Water and soil pollution as determinant of water and food quality/contamination and its impact on female fertility

BPA has been shown to affect female fertility in humans. Several cohort studies have examined BPA levels at different reproductive endpoints in women undergoing fertility treatments. In one study, Ehrlich et al. [24] prospectively measured urinary BPA concentrations in 174 women, aged between 18 and 45, who underwent a total of 237 in vitro fertilization (IVF) cycles. After adjusting for age, body mass index (BMI), day 3 follicle-stimulating hormone (FSH) and smoking, the authors reported that elevated BPA levels were associated with lower number of oocytes retrieved, fewer mature metaphase II oocytes, fewer normally fertilized oocytes, lower serum E2 levels, and a trend for having lower blastocyst formation [24]. These results suggest that BPA is associated with poorer reproductive outcome in infertile women undergoing IVF. Similarly in another study by Mok-Lin et al. [25], the authors measured urinary BPA levels in 84 women who underwent a total of 112 IVF cycles and showed that higher total urinary BPA was significantly correlated with poorer ovarian response, as reflected by fewer oocytes retrieved per cycle and lower serum E2 levels. Higher urinary BPA also correlated with reduced oocyte maturation and lower fertilization rates [25]. In another study by Bloom et al. [26], the authors measured fasting serum BPA levels in 44 women who underwent IVF. Although that study showed that higher BPA levels were significantly associated with lower serum E2 levels per mature follicle, it did not find a significant correlation between BPA and the number oocytes retrieved per IVF cycle [26].

On the other hand, a large and well-designed study by Minguéz-Alarcón et al. [27] found no association between urinary BPA concentrations and IVF outcomes. That prospective cohort study (from 2004 to 2012) was performed at the Massachusetts Fertility Center and included 256 women who underwent 375 IVF cycles. Each woman provided two urine samples prior to oocyte retrieval. General linear mixed models with random intercepts were used to evaluate the association between BPA and IVF outcomes, based on data abstracted from electronic medical records on intermediate and clinical end-points of IVF treatments. The results specifically showed no association between BPA concentrations and peak E2 levels, proportion of high quality embryos, fertilization rates, implantation, clinical pregnancy, or live birth rates per initiated cycle or per embryo transfer. The only significant finding was that there was a relationship between BPA and endometrial wall thickness that was modified by age. Although this was a well-designed study on a large group of women, the authors do agree with the general consensus that data on the relation between BPA exposure and reproductive outcomes remain scares, and that additional research is necessary to clarify BPA’s role in human reproduction.

The adverse impact that BPA exposure has on fertility in women could be attributed to changes in sex hormone concentrations. Many studies have shown a relationship between BPA and PCOS, one of the most common endocrine disorders in reproductive-aged women associated with hyperandrogenemia. Kandaraki et al. [28] performed a cross sectional study on women with (n = 71) or without (n = 100) PCOS that were matched by age and BMI. The authors reported that blood BPA levels in the PCOS group were significantly higher than BPA levels in the control group. Even when women were categorized into lean and overweight subgroups, PCOS women in both lean and overweight groups had significantly higher serum BPA levels than the control group. There was also a significant association between serum BPA levels and both testosterone and androstenedione, as well as positive correlation with insulin resistance in the PCOS group. The results of that study suggest that environmental exposure to BPA may play a role in the complex pathophysiology of PCOS [28].

Another study by Takeuchi et al. [29] found similar results. The authors measured serum BPA levels in 26 women with normal menstrual cycles (control group) (17 of whom were obese); 19 women with PCOS (6 of whom were obese); 7 women with hyperprolacinemia, and 21 women with hypothalamic amenorrhea. Both obese and normal weight women with PCOS had significantly higher BPA levels than normal weight controls. The study also reported that serum BPA levels were positively correlated with serum testosterone (total and free), androstenedione, and dihydroepiendrosterone (DHEA) concentrations in all study participants. Therefore, this study further shows a relationship between BPA and androgen levels further implying that BPA may play a role in the pathophysiology of PCOS [29]. Interestingly, in vitro exposure to BPA at low doses does not affect granulosa cell steroidogenesis, while at supra-physiological concentrations, BPA alters progesterone and estradiol synthesis and significantly reduces the mRNA and protein expression levels of 3β-HSD, CYP11A1 and CYP19A1 [30]. Additionally, in one study, urinary BPA concentrations did not correlate with CYP19 mRNA expression levels in granulosa cells of women who underwent IVF [31].

