Bisphenol A Affects on the Functional Properties and Proteome of Testicular Germ Cells and Spermatogonial Stem Cells in vitro Culture Model

The endocrine disruptor bisphenol A (BPA) is well known for its adverse effect on male fertility. Growing evidence suggests that BPA may interact with testicular germ cells and cause infertility as a result of its estrogenic activity. Objective of current in vitro study was to investigate the proliferation, survivability and stemness properties of mouse testicular germ cells exposed to BPA, and to evaluate possible expression of cellular proteome. Our results showed that germ cell viability and proliferation were not affected by low concentrations (0.01, 0.1, 1, and 10 µM) although significant reduction observed at 100 µM BPA. Germ cell self-renewal and differentiation related marker proteins expression found unchanged at those concentrations. When BPA-exposed germ cells were transplanted into recipient testes, we observed fewer colonies at higher concentrations (10 and 100 µM). Additionally, a significant frequency of recombination failure during meiosis was observed in 10 µM BPA-exposed germ cell transplanted recipient. Moreover, experiment on continuous BPA-exposed and 100 µM BPA-recovered germ cells suggested that spermatogonial stem cells are more potential to survive in adverse environment. Finally, scrutinizing differentially expressed cellular proteins resulted from our proteomic analysis, we conclude that BPA exposure might be associated with several health risks and infertility.


Results
BPA hinders testicular germ cell proliferation. Firstly, we used germ cell lines from ICR (CD-1) and C57 GFP transgenic mice for the visual comparison of cultured cells under brightfield and fluorescent microscope (Fig. 1A). BPA was administrated to CD-1 and C57 GFP germ cell lines ranging from 0.01 to 100 µM in a 10-fold increasing pattern and cells were cultured for 1 week to examine cell proliferation and viability. There was a sharp decline in germ cell number (Fig. 1B) and remarkable decrease in viability at highest BPA concentration (100 µM) following the decline starting point at 1 µM BPA (Fig. 1C). We observed similar patterns of cell proliferation and viability for both wild-type (CD-1) and transgenic (C57 GFP) mice. So, we planned to use transgenic cell line for the subsequent experiments as it is easily visualized in recipient testis after germ cell transplantation. For every set of BPA-treated cultures, we also prepared control cultures where cell count and viability were optimum which indicated the utmost culture conditions. BPA induces apoptosis of cultured SSCs. We observed that the highest concentration of BPA caused damage to germ cell viability and proliferation. Therefore, we intended to observe the percentage of apoptotic cells among the treatment groups. Harvested germ cells were used immediately for staining of Annexin V and propidium iodide (PI) and we detected significantly higher percentages of apoptotic cells in 10 µM BPA and 100 µM BPA treatment groups. 0.01, 0.1, and 1 µM BPA-exposed cells showed that the percentage of apoptotic cells were similar to control. Highest proportion of apoptotic germ cells were found in 100 µM BPA treatment ( Fig. 2A and B).

Evaluation of spermatogonia related marker proteins.
To confirm undifferentiated spermatogonia related marker proteins PLZF, GFRα1 and RET, we investigated their expression patterns in BPA-treated cultured germ cells. In immunocytochemistry, there were no significant differences observed in the percentages of PLZF, GFRα1 and RET positive germ cells (Fig. 3A, B, C, E, F, and G). Differentiated spermatogonia related marker protein C-KIT was also examined. C-KIT was expressed in a very small percentage of cells with no significant differences among BPA treatments and control ( Fig. 3D and H).

Effects of BPA on SSCs stemness properties. Germ cells treated with different concentrations of
BPA were transplanted into recipient C57/BL6 mice. After 2.5 months, the transplanted testes were collected, and donor cell derived C57 GFP positive colonies were counted under Nikon AZ100 microscope (Fig. 4A). Quantification of donor cell derived colonies indicated that SSCs among the cultured germ cells treated with low BPA concentrations (0.01, 0.1, and 1 µM) could not alter their stemness properties. At relatively higher BPA concentrations (10 and 100 µM), SSCs could produce significantly less colonies compared to the number of transplanted germ cells and total number of cultured germ cells ( Fig. 4B and C). This result indicated that stem cell like characteristics of SSCs could reduce significantly with BPA exposure.
