C28-induced autophagy of female germline stem cells in vitro and its potential mechanisms

There are few studies indicating that small molecular compounds affect the proliferation, differentiation, apoptosis, and autophagy of female germline stem cells (FGSCs). However, the epigenetic regulatory mechanism of small molecular compounds that induce autophagy in FGSCs remains unknown. In this study, we found that C28 reduced the viability and proliferation of FGSCs, respectively. Additionally, western blotting showed that the expression of autophagy marker light chain 3 beta II (LC3B-II) was significantly increased and expression of sequestosome-1 (SQSTM1) was significantly reduced in C28-treated groups. Immunofluorescence showed that, in C28-treated groups, the number of LC3B-II-positive puncta was increased significantly. These results indicated that C28 induced autophagy of FGSCs in vitro. ChIP-seq data showed that autophagy-related biological processes such as regulation of mitochondrial membrane potential, Golgi vesicle transport, and cellular response to reactive oxygen species were enriched. In addition, RNA-Seq showed that the expression of genes ( Trib3, DDIT3 , and ATF4 ) related to endoplasmic reticulum (ER) stress was enhanced by C28. describe potential epigenetic mechanisms of autophagy


Background
Female germline stem cells (FGSCs) have been isolated from ovarian tissues of neonatal and adult mice to successfully establish cell lines in vitro [1][2][3][4][5]. FGSCs can restore the 4μM C28-treated groups at 24h compared with the controls (DMSO treatment) (p < 0.05) ( Fig. 1B and C). Subsequently, cell activity and proliferation were detected by CCK8 and EdU assays. Exposure for 24 and 48h to 0.5, 1, 2, and 4μM C28 significantly decreased cell viability in a concentration-dependent manner according to CCK8 assay results ( Fig. 1D and E). EdU assays also revealed a statistically significant decrease in the proliferation of FGSCs after exposure to 0.5, 1, 2, and 4μM C28 for 24 and 48h ( Fig. 1F and G). These results showed that C28 reduced the number, viability, and proliferation of FGSCs in vitro.
In all subsequent experiments, 0.5μM C28 was used to treat FGSCs.

C28 has no effect on differentiation or apoptosis of FGSCs in vitro
To explore the mechanisms of C28 decreasing the number of FGSCs, differentiationrelated marker genes of FGSCs, stra8 and sycp3, were detected by RT-PCR. There was no expression of stra8 or sycp3 in C28-treated groups ( Fig. 2A). Next, we examined the apoptosis rate by flow cytometry and found no significant difference in the percentage of apoptotic cells between C28-treated groups and the controls at 3 and 24h ( Fig. 2B-E).
These results indicated that C28 had no effect on differentiation or apoptosis of FGSCs in vitro.

C28 induces FGSC autophagy in vitro
Next, FGSC autophagy was investigated after treatment with C28. Western blot results showed that expression of LC3B-II was significantly increased and SQSTM1 expression was significantly decreased compared with the control at 3h (Fig. 3A-D). Furthermore, immunofluorescence showed that the number of LC3B-II-positive vesicles was significantly increased in FGSCs exposed to C28 for 3h compared with the control (Fig. 3E, F). These results showed that C28 induced FGSC autophagy at 3h. Therefore, 3h was selected to treat FGSCs with C28 for the following experiments. Taken together, the results indicated that C28 induced autophagy of FGSCs in vitro.

Genome wide profiling of H3K27ac indicates its role in C28-induced autophagy of FGSCs
To explore the epigenetic mechanism of C28-induced FGSC autophagy, H3K27ac was investigated by ChIP-Seq in C28-treated FGSCs and the control (Fig. 4). Before analyzing the ChIP-seq data, we applied FastQC v0.11.5 software to determine the data quality and analyzed several related variables [42]. The analysis indicated that the sequencing quality was very high (Additional file 2 Fig. S1 and S2). As shown in Fig. 4a, we found that three H3K27ac-marked regions were clustered by analyzing ChIP-Seq data with cluster 2 and 3 marked by H3K27ac less in C28-treated FGSCs. To understand how the different enrichments of enhancer signatures contributed to FGSC phenotypes induced by C28, we performed GO analysis with GREAT. The results showed that enrichment terms in clusters 2 and 3 were related to autophagy, such as regulation of mitochondrial membrane potential, Golgi vesicle transport, and cellular response to reactive oxygen species (Fig.   4b, c). These results suggested that H3K27ac played an important role in C28-induced FGSC autophagy.

ER stress is involved in C28-induced FGSC autophagy
The quality of RNA-seq reads was examined using FastQC. High quality clean reads were obtained from raw reads by removing the adaptor sequences, reads with >5% ambiguous bases, and low quality reads. The clean reads were then aligned to the mouse genome (version: mm10_GRCm38) using the Hisat2 program. DEG analysis was performed using DESeq2. Significance and false discovery rate (FDR) analyses were conducted according to the following criteria: i) fold change >1.5, <1.5 or P< 0.05 and ii) FDR<0.05 (Additional file 2 Fig. S3A). Moreover, the correlation coefficient of each replicated sample showed high consistency (Additional file 2 Fig. S3B). Transcriptome sequencing was performed in C28-treated groups and the control. DESeq2 was used to analyze differentially expressed genes (DEGs). The DEGs between the C28-treated group and control were directly reflected by hierarchical clustering (Fig. 5a) and a volcano plot (Fig. 5B). Principal component analysis (PCA) indicated that similar within-sample and different betweensample expression patterns (Fig. 5C). Functional annotation analysis revealed upregulation of some ER stress-and autophagy-related genes (Trib3, DDIT3, and ATF4 ) ( Fig. 5D). qPCR results confirmed that the expression of these was genes significantly increased in the C28-treated group compared with the control and were consistent with RNA-Seq data (