Although studies have shown consistency in the relationship between BPA and PCOS, the association is still vague since PCOS is a complex endocrine problem associated with elevated androgen and insulin resistance. Whether BPA has a causal relationship rather than only correlation with PCOS or one of its features, such as elevated androgens or hyperinsulinemia, remains to be determined.

Not only BPA affects steroid production (as mentioned above in the animals section), but the reverse could be true, i.e., steroids, such as androgens, can affect BPA levels. Since high levels of BPA are observed in hyperandrogenemic women with PCOS, a study [32] investigated the effect of testosterone on BPA metabolism. Ovariectomized female rats were injected subcutaneously with increasing doses of testosterone propionate daily for 2 weeks after which serum BPA levels were measured. The results showed that serum BPA increased with testosterone propionate administration in dose-dependent manner. The authors also quantified the enzyme reaction of BPA glucuronidation in the rat liver. Their results showed that the ratio of glucuronide in the rats was significantly reduced in a testosterone dose-dependent manner. Additionally, the relative mRNA expression of UDP-glucuronosyltransferase 2B1 (UGT2B1) showed a testosterone dose-dependent decrease. The ratio of BPA glucuronidation and UGT2B1 mRNA levels were significantly lower in environment with elevated testosterone. That study concluded that the clearance of BPA might be slowed down in the presence of high levels of testosterone, thus potentially explaining the elevated serum BPA levels in hyperandrogenic women with PCOS.

The relationship between BPA levels/exposure and puberty in humans has been evaluated in several studies and the results have been controversial. In a cross sectional study, Wolff et al. [33] did not find an association between urinary BPA levels and premature puberty in 9-year-old girls (n = 192). In another prospective cohort study of girls aged between 6 and 8, Wolff et al. [34] did not find an association between urinary BPA levels and breast or pubic hair development. On the other hand, Qiao et al. [35] showed that serum BPA levels were significantly elevated in girls with precocious puberty compared to controls and higher serum BPA levels were positively correlated with increased uterine and ovarian volume. A recent review article [36] reported that out of 19 studies, only 7 showed a correlation between BPA and puberty. Taken altogether [36], although data in animal models showed a relationship between BPA exposure and early puberty (above section on animals), available data to date in humans do not show a clear role for BPA in pubertal development in humans due to the conflicting results among all studies examined.

Exposure to BPA may be associated with recurrent pregnancy loss. A study [37] showed that serum BPA levels in women with a history of three or more consecutive first-trimester miscarriages (n = 45) were significantly higher compared to serum BPA levels in 32 healthy women (no history of live birth and infertility). In a case-control study in eastern China [38], total urinary BPA concentrations were measured in 102 women with recurrent miscarriages and 162 control women (all participants aged 20–40 years). The creatinine-adjusted BPA levels were significantly higher in women with recurrent miscarriages compared to control women. Additionally, higher level of urinary BPA was significantly associated with a 3–9 times increased risk of recurrent miscarriages [38]. Clearly, more prospective well-designed studies are needed to better assess the relationship between BPA and recurrent pregnancy loss.