BPA can induce meiotic abnormalities in SSCs. At a concentration of 10 µM, BPA did not affect the viability and proliferation of germ cell in vitro but showed significantly higher proportion of apoptotic cells.
Additionally, we observed that treatment of SSCs with 10 µM BPA resulted in fewer colonies in the recipient testes after transplantation. Therefore, we intended to examine some proteins expression pattern related to meiotic stages and that was why DMSO (control) and 10 µM BPA-treated germ cell transplanted mice were analyzed for this experiment. Among the nuclei of pachytene spermatocyte, we did not observe any difference in synaptonemal complex (SC) length of autosomes between control and 10 µM BPA treatment group (Fig. 5C); however, chromosomes that lacked MutL homolog 1 (MLH1, crossover-associated protein) foci were found significantly high in 10 µM BPA treatment group ( Fig. 5A and B). This result indicates that BPA exposure can trigger germ cells to produce "crossover-less" SCs and can increase the frequency of recombination failure. We made our next move to examine the number of RAD51 (assists in repair of DNA double strand breaks) foci among zygotene spermatocyte nuclei. We did not find any remarkable difference between control and 10 µM BPA treatment group ( Fig. 5D and E). Finally, we checked the expression of meiotic DNA double-strand breaks (mDSB) indicating protein γH2AX in pachytene nuclei. γH2AX is spatially and temporally linked to DSB in leptotene and zygotene spermatocytes stages, disappears from synapsed autosomes in the pachytene stage but remains on the sex chromosomes 31 . Consistent with these results, we observed expression of γH2AX only in the unpaired regions of the X and Y chromosomes (Fig. 5F).
Effects of long term BPA exposure on germ cell enriched for SSCs. To investigate germ cell proliferation strategy in case of continuous BPA exposure, we conducted a long term (at least for 3 weeks) exposure of different concentrations of BPA (Fig. 6A). The cultured cell proliferation rate was evaluated and we chose 1, 10, and 100 µM BPA concentrations according to their effects found at cellular level. As shown in Fig. 6B, proliferation of 1 and 10 µM BPA exposed germ cells increased gradually from week 1 to week 3 and cell numbers and are analyzed by one-way ANOVA ( #, *compared with control, P < 0.05; ###, ***compared with control, P < 0.001).
were found significantly different between 1 and 10 µM BPA at week 3. A dramatic decrease of cell proliferation was observed at 100 µM BPA treatment; only a small percentages of cells seeded at week 1 survived until week 3 (Fig. 6B). Therefore, to investigate the functional status of SSCs after long exposure to BPA, transplantation to recipient mice were performed with germ cells exposed to control, 1, and 10 µM BPA for 3 weeks. 100 µM BPA-treated cultured cells were excluded due to the small number of cells harvested, which was insufficient for transplantation. Colonies were counted 2.5 months after transplantation, and the results indicated that the activity of SSCs subjected to long-term BPA exposure decreased in a dose-dependent manner, but no significant difference was observed compared with control group (Fig. 6C). However, analyzing colony number among the total number of cultured germ cells, we observed a significant reduction in colony number in case of continuous exposure to 10 µM BPA compared with 1 µM BPA (Fig. 6D).
