal exposure to hormone, pharmaceuticals, pol- concentrations in the early second trimester. Reprod Toxicol 16:735–739 lutants and other agents or conditions. Med Sci Res 20:391–397 43. Rhind SM, Rae MT, Brooks AN 2001 Effects of nutrition and environmental 17. Heindel JJ 2003 Endocrine disruptors and the obesity epidemic. Toxicol Sci factors on the fetal programming of the reproductive axis. Reproduction 122: 76:247–249 205–214 18. Newbold RR, Padilla-Banks, Snyder RJ, Jefferson WN 2005 Developmental 44. Sawyer HR, Smith P, Heath DA, Juengel JL, Wakefield SJ, McNatty KP 2002 exposure to estrogenic compounds and obesity. Birth Defects Res (Part A) Formation of ovarian follicles during fetal development in sheep. Biol Reprod 73:478–480 66:1134–1150 19. Alonso-Magdalena P, Morimoto S, Ripoll C, Fuentes E, Nadal A 2006 The 45. Phillips DJ, Smith PR, Heath DA, Condell LA, McNatty KP 1992 Bioactive estrogenic effect of bisphenol A disrupts pancreatic ␤-cell function in vivo and immunoreactive FSH and immunoreactive inhibin concentrations in the induces insulin resistance. Environ Health Perspect 114:106–112 ovine fetus. J Endocrinol 134:287–295 20. EPA 1997 Special report on environmental endocrine disruption: an effects 46. Upmeier A, Degen GH, Diel P, Michna H, Bolt HM 2000 Toxicokinetics of assessment and analysis. Washington, DC: U.S. Environmental Protection bisphenol A in female DA/Han rats after a single i.v. and oral administration. Agency, Risk Assessment Forum. EPA630R96012 Arch Toxicol 74:431–436 21. NTP 2001 Final report of the endocrine disruptors low dose peer review panel. 47. Clarkson TW 1995 Environmental contaminants in the food-chain. Am J Clin In: Endocrine disruptors low dose peer review. Research Triangle Park, NC: Nutr 61(Suppl):S682–S686 National Toxicology Program (http://ntp.server.niehs.nih.gov/htdocs/lia- 48. Nunez AA, Kannan K, Giesy JP, Fang J, Clemens LG 2001 Effects of bisphenol son/LowDoseWebPage.html) A on energy balance and accumulation in brown adipose tissue in rats. Che- 22. Honma S, Suzuki A, Buchanan DL, Katsu Y, Watanabe H, Iguchi T 2002 Low mosphere 42:917–922 dose effect of in utero exposure to bisphenol A and diethylstilbestrol on female 49. Kannan K, Kajiwara N, Le Boeuf BJ, Tanabe S 2004 Organochlorine pesticides mouse reproduction. Reprod Toxicol 16:117–122 and polychlorinated biphenyls in California sea lions. Environ Pollut 131:425– 23. vom Saal FS, Hughes C 2005 An extensive new literature concerning low-dose 434 effects of bisphenol A shows the need for a new risk assessment. Environ 50. Niswender GD, Reichert Jr LE, Midgley Jr AR, Nalbandov AV 1969 Radio- Health Perspect 113:926–933 immunoassay for bovine and ovine luteinizing hormone. Endocrinology 84: 24. Wood RI, Foster DL 1998 Sexual differentiation of reproductive neuroendo- 1166–1173 crine function in sheep. Rev Reprod 3:130–140 51. Padmanabhan V, McFadden K, Mauger DT, Karsch FJ, Midgley Jr AR 1997 25. Crespi EJ, Steckler T, Mohankumar PS, Padmanabhan V 2006 Prenatal tes- Neuroendocrine control of follicle-stimulating hormone (FSH) secretion. I. tosterone excess and IGF bioavailability during intrauterine growth retarda- Direct evidence for separate episodic and basal components of FSH secretion. tion and postnatal catch-up growth in sheep. J Physiology 572:119–130 Endocrinology 138:424–432 26. Manikkam M, Crespi EJ, Doop DD, Herkimer C, Lee JS, Yu S, Brown MB, 52. Padmanabhan V, Evans NP, Dahl GE, McFadden KL, Mauger DT, Karsch FJ Foster DL, Padmanabhan V 