High-Content analysis of testicular toxicity of BPA and its selected analogs in mouse spermatogonial, Sertoli cells, and Leydig cells revealed BPAF induced unique multinucleation phenotype associated with the increased DNA synthesis

Abstract

Bisphenol A is an endocrine disruptor that has been shown to have testicular toxicity in animal models. Its structural analog, including bisphenol S (BPS), bisphenol AF (BPAF), and tetrabromobisphenol A (TBBPA) have been introduced to the market as BPA alternatives. Previously, we developed high-content analysis (HCA) assays and applied machine learning to compare the testicular toxicity of BPA and its analogs in spermatogonial cells and testicular cell co-culture models. There are diverse cell populations in the testis to support spermatogenesis, their cell type-specific toxicities are still not clear. The purpose of this study is to examine the selective toxicity of BPA, BPS), BPAF, and TBBPA on these testicular cells, including Sertoli cell, Leydig cells, and spermatogonia cell. We developed a high-content image-based single-cell analysis and measured a broad spectrum of adverse endpoints related to the development of reproductive toxicology, including cell number, nuclear morphology, DNA synthesis, cell cycle progression, early DNA damage response, cytoskeleton structure, DNA methylation status, and autophagy. We introduced an HCA index and spectrum to reveal multiple HCA parameters and observed distinct toxicity profiling of BPA and its analogs among three testicular types. The HCA spectrum reveals the dynamic, chemical-specific, dose-dependent changes of each HCA parameter. Each chemical displayed a unique dose-dependent profile within each type of cell. All three types of cells showed the highest response to BPAF at 10 μM across all endpoints measured. BPAF targeted spermatogonial cell (C18) more significantly at 5 μM. BPS more likely targeted Sertoli cell (TM4) and Leydig cell (TM3) and less at spermatogonia cells. TBBPA targeted spermatogonia, Sertoli cells, and less at TM3 cells. BPA is mainly targeted at TM4, followed by TM3 cells, and less at spermatogonial cells. Most importantly, we observed BPAFR induced a dose-dependent increase in spermatogonia cell, not in Sertoli cells and Leydig cells. germ In summary, our current HCA assays revealed the cell-type-specific toxicities of BPA and its analogs in different testicular cells. Multinucleation induced by BPAF along with increased DNA damage and synthesis at low doses, could possibly have a profound long-term effect on reproductive systems.

Introduction

The incidence of male reproductive disorders has increased in recent decades. Exposure to environmental toxicants and pharmaceuticals has been proposed to impact male reproductive health (Nordkap, Joensen et al. 2012, Knez 2013, Jeng 2014, Jenardhanana, Panneerselvama et al. 2016, Leung, Phuong et al. 2016, Skakkebaek, Rajpert-De Meyts et al. 2016). The decline of human semen quality, such as sperm count and motility, has been reported to be linked to exposure to environmental contaminants (Nordkap, Joensen et al. 2012, Knez 2013, Jeng 2014, Skakkebaek, Rajpert-De Meyts et al. 2016). Bisphenol A (BPA) is a high-production volume chemical used extensively in various consumer products (Rochester 2013, Siracusa, Yin et al. 2018). BPA in urine was detectable in over 90% of the United States population (Calafat, Ye et al. 2008, Lakind and Naiman 2011, CDC 2022). As an endocrine-disrupting chemical, BPA has shown various adverse health effects, such as reproduction and neurotoxicity (Rubin 2011, Peretz, Vrooman et al. 2014, Santoro, Chianese et al. 2019). BPA exposure was found to affect oocyte quality and maturation adversely, decrease sperm production and quality, damage testicular cells, perturb hormone levels, and disrupt ovary function and uterine morphology in animal models (Delclos, Camacho et al. 2014, Qi, Fu et al. 2014, Moore-Ambriz, Acuna-Hernandez et al. 2015, Vigezzi, Bosquiazzo et al. 2015, Barbonetti, Castellini et al. 2016, Berger, Ziv-Gal et al. 2016, Ferris, Mahboubi et al. 2016, Wang, Han et al. 2016). Human exposure to BPA leaches out of consumer products mainly through ingestion, inhalation, and dermal absorption (Kang, Kondo et al. 2006). Although FDA determined the use of BPA at current level is safe as supported by the CLARITY-BPA studieas (Heindel, Newbold et al. 2015, Camacho, Lewis et al. 2019) (FDA. 2012, FDA. 2013, EU. 2016), widespread exposure and potential hazards of BPA in humans drove manufacturers to abandon the use of BPA and introduce analogous chemicals as crosslinking reagents and flame retardants in the plastics industry to produce “BPA-free” products. There is a lack of production data for these analogs, and the usage of these chemicals is rising globally since these are not all strictly used as replacements for BPA, but have a variety of commercial uses. BPA analogs are found in the environment, foods, consumer products, and human urine samples (Geens, Roosens et al. 2009, Liao, Liu et al. 2012, Liao, Liu et al. 2012, Shi, Jiao et al. 2013, Ye, Wong et al. 2015). With high degrees of structural similarities to BPA, these analogs potentially have a similar endocrine-disrupting capacity and exert adverse effects on the reproductive system. Emerging evidence suggests that BPA analogs act with various physiological receptors (Kitamura, Suzuki et al. 2005, Stossi, Bolt et al. 2014). Recently, the National Toxicology Program reported on the biological activity of BPA and several of its analogs, Bisphenol S (BPS), Bisphenol F (BPF), and Bisphenol AF (BPAF), and suggested BPA analogs causing mammalian reproductive toxicity (Pelch, Wignall et al. 2017, Pelch, Wignall et al. 2019). So far, 16 bisphenol analogs have been detected in house dust, drinking water, serum, and human milk (Deceuninck, Bichon et al. 2015, Owczarek, Kubica et al. 2018, Zhang, Zhang et al. 2019, Yang, Shi et al. 2022).

Compared to BPA, little is known about the reproductive toxicity of these analogs. In our previous study, we applied the testicular cell co-culture model and employed a high-content image (HCA) based single-cell analysis to compare the testicular toxicities of BPA and its analogs (Yin, Siracusa et al. 2020). We found dose and time-dependent changes in a broad spectrum of adverse endpoints, including nuclear morphology, DNA synthesis, DNA damage, and cytoskeletal structure in the testicular co-culture model. Furthermore, the co-cultured testicular model was more sensitive than C18 spermatogonial cells in response to BPA and its analogs (Liang, Yin et al. 2017). Sertoli, Leydig, and germ cells have been reported to respond differently compared to BPA, and it is unclear if Sertoli and Leydig cells are more sensitive than germ cells to these BPA analogs. The purpose of this study is to examine the selective toxicity of BPA, bisphenol S (BPS), bisphenol AF (BPAF), and tetrabromobisphenol A (TBBPA) on these testicular cells, including Sertoli cell, Leydig cells, and spermatogonia cell. This study employed a high-content image-based single-cell analysis and measured a broad spectrum of adverse endpoints related to the development of reproductive toxicology, including cell number, nuclear morphology, DNA synthesis, cell cycle progression, early DNA damage response, cytoskeleton structure, DNA methylation status, and autophagy. We introduced an HCA spectrum to reveal multiple HCA parameters and observed distinct toxicity profiling of BPA and its analogs among three testicular types. These findings leveraged the specific-cell target and toxicity mechanism action for each BPA and its analogs, which provided a novel methodology to efficiently identify cell targets and valuable mechanical information in interpreting in vivo reproductive toxicity of BPA and its selected analogs.

Material and Methods:

Chemicals

Dulbecco’s Modified Eagle Medium (DMEM), Modified Eagle’s Medium/Nutrient Mixture F-12 (DMEM/F12), fetal bovine serum (FBS), horse serum, and penicillin-streptomycin were purchased from Thermo Scientific (Waltham, MA). 4,4’-(propane-2,2-diyl) diphenol (BPA, ≥99%), 4,4’-sulfonyldiphenol (BPS, 98%), and 2,2’,6,6’-Tetrabromo-4,4’-isopropylidenediphenol (TBBPA, 97%) were obtained from Sigma-Aldrich (St Louis, MO). 4-[1,1,1,3,3,3-Hexafluoro-2-(4-hydroxyphenyl)propan-2-yl]phenol (BPAF, 98%) was obtained from Alfa Aesar (Ward Hill, MA). Nu-Serum was purchased from BD BioScience (Redford, MA). 5-Bromo-2’-deoxyuridine (BrdU, 99%) was purchased from Thermo Scientific (Waltham, MA). 4% Paraformaldehyde was purchased from Boston Bioproducts (Ashland, MA).

Cell Culture and treatment:

C18-4 mouse spermatogonial cell line was established from germ cells isolated from the testes of 6-day-old Balb/c mouse as reported previously (Hofmann et al., 2005). This cell line exhibits morphological features of type A spermatogonia and expresses testicular germ cell-specific genes such as GFRA1, Dazl, and Ret (Hofmann et al., 2005). Mouse TM3 Leydig and TM4 Sertoli cell lines were purchased from ATCC (Manassas, VA). All the cell culture procedures were conducted as reported previously (Liang, Yin et al. 2017). Briefly, C18-4 spermatogonial cells were maintained in DMEM composed of 5% FBS, 100 U/ml streptomycins/penicillin in a 33°C, 5% CO2 humidified environment in a sub-confluent condition with passaging every 3-4 days. Leydig and Sertoli cells were cultured in DME/F12 composed of 2.5% FBS, 5.0 % horse serum, 100 U/ml streptomycin/penicillin at 37°C, 5% CO2 in a sub-confluent condition with passaging every 2-3 days. When cells reached 70-80% confluence, the spermatogonial cells were inoculated with 1.2 ×10⁴ cells per well, and Leydig and Sertoli cells were inoculated with 1.5 ×10⁴ cells per well into a 96-well plate. Cells were cultured overnight reaching 70-80% confluence, then treated with doses of BPA, BPS (0, 5, 50 and 100 μM), BPAF and TBBPA (0, 1, 5 and 10 μM) at for 24 and 48 h. The experiment was completed with technical triplicates for each concentration, and experiments were repeated 3 or 4 times.