Recovered SSCs from high BPA exposure increases colonization. A minute proportion of cells was observed to survive in vitro against 1 week 100 µM BPA treatment. We tried to scrutinize the proliferation and stem cell like abilities of the surviving germ cells. Therefore, 100 µM BPA was applied to germ cell culture  for 1 week and then survived cells (recovered germ cells) were used to set a new BPA-free culture panel for three consecutive weeks (Fig. 6E). We did not observe any remarkable difference in the number of recovered germ cell proliferation between control group and recovered group during the first 2 weeks; however, during 3 rd week of culture, recovered group proliferated more slowly than control with a significant difference in cell count (Fig. 6F). We performed germ cell transplantation into recipient mice to check the stemness properties of 100 µM BPA-recovered SSCs. We counted significantly higher number of SSC colonies in case of 100 µM BPA-recovered germ cells than control ( Fig. 6G) but the total number of colonies of recovered cells was not significantly different compared with control (Fig. 6H).

2-DE analysis of BPA-induced proteome alterations in germ cells.
A two-dimentional gel electrophoresis (2-DE) based method was applied for comparative proteomic analysis and to observe the changes in protein expression patterns after BPA treatment in germ cell culture. On average, 227 protein spots were detected in each gel among which 45 different spots showed dose dependent expression profile. Significant (P < 0.05) changes were noticed in 16 spots in case of control and 0.01, 0.1, 1, 10, and 100 µM BPA-treated groups and 9 of them were characterized as important proteins for cellular functions (Table 1). Peptidylprolyl isomerase A (PPIA), which is involved in inflammation, apoptosis, and other intracellular signaling was expressed significantly high in 100 µM BPA-treated germ cells compared with control. Similarly, expression of annexin A4 (ANXA4), which is involved in exocytic and endocytic pathways, was up-regulated in 100 µM BPA-treated germ cells compared with control. The expression of oxidative stress related protein peroxiredoxin 4 (PRDX4) and antioxidant defense initiated enzyme superoxide dismutase (SOD2) were also found as significantly high in 100 µM BPA-treated cells. Additionally, expression of heat shock proteins (HSPs) [10 kDa HSP (HSPE1), 60 kDa HSP (HSPD1) and 70 kDa HSP8 (HSPA8)] differed markedly in the 100 µM BPA-treated group; HSPE1 expression was up-regulated, while HSPD1 and HSPA8 expression were down-regulated. Glycosyl transferase group 1 (GLT1D1), an enzyme capable of producing several types of polysaccharides, was also down-regulated as a result of 100 µM BPA treatment. However, significantly higher expression of cell proliferation regulator protein prohibitin (PHB) was observed in 0.01 µM BPA-treated germ cells compared with all other groups including control (Table 1).
Based on the proteins detected by 2-DE, including proteins that were similarly or differentially expressed in germ cells treated with different BPA concentrations, we plotted them to observe their interrelationship. Thus, a co-relation among these proteins was observed regarding cell signaling, growth, apoptosis, and disease related pathways (Fig. 7A). Moreover, prediction of protein-protein interactions among the identified proteins based on the STRING system showed that oxidative stress and apoptosis related proteins were connected with HSPs, while HSPE1 is also associated with PHB (Fig. 7B). These protein-protein interactions were observed as gene co-occurrence, gene neighborhood and co-expression. These results suggest that even minor alterations in protein expression as a result of BPA exposure could result in up-or down-regulation of associated proteins, with wide-ranging impacts on cellular physiology.

Discussion
The xenoestrogenic compound BPA has become an important topic nowadays because of its ability to increase the risk of several diseases and health related issues throughout 50 years. Many studies have reported that BPA has vicious effects on reproductive system and sexual behavior of both males and females, following exposure during childhood or even adulthood [32][33][34] . BPA can also simulate the risk of breast cancer of female if exposure occurs during perinatal or fetal development 35 , and can promote the proliferation, invasiveness and metastasis of neuroblastoma 36 . Additionally, this EDC has been discovered as one of the risk factors of thyroid disfunction 37 , asthma 38 , heart disease 39 , and obesity 40 . Therefore, researchers have conducted their experiments to characterize its involvement in the pathogenesis of these diseases along with its chemical kinetics that can disrupt or alter normal physiological and neural development, especially within the tolerance range proposed by the Food and Drug Administration (FDA). Although the commercial use of BPA is progressively substituted by other bisphenols such as BPS (alternative in BPA-free thermal paper), BPF, and BPAF (in BPA-free plastics) 41 , some recent studies have also described them as hazardous components by characterizing their estrogenic activities 42 , negative effects on meiotic maturation of pig oocytes 43 , and above all, their presence in human plasma 41 .