2004 Fetal programming: prenatal testosterone 1995 Evidence for short or ultrashort loop negative feedback of gonadotropin- excess leads to fetal growth retardation and postnatal catch-up growth in releasing hormone secretion. Neuroendocrinology 62:248–258 sheep. Endocrinology 145:790–798 53. Berrie RA, Hallford DM, Galyean ML 1995 Effects of zinc source and level 27. Sharma TP, Herkimer C, West C, Ye W, Birch R, Robinson, JE, Foster DL, on performance and metabolic hormone concentrations of growing and fin- Padmanabhan V 2002 Fetal programming: prenatal androgen disrupts posi- ishing lambs. Prof Anim Sci 11:149–156 5966 Endocrinology, December 2006, 147(12):5956–5966 Savabieasfahani et al. • Prenatal EDC Exposure and Reproductive Dysfunction

54. Herbosa CG, Foster DL 1996 Defeminization of the reproductive response to without affecting pituitary size or LH and prolactin secretion in male rats. photoperiod occurs early in prenatal development in the sheep. Biol Reprod Toxicol Health 15:37–47 54:420–428 73. Vom Saal FS, Timms BG, Montano MM, Palanza P, Thayer KA, Nagel SC, 55. Veldhuis JD, Johnson ML 1986 Cluster analysis: a simple versatile, and robust Dhar MD, Ganjam VK, Permigiani S, Welshons WV 1997 Prostate enlarge- algorithm for endocrine pulse detection. Am J Physiol 250:E486–E493 ment in mice due to fetal exposure to low doses of estradiol or diethylstil- 56. Goodman RL 1994 Neuroendocrine control of the ovine estrous cycle. In: besterol and opposite effects at high doses. Proc Natl Acad Sci USA 94:2056– Knobil E, Neill JD, eds. The physiology of reproduction,. New York: Raven 2061 Press; 659–709 74. Cummings AM 1997 Methoxychlor as a model for environmental estrogens. 57. Brotons JA, Olea-Serrano MF, Villalobos M, Pedraza V, Olea N 1995 Xe- Crit Rev Toxicol 27:367–379 noestrogens released from lacquer coatings in food cans. Environ Health 75. Nikaido Y, Yoshizawa K, Danbara N, Tsujita-Kyutoku M, Yuri T, Uehara N, Perspect 103:608–612 Tsubura A 2004 Effects of maternal xenoestrogen exposure on development 58. Olea N, Pulgar R, Perez P, Olea-Serrano F, Rivas A, Novillo-Fertrell Am of the reproductive tract and mammary gland in female CD-1 mouse offspring. Pedraza V, Soto AM, Sonnenschein C 1996 Estrogenicity of resin-based com- Reproductive Toxicology 18:803–811 posites and sealants used in dentistry. Environ Health Perspect 104:298–305 76. Laws SC, Carey SA, Ferrell JM, Bodman GJ, Cooper Rl 2000 Estrogenic 59. Yamamoto T, Yasuhara A 1999 Quantities of bisphenol A leached from plastic activity of octylphenol, nonylphenol, bisphenol A and methoxychlor in rats. waste samples. Chemosphere 38:2569–2576

Toxicol Sci 54:154–167 Downloaded from https://academic.oup.com/endo/article/147/12/5956/2501261 by guest on 27 September 2021 60. Yoshida T, Horie M, Hoshino Y, Nakazawa H 2001 Determination of bis- 77. Herbosa CG, Wood RI, I’Anson H, Foster DL 1994 Prenatal photoperiod and phenol A in canned vegetables and fruit by high performance liquid chro- the timing of puberty in the female lamb. Biol Reprod 50:1367–1376 matography. Food Addit Contam 18:69–75 78. Foster DL, Ryan KD, Goodman RL, Legan SJ, Karsch FJ, Yellon SM 1986 61. Mountfort KA, Kelly J, Jickells SM, Castle L 1997 Investigations into the Delayed puberty in lambs chronically treated with estradiol. J Reprod Fertil potential degradation of polycarbonate baby bottles during sterilization with 78:111–117 consequent release of bisphenol A. Food Addit Contam 14:737–740 79. Malcolm KD, Jackson LM, Bergeon C, Lee TM, Padmanabhan V, Foster DL 62. Mulkey ME 2002 Methoxychlor: tolerance revocations. Fed Regist 46906– 2006 Chronic exposure to estradiol during gestation and suckling alters LH 46909 surge dynamics, increases aggressive behavior and delays puberty in the 63. Skinner MK, Anway MD 2005 Seminiferous cord formation and germ-cell female sheep. Biol Rep Special issue, Abstract 422, p 166 programming: epigenetic transgenerational actions of endocrine disruptors. 