Fluorescence staining and image acquisition

For BrdU incorporation, cells were incubated with BrdU (40 μM) for three hours before fixation with 4% paraformaldehyde for 30 min at room temperature, followed by three times washing with phosphate-buffered saline (PBS) (Liang, Yin et al. 2017). After fixation, cells were permeabilized by 0.1% Triton X-100 in PBS, blocked with PBS/3% BSA, and incubated with a mouse anti-BrdU antibody (Thermo Scientific, MA) in PBS/BSA/0.5% Tween 20 overnight at 4 °C. After washing twice with PBS/BSA, the cells were incubated with goat anti-mouse DyLight 488 and Hoechst 33342 (Thermo Scientific, MA) in PBS/BSA solution for 90 mins at room temperature.

DNA damage responses was detected by anti-phospho-histone-H2AX (Ser139) (γ-H2AX) and cytoskeleton analysis was performed by F-actin staining with AlexaFluor 488 Phalloidin (Cell Signaling 8878) following the protocol reported previously (Liang, Yin et al. 2016). For epigenetic or autophagy evaluation, antibodies against Methyl CpG Binding Protein 1 (MBD1) (Developmental Studies Hybridoma Bank, http://dshb.biology.uiowa.edu, Iowa) and LC-3B (ThermoFisher Scientific, MA) were used to incubate with the fixed cells and followed by the corresponding secondary antibodies for imaging acquisition.

Multi-channel images were automatically acquired using an Arrayscan^™ VTI HCS reader (Thermo Scientific, MA). Forty-nine fields per well were acquired at 20x and 40X magnification using Hamamatsu ROCA-ER digital camera in combination with 0.63x coupler and Carl Zeiss microscope optics in auto-focus mode. Channel one (Ch1) applied the BGRFR 386_23 for Hoechst 33342 auto-focus, object identification, and segmentation. For MBD1 and LC3B staining, channel two (Ch2) was assigned to BGRFR 485_20 for MBD1 detection, and channel three (Ch3) was assigned to BGRFR 650_13 for LC3B detection. Channel one was set to BGRFR 386_23 for Hoechst 33342 for all images and used for auto-focus, object identification and segmentation.” For F-actin and γ-H2AX staining, Ch2 was assigned to BGRFR 485_20 for F-actin, and Ch3 was set to BGRFR 650_13 for γ-H2AX visualization.

High-content images analysis

Multi-channel images were analyzed following the procedure described in the previous study (Yin, Siracusa et al. 2020) using HCS Studio^™ 2.0 TargetActivation BioApplication. Multi-parameters of nuclei were characterized in HCA, including the nuclei number, nuclear area, shape, and total DNA intensity. The total intensity of BrdU, γ-H2AX, and F-actin, MBD1, and LC3B of the individual cell ere also quantified from HCA. Nuclear shape measurement included P2A (nuclear perimeter²/4 π * nuclear area) to evaluate nuclear smoothness and LWR (nuclear length/width) to measure nuclear roundness. For a fairly round and smooth object, the values for P2A and LWR are close to 1.0. The total intensity was defined as the total pixel intensities within a cell in the respective channel. With forty-nine fields of each well, at least 1000 cells were analyzed per well, and single-cell-based data were exported for further analysis. The experiments were performed with at least four biological replicates and repeated twice.

HCA-based cell cycle analysis was conducted as previously described (Liang, et al., 2016; Roukos et al., 2015). Briefly, the histogram of total DNA intensity of each replicate in various experimental conditions was constructed in a custom script written in Python 2.7.12 (Python Software Foundation, OR; this script is freely available from the authors upon request). The input data were gated using the nuclear area and total DNA intensity to exclude cell debris and clumps. Cell populations in sub-G1 (apoptotic cells), G0/1, S, and G2/M in the controls and treatments were quantified by the appropriate selection of gating threshold.

Statistical Analysis

Data obtained from the HCS StudioTM 2.0 TargetActivation BioApplication were exported and further analyzed using the JMP statistical analysis package (SAS Institute, NC). A nucleus with an area larger than 1000 μm² in a 20X image or 2000 μM² in a 40X image was excluded as cell clumps. For each plate, the vehicle control showed consistent measurements for all endpoints tested. For intra-plate normalization, data were normalized to the overall scaling factors, which was the mean of medians of vehicle controls in each plate. The single cell-based data were averaged for the well-based condition. BrdU-positive cells were set by the total intensity of BrdU in the Control over 25,000 pixels. γ-H2AX positive cells were set by the total intensity of γ-H2AX in the Control over 140,000 pixels. Data were presented as the mean ± standard deviation (SD). Statistical significance was determined using one-way ANOVA followed by Tukey-Kramer all-pairs comparison. A P value less than 0.05 denoted a significant difference compared to the vehicle control (*). To compare the multiplexed HCA parameters among BPA and its analogs between the three cell types, C18-4 spermatogonial, TM3 Leydig, and TM4 Sertoli cells, we introduced the HCA spectrum to reveal the dynamic, chemical-specific, dose-dependent changes of each HCA assay. We further developed an HCA index to integrate these parameters from diverse HCA assays to show dose-dependent dynamic changes in these cellular markers. HCA spectrum reveals the deviation (fold changer from the Control) of each assay, including cell number, nuclear area, the marker for shape measurements (P2A and LWR), DNA synthesis (BrdU positive cells), cell cycle phases (cell population in subG1 phase, G0/G1 phase, S phase, and G2/M phase), DNA damage response (γ-H2AX positive cells), F-actin total intensity, MBD1 total intensity and LC3B total intensity. For each feature parameter, deviation from Control as reflected by the fold change over the mean of the Control was calculated and then log2-transformed. HCA index is the sum of the absolute value of each log2 transformed fold change.

Results:

Cell viability and nuclear morphological alterations

Image-based HCA was used to measure multiple parameters, such as cell number and nuclear morphology changes after 24 and 48h of BPA analogs treatment (O’Brien, Irwin et al. 2006, Yin, Siracusa et al. 2020). We treated three types of testicular cells with concentrations of 0, 5, 50 and 100 μM for BPA and BPS, and 0, 1, 5, and 10 μM for BPAF and TBBBPA based on a previous study (Liang, Yin et al. 2017, Yin, Siracusa et al. 2020). Statistical analysis with doses, cell types, and chemicals was conducted (ANOVA) at each time-point. Dose-dependent changes in cell number were observed in all three types of testicular cells at both 24 and 48 h (Figure 1). BPA induced a dose-dependent decrease in cell number in all testicular cell types. The reduction of cell number is most evident in C18-4 spermatogonial cells at 24h. However, TM4 Sertoli cells lost more cells than C18-4 or TM3 at 48h treatment of BPA. BPS also induced a dose-dependent decrease in cell number at 24h and 48h, and there was no difference among the three types of testicular cells at 24h. Significant reductions in TM3 and TM4 cells were observed compared to spermatogonia cells at 48h. BPAF induced the most considerable cell number decrease among all tested compounds at both time points. The spermatogonial cells decreased more at 5 μmM compared to other types of cells at both 24 and 48h. TBBPA treatment showed distinct responses among the three types of cells, and a dose-dependent decrease of cell number in both spermatogonia cell and TM4 cells was observed but not in TM3 cells. TBBPA induced a more significant reduction in cell number in spermatogonia cells than in TM4 cells. Figure 2A shows the representative nuclear morphology following 24 or 48h treatments. Significant induction of multinucleated cells was only observed in spermatogonial cells treated with BPAF at 5 μM or more. Quantification of nuclear morphology demonstrated exposure to BPA and its analogs induced significant changes in the nuclear area and nuclear shape measurement parameters, P2A, and LWR (Figure 2B–D). As reflected in the nuclear area, the nuclear size decreased at 48h as compared to 24h for the three types of cells (Figure 2B). BPA treatment at 100 μM reduced the size of C18-4 cells at both timepoints, reduced the size of TM3 at 100 μM and TM4 at 50 and 100 μM at 24 h, but increased the size of TM3 and TM4 at both 50 and 100 μM. BPS treatment only increased the nuclear size of C18-4 at 100 μM, TM3 at doses above 5 μM, and TM4 at 50 and 100 μM. BPAF treatment at 100μM increased the nuclear size of C18-4 cells at 48h, increased the size of TM3 at 100 μM and TM4 at 50 and 100 μM at 24 h, and 5, 50 and 100 mM at 48h. TBBPA treatment at 100 μM reduced the size of C18-4 cells at both time points, reduced the size of TM3 at 100 μM and TM4 at 50 and 100 μM at 24 h, increased the size of TM3 and TM4 at both 50, and 100 μM at 48h and increased the size of TM3 and TM4 at both 50, and 100 μM. Nuclear shape measurements included nuclear smoothness P2A and nuclear roundness LWR were quantified (Figure 2C–D). BPA treatment at 100 μM increased the P2A and LWR in C18-4 cells at 48h, reduced P2A and LWR in TM3 cells at 50, 100 μM at 24h and at 5, 50, 100 μM at 48h, and TM4 cell only at 24 h. BPS treatment reduced P2A and LWR in TM3 and TM4 cells at 50, 100 μM at both time points. BPAF treatment significantly increased P2A and LWR in C18-4 cells at 5 μM at 24h and 5 and 10 μM at 48 h, in TM3 at 100 μM at both timepoints, TM4 at 50 and 100 μM at 48h. TBBPA treatment significantly decreased LWR in C18-4 cell at 5 μM at 48 h and decreased P2A and LWR in TM3 at concentration 5 μM or over at both 24h and 48 h, and P2A in TM4 at 100 μM at both time points, TM4 at 50 and 100 μM at 48h. No changes in P2A in BPA, BPS, and TBBPA treatment in C18-4 cells at both time points were observed. Overall, BPA and its analogs have different in vitro adverse effects among these testicular cells and induce distinct changes in nuclear morphology. BPAF was the most cytotoxic compound in all three testicular cells and showed unique multinucleation in spermatogonial cells.

Figure 1. High-content image-based analysis (HCA) of single cell numbers changes in C18-4 spermatogonial cells, TM3 Leydig cells, and TM4 Sertoli cells.