BPA closely mimics the structure and function of the hormone estradiol, with the ability to bind to and activate the same estrogen receptor as the natural hormone 9 , which interferes with the function of androgens or male hormones. To clarify this, we used different concentrations of BPA in germ cell culture along with all required growth factors and maintained all necessary environmental factors. We observed devastating cell loss and a significant decline in viability at the highest BPA concentration (100 µM). High concentration of any substrates in cell culture media can hamper cell growth, and this effect can be severe for toxic compounds; thus, the profound cell loss observed in the presence of 100 µM BPA indicates that BPA is toxic to testicular germ cells at high doses.

Symbol
Protein description gi no.  Additionally, we observed a decline in germ cell number at micromolar (µM) concentration of BPA which suggests that this concentration level may damage the male reproductive system. Another study showed that exposure of BPA over 100 μM for 48 h was harmful to mouse Sertoli cells 30 . We found similar effect of this chemical on testicular germ cells. As BPA has estrogenic characteristic, low concentrations of BPA administration to germ cells and continuous BPA exposure in consecutive cultures were also considered as our further investigations. Considering the toxic nature of high doses (100 µM) of BPA on cell proliferation and viability, we measured the percentage of apoptotic cells in all treatment groups of cultured germ cells. We observed the highest percentage of apoptotic cells in the 100 µM BPA-treated group. However, 10 µM BPA-treated germ cells also showed a significantly higher proportion of apoptotic cells compared to control and cells treated with lower BPA doses, although significant differences in cell viability and proliferation were not observed at this concentration. These results suggest that 10 µM BPA might impair cell health and cause cell death in testicular germ cells, even though cell proliferation appeared normal in culture.
Our next move was to determine whether there was any BPA-induced change in cultured germ cell characteristics. In the process of self-renewal, cultured germ cells normally undergo numerous mitotic cell divisions, thereby increasing their number. We used specific marker proteins for self-renewal such as PLZF, GFRα1, and RET, as well as the differentiation marker protein C-KIT. Our immunocytochemistry (ICC) data indicated that there was no significant difference in the percentage of SSCs with self-renewal characteristic following treatment with various concentrations of BPA. Usually very few cultured germ cells exhibit expression of differential marker C-KIT and we observed no difference in C-KIT expression due to BPA exposure. Accumulation of these data indicate that BPA provides toxic environment to germ cell culture regarding their survivability without altering the expression of their characterized marker proteins.
Transplantation is the most accurate method available to examine the stemness properties of SSCs 44,45 and the actual number of stem cells among cultured germ cells. Therefore, germ cell transplantation was performed in view of observing BPA induced cellular defects in SSCs. We used C57BL/6 mice in which endogenous germ cells were removed by busulfan treatment, to enable C57 GFP cells to occupy the new niche. As shown in Fig. 4B and C, a slight increase in the number of SSCs was observed at lower BPA doses (0.01 and 0.1 µM) compared to control; however, this difference was not statistically significant. This result indicates that stem cells are able to tolerate BPA at a certain concentration and continue to proliferate. Additionally, considering10 5 transplanted germ cells and total number of cultured germ cells, dose dependent decreases in donor derived colony number were observed which indicates a decrease of functional SSCs number among BPA treated cultured germ cells, especially at higher doses. Interestingly, we observed a remarkable reduction in colony number among total number of cultured cells treated with 10 µM BPA which provided further evidence for a large population of apoptotic cells at this concentration. Therefore, based on the effects of 10 µM BPA on germ cell colonization and apoptosis, we considered this dose for further experiment related to meiotic abnormalities during spermatogenesis using spermatocytes from transplanted recipients. As shown in Fig. 5, we did not observe any change in autosomal SC length in pachytene spermatocytes (PS) between control and BPA-treated group, but did observe a significantly higher number of PS with SCs lacking at least one MLH1 focus. This could explain the result of recombination failure in unpartnered univalents at metaphase I which can promote cell death 25,46 . DNA double strand break repair protein RAD51 plays a major role in homologous recombination of DNA, we examined its expression in zygotene spermatocyte nuclei; however, no reduction in the number of RAD51 foci was observed in the BPA-treated group. Finally, we observed no expression of γH2AX in the synapsed autosomes of PS, which suggests that 10 µM BPA did not cause persistent mDSBs. Therefore, our results reveal that BPA exposure is not accompanied by a reduction in either SC length or the number of RAD51 foci, but may cause recombination failure during metaphase I of meiosis without causing any mDSB-related abnormalities.