80. Foster DL, Lemons JA, Jaffe RB, Niswender GD 1975 Sequential patterns of Ann NY Acad Sci 1061:18–32 circulating luteinizing hormone and follicle stimulating hormone in female 64. Sutherland JE, Costa M 2003 Epigenetics and the environment. Ann NY Acad Sci 983:151–160 sheep from early postnatal life through the first estrous cycles. Endocrinology 65. Watson RE, Goodman JI 2002 Epigenetics and DNA methylation come of age 97:985–994 in toxicology. Toxicol Sci 67:11–16 81. Rubin BS, Murray MK, Damassa DA, King JC, Soto AM 2001 Perinatal 66. Barker DJP 1994 Programming the baby. In: Mothers, babies, and disease in exposure to low doses of bisphenol A affects body weight, patterns of estrous later life. London: BMJ Publishing Group; 14–36 cyclicity, and plasma LH levels. Environ Health Perspect 109:675–680 67. Nathanielz PW 1999 Life in the womb. Origin of adult diseases. Ithaca, NY: 82. Bowen JM, Dahl GE, Evans NP, Thrun LA, Wang Y, Brown MB, Karsch FJ Promethean Press 1998 Importance of the gonadotropin-releasing hormone (GnRH) surge for 68. Padmanabhan V, Manikkam M, Recabarren S, Foster DL 2006 Prenatal induction of the preovulatory luteinizing hormone surge of the ewe: dose- testosterone programs reproductive and metabolic dysfunction in the female. response relationship and excess of GnRH. Endocrinology 139:588–595 Mol Cell Endocrinology 246:165–174 83. Skinner DC, Harris TG, Evans NP 2000 Duration and amplitude of the luteal 69. Ozanne SE 2001 Metabolic programming in animals. Br Med Bull 60:143–152 phase progesterone increment times the estradiol-induced luteinizing hor- 70. Gaido KW, Maness SC, McDonnell DP, Dehal SS, Kupfer D, Safe S 1999 mone surge in ewes. Biol Reprod 63:1135–1142 Differential interaction of the methoxychlor metabolite 2,2-bis-(p-hydroxy- 84. Miller KP, Gupta RK, Greenfeld CR, Babus JK, Flaws JA 2005 Methoxychlor phenyl)1,1,1-trichloroethane with estrogen receptors ␣ and ␤. Endocrinology directly affects ovarian antral follicle growth and atresia through Bcl-2- and 140:5746–5753 Bax-mediated pathways. Toxicol Sci 88:213–221 71. Gaido KW, Maness SC, McDonnell DP, Dehal SS, Kupfer D, Safe S 2000 85. Evans NP, Dahl GE, Mauger DT, Padmanabhan V, Thrun LA, Karsch FJ 1995 Interaction of methoxychlor and related compounds with estrogen receptor ␣ Does estradiol induce the preovulatory gonadotropin-releasing hormone and ␤, and androgen receptor: structure-activity studies. Mol Pharmacol 58: (GnRH) surge in the ewe by inducing a progressive change in the mode of 852–858 operation of the GnRH neurosecretory system. Endocrinology 136:5511–5519 72. Gray Jr LE, Ostby J, Cooper RL, Kelce WR 1999 The estrogenic and antian- 86. Butcher RL, Fugo NW 1967 Overripeness and the mammalian ova. II. Delayed drogenic pesticide methoxychlor alters the reproductive tract and behavior ovulation and chromosome anomalies. Fertil Steril 18:297–302

Endocrinology is published monthly by The Endocrine Society (http://www.endo-society.org), the foremost professional society serving the endocrine community.