Cells were treated with various concentrations of BPA and BPS (5, 50, and 100 μM), and BPAF and TBBPA (1, 5, and 10 μM) for 24 and 48 h. Data were presented as mean ± SD, n = 10. Five bilogical replicates in two independent experiments were included. One-way ANOVA conducted statistical analysis followed by Tukey-Kramer multiple comparisons (*P < .05).

Figure 2. Effect of BPA and its analogs on nuclear morphology of C18-4 spermatogonial, TM3 Leydig and TM4 Sertoli cells.

A. Representative images of nuclear morphology in spermatogonial, Leydig, and Sertoli cells treated with BPA, BPS, BPAF, and TBBPA for 24h. The nuclei were stained with Hoechst 33342, and images were automatically obtained with a 40× objective, 49 fields per well. Yellow arrows indicate the multinucleated cells. Scale bar = 50 μm. B–D, Quantification of absolute nuclear area (μm2) (B), nuclear shape parameters, including P2A for smoothness (C) and LWR for nuclear roundness (D) of the 3 testicular cell types treated with BPA and BPS (5, 50, and 100 μM) and BPAF and TBBPA (1, 5, and 10 μM) for 24 and 48 h. Data were presented as mean ± SD, n = 9. Three replicates in 3 separate experiments were included. Statistical analysis was conducted by 1-way ANOVA followed by Tukey-Kramer multiple comparisons (*P < .05).

BPA and its analogs altered cell cycle progression

The formation of the multinucleated cell was potentially due to asynchronous nuclear division and dysregulation of the cell cycle control network (Gladfelter, Hungerbuehler et al. 2006). We previously developed multiparametric HCA assays to measure cell cycle progression (Liang, et al., 2017). DNA total intensity histogram was constructed, and the proportion of cells in SubG1, G0/1, S, and G2/M phases of the cell cycle were quantified (Figure 3). Among the three types of testicular cells, alterations in the cell cycle were observed with all four treatments. Among the treated compounds, BPAF induced dose-dependent alterations in all three types of cells at doses of 5 and 10 μM at both time points. In C18-4 cells, both BPA at 100 μM and TBBPA at 10 μM, significantly increased cells in G2/M, and reduced cells in G0/1 and S phases at both 24 and 48 h time points. In TM3 Leydig cells (Figure 3), significant increases in cells in the G0/1 phase and decreases in the G2/M phase were observed in the BPA and BPS treatment at 50 and 100 μM at both 24 and 48 h. In comparison, reductions of cells in the G0/1 phase and increases of cells in the G2/M phase were observed in BPAF treatment at a dose of 10 μM for 24 and 48 h. A significant decrease of cells in the S phase was observed in BPA treatment at 100 μM for 24 h, 50 and 100 μM for 48 h, BPS treatment at 50 and 100 μM for 24 and 48 h, and TBBPA treatment at 10 μM for 24 h. In contrast, accumulation of cells in the S phase was observed in BPAF treatment at a dose of 10 μM for 48 h, accompanied by a significant increase of cells in the sub-G1 phase. In TM4 Sertoli cells (Figure 3), significant decreases in cell proportion in G0/1 phase and increases in the proportion of cells in the G2/M phase were observed in BPA and BPS treatment at 100 μM for 24 h, 50, and 100 μM for 48 h BPAF treatment at 10 μM for 24 h, 5 and 10 μM for 48 h, and TBBPA treatment at 10 μM for 24 and 48 h. Significant decreases in the proportion of cells in S phase were observed in BPA and BPS treatment at 100 μM for 24 h, 50 and 100 μM for 48 h, and TBBPA treatment at 10 μM for 48 h. In contrast, the accumulation of cells in S phase was observed in BPAF treatment at a dose of 10 μM. Increases of cells in the sub-G1 phase were observed in BPAF treatment at 5 μM for 24 and 48 h, and TBBPA treated at 5 and 10 μM for 24 h. Therefore, BPA and its analogs caused differential cell cycle arrest among different types of testicular cells.

Figure 3. Cell cycle analysis of C18-4 spermatogonial cells, TM3 Leydig cells, and TM4 Sertoli cells.

Cells were treated with various concentrations of BPA and BPS (5, 50, and 100 μM), and BPAF and TBBPA (1, 5, and 10 μM) for 24 and 48 h. The percentage of cells in each cell cycle stage was determined by creating gating parameters on histograms of total DNA intensity per cell (by Hoechst 33342 staining) per well at sub-G1, G0/G1, S, and G2M phases using Python script.

Data were presented as mean ± SD, n = 10. Five replicates in two independent experiments were included. One-way ANOVA conducted statistical analysis followed by Tukey-Kramer multiple comparisons (*P < .05).

BPA and its analogs induced alterations in DNA synthesis

BrdU labeling is a marker of DNA synthesis and not S-phase of the cell cycle or cell division. Representative morphology (Figure 4A) at 24h and HCA quantification (Figure 4B) of BrdU incorporation at both time points revealed differential effects of BPA and its analog in these cells. The percentage of BrdU positive cells was 40%, 32%, 33% in C18-4, TM3 and TM4 cells individually at 24h and 30%, 20%, 15% in C18-4, TM3 and TM4 cells, respectively at 48h. BPA dose-dependently reduced the number of BrdU positive cells in C18-4, TM3 and TM4 at 24h, significantly at 50 and 100 μM. A significant decrease in the number of BrdU positive cells at 50 and 100 μM in C18-4 cells and TM3 at 100 μM at 48h was observed. BPS dose-dependently reduced the number of BrdU positive cells in C18-4, TM3 and TM4 at 24h, significantly at 50 and 100 μM. A significant decrease in the percentage of BrdU positive cells at 50 and 100 μM in C18-4 cells, TM3 at 100 μM, and TM4 at 100 μM at 48h were observed. BPS dose-dependently reduced DNA synthesis in all three types of testicular cells at 24h, significantly at 50 and 100 μM. At 48h, significant reductions were observed in C18-4 and TM3 at 50 and 100 μM and TM4 at 100 μM for 48 h. TBBPA reduced the percentage of BrdU positive cells in spermatogonial C18-4 cells at 5 and 10 μM at 24h and 10 μM at 48h. Interestingly, a decrease in the number of BrdU positive cells was only observed in TM4 Sertoli cells at 10 μM at 24h. In contrast to BPA, BPS and TBBPA, BPAF induced a different change pattern of in BrdU synthesis. In C18-4 spermatogonia cells, a dose-dependent decrease of BrdU positive cells was observed, significantly at 10 μM at 24h. However, a dose-dependent increase of BrdU positive cells was observed at 48, significantly at both 5 and 10 μM. Decreases of BrdU positive cells in TM3 and TM4 at 10 μM were observed at 24h; however, a significant increase of BrdU positive cells at 10 μM TM3 and TM4 cells at 48 were found.

Figure 4. Effect of BPA and its analogs on DNA synthesis of C18-4 spermatogonial, TM3 Leydig and TM4 Sertoli cells.

A shows the representative images of BrdU incorporation in spermatogonial (C18-4), Leydig (TM3), and Sertoli (TM4) cells treated with BPA, BPS, BPAF, and TBBPA at various concentrations for 24h. The nuclei were stained with Hoechst 33342, BrdU positive labeling was pseudo-colored in green. Images were automatically obtained by ArrayScan HCA Reader with a 40× objective, 49 fields per well. Scale bar = 50 μm. B Quantification of BrdU incorporation (% BrdU positive cells vs Control) in the 3 testicular cell types treated with BPA and BPS (5, 50, and 100 μM) and BPAF and TBBPA (1, 5, and 10 μM) for 24 and 48 h. Single cell BrdU labeling data were quantified by HCS Studio^™ 2.0 TargetActivation BioApplication. Data were presented as mean ± SD, n = 9. Three replicates in 3 separate experiments were included. Statistical analysis was conducted by 1-way ANOVA followed by Tukey-Kramer multiple comparisons (*P < .05).

BPA and its analogs perturbed cytoskeleton and induced early DNA damage responses

Dose-dependent alterations in the cytoskeleton were observed in the treatment with BPA and its analogs (Figure 5). BPA induced dose-dependently increase of F-actin total intensity in all three types of cells, with statistical significance in C18-4 spermatogonial cells at 100 μM at 24h and 50 and 100 μM at 48h, TM3 and TM4 at 50 and 100 μM at both timepoints (Figure 5). BPS significantly increased F-actin total intensity in spermatogonial C18-4 cells at 100 μM at both timepoints, in TM3 Leydig cells at 50 and 100 μM at 24h and at 5 μM or over at 48h, in TM4 Sertoli cells at 50 and 100 μM at both timepoints. BPAF significantly increased F-actin total intensity observed in spermatogonial cells at 10 μM for both timepoints, in TM3 Leydig cells at 5 and 10 μM at 24 h and at 1 or over at 48 h, in TM4 Sertoli cells at 10 μM at 24h and at 5 and 10 μM at 48 h. TBBPA treatment caused F-actin total intensity induction observed in C18-4 spermatogonial and TM4 Sertoli cells at 10 μM for both timepoints, and in TM3 Leydig cells at 1, 5, and 10 μM at 48 h.

Figure 5. BPA and its analogs differentially perturbed cytoskeleton (F-actin) and induced early DNA damage responses (γH2AX) in spermatogonial (C18-4), Leydig (TM3), and Sertoli cells (TM4).

Cells treated with vehicle (0.01% DMSO) were used as negative controls. The nuclei were stained with Hoechst 33342 (blue), F-actin with Phalloidin staining (green), and γ-H2AX with a combination of primary phosphorylated γ–H2AX and secondary Dylight 650 conjugated antibody (red). A shows the representative images of cells treated with BPA and BPS (100 μM), BPAF and TBBPA (10 μM) for 24 h. Arrows indicate dot-like structures. Scale bar = 50 μm. B-C shows the single-cell quantification of the log-transformed total intensity of F-actin (B) and positive γ-H2AX cells (C) of the three testicular cell types treated with various BPA, BPS, BPAF and TBBPA concentrations for 24 and 48h. Data were presented as mean ± SD, n = 6. Three replicates in two separate experiments were included. Statistical analysis was conducted by 1-way ANOVA followed by Tukey-Kramer multiple comparisons (*P < 0.05).

