Gestational vinclozolin exposure suppresses fetal Leydig cell development in rats

Background: Vinclozolin is not only a common dicarboximide fungicide used to protect crops against diseases but also an endocrine disruptor. This study aimed to investigate the effects of gestational vinclozolin exposure on the development of rat fetal Leydig cells. Methods: Female pregnant Sprague-Dawley rats were exposed to vinclozolin (0, 25, 50, and 100 mg/kg body weight/day) by oral gavage from gestational day 14 to 21. Results: Vinclozolin dose-dependently depressed serum testosterone levels at doses of 50 and 100 mg/kg and anogenital distance at 100 mg/kg. RNA-seq, qPCR, and Western blot showed that vinclozolin down-regulated the expression of Nr5a1 , Sox9 , Lhcgr , Cyp11a1 , Hsd3b1 , Hsd17b3 , Amh , Pdgfa , and Dhh and their encoded proteins. Vinclozolin depressed NR2F2-positive stem Leydig cell number at a dose of 100 mg/kg and also enhanced autophagy in the testis. Conclusion: Vinclozolin disrupts fetal Leydig cell development via several pathways.

VCZ has been thought to be an environmental contaminant that binds to the androgen receptor (NR3C4, a nuclear receptor) to cause anti-androgenic effects as an antagonist [3]. The transient administration of VCZ to pregnant female rats at the time of embryonic sex determination promoted defects in the reproductive tract and subfertility of the F1 male offspring [4]. The in utero VCZ exposure to the pregnant dams caused abnormalities of the androgen-regulated sexual differentiation in the male offspring, including shortening of the anogenital distance (AGD) [5][6][7]. VCZ-exposed male offspring during gestation developed the retained nipples, cleft phallus, hypospadias, cryptorchidism, small ventral prostate, seminal vesicles, and epididymis, as well as reduced sperm counts [5][6][7]. However, the underlying mechanism is poorly understood on a molecular level.

Chemicals
The detailed materials and methods were described in the Supplementary material S1.
General toxicological parameters after treatment of vinclozolin were described in Table1.
Chemicals, reagents, kits, software, and equipment were described in Table2. Antibody sources and information were described in Table3. Primer information was described in Table4.

Animals and experimental design
The animal model used in the current study was the Sprague-Dawley rat. Adult female and male Sprague-Dawley rats (56 days of age, weighed at 220 ± 10 g) were purchased from the Shanghai Laboratory Animal Center (Shanghai, China). Rats were allowed one week to acclimate to the experimental environment. Time-pregnant female rats were randomly assigned into different treatment groups: 0, 25, 50, and 100 mg/kg/day VCZ (n = 8 rats/group). The experimental protocol for animal toxicity experiment was approved by Wenzhou Medical University Laboratory Animal Ethics Committee. The experiment was performed following the procedures described in the Guide for the Care and Use of Laboratory Animals published by the United States National Institutes of Health.
VCZ was suspended in corn oil (vehicle control). Daily gavage dosing of the dams started on GD 14 and continued until the initiation of parturition at doses of 0, 25, 50, or 100 mg/kg. Dose selection is based on the observation that VCZ significantly caused anomalies of the male reproductive tracts of rats at 200-400 mg/kg [12]. Body weights of dams were daily recorded. Male pups were euthanized by CO 2 . Blood and testis of each male pup was collected for the following studies.

Measurement of serum T level
Serum T level was detected by the Siemens Healthcare Diagnostics Total Testosterone Kit as previously described [13]. The inter-assay and intra-assay coefficients of variation were within 15%.

Histochemical hematoxylin staining
The incidence of multinucleated gonocytes (MNGs) in the fetal testis can be increased by some endocrine-disrupting chemicals [14][15][16]. The occurrence rate of MNGs after VCZ treatment was counted as previously described [13]. In brief, cross-sections were sliced at 6 µm from a testis tissue-array by microtome and stained with hematoxylin solution as previously described [17]. The percentage of the seminiferous cord containing MNGs was calculated.