Considering these experimental data, we decided to evaluate the proliferation and functional activities of cultured germ cells subjected to prolonged BPA exposure at three relatively high concentrations (1, 10, and 100 µM) for 3 weeks. Interestingly, 1 µM BPA-treated germ cells showed the highest proliferation at the end of week 3, suggesting that cell growth might be boosted by low concentrations of BPA 29 ; however, the difference was not significantly higher than control. These findings led us to think about the stemness status of SSCs among BPA-treated germ cells. However, our transplantation data did not show any significant reduction in stemness properties or proportion of SSCs among long-term BPA treated cells. Moreover, recovered germ cells from 100 µM BPA treatment showed acceleration in cell proliferation over the weeks. It is needless to say that BPA-exposed germ cells need a period of time to be cured from the toxic stress. So, proliferation of BPA-exposed cells was remarkably lower than control even at the end of week 3. Unexpectedly, we observed a significantly higher number of donor derived colonies into recipient per 10 5 BPA-recovered germ cells (Fig. 6G), suggesting that a higher proportion of SSCs among germ cells survived in 100 µM BPA treatment. Therefore, SSCs may be more robust to BPA administration in vitro among cultured germ cells and BPA-exposed SSCs can retain or regain their stemness properties during the subsequent BPA-free culture periods. However, as the total number of BPA-recovered germ cells was lower than control, we did not observe any change in colony count per total number of cultured germ cells.
Next, two dimensional gel electrophoresis (2-DE) was performed in order to determine the number and level of expressed proteins related to several cellular functions. It was observed that proteins related to apoptosis and oxidative stress, as well as heat shock proteins exhibited significantly high responses at 100 µM BPA. Expression of inflammation and apoptosis related protein PPIA, which has roles in intracellular signaling, protein folding and assembly, and cell cycle regulation 47 , was elevated in germ cells cultured with 100 µM BPA. One of the members of annexin family, ANXA4, has already been reported as subcellular localization from the nucleus to the cytoplasm in the time of apoptosis 48 . Evaluating the elevated expression of this protein, we hypothesize that higher concentration of BPA triggers apoptosis. Similarly, antioxidant defense protein PRDX4 and oxidative stress related mitochondrial enzyme SOD2 were highly expressed in germ cells treated with the highest BPA dose. PRDX4 controls cytokine-induced peroxide levels and thereby mediates signal transduction in mammalian cells. It is known that spermatogenic cells lacking PRDX4 are more susceptible to cell death via oxidative damage than their wild-type counterparts 49 , and that PRDX4 exerts its protective function against oxidative damage by scavenging reactive oxygen species (ROS) in the extracellular space 50 . On the other hand, studies have proven that SOD2 is a biomarker of several types of cancer 51,52 and that BPA has the ability to modify SOD2 levels in different cell types 53,54 . Therefore, up-regulation of these proteins in 100 µM BPA-treated germ cells could be associated with abnormal cell growth, cellular damage, and cancer. Unlike the other proteins, PHB showed maximal expression at 0.01 µM BPA treatment and relatively higher expressions at other BPA doses than control. Heat shock proteins (HSPs) are generally responsible for structural formation of several cellular proteins and the prevention of protein misfolding. HSPE1 level increased at 100 µM BPA treatment, whereas HSPD1 and HSPA8 level decreased following the same treatment, indicating the possibility of the formation of abnormal decorating spindles and centrosomes during cell division, as reported in