As mentioned earlier, evidence from toxicological studies in animals has demonstrated that phthalates could adversely impact fertility through effects on folliculogenesis, steroidogenesis, oocyte maturation and embryonic development, but human data are scarse. Concentrations of eight phthalate metabolites in 110 follicular fluid and urine samples were collected from women (n = 112) attending an infertility clinic in China were quantified and the results showed that follicular fluid and urinary metabolite concentrations of MEHP were not associated with any IVF parameters such as peak E2 levels, number of oocytes retrieved, number of mature oocytes, fertilization rates, number of good-quality embryos and rate of blastocyst formation [50]. However, that study was limited by the small sample size, which may not have adequate power to detect significant associations. Interestingly, among women with a history of infertility, the molar sum of urinary DEHP was significantly lower in women who conceived after IVF compared to those who did not [51]. Whether women pursuing fertility treatments take precautions to avoid exposure to environmental toxins, to improve treatment outcomes needs to be explored. Thus, given the prevalence of phthalates exposure, further larger studies are needed to elucidate the potential hazard on female reproduction in the setting of infertility treatment.

In a prospective study, Messerlian et al. [52] evaluated the association between 11 urinary phthalate metabolites and antral follicle growth in a study that included 215 infertile women. The higher levels of urinary phthalate concentrations negatively correlated with antral follicle count indicating that phthalates are associated with lower ovarian reserve in infertile women [52]. Interestingly, among women with a history of infertility, urinary DEHP levels were significantly lower in women who conceived after assisted reproductive technology compared to women who did not get pregnant [51]. Mural granulosa cells from 48 patients undergoing IVF were treated with increasing concentrations of dibutyl-phthalate in vitro following which gene microarray analysis was performed [53]. When compared with untreated cells, exposure to high doses of dibutyl-phthalate resulted in significant differences in expression of 346 annotated genes (151 were upregulated and 195 were downregulated). The main functional annotations affected were associated with cell cycle and mitosis thus indicating that acute treatment with high concentrations of dibutyl-phthalate alters gene expression pathways mainly associated with the cell cycle. Reinsberg et al. [54] collected human luteinized granulosa cells from women undergoing IVF and cultured them with varying concentrations of MEHP in the presence of FSH, hCG and cAMP after which they assessed steroidogenesis. MEHP suppressed aromatase expression and E2 production in a dose-dependent manner. MEHP did not, however, change the production of progesterone [54].

A study [55] assessed whether urinary concentrations of metabolites of phthalates and phthalate alternatives were associated with IVF outcome in 136 women where participants provided one to two urine samples per cycle during controlled ovarian stimulation and then before oocyte retrieval. Urinary concentrations of the sum of DEHP and other phthalate metabolites were negatively associated with the number of total oocytes retrieved, total number of mature oocytes, total number of fertilized oocytes, and good quality embryos but none of the urinary phthalate metabolite concentrations were associated with a reduced implantation, lower clinical pregnancy or lower live birth rates [55].

Literature to date indicates that phthalates could inhibit the size of the growing antral follicle pool and could potentially impair fertility and fecundity. There is a need of further investigations pertaining to the impact of phthalates on the human oocyte and follicular development.

As of today, there are no studies focusing on the relationship between DEHP and MEHP with PCOS. In one study, 52 subjects with PCOS had lower urine concentrations of phthalate metabolites compared to subjects without PCOS [56]. In another study including 244 girls, the sum of phthalate metabolites was protective against PCOS in adolescence where there was a negative association of phthalate with PCOS and of phthalate with serum anti-Mullerian hormone [57]. Future studies are needed to confirm these preliminary findings and determine if DEHP and MEHP may have a role in the pathogenesis of PCOS.

There is preliminary contentious evidence showing that early pregnancy may be adversely affected by DEHP exposure. The first study to show this association included Danish women (n = 128) who reported increased risk of early pregnancy loss with higher urinary concentrations of the DEHP metabolite MEHP [58]. In another study that included women (n = 256) undergoing assisted reproduction, there was an increased conception cycle-specifc urinary concentrations of total sum of DEHP and individual DEHP metabolites were associated with biochemical pregnancy loss [59]. On the other hand, menstrual cycle–specific estimates of urinary phthalate metabolites in 221 women were not associated with detrimental alterations in follicular-phase length, time to pregnancy, or early pregnancy loss; rather DEHP metabolites were associated with reduced early loss [60]. Thus there is no clear consensus on whether DEHP/MEHP are related to early pregnancy loss and there is a need for such studies.