Immunohistochemical staining
CYP11A1 is used as the biomarker of the FLC [18]. Transcription factor SOX9 is used as the biomarker of the SC [19]. The total numbers of CYP11A1-positive cell and SOX9positive cells were counted by a fractionator technique as previously described [20]. In brief, fetal testes were embedded in paraffin block in a tissue-array block, cross-sections were sliced and stained immunohistochemically with CYP11A1 or SOX9.
Immunohistochemical staining method was used as previously described [21].

Computer-assisted analysis of FLC metrics
The FLC metrics include cell, nuclear, and cytoplasmic sizes, which were measured as previously described [22]. In brief, cell and nuclear sizes were measured by the Media Cybernetics Image-Pro 6 Plus software, with the measurement parameter of the average area.

Semi-quantitative immunohistochemical measurement of CYP11A1 and SOX9
Immunohistochemical staining of CYP11A1 and SOX9 were performed as described [22].
Target protein density and surrounding area background density were measured by the Media Cybernetics Image-Pro 6Plus with the measurement parameter of the main density.

Measurement of FLC proliferation
The proliferation of FLCs was measured by immunofluorescent staining of proliferating cell nuclear antigen (PCNA) and CYP11A1 as previously described [23]. In brief, the crosssections in the tissue-array congregated above were utilized. Sections were double-labeled with the primary antibodies of CYP11A1 and PCNA followed by the fluorescent secondary antibody.

Measurement of stem Leydig cell number
Previous studies demonstrated that FLCs were developed from stem Leydig cells [8,24].
Stem Leydig cells in the fetal testis are spindle-shaped and express NR2F2, also called COUP-TFII [8, 24-26]. We used NR2F2 as the biomarker to identify the stem Leydig cell and SOX9 as the biomarker to draw a boundary for the seminiferous cord. Double immunofluorescent staining was performed as above. The number of NR2F2-positive cells per square millimeter was calculated.

Preparation of RNA-seq library
Eight testes per dam were randomly selected for total RNA isolation. Total RNAs were extracted from testes using Invitrogen Trizol Kit as previously described [27].

RNA-seq analysis
We performed RNA-seq of the testis and biological pathway analysis to address the pathway of VCZ-mediated action in vivo as previously described [28]. Gene expression level, Gene Ontology, Pathway analysis, scatter plots, and volcano plots were performed.

Biological pathway analysis
Biological pathway analysis was performed as previously described [29]. In brief, the GenMAPP2.1 software was used to create a map of signal pathways for the potential pathways.

Quantitative real-time PCR (qPCR)
We performed the qPCR of VCZ-treated samples to verify the sequencing data of the testes and to measure the levels of some mRNAs that were not detectable by RNA-seq as previously described [30]. The mRNA levels of Lhcgr, Scarb1, Star, Cyp11a1, Hsd3b1, Cyp17a1, Hsd17b3, Insl3, Nr5a1, Pcna, Pdgfa, Amh, Dhh, and Sox9 were analyzed using the SYBR Green qPCR Kit. Lhcgr, Scarb1, Star, Cyp11a1, Hsd3b1, Cyp17a1, Hsd17b3, were Insl3 were FLC genes. Pdgfa, Amh, Dhh, and Sox9 are SC genes, and Nr5a1 is the critical transcription factor for the development of both FLCs and SCs. Pcna is the proliferating cell biomarker. The mRNA levels were determined by a standard curve method.

Western blot
Western blot was performed as previously described [28]. In brief, testis is lysed, denatured, electrophoresed and blotted with the primary antibody followed by the secondary antibody. Blots were stripped and incubated with a monoclonal GAPDH antibody served as the internal control.

Measurement of serum IGF-1
Insulin-like growth factor 1 (IGF-1) plays important roles in cell proliferation, differentiation and growth of many cells, including Leydig cells [11,31]. The circulatory IGF-I is mainly produced by mammalian livers [31]. IGF-1 ELISA kit was used for the measurement of serum IGF-I according to the manufacturer's instruction.

Statistical analysis
Values for all intents and purposes are expressed as mean ± S.E.M., and data were analyzed by one-way ANOVA followed ad hoc Turkey's analysis to specifically compare values from VCZ-treated rats to the control. GraphPad Prism was used.