BPA showed significant increases in γ-H2AX positive cells at 100 μM for 24 and 48 h in C18-4 spermatogonial cells, 24 h in TM4 Sertoli cells, 50 and 100 μM for 48 h in TM3 Leydig and TM4 Sertoli cells (Figure 5B). BPS showed accumulation of γ-H2AX positive cells at 100 μM for 24 and 48 h in C18-4 spermatogonial cells, 50 and 100 μM for 24 and 48 h in TM4 Sertoli cells, and 48 h in TM3 Leydig cells. BPAF significantly induced γ-H2AX positive cells at 10 μM for 24 h in TM3 Leydig and TM4 Sertoli cells, 5 and 10 μM for 24 and 48 h in C18-4, and 48 h at TM3 Leydig and TM4 Sertoli cells. TBBPA significantly increased the number of γ-H2AX positive cells at 10 μM for 24 and 48 h in C18-4 spermatogonial and TM4 Sertoli cells.

BPA and its analogs differentially induced DNA methylation and activated autophagy

We used Methyl-CpG binding domain 1 (MBD1), a transcriptional repressor that binds to single methylated CpGs as a methylation marker, to quantify the methylation status of cells treated with BPA and its analogs. Figure 6A shows the representative images for MBD1 and LC3B co-staining in three cell types with various treatment conditions after 24h exposure. As shown in Figure 6B, BPA significantly increased MBD1 total intensity at 50 and 100 μM for 24 h in C18-4 spermatogonial, 48 h in TM4 Sertoli cells, 100 μM for 24 h in TM3 Leydig and TM4 Sertoli cells, and 48 h in C18-4 spermatogonial and TM3 Leydig cells.BPS significantly induced MBD1 total intensity at 100 μM in spermatogonial cells, 50 and 100 μM in Leydig and Sertoli cells for 24 and 48 h. BPAF significantly induced MBD1 total intensity at a dose of 10 μM for 24 h, 5 and 10 μM for 48 h in all three cell types. TBBPA increased MBD1 total intensity at 10 μM for 24 h and 48 h in spermatogonial cells and 48 h in Sertoli cells.

Figure 6. BPA and its analogs differentially induced DNA methylation and activated autophagy in spermatogonial (C18-4), Leydig (TM3), and Sertoli cells (TM4).

Cells treated with vehicle (0.01% DMSO) were used as negative controls. A shows the representative images of cells treated with BPA and BPS (100 μM), BPAF and TBBPA (10 μM) for 24 h. The nuclei were stained with Hoechst 33342 (blue). Double immunofluorescence in the cell lines using the LC-3B (red) and MBD-1 (green) antibodies. Noted a clearly distinct expression of LC3A in red vacuoles with perinuclear/nuclear localization and of MBD-1 green that has a diffuse, throughout the cytoplasm, localization. Scale bar = 50 μm. B-C shows the single-cell quantification of the log-transformed total intensity of MBD1 (B) and LC3B (C) immuno-labeling of the three testicular cell types treated with various BPA, BPS, BPAF and TBBPA concentrations for 24 and 48h. Data were presented as mean ± SD, n = 6. Three replicates in two separate experiments were included. Statistical analysis was conducted by 1-way ANOVA followed by Tukey-Kramer multiple comparisons (*P < 0.05).

Autophagy plays an essential role in regulating cell survival by degrading and recycling cytoplasmic components to maintain cellular homeostasis (Baehrecke 2005, Gallagher, Williamson et al. 2016, Hargarten and Williamson 2018). Here we employed an autophagosome membrane protein LC3B to monitor autophagy activity, as shown in Figures 6A and 6C. BPA significantly increased LC3B total intensity at 50 and 100 μM for 24 h in spermatogonial cells, 48 h for Leydig and Sertoli cells, 100 μM for 24 h in Leydig and Sertoli cells, and 48 h in spermatogonial cells. BPS significantly induced LC3B total intensity at 100 μM for 24 h in spermatogonial cells, and 50 and 100 μM for 24 and 48 h in Leydig and Sertoli cells. BPAF induced LC3B expression at 10 μM for 24 h in spermatogonial and Sertoli cells, 5 and 10 μM for 24 h in Leydig cell, and 48 h for all cell types. TBBPA significantly increased LC3B level at 10 μM for 24 h in spermatogonial cells, and 48 h in spermatogonial and Sertoli cells.

BPA and its analogs demonstrated differential toxicity responses

Figure 7A illustrates the differential toxicity spectrum profile of tested compounds in spermatogonial, Sertoli, and Leydig cells (cell types with the indicated concentration of tested chemicals on the left). The HCA spectrum reveals the dynamic, chemical-specific, dose-dependent changes of each HCA parameter. Figure 7A compares multiplexed HCA parameters among BPA and its analogs from three types of testicular cells. Each chemical displayed a unique dose-dependent profile within each type of cell. The aptitudes of these spectra were more prominent at the 48h than at the 24h. Combined with the HCA index (Figure 7B), we found all three types of cells showed the highest response to BPAF at 10 μM across all variables measured. BPAF targeted spermatogonial cell more significantly (C18) at 5 μM. BPS more likely targeted Sertoli cell (TM4) and Leydig cell (TM3) and less at spermatogonia cells. TBBPA targeted spermatogonia, Sertoli cells, and less at TM3 cells. BPA is mainly targeted at TM4, followed by TM3 cells, and less at spermatogonial cells. Finally, we examined the statistical correlations among these HCA parameters, as shown in Table 1. These correlation-coefficient matrices indicate cell type-specific and chemical-specific. For example, cell viability was positively correlated with BrdU and negatively correlated in MBD1, LC3B, and H2AX in BPA, BPS, TBBPA treated C18 cells except for BPAF treatment in which cell viability was negatively associated with BrdU. We also observed that BPAF was negatively associated with BrdU in TM3 and TM4 cells, different from other treatments. Also, we observed cell viability negatively associated with the cell cycle G2M phase in all treatments in C18-4 and TM4 cells but positively associated with G2M in BPA, BPS, and TBBPA in TM3 cells, not BPAF. Further studies would be needed to clarify the mechanisms involved in differential sensitivity among three cell types in response to BPA and its analogs, especially BPAF.

Figure 7. Differential toxic responses of BPA and its analogs on C18 spermatogonial, TM3 Leydig, and TM4 Sertoli cells.

A. HCA toxicity spectrums of BPA and its analogs. Multiplexed HCA parameters among BPA and its analogs among three types of testicular cells. The deviation (fold changer from the Control) of each assay, including cell number, nuclear area, nuclear shape measurements (P2A and LWR), DNA synthesis (BrdU positive cells), cell cycle phases (cell population in subG1 phase, G0/G1 phase, S phase, and G2/M phase), DNA damage response (γ-H2AX positive cells), F-actin total intensity, MBD1 total intensity and LC3B total intensity were calculated, log2-transformed and represented as color-spectrums (by cellular functions). Each chemical displayed a unique dose-dependent profile within each type of cell. The aptitudes of these spectra were more prominent at the 48h time point than at the 24h time point. B. HCA index of BPA and its analogs. HCA index is calculated as the sum of the absolute value of each log2 transformed fold change, representing each chemical’s accumulative toxicity.

Table 1.

Corelation matrix of 13 parameters from the high conetent analsysis in spermatogonial cells, TM4 Setoli cells and TM3Leydig cells. The correlations are estimated by Row-wise method.