There is a possible link between phthalate esters and endometriosis. Cobellis et al. [61] collected blood and peritoneal fluid samples from 55 women with endometriosis and 24 age-matched women without endometriosis. Women with endometriosis had significantly higher plasma DEHP concentrations than control women and the majority of women with endometriosis had detectible levels of DEHP and/or MEHP in the peritoneal fluid. Similarly, another study by Kim et al. [62] showed that the urinary concentration MEHP, mono (2-ethyl-5-oxohexyl) phthalate and mono (2-ethyl-5-carboxyphentyl) phthalate, were significantly higher in women with endometriosis compared to women without endometriosis. Another prospective study by Kim et al. [63] showed that 97 women with advanced-stage endometriosis had significantly higher plasma levels of MEHP and DEHP compared to 169 control women [63]. On the other hand, Itoh et al. [64] found no significant association between endometriosis and 6 urinary different phthalate monoesters in infertile Japanese women who had laparoscopy for diagnosis of endometriosis after adjusting for urinary creatinine. In that study, the control group subjects were categorized as stages 0–1 endometriosis (n = 80) and the experimental group subjects were categorized as stages 2–4 endometriosis (n = 57).

Interestingly, a study [65] treated human endometrial stromal cells with different concentrations of DEHP and assessed ROS generation, expression levels of antioxidant enzymes, alteration of MAPK/NF-κB signaling and hormonal receptors. DEHP increased ROS generation and decreased expression of superoxide dismutase (SOD), glutathione peroxidase (GPX), heme oxygenase (HO), and catalase (CAT). DEHP induced p-ERK/p-p38 and NF-κB mediated transcription and induced estradiol receptor-α expression in a dose-dependent manner. That study suggests that DEHP may be associated with the development of endocrine-related disease such as endometriosis.

Finally, consistent with the substantial body of evidence from animal studies, more human studies are emerging linking phthalates with altered female reproductive system. Considering that phthalate exposure is nearly universal, these results may have important clinical and public health relevance.

A number of studies have suggested that PFAS exposure can seriously impair the reproductive health in humans, both in fertility and infertility setting. A study by Fei et al. [71] showed that women with higher serum PFOA levels had higher rates of subfecundity and longer time to achieve a pregnancy. They also reported that women with the highest PFOA exposure had increased rates of menstrual cycle irregularities. In the Maternal-Infant Research on Environmental Chemicals (MIREC) Study, a cohort study of 2001 women recruited before 14 weeks of gestation in 10 cities across Canada, the investigators reported that, after adjustment for potential confounders, PFOA and PFHxS were associated with approximately 10% reduction in fecundability per one standard deviation increase; however, no significant association was observed for PFOS [72]. In addition, the odds of infertility increased by 31% per one standard deviation increase of PFOA and by 27% per one standard deviation increase of PFHxS, while no significant association was observed for PFOS [72].

Studies on the exposure to PFAAs and female fertility have provided conflicting results. Jorgensen et al. [73] evaluated human fecundity by measuring time to pregnancy in women from different geographical populations (Greenland, Poland and Ukraine) representing varying PFAS exposure and pregnancy planning behaviors. They assessed the association between serum levels of PFOA, PFOS, PFHxS and PFNA in those women and rates of infertility (defined as time to pregnancy more than 13 months). They found that higher PFNA levels were associated with infertility in the pooled sample and specifically in women from Greenland. PFNA effect on infertility was weaker for women from Poland and Ukraine. Although they found that PFNA levels could be associated with infertility, they did not find this association for other PFAS such as PFOS, PFOA or PFHxS. In a follow up study, Bach et al. [74] investigated the association between PFAS and infertility in additional populations. In a pooled analysis including parous and nonparous women, they found that the fecundibility rates were lower in women with higher levels of PFOS and PFOA. PFOS was not associated with higher rates of infertility but there was a trend for an association between infertility and PFOA in parous women. Bach et al. [75] found no association between PFAA levels in maternal serum prior to 20 weeks gestation and diagnosis of infertility in nulliparous women (n = 1372). This is consistent with their previous findings [74] where only a trend for an association between infertility and PFOA in parous but not nulliparous women was reported. Interestingly, there is evidence showing that follicular fluid levels of perfluorinated compounds in women undergoing IVF have a detrimental effect of oocytes fertilization capacity with subsequent decrease in the number of embryos transferred [76].