VCZ shortens AGD and depresses serum T levels in male pups
Eight dams per group were orally administered via gavage of 0, 25, 50, and 100 mg/kg/day VCZ for 10 days from GD 14 to 21 (Fig.1A). Body weights of dams and pups after VCZ treatment were without change (Supplementary material S5). Birth rate, pup number per dam, male birth weights and percent male ratio were without change either (Supplementary material S5). However, VCZ significantly depressed AGD at 100 mg/kg ( Fig.1B), indicating that VAC is an anti-androgen. We measured serum T levels in male pups. VCZ significantly depressed serum T levels in male pups at doses of 50 and 100 mg/kg (Fig.1C), suggesting that VCZ disrupts FLC steroidogenesis.

VCZ does not alter incidence of MNGs
Some endocrine disruptors, such as phthalates, can disrupt germ development, increasing the incidence of MNGs in the fetal testis [14,32]. The rate of MNGs after VCZ treatment was counted. VCZ did not alter MNG incidence in the fetal testis ( Supplementary Fig.S1).

VCZ deceases FLC cluster size
FLCs exist as single cell or a cluster of FLCs with two or more cells [33]. In the current study, we defined the cluster size for FLCs as single (one cell per cluster), small (2-4 cells per cluster), medium (5-16 cells per cluster), and large (>16 cells per cluster). VCZ showed slight increase of FLC number at 100 mg/kg, but no significance was observed when compared to the control ( Fig.2A-E). VCZ significantly depressed the percentage of medium-size FLC cluster (Fig.2F). This indicates that VCZ might disrupt the growth of FLC clusters. We also analyzed FLC metrics (cell and cytoplasmic sizes) and did not find any change of these two parameters ( Fig.2G and 2H).

VCZ does not change PCNA labeling index of FLCs
We performed double-staining of CYP11A1 for FLCs and PCNA for the proliferating cell.
VCZ did not alter the percentage of PCNA labeling index of FLCs ( Supplementary Fig.S2), confirming the above finding that unaltered FLC number was observed after VCZ treatment.

VCZ lowers NR2F2-positive stem Leydig cell number
Previous studies indicated that NR3C4 is critical for Leydig cell development [34,35] because stem Leydig cells express NR3C4 [36]. VCZ is an NR3C4 antagonist. We used NR2F2 as a biomarker of stem Leydig cells to identify their number. Indeed, NR2F2 is expressed in the spindle-shaped cells in the interstitial area ( Fig.3A-B). VCZ significantly lowered NR2F2-positive cell number in the interstitial compartment (Fig.3C). This indicates that VCZ disrupts stem Leydig cell number.

VCZ does not affect SC number
SCs were marked by biomarker SOX9. VCZ did not affect SC number ( Supplementary   Fig.S3). This indicates that the proliferation of SCs is not influenced by VCZ.

RNA-seq analysis reveals VCZ-induced fetal testis gene expression
We examined the effects of VCZ on testicular gene expression using RNA-Seq analysis. We sequenced the transcripts and 8848 transcripts were identified in the testis between control and 25-100 mg/kg VCZ groups. We particularly analyzed the difference between 100 mg/kg VCZ and the control. As shown in Supplementary Fig.S4A (volcanoplot), of these transcripts, 107 transcripts were significantly up-regulated (P < 0.05) and 101 transcripts were significantly down-regulated (P < 0.05) in the 100 ng/testis VCZ group ( Supplementary Fig.S4A). GO analysis identified that most of down-regulated genes were within regulation of protein catabolism, protein acylation, peptidyl-lysine acetylation and that most of up-regulated genes were within regulation of cellular protein catabolism, protein amino acid acetylation, peptidyl-lysine acetylation, inner cell mass proliferation,

Pathway analysis reveals VCZ-induced down-regulation of steroidogenesis and regulation pathway
In the steroidogenic pathway, we found Cyp11a1 was down-regulated by ³2 folds at 25, 50 and 100 mg/kg VCZ (Fig.4). We also found that two critical transcription factors of regulating FLC development and steroidogenesis, Nr5a1 and Nr4a1, were significantly down-regulated at doses of 25, or 50 and 100 mg/k VCZ. Further pathway analysis showed that Gli transcription factor in the hedgehog pathway was down-regulated at 100 mg/kg VCZ (Fig.4). We used qPCR to measure other FLC and SC gene expression. We found that VCZ down-regulated expression of FLC steroidogenesis-related genes (Lhcgr, Hsd3b1, and Hsd17b3) at 100 mg/kg (Fig.5) (Fig.5). This indicates that the FLC differentiation is blocked, in part by the down-regulation of SCsecreted growth factors.