several studies, especially on 70 kDa HSP 55 . Furthermore, expression of the enzyme GLT1D1, which is responsible for the formation of different types of polysaccharides, was remarkably down-regulated by 100 µM BPA treatment. Additionally, as shown in Fig. 7B, almost all of the identified proteins are interrelated via regulatory networks, suggesting simultaneous up-or down-regulation on the expression of these proteins in germ cells due to BPA administration. As the proteins are interlinked, BPA induced alteration of the expression of oxidative stress (SOD2, PRDX4) and apoptosis (PPIA) related proteins can regulate HSPs and may affect fertility associated cell processes by triggering fertility related proteins such as PHB. Therefore, based on these findings, we conclude that BPA can cause effects on differentially expressed cellular proteins related to cell survival, growth and apoptotic pathways, including oxidative stress initiation, cellular immune response, and mitochondrial survivability (Fig. 7A). In other word, these findings indicate the possibility that BPA exposure may lead to the initiations of some latent diseases (shown in Fig. 7A) such as neoplasm and necrosis, together with delayed sex maturation (due to reduced expression of HSPA8) and infertility.

Conclusion
To the best of our knowledge, this is the first comprehensive study on the effect of BPA exposure on the-proliferation, survivability, stemness properties, and cellular proteome of testicular germ cells, based on in vitro culture method. Low concentrations of BPA (0.01, 0.1, and 1 µM) did not produce significant or partial toxic effects in cultured germ cells, whereas 10 and 100 µM BPA doses resulted in apoptosis, altered stem cells functional abilities possibly triggering genetic abnormalities. This study also provides information about germ cell survivability and stemness properties under continuous BPA exposure. Finally, the up-regulation of proteins related to cancer, oxidative stress, and apoptosis in BPA-treated germ cell cultures suggests the possibility that BPA might be associated with various health risks and infertility. From this perspective, this study could be used as a starting point of examining the effects of other environmentally abundant EDCs on human health issues including reproduction. Magnetic activated cell sorting (MACS) and germ cell culture with BPA. MACS, which is generally used for isolation and culture of testis cells from mouse pup (6-8 days old) for enrichment of SSCs, was performed as previously described 56 . Germ cell culture was then set to enhance the number of cells stable enough for further experiment progressing. The detailed procedure is described in Supplementary Methods section.

Reagents. Reagent related information is described in
BPA (239658, Sigma, St. Louis, MO, USA) was dissolved in 100% dimethyl sulfoxide (DMSO; D2650, Sigma, St. Louis, MO, USA), a 10-fold serial dilution series, ranging in concentration from 0.01 to 100 mM, was prepared in DMSO. 0.1% (v/v) from each of the BPA concentration was mixed with mSFM so that the final concentration reached to µM level. Thus, germ cell culture of CD-1 and C57 GFP mice were prepared with BPA-mixed medium and media were changed every 24 h. Control groups received medium containing 0.1% (v/v) DMSO. After 1 week culture with different BPA concentrations, harvested germ cells were counted by hemocytometer. Germ cell proliferation was measured by counting total number of harvested cells from culture and viability was assessed by calculating percentage of viable cells with trypan blue exclusion test 57 . Four replicates of each germ cell culture were used for data analysis.

Analysis of germ cell apoptosis.
To observe the effect of BPA on germ cell apoptosis, the percentage of apoptotic cells was examined immediately after harvesting cultured cells. The detailed procedure is described in Supplementary Methods section.