	C18															TM3															TM4
		CellViab	G0/1	S phase	G2/M	A	Area	P2A	LWR	F-actin	BrdU	MBD1	LC3B	H2AX			CellViab	G0/1	S phase	G2/M	A	Area	P2A	LWR	F-actin	BrdU	MBD1	LC3B	H2AX			CellViab	G0/1	S phase	G2/M	A	Area	P2A	LWR	F-actin	BrdU	MBD1	LC3B	H2AX
BPA	CellViab	1	0.79	0.73	−0.9	−0.8	0.82	−0.8	−0.6	−0.9	0.84	−1	−0.83	−0.95		CellViab	1	−0.9	0.96	0.83	−0.3	−0.6	0.44	0.37	−1	0.51	−0.97	−0.96	−0.84		CellViab	1	0.79	0.82	−0.8	0.42	−0.6	0.5	0.17	−1	−0.3	−1	−0.97	−0.99
	G0/1	0.79	1	0.3	−1	−1	0.56	−1	−0.9	−0.5	0.5	−0.6	−0.47	−0.89		G0/1	−0.86	1	−1	−1	0.06	0.32	−0.7	−0.7	0.92	−0.8	0.87	0.85	0.67		G0/1	0.79	1	0.97	−1	0.58	−0.5	0.69	0.22	−0.9	−0	−0.9	−0.9	−0.85
	S phase	0.73	0.3	1	−0.4	−0.4	0.8	−0.4	−0.2	−0.9	0.87	−0.9	−0.93	−0.63		S phase	0.96	−1	1	0.95	−0.3	−0.4	0.57	0.47	−1	0.71	−0.92	−0.89	−0.74		S phase	0.82	0.97	1	−1	0.52	−0.4	0.75	0.41	−0.9	0.12	−0.9	−0.9	−0.86
	G2/M	−0.9	−1	−0.4	1	0.97	−0.7	0.96	0.87	0.61	−0.6	0.72	0.55	0.92		G2/M	0.83	−1	0.95	1	−0.2	−0.2	0.62	0.5	−0.9	0.89	−0.8	−0.76	−0.53		G2/M	−0.8	−1	−1	1	−0.6	0.42	−0.71	−0.3	0.86	−0.1	0.89	0.88	0.83
	A	−0.8	−1	−0.4	0.97	1	−0.6	0.93	0.85	0.6	−0.6	0.68	0.52	0.9		A	−0.28	0.06	−0.3	−0.2	1	−0.1	0.34	0.43	0.15	−0.3	0.07	0.05	0.03		A	0.42	0.58	0.52	−0.6	1	−0.5	−0.05	−0.2	−0.4	−0.5	−0.4	−0.42	−0.46
	Area	0.82	0.56	0.8	−0.7	−0.6	1	−0.6	−0.5	−0.9	0.84	−0.9	−0.94	−0.73		Area	−0.64	0.32	−0.4	−0.2	−0.1	1	−0.3	−0.3	0.63	0.31	0.7	0.75	0.9		Area	−0.6	−0.5	−0.4	0.42	−0.5	1	0.09	0.62	0.56	0.79	0.51	0.61	0.68
	P2A	−0.8	−1	−0.4	0.96	0.93	−0.6	1	0.95	0.49	−0.5	0.66	0.53	0.87		P2A	0.44	−0.7	0.57	0.62	0.34	−0.3	1	0.99	−0.6	0.43	−0.54	−0.57	−0.58		P2A	0.5	0.69	0.75	−0.7	−0.1	0.09	1	0.77	−0.7	0.63	−0.7	−0.65	−0.5
	LWR	−0.6	−0.9	−0.2	0.87	0.85	−0.5	0.95	1	0.25	−0.3	0.43	0.34	0.69		LWR	0.37	−0.7	0.47	0.5	0.43	−0.3	0.99	1	−0.5	0.28	−0.49	−0.54	−0.59		LWR	0.17	0.22	0.41	−0.3	−0.2	0.62	0.77	1	−0.3	0.82	−0.4	−0.24	−0.12
	F-actin	−0.9	−0.5	−0.9	0.61	0.6	−0.9	0.49	0.25	1	−1	0.93	0.93	0.81		F-actin	−0.98	0.92	−1	−0.9	0.15	0.63	−0.6	−0.5	1	−0.5	0.98	0.98	0.87		F-actin	−1	−0.9	−0.9	0.86	−0.4	0.56	−0.66	−0.3	1	0.1	0.97	0.99	0.97
	BrdU	0.84	0.5	0.87	−0.6	−0.6	0.84	−0.5	−0.3	−1	1	−0.9	−0.92	−0.81		BrdU	0.51	−0.8	0.71	0.89	−0.3	0.31	0.43	0.28	−0.5	1	−0.45	−0.39	−0.08		BrdU	−0.3	−0	0.12	−0.1	−0.5	0.79	0.63	0.82	0.1	1	0.07	0.15	0.28
	MBD1	−1	−0.6	−0.9	0.72	0.68	−0.9	0.66	0.43	0.93	−0.9	1	0.93	0.86		MBD1	−0.97	0.87	−0.9	−0.8	0.07	0.7	−0.5	−0.5	0.98	−0.5	1	0.99	0.88		MBD1	−1	−0.9	−0.9	0.89	−0.4	0.51	−0.69	−0.4	0.97	0.07	1	0.99	0.96
	LC3B	−0.8	−0.5	−0.9	0.55	0.52	−0.9	0.53	0.34	0.93	−0.9	0.93	1	0.75		LC3B	−0.96	0.85	−0.9	−0.8	0.05	0.75	−0.6	−0.5	0.98	−0.4	0.99	1	0.92		LC3B	−1	−0.9	−0.9	0.88	−0.4	0.61	−0.65	−0.2	0.99	0.15	0.99	1	0.98
	H2AX	−1	−0.9	−0.6	0.92	0.9	−0.7	0.87	0.69	0.81	−0.8	0.86	0.75	1		H2AX	−0.84	0.67	−0.7	−0.5	0.03	0.9	−0.6	−0.6	0.87	−0.1	0.88	0.92	1		H2AX	−1	−0.9	−0.9	0.83	−0.5	0.68	−0.5	−0.1	0.97	0.28	0.96	0.98	1
		CellViab	G0/1	S phase	G2/M	A	Area	P2A	LWR	F-actin	BrdU	MBD1	LC3B	H2AX			CellViab	G0/1	S phase	G2/M	A	Area	P2A	LWR	F-actin	BrdU	MBD1	LC3B	H2AX			CellViab	G0/1	S phase	G2/M	A	Area	P2A	LWR	F-actin	BrdU	MBD1	LC3B	H2AX
BPAF	CellViab	1	0.98	−0.1	−1	0.17	−0.8	−0.5	−0.6	−0.9	−0.4	−1	−0.95	−0.93		CellViab	1	0.96	−0.9	−0.9	−0.8	−1	−0.9	−0.9	−1	−0.8	−0.99	−0.98	−0.99		CellViab	1	0.99	−0.7	−1	0.07	−1	−0.97	−1	−1	−0.9	−1	−0.99	−0.94
	G0/1	0.98	1	−0	−1	0	−0.7	−0.6	−0.7	−0.9	−0.4	−0.9	−0.88	−0.97		G0/1	0.96	1	−0.9	−1	−0.8	−0.9	−1	−1	−0.9	−0.7	−0.94	−0.89	−0.95		G0/1	0.99	1	−0.8	−1	0.11	−0.9	−0.94	−1	−1	−0.9	−1	−0.97	−0.93
	S phase	−0.1	−0	1	−0	0.46	0.01	0.32	0.22	−0.2	0.83	−0.1	−0.11	0.18		S phase	−0.94	−0.9	1	0.86	0.87	0.94	0.77	0.85	0.9	0.85	0.97	0.96	0.88		S phase	−0.7	−0.8	1	0.75	−0	0.54	0.64	0.64	0.77	0.4	0.71	0.64	0.77
	G2/M	−1	−1	−0	1	0.01	0.68	0.63	0.76	0.86	0.29	0.93	0.87	0.96		G2/M	−0.93	−1	0.86	1	0.7	0.88	0.91	0.94	0.85	0.74	0.91	0.85	0.93		G2/M	−1	−1	0.75	1	−0.2	0.96	0.94	0.93	0.98	0.87	0.97	0.96	0.94
	A	0.17	0	0.46	0.01	1	−0.6	0.78	0.64	−0.5	0.07	−0.3	−0.43	0.21		A	−0.82	−0.8	0.87	0.7	1	0.72	0.75	0.84	0.79	0.51	0.85	0.83	0.75		A	0.07	0.11	−0	−0.2	1	−0.2	0.07	−0	−0.1	−0.1	−0.1	−0.1	0.07
	Area	−0.8	−0.7	0.01	0.68	−0.6	1	−0.1	0.1	0.93	0.45	0.86	0.9	0.6		Area	−0.96	−0.9	0.94	0.88	0.72	1	0.73	0.8	0.95	0.88	0.97	0.98	0.95		Area	−1	−0.9	0.54	0.96	−0.2	1	0.95	0.94	0.94	0.96	0.95	0.97	0.87
	P2A	−0.5	−0.6	0.32	0.63	0.78	−0.1	1	0.97	0.16	0.22	0.34	0.22	0.76		P2A	−0.88	−1	0.77	0.91	0.75	0.73	1	0.98	0.82	0.44	0.85	0.79	0.88		P2A	−1	−0.9	0.64	0.94	0.07	0.95	1	0.97	0.97	0.9	0.96	0.96	0.95
	LWR	−0.6	−0.7	0.22	0.76	0.64	0.1	0.97	1	0.33	0.19	0.51	0.4	0.86		LWR	−0.93	−1	0.85	0.94	0.84	0.8	0.98	1	0.87	0.55	0.91	0.86	0.91		LWR	−1	−1	0.64	0.93	−0	0.94	0.97	1	0.97	0.94	0.98	0.99	0.88
	F-actin	−0.9	−0.9	−0.2	0.86	−0.5	0.93	0.16	0.33	1	0.29	0.97	0.98	0.75		F-actin	−0.98	−0.9	0.9	0.85	0.79	0.95	0.82	0.87	1	0.7	0.97	0.98	0.98		F-actin	−1	−1	0.77	0.98	−0.1	0.94	0.97	0.97	1	0.89	0.99	0.98	0.95
	BrdU	−0.4	−0.4	0.83	0.29	0.07	0.45	0.22	0.19	0.29	1	0.31	0.32	0.4		BrdU	−0.75	−0.7	0.85	0.74	0.51	0.88	0.44	0.55	0.7	1	0.78	0.79	0.71		BrdU	−0.9	−0.9	0.4	0.87	−0.1	0.96	0.9	0.94	0.89	1	0.92	0.96	0.78
	MBD1	−1	−0.9	−0.1	0.93	−0.3	0.86	0.34	0.51	0.97	0.31	1	0.99	0.85		MBD1	−0.99	−0.9	0.97	0.91	0.85	0.97	0.85	0.91	0.97	0.78	1	0.99	0.97		MBD1	−1	−1	0.71	0.97	−0.1	0.95	0.96	0.98	0.99	0.92	1	0.99	0.92
	LC3B	−1	−0.9	−0.1	0.87	−0.4	0.9	0.22	0.4	0.98	0.32	0.99	1	0.79		LC3B	−0.98	−0.9	0.96	0.85	0.83	0.98	0.79	0.86	0.98	0.79	0.99	1	0.96		LC3B	−1	−1	0.64	0.96	−0.1	0.97	0.96	0.99	0.98	0.96	0.99	1	0.89
	H2AX	−0.9	−1	0.18	0.96	0.21	0.6	0.76	0.86	0.75	0.4	0.85	0.79	1		H2AX	−0.99	−1	0.88	0.93	0.75	0.95	0.88	0.91	0.98	0.71	0.97	0.96	1		H2AX	−0.9	−0.9	0.77	0.94	0.07	0.87	0.95	0.88	0.95	0.78	0.92	0.89	1