Altogether, there is a mild evidence that exposure to PFAS, even at low levels, may reduce fecundability and that environmental exposure to PFAS impairs female fecundity by delaying time taken to conceive.

A number of studies have suggested that the effects of PFAS on reproductive health and development are mediated by their effect on the hormonal milieu. In a case control study, subjects with PCOS (n = 52) had significantly higher geometric mean serum concentrations of PFOA and PFOS compared to controls (n = 50) [56]. That study suggest that women with PCOS could have a different environmental contaminant profile. Barrett et al. [77] found that certain PFAS are associated with ovarian hormonal changes in certain populations of reproductive-aged women. They measured daily salivary levels of E2 (calculated mean follicular levels) and progesterone (calculated mean luteal levels) as well as daily serum levels of PFAS (including PFOS and perfluoroctanoic acid) in young healthy regularly cycling women (n = 178) in a single menstrual cycle. They found that in nulliparous, but not parous, women that PFOS and perfluorooctanesulfonic acid levels were inversely associated with E2 and progesterone levels. Tsai et al. [78] evaluated the association between PFAS serum concentrations and reproductive hormones in young Taiwanese adults and adolescents (between 12 and 30 years old) and found that serum POFA, PFOS and PFDA levels were negatively associated with serum levels of SHBG, FSH and testosterone--associations that were strongest in females aged between 12 and 17. Maissonet et al. [79], in Avon Longitudinal Study of Parents and Children (ALSPAC), found that prenatal exposure to PFASs can affect the hormonal milieu even later in life. They assessed pregnant women (n = 72) at 16 weeks gestation for serum levels of PFAA and then measured total testosterone and SHBG in their daughters at 15 years of age. They found that total testosterone concentrations were higher in daughters with prenatal exposure to PFOS or PFOA but not PFNA. SHBG was not affected by exposure to any PFAAs in utero. These results indicate that exposure to certain PFAA (PFOS, PFOA, PFHxS) in utero can lead to alterations in a woman’s testosterone levels later in life.

In brief, PFAS seem to be related to parity and could affect steroidogenesis. These potential alterations could lead to abnormally elevated androgens and might theoretically contribute to the complex pathogenesis of PCOS.

Equivocal findings have been reported for the association between PFAS and miscarriages. A prospective study assessed PFAS and pregnancy loss in couples (n = 501) that were followed daily from preconception through the 7th week post-conception. There was no significant association between pregnancy loss and any of the 7 PFASs that were quantified: 2-N-ethyl-perfluorooctane sulfonamide acetate (Et-PFOSA-AcOH); 2-N-methyl-perfluorooctane sulfonamido acetate (Me-PFOSA-AcOH); perfluorodecanoate (PFDeA); PFNA; perfluorooctane sulfonamide (PFOSA); PFOS; and PFOA. Limitations of that study were that women used home pregnancy test kits, and that pregnancy loss was documented by conversion from a positive to a negative pregnancy test, onset of menses or clinical confirmation. Similarly, another study showed no association between serum PFOA or serum PFOS levels with miscarriage rate [80]. In a prospective study of miscarriage in a population exposed to high levels of PFOA and PFOS, there was little evidence of association with serum levels of PFOA and limited evidence of association with serum levels of PFOS [81]. As of today, it is hard to draw from the evidence to date a clear conclusion between the relationship between PFAS and pregnancy loss.