VCZ lowers FLC and SC protein levels
We further performed Western blot for FLC proteins (LHCGR, CYP11A1, HSD3B1, and HSD17B3), SC proteins (PDGFA, AMH, DHH, SOX9) and FLC/SC common protein (NR5A1). It was found that VCZ lowered their levels at 25 or higher doses (Fig.6). IGF-1 was primarily secreted by liver and also was critical for the regulation of FLC development [41]. We performed ELISA to measure IGF-1 level in the serum of male offspring and found that VCZ decreased IGF-1 level at 100 mg/kg. Semi-quantitative immunohistochemical staining for CYP11A1 density per FLC and SOX9 density per SC showed that VCZ lowered CYP11A1 and SOX9 densities at doses of 50 and 100 mg/kg ( Supplementary Fig.S5). These data further confirm qPCR data.

VCZ induces autophagy in fetal testis in vivo
Previous studies indicated that many environmental chemicals induced autophagy to disrupt Leydig cell function [42,43]. Western blot showed that VCZ significantly induced autophagy in the testis tissues with increases in LC3-II and beclin-1 proteins and decrease in p62 at 100 mg/kg (Fig.7). This indicates that VCZ can induce autophagy in the testis as a protective mechanism.

Discussion
Here, we found that gestational exposure of VCZ to male fetuses inhibited T synthesis, decreased expression of several genes linked with the steroidogenesis of the testis.
We reported that VCZ shortened AGD at doses of 50 and 100 mg/kg. This is consistent with the observations from the exposure to 100 or 200 mg/kg/day VCZ from GD  Sox9 as well as Nr5a1 (Fig. 5) and their protein (Fig. 6)  Interestingly, the promoter analysis also showed that Cyp11a1 and Hsd3b1 were the targets of SOX9 [56]. Therefore, the significant down-regulation of SOX9 also led to the reduction of expression of Cyp11a1 and Hsd3b1.
Nr5a1, which is a gene expressed downstream of Sry, has a significant role in gonadal being a transcriptional activator of steroidogenic enzymes and other genes that are essential for androgen biosynthesis (Parker and Schimmer, 1997) and it also promotes SC  60]. Autophagy is characterized by increased expression of autophagy-related proteins, such as LC3-I to LC3-II conversion and beclin-1 as well as a decreased expression of p62 [61]. Adequate autophagy is important for maintaining the Leydig cell function [62] because it is able to reduce ROS accumulation by clearing damaged mitochondria [63]. Indeed, in the current study, we demonstrated that autophagy of testis was increased as shown by the increase of LC3-II and beclin-1 levels and a decrease of p62 levels (Fig. 7). This could be due to the decrease of growth factors such as circulatory IGF-1 (Fig. 6) or PDGFA, AMH, and DHH locally in the testis (Fig. 6).
This indicates that the increased autophagy after VCZ exposure at the highest dose is the protective mechanism for fetal testis function.

Conclusion
Our data show that VCZ exposure inhibits the secretion of growth factors such as PDGFA, AMH, and DHH after down-regulating expression of SOX9, thus blocking the differentiation of FLCs, leading to low T synthesis (Supplementary FigS6). At the high dose, VCZ also activates the autophagy pathway to count its effects.

Consent for publication
Not applicable.

Availability of data and materials
All data are included in this published article and its supplementary information files or available from the corresponding author on reasonable request.

Funding
The research is supported by NSFC (81730042) and Department of Health of Zhejiang Province (11-CX29).

Competing interests
The authors declare that they have no competing interests.           respectively. Mean ± SEM, n = 3. Identical letters designate no significant difference between the two groups at p < 0.05.