BPS		CellViab	G0/1	S phase	G2/M	A	Area	P2A	LWR	F-actin	BrdU	MBD1	LC3B	H2AX			CellViab	G0/1	S phase	G2/M	A	Area	P2A	LWR	F-actin	BrdU	MBD1	LC3B	H2AX			CellViab	G0/1	S phase	G2/M	A	Area	P2A	LWR	F-actin	BrdU	MBD1	LC3B	H2AX
	CellViab	1	−0.6	0.92	−0.5	−0.4	−0.6	0.58	0.84	−1	0.63	−0.9	−0.87	−0.89		CellViab	1	−0.9	0.96	0.83	−0.3	−0.6	0.44	0.37	−1	0.51	−0.97	−0.96	−0.84		CellViab	1	0.89	0.91	−0.8	0.19	−0.9	0.74	0.39	−1	0.7	−1	−0.97	−0.97
	G0/1	−0.6	1	−0.7	−0.2	0.14	0.92	−0.7	−0.7	0.5	−0.6	0.27	0.12	0.35		G0/1	−0.86	1	−1	−1	0.06	0.32	−0.7	−0.7	0.92	−0.8	0.87	0.85	0.67		G0/1	0.89	1	0.97	−1	0.27	−0.8	0.78	0.5	−0.9	0.79	−0.9	−0.93	−0.84
	S phase	0.92	−0.7	1	−0.5	−0.4	−0.7	0.55	0.79	−0.9	0.71	−0.7	−0.75	−0.78		S phase	0.96	−1	1	0.95	−0.3	−0.4	0.57	0.47	−1	0.71	−0.92	−0.89	−0.74		S phase	0.91	0.97	1	−1	0.12	−0.9	0.75	0.43	−0.9	0.76	−0.9	−0.93	−0.9
	G2/M	−0.5	−0.2	−0.5	1	−0.1	−0.2	−0.1	−0.4	0.46	0.02	0.48	0.74	0.56		G2/M	0.83	−1	0.95	1	−0.2	−0.2	0.62	0.5	−0.9	0.89	−0.8	−0.76	−0.53		G2/M	−0.8	−1	−1	1	−0.3	0.76	−0.75	−0.5	0.85	−0.8	0.89	0.89	0.79
	A	−0.4	0.14	−0.4	−0.1	1	0.24	0.17	0.08	0.55	−0.3	0.59	0.44	0.58		A	−0.28	0.06	−0.3	−0.2	1	−0.1	0.34	0.43	0.15	−0.3	0.07	0.05	0.03		A	0.19	0.27	0.12	−0.3	1	−0.2	−0.19	−0.3	−0.2	−0.2	−0.2	−0.16	−0.15
	Area	−0.6	0.92	−0.7	−0.2	0.24	1	−0.8	−0.7	0.62	−0.8	0.21	0.14	0.36		Area	−0.64	0.32	−0.4	−0.2	−0.1	1	−0.3	−0.3	0.63	0.31	0.7	0.75	0.9		Area	−0.9	−0.8	−0.9	0.76	−0.2	1	−0.49	−0.1	0.84	−0.5	0.83	0.83	0.94
	P2A	0.58	−0.7	0.55	−0.1	0.17	−0.8	1	0.89	−0.5	0.82	−0.2	−0.21	−0.25		P2A	0.44	−0.7	0.57	0.62	0.34	−0.3	1	0.99	−0.6	0.43	−0.54	−0.57	−0.58		P2A	0.74	0.78	0.75	−0.8	−0.2	−0.5	1	0.9	−0.8	0.99	−0.8	−0.86	−0.72
	LWR	0.84	−0.7	0.79	−0.4	0.08	−0.7	0.89	1	−0.7	0.72	−0.5	−0.55	−0.56		LWR	0.37	−0.7	0.47	0.5	0.43	−0.3	0.99	1	−0.5	0.28	−0.49	−0.54	−0.59		LWR	0.39	0.5	0.43	−0.5	−0.3	−0.1	0.9	1	−0.5	0.91	−0.5	−0.56	−0.35
	F-actin	−1	0.5	−0.9	0.46	0.55	0.62	−0.5	−0.7	1	−0.7	0.79	0.84	0.88		F-actin	−0.98	0.92	−1	−0.9	0.15	0.63	−0.6	−0.5	1	−0.5	0.98	0.98	0.87		F-actin	−1	−0.9	−0.9	0.85	−0.2	0.84	−0.82	−0.5	1	−0.8	0.98	0.98	0.95
	BrdU	0.63	−0.6	0.71	0.02	−0.3	−0.8	0.82	0.72	−0.7	1	−0.4	−0.35	−0.36		BrdU	0.51	−0.8	0.71	0.89	−0.3	0.31	0.43	0.28	−0.5	1	−0.45	−0.39	−0.08		BrdU	0.7	0.79	0.76	−0.8	−0.2	−0.5	0.99	0.91	−0.8	1	−0.8	−0.84	−0.67
	MBD1	−0.9	0.27	−0.7	0.48	0.59	0.21	−0.2	−0.5	0.79	−0.4	1	0.92	0.85		MBD1	−0.97	0.87	−0.9	−0.8	0.07	0.7	−0.5	−0.5	0.98	−0.5	1	0.99	0.88		MBD1	−1	−0.9	−0.9	0.89	−0.2	0.83	−0.84	−0.5	0.98	−0.8	1	1	0.94
	LC3B	−0.9	0.12	−0.8	0.74	0.44	0.14	−0.2	−0.6	0.84	−0.4	0.92	1	0.88		LC3B	−0.96	0.85	−0.9	−0.8	0.05	0.75	−0.6	−0.5	0.98	−0.4	0.99	1	0.92		LC3B	−1	−0.9	−0.9	0.89	−0.2	0.83	−0.86	−0.6	0.98	−0.8	1	1	0.95
	H2AX	−0.9	0.35	−0.8	0.56	0.58	0.36	−0.3	−0.6	0.88	−0.4	0.85	0.88	1		H2AX	−0.84	0.67	−0.7	−0.5	0.03	0.9	−0.6	−0.6	0.87	−0.1	0.88	0.92	1		H2AX	−1	−0.8	−0.9	0.79	−0.2	0.94	−0.72	−0.4	0.95	−0.7	0.94	0.95	1
TBBPA		CellViab	G0/1	S phase	G2/M	A	Area	P2A	LWR	F-actin	BrdU	MBD1	LC3B	H2AX			CellViab	G0/1	S phase	G2/M	A	Area	P2A	LWR	F-actin	BrdU	MBD1	LC3B	H2AX			CellViab	G0/1	S phase	G2/M	A	Area	P2A	LWR	F-actin	BrdU	MBD1	LC3B	H2AX
	CellViab	1	0.74	0.87	−0.9	−0.4	−0.9	−0.1	−0.3	−0.9	0.71	−0.9	−0.99	−0.99		CellViab	1	−0.5	0.39	0.47	0.76	−0.5	0.47	0.53	−0.5	0.09	−0.02	−0.04	−0.33		CellViab	1	0.96	0.8	−0.9	−0.2	−0.6	0.91	0.1	−1	−0.1	−0.9	−0.86	−0.96
	G0/1	0.74	1	0.93	−0.9	−0.3	−0.5	−0.3	−0.6	−0.9	0.97	−0.9	−0.79	−0.79		G0/1	−0.5	1	−0.2	−0.9	−0.5	0.62	−0.6	−0.7	0.84	−0.7	−0.33	−0.33	0.42		G0/1	0.96	1	0.62	−1	−0	−0.8	0.79	0.08	−1	−0.3	−0.9	−0.86	−0.98
	S phase	0.87	0.93	1	−1	−0.2	−0.7	−0.1	−0.4	−1	0.95	−1	−0.93	−0.92		S phase	0.39	−0.2	1	0.6	0.75	0.36	−0.3	−0.3	0.2	−0.4	−0.63	−0.67	0.02		S phase	0.8	0.62	1	−0.7	−0.6	−0.1	0.9	0.28	−0.6	0.35	−0.8	−0.53	−0.64
	G2/M	−0.9	−0.9	−1	1	0.32	0.75	0.18	0.46	0.97	−0.9	0.95	0.93	0.93		G2/M	0.47	−0.9	0.6	1	0.64	−0.3	0.36	0.39	−0.6	0.38	0.02	−0.01	−0.33		G2/M	−0.9	−1	−0.7	1	0.07	0.61	−0.85	−0.2	0.89	0.06	0.8	0.75	0.92
	A	−0.4	−0.3	−0.2	0.32	1	0.13	0.77	0.76	0.1	−0.1	0.17	0.28	0.3		A	0.76	−0.5	0.75	0.64	1	−0.1	0.16	0.18	−0.2	−0.2	−0.44	−0.43	0.03		A	−0.2	−0	−0.6	0.07	1	−0.2	−0.37	0.22	0.09	−0.4	0.32	0.12	0.1
	Area	−0.9	−0.5	−0.7	0.75	0.13	1	−0.4	−0.1	0.77	−0.5	0.79	0.82	0.83		Area	−0.53	0.62	0.36	−0.3	−0.1	1	−1	−1	0.93	−0.9	−0.8	−0.8	0.68		Area	−0.6	−0.8	−0.1	0.61	−0.2	1	−0.31	0.42	0.83	0.78	0.68	0.87	0.8
	P2A	−0.1	−0.3	−0.1	0.18	0.77	−0.4	1	0.93	0	−0.2	0.03	0.05	0.04		P2A	0.47	−0.6	−0.3	0.36	0.16	−1	1	0.99	−0.9	0.88	0.76	0.78	−0.61		P2A	0.91	0.79	0.9	−0.9	−0.4	−0.3	1	0.2	−0.8	0.28	−0.8	−0.62	−0.78
	LWR	−0.3	−0.6	−0.4	0.46	0.76	−0.1	0.93	1	0.29	−0.5	0.27	0.27	0.27		LWR	0.53	−0.7	−0.3	0.39	0.18	−1	0.99	1	−1	0.86	0.76	0.77	−0.63		LWR	0.1	0.08	0.28	−0.2	0.22	0.42	0.2	1	0.1	0.49	0.14	0.3	0.06
	F-actin	−0.9	−0.9	−1	0.97	0.1	0.77	0	0.29	1	−0.9	0.98	0.93	0.93		F-actin	−0.45	0.84	0.2	−0.6	−0.2	0.93	−0.9	−1	1	−0.9	−0.73	−0.74	0.63		F-actin	−1	−1	−0.6	0.89	0.09	0.83	−0.76	0.1	1	0.39	0.93	0.95	0.99
	BrdU	0.71	0.97	0.95	−0.9	−0.1	−0.5	−0.2	−0.5	−0.9	1	−0.9	−0.79	−0.78		BrdU	0.09	−0.7	−0.4	0.38	−0.2	−0.9	0.88	0.86	−0.9	1	0.9	0.91	−0.67		BrdU	−0.1	−0.3	0.35	0.06	−0.4	0.78	0.28	0.49	0.39	1	0.29	0.56	0.36
	MBD1	−0.9	−0.9	−1	0.95	0.17	0.79	0.03	0.27	0.98	−0.9	1	0.97	0.97		MBD1	−0.02	−0.3	−0.6	0.02	−0.4	−0.8	0.76	0.76	−0.7	0.9	1	0.99	−0.74		MBD1	−0.9	−0.9	−0.8	0.8	0.32	0.68	−0.8	0.14	0.93	0.29	1	0.95	0.94
	LC3B	−1	−0.8	−0.9	0.93	0.28	0.82	0.05	0.27	0.93	−0.8	0.97	1	1		LC3B	−0.04	−0.3	−0.7	−0	−0.4	−0.8	0.78	0.77	−0.7	0.91	0.99	1	−0.69		LC3B	−0.9	−0.9	−0.5	0.75	0.12	0.87	−0.62	0.3	0.95	0.56	0.95	1	0.95
	H2AX	−1	−0.8	−0.9	0.93	0.3	0.83	0.04	0.27	0.93	−0.8	0.97	1	1		H2AX	−0.33	0.42	0.02	−0.3	0.03	0.68	−0.6	−0.6	0.63	−0.7	−0.74	−0.69	1		H2AX	−1	−1	−0.6	0.92	0.1	0.8	−0.78	0.06	0.99	0.36	0.94	0.95	1

Discussion:

Spermatogenesis is regulated and supported by Sertoli cells, Leydig cells, and other cell types in the testis. Injury of these cell types will result in testicular damage and reproductive dysfunction. Intercellular communication among spermatogonial, Sertoli and Leydig cell in the various compartments of the testis are essential to sustain the normal spermatogenic process. The animal models showed that these cells could be selectively targeted by specific toxicants, resulting in germ-cell apoptosis and spermatogenesis dysfunction (Boekelheide 2005). For example, ubiquitous environmental toxicants, and phthalates can produce testicular atrophy by targeting Sertoli cells in laboratory animals (Lee, Richburg et al. 1999). Ethane-1,2-dimethanesulfonate is a cytotoxic alkylating agent that selectively impacts adult Leydig cells, while X-irradiation targets the germ cell resulting in alteration of cell cycle and apoptosis (Hasegawa, Wilson et al. 1997). Although these animal models have generated valuable evidence to support toxicology assessment, advanced in vitro models allow us to elucidate the toxic mechanisms, discern the cell-specific sensitivity in the testis at a low cost, and broaden biological pathways. Previous studies demonstrated that BPA and its analogs induced a broad spectrum of adverse effects on the reproductive system in animal models (Yamasaki, Noda et al. 2004, Feng, Yin et al. 2012, Peretz, Vrooman et al. 2014, Cope, Kacew et al. 2015, Chen, Shu et al. 2016). However, cell-specific sensitivity to a particular compound remains unclear. The co-culture system with germ cells, Sertoli cells, and Leydig cells allows for evaluating cellular interactions. However, it is technically challenging to assess the effect of BPAs on these individual types of cells. Thus, comparing chemical toxicity in different testicular cell types could help elucidate cell type-specific toxicity and the functional significance of each cell type although this approach loses the cell-cell interactions that were present in the co-culture system . We selected C18-4 spermatogonial cells, TM3 Leydig, and TM4 Sertoli cells since these cells have been widely used to examine cell-specific toxicity in the testis (Mather 1980, Braydich-Stolle, Hussain et al. 2005, He, Jiang et al. 2008, Golestaneh, Beauchamp et al. 2009, Braydich-Stolle, Lucas et al. 2010, Zhang, Gong et al. 2013, Reis, Moreira et al. 2015, Liang, Yin et al. 2017). C18-4 cell line exhibited the morphological features of type A spermatogonia and expressed the germ cell-specific genetic markers, and has been applied in multiple studies to examine signaling pathways involved in germ cell development and characterize reproductive toxicity of nanoparticles (Braydich-Stolle, Hussain et al. 2005, Hofmann, Braydich-Stolle et al. 2005, He, Jiang et al. 2008, Golestaneh, Beauchamp et al. 2009, Braydich-Stolle, Lucas et al. 2010, Zhang, Gong et al. 2013). TM4 (Sertoli) and TM3 (Leydig) cells were isolated from pre-pubertal mouse gonads and expressed the various hormone receptors (Mather 1980). TM4 Sertoli cells express functional receptors for sollicle-stimulating hormone (FSH) in supporting spermatogenesis, whereas TM3 Leydig cells express luteinizing hormone (LH) receptors and stimulate androgen production. TM3 Leydig cells were used to assess chemical effects on steroidogenic activity (Komatsu, Tabata et al. 2008, Chen, Zhang et al. 2015). TM4 Sertoli cells have been used as a testicular toxicity model to investigate adverse effects of chemical exposure on blood-testis-barrier (BTB) stability (Reis, Moreira et al. 2015). Therefore, the cell-specific response within testis in response to specific compounds could reflect regulated differences in signaling molecules that underlie the discrete functional roles among these three cell types.

In this study, we selected different dose ranges for the tested compound, 0, 5, 50, and 100 μM for BPA and BPS, and 0, 1, 5, and 10 μM for BPAF and TBBPA since our previous study revealed about tenfold difference in LC50 between BPAF and BPA in the testicular co-culture model (Yin, Siracusa et al. 2020). Consistent with our previous study with the testicular cell co-culture model, we found BPAF induced cytotoxicity in all three types of testicular cells, showing significant dose-dependent decreases in cell number at both 24 and 48h. BPAF showed a significant reduction in cell number at a low dose of 5 μmM in spermatogonia cells compared to TM3 and TM4 cells, suggesting BPAF may target spermatogonial cells (C18-4). BPA mainly targeted at TM4, followed by TM3 cells, and less at spermatogonial cells. BPS was revealed to have similar cytotoxicity as BPA, more likely targeting Sertoli cell (TM4) and Leydig cell (TM3) and less at spermatogonial cells. TBBPA targeted spermatogonia, TM4 Sertoli cells, and less at TM3 Leydig cells. Toxicity and teratogenic effects of the bisphenols BPA, BPS, BPF, and BPAF were compared in zebrafish embryo-larvae on their estrogenic mechanisms in an estrogen-responsive transgenic fish (Moreman, Lee et al. 2017). Developmental deformities effects were assessed, including cardiac edema, spinal malformation, and craniofacial deformities, and the rank order for toxicity was BPAF > BPA > BPF > BPS (Moreman, Lee et al. 2017). The National Toxicology Program (Tox21) reported on the biological activity of BPA and several of its analogs, BPS, Bisphenol F (BPF), and BPAF, and suggested BPA analogs causing mammalian reproductive toxicity (Program 2021). Furthermore, BPAF was reported to have significantly more effects on the reproductive system than other BPA-related compounds (Program 2021). Like BPA, the BPA analogs were shown to have varying levels of estrogenic activity in the testis. Still, the majority had activity within the same order of magnitude as BPA and were less potent than positive control reference agonists such as -estradiol. However, BPAF was consistently one of the most potent BPA analogs in the estrogen agonist and antagonist assays. In Tox21, BPAF was the most potent (estrogen receptor) ER agonist of the BPA analogs and was more potent than BPA (Pelch, Wignall et al. 2017, Pelch, Wignall et al. 2019). It is interesting to explore further if observed BPAF toxicity is associated with the ER activity.

The underlying mechanism of the cell-specific cytotoxicity among BPA and its analogs are unclear. Therefore, we developed a multi-endpoint high-content analysis (HCA), including nuclear morphology, cell cycle, DNA synthesis, DNA damage, cytoskeletal structure and DNA methylation and phagocytosis based on our previous studies (Liang, Yin et al. 2017). These HCA-based endpoint assays are associated with adverse outcome pathways (AOPs), including 13 parameters from three types of testicular cells at two time periods. Multiplexed HCA parameters spectrum (Figure 7A) demonstrated differential profiles among BPA and its analogs, and revealed dynamic, dose-dependent, and cell-specific changes in these HCA markers. Each chemical displayed a unique dose-dependent profile within each type of cell. HCA index, the sum of the absolute fold changes of 13 endpoints, reflects dose and time dependent alterations. BPAF at 10 μM resulted in similar HCA in all three types of cells, but the spermatogonial cell has the highest HCA index value at 5 μM, indicating that C18-4 cells are more susceptible to BPAF than other cell types. types of cells. Our current observation was consistent with the previous study showed that BPAF was most cytotoxic compared to other analogs (Liang, Yin et al. 2017). We observed that BPAF was more toxic to C18-4 cells than Sertoli or Leydig cells, especially in nuclear structure (area, shapes) and cell cycle alteration (G2M phase). Most importantly, tthrough the comparison of these multi-endpoint parameters, we found a unique toxic signature for BPAF. We found cell viability was positively correlated with BrdU and negatively correlated with MBD1, LC3B, and H2AX in BPA-, BPS-, TBBPA-treated C18 cells. However, cell viability in BPAF treatment was negatively associated with BrdU in all testicular cells.

In our previous study, we developed an approach to constructing a cell cycle profile with discrete SubG1, G0/1, S, and G2/M phases in the spermatogonial cell line, which showed accurate segmentation of cell cycles with high-content image analysis (Liang, et al., 2017). In this study, we employed the same approach to examine the chemical effects on cell cycle progression in different cell lines. The data revealed that BPA and its analogs perturbed DNA synthesis and cell cycle progression in a cell- and chemical-specific manner. Our results demonstrated that BPA and its analogs induced G2/M phase arrest in both spermatogonial cells and TM4 Sertoli cells. However, in TM3 cells, BPA and BPS treatment induced G0/1 phase arrest. It has been reported that chemical modulation of cyclin-dependent kinase complexes was able to result in G2/M phase arrest. In contrast, for chemicals affecting the calmodulin-dependent protein kinase-II, the G1 phase arrest was observed in the cell population (Morris, DeLorenzo et al. 1998, Malumbres and Barbacid 2005). Thus, the differential alteration of cell cycle progression in testicular cells might suggest the distinct cellular mechanism underlying chemical toxicity.

Nuclear morphological features have been suggested as valuable biomarkers in various adverse cellular events (M, T et al. 1999, Eidet, Pasovic et al. 2014). Multinucleated germ cells formed from round spermatids, or occasionally spermatocytes were reported due to failed integrity of the intercellular bridges serving as partitions (Miething 1993, Miething 1995, Greenbaum, Iwamori et al. 2011, McClusky 2011).Our current study confirmed that BPAF induced unique multinucleated germ cells (MNGs) in the spermatogonia cell, not in the TM3 and TM4 cell. The above results are consistent with our previous studies in C18 spermatogonial cell culture (Liang, Yin et al. 2017) and the testicular cell co-culture model (Yin, Siracusa et al. 2020). It has been reported that MNGs occurred in cryptorchidism and may be linked to testicular cancers (Cho, Ahn et al. 1989, Nacov 1990, D, J et al. 2003, Nistal, Gonzalez-Peramato et al. 2006, Zhang, Lin et al. 2011) and decreased sperm count (D, J et al. 2003, DJ, SJ et al. 2014). The induction of multinucleated gonocyte was reported as a reproductive toxicity hallmark in animal models exposed to di-(n-butyl) phthalate (DBP) (T, W et al. 1999, Akbarsha and Murugaian 2000, Mylchreest, Sar et al. 2002, NJ, BS et al. 2004, Faridha, Faisal et al. 2007, G, J et al. 2010). Having anti-androgenic effects, DBP was found to induce high levels of MNG in correlation with the decrease of testosterone production in fetal rat testis and human fetal testis xenograft (Kleymenova, Swanson et al. 2005, Spade, Hall et al. 2015). DBP targets testicular cells by altering cytoskeletal features and creating aberrant mitosis at the basal lamina, which causes the disorganized localization of actin (Kleymenova, Swanson et al. 2005, Spade, Hall et al. 2015). Likewise, DBP created MNGs and disrupted testicular development in fetal rats (Gaido, Hensley et al. 2007). While Spade et al. discovered a nonproliferative mechanism of DBP in MNGs, our study using a single cell-based quantification of BrdU further demonstrated that multinucleated germ cells induced by BPAF were positively correlated with cell proliferation (BrdU positive staining The irregularity of multinucleation in the spermatogonial cells can result in genomic instability that affects various proteins, cell cycle checkpoints, and DNA responses that are necessary for proper cellular division (Hanahan and Weinberg 2000, Armit, O’Dea et al. 2002, Ariizumi, Ogose et al. 2009). Polyploidy and multinucleation have been identified as hallmarks of several malignancies, including testicular cancer (Makarovskiy, Siryaporn et al. 2002, D, J et al. 2003, Storchova and Pellman 2004, Nistal, Gonzalez-Peramato et al. 2006, Zhang, Lin et al. 2011). The molecular pathway of MNG formation is still poorly known (Spade, Hall et al. 2015), and the examination of the underlining mechanism of BPAF-induced multinucleation of germ cells will fill the knowledge gap and gain insight into the link between the formation of multinucleated spermatogonial cells and fertility and possibly testicular cancer cell growth.

Spermatogenesis encompasses multiple cellular events for generating mature sperm, including dramatic cell morphology change and cell movement within seminiferous tubules. As one of a major components of the cytoskeleton, F-actin has been shown to play an important role in cell movement, cell connection, cargo transportation, acrosome reaction, and nuclear modification during spermatogenesis (Vogl 1989, Sun, Kovacs et al. 2011). In Sertoli cells, F-actin forms parallel actin bundles in ectoplasmic specialization that are located between adjacent Sertoli cells and adhesion between Sertoli cells and germ cells, which regulate the movement and release of spermatids (Setchell 2008, Cheng and Mruk 2009, Sun, Kovacs et al. 2011). We observed the distinct pattern of F-actin cytoskeleton among three cell types and alteration of these structures in a dose-dependent manner. The significant induction of F-actin total intensity further suggested the aberrant F-actin accumulation within cells. In an in vivo study, BPA induced aberrant actin distribution, contributing to the cytokinesis failure in the porcine oocyte (Wang, Han et al. 2016). Exposure to BPA causes the truncation and depolymerization of actin via perturbing localization of actin regulatory proteins in Sertoli cells, which provide a cellular mechanism for BPA-induced BTB damage (Xiao, Mruk et al. 2014). DNA damage has been suggested to affect BPA reproductive toxicity (Meeker, Ehrlich et al. 2010, Ferris, Mahboubi et al. 2016) (Ganesan and Keating 2016). γ-H2AX, an early cellular DNA damage response marker, has recently been applied in HCA to screen genotoxicity of chemicals (Garcia-Canton, Anadon et al. 2013, Garcia-Canton, Anadon et al. 2013, Ando, Yoshikawa et al. 2014). In this study, we detected γ-H2AX expression level and found that BPA and its analogs differentially induced the DNA damage response in a cell type-specific manner. It could be explained by differential biotransformation of BPA and its analogs and DNA damage repair capacity among different cell models. In in vitro study BPA was able to induce γ-H2AX at a low dose of 10 nM in human breast cancer cells, while in HepG2 cell line, BPA did not induce γ-H2AX even at a dose of 100 μM, but bisphenol F did (Audebert, Dolo et al. 2011, Pfeifer, Chung et al. 2015). In addition, BPAF did not induce γ-H2AX in HepG2 and chicken DT40 cells, while it significantly increased γ-H2AX in spermatogonial cell lines and co-culture model (Fic, Zegura et al. 2013, Lee, Liu et al. 2013, Liang, Yin et al. 2017).

This study aimed to investigate how cell cycle and DNA damage are related to autophagy or epigenetic regulation in BPA and its analog exposure. DNA methylation, one of the significant processes of epigenetic alterations, has been reported to show a unique pattern in testis development (Oakes, La Salle et al. 2007). There is increasing evidence to indicate epigenetic alterations could be one of the possible mechanisms of BPA-induced reproductive dysfunction (Bromer, Zhou et al. 2010, Doshi, Mehta et al. 2011, Tang, Morey et al. 2012). It has been reported that neonatal exposure to BPA induced hypermethylation of the estrogen receptor promoter region in rat testis (Doshi, Mehta et al. 2011). In the current study, BPA and its analogs induced the expression level of MBD1, a transcriptional repressor that could bind to single methylated CpGs, in three testicular cell types, suggesting the aberrant methylation occurred. In an epidemiological study, DNA methylation on multiple loci has been associated with decreased sperm concentration and quality (Houshdaran, Cortessis et al. 2007). Autophagy and the DNA damage response are biological processes essential for cellular homeostasis. Notably, environmental chemical exposure could cause cellular damage, including DNA damage responses, resulting in numerous cellular processes, such as cell cycle checkpoint activation and DNA repair machinery. Along with the measurement of MBD1, we also examined the expression level of an autophagosome membrane protein, LC3B, to monitor autophagic activity. We found BPA and its analogs differentially activated autophagy in spermatogonial, Sertoli and Leydig cells. As a highly conserved process to degrade and recycle cytoplasmic components, autophagy is essential for cell proliferation and differentiation and serves as a cell survival mechanism in stressed cells (Baehrecke 2005, Mizushima and Levine 2010, Mizushima and Komatsu 2011). The epigenetic modulation regulates the autophagy process by altering the activity of transcriptional factors involved in signaling pathways (Yin, Ma et al. 2021). It has been reported that autophagy was activated by environmental chemical exposure during spermatogenesis (Liu, Zhang et al. 2014, Liu, Wang et al. 2015, Zhang, Ling et al. 2016). Activation of autophagy acted as a protective role against BPA-induced neurodegeneration via AMPK and mTOR pathways at doses of 40 and 400 μg (Agarwal, Tiwari et al. 2015). In contrast, BPA treatment induced autophagy and apoptosis concurrently in rat testis at a dose of 50 mg/kg (Quan, Wang et al. 2016). Uncovering the mechanisms by which autophagy, DNA damage response, and epigenetic regulation provide novel insight into the pathobiology of conditions associated with dynamic spermatogenesis, including cancer and aging, and novel concepts for the development of improved therapeutic strategies against these pathologies.

As compared to the previous study, a similar toxicity ranking of BPA and its analogs was observed in spermatogonial cells, Sertoli cells, and Leydig cells, respectively. However, BPAF and TBBPA exhibited toxicity at lower doses (Liang, Yin et al. 2017). Fetal bovine serum (FBS) contains multiple hormones, including the FSH, testosterone, progesterone, and LH, to support cell growth, which could potentially exert agonist or antagonist effects on BPA or its analogs toxicity in testicular cells. Reduced serum or serum-free cell culture conditions are increasingly required in current studies to remove the effect of the hormone. In this study, we employed Nu-serum, a cost-effective FBS alternative, in the cell culture system. As compared to FBS serum, Nu-serum reduced the various hormone levels and protein content and has successfully supported a large variety of cell types (Culty, Nguyen et al. 1992, Medh, Santell et al. 1992). Interestingly, among all three cell types, spermatogonial cells appear to be the target for BPAF, and Sertoli cell is highly resistant to TBBPA toxicity. Although we previously observed high testicular toxicity of TBBPA in in vitro spermatogonial cells and testicualr co-culture model (Liang, Yin et al. 2017), its in vivo toxicity still remains controversial (Cope, Kacew et al. 2015). Cope, et al. only observed the decrease in thyroid hormone T4 levels in SD rats exposed to TBBPA at a dose of 1000 mg/kg. In contrast, no changes in sperm motility, concentration, or abnormal sperm percentage were observed in multiple generations (Cope, Kacew et al. 2015). Furtehr study is needed to undersatnd the mechisim of TBBPA induced toxicity.

In summary, by utilizing HCA assays in spermatogonial cells, Sertoli cells, and Leydig cells, we characterized multiple adverse effects of BPA and its analog on different testicular cells . The HCA toxicity specttrum and HCA index generated from multiparametric HCA data were able to evaluate cell-type specific toxicity profiling and identify targeted cell types, which provide substantial evidence for chemical testicular toxicity and multidimensional toxicity profiling for toxicity classification. In the future, we will include more endpoints that are closely related to in vivo reproductive function, such as steroidogenesis and hormone receptor expression, and more compounds to construct reproductive toxicological profiles and provide mechanistic information for the prediction of biological targets of reproductive toxins.