Immunocytochemistry. SSC self-renewal related marker protein (PLZF), undifferentiated spermatogonia surface receptors (GFR α1 and RET), and differentiated spermatogonia marker protein (C-KIT) were selected to evaluate their expression in BPA-exposed cultured germ cells. Immunocytochemistry was performed following the procedure described preciously 58 . Briefly, BPA-treated cells were fixed with 4% paraformaldehyde (PFA) for 20 min and permeabilized with 0.1% Triton X-100 in PBS for 10 min at room temperature (RT). DPBS containing 5% (w/v) bovine serum albumin (BSA) was applied for 30 min at RT to avoid nonspecific antibody binding. Then germ cell fixed slides were separated into four groups, for four different marker proteins. Cells were incubated with monoclonal anti-mouse promyelocytic leukemia zinc finger (PLZF, 1:100; OP128, Calbiochem, USA), rabbit anti-human glial-derived neurotrophic factor family receptor alpha 1 (GFRα1, 1:100; sc-10716, Santacruz, Texas, USA), goat anti-mouse receptor tyrosine kinase (RET, 1:100; AF482, R&D, Minneapolis, MN, USA) and goat anti-mouse C-KIT (1:100; sc-1494, Santacruz, Texas, USA) primary antibodies at 4 °C overnight. Cells were washed in DPBS for 3 times, and incubated with Alexa fluor 568-conjugated Goat anti-mouse IgG Germ cell transplantation for evaluating SSC activity after BPA exposure. The transplantation was carried out as described by Brinster et al. 45,56 , to investigate the stemness state and changed characteristics of cultured germ cells with the presence of endocrine disruptor BPA. A similar mouse strain (C57BL/6) was used as recipient because the cultured germ cells were also derived from C57 GFP strain. The detailed procedure is described in Supplementary Methods section. Germ cells were injected in both testes and overall 8-10 recipient mice were used for control and each BPA group.
The Spermatocyte preparation and immunostaining. BPA-treated germ cell transplanted recipient mice were used to investigate the presence of meiotic abnormalities during spermatogenesis. Testes were collected from the recipient mice. Preparations of spermatocytes were processed according to the protocol of Peters et al. 59 with a little modification where 1% PFA was applied as a thin layer over a clean slide rather than by dipping it into PFA. A drop of nuclear suspension was spread over the slides and then kept in a humid chamber for overnight, dried and washed with a solution of 0.4% Photo-Flo 200 (Kodak Professional, NY, USA). Finally slides were kept to air-dry.
Long-term culture with BPA and recovery of germ cells from BPA exposure. To test the effects of long-term BPA exposure, germ cells were treated with BPA concentrations ranging from 1 to 100 µM for 3 weeks. Subculture was performed every week on fresh STO feeder layer. Cells were seeded at a density of 0.2 × 10 6 cells/ well each week. If any BPA treated group showed cell number lower than 0.2 × 10 6 , all of the harvested cells from each wells were used for subculture.
In case of recovery test, germ cells were treated with 100 µM BPA for 1 week and then subcultured for the following 3 weeks without BPA treatment. Both the long-term cultured germ cells and recovered germ cells from BPA exposure were transplanted into recipient mice and 8-10 mice were used per group.

Two dimensional gel electrophoresis (2-DE) analysis of BPA-induced proteome alterations in germ cells and pathway studio. A 2-DE-based comparative proteomic analysis was performed to
investigate the changes of protein expressions on BPA effect and other procedures were followed according to described 20,60 . The detailed procedure is described in Supplementary Methods section.
Statistics. Data were analyzed by one-way analysis of variance (ANOVA) using SPSS (version 12.0, IBM, USA) and Prism software (version 5.03; GraphPad, La Jolla, CA, USA). Analysis of variance was used to test for differences among treatment groups and each assay was run at least in triplicate. For identification of differences in the recovery test, student's two-tailed t test was used. Significant differences between means were determined using Duncan and Tukey's Honestly Significant Difference test where P < 0.05 was considered as significant.
Data availability. The datasets generated and/or analyzed in the current study are available from the corresponding author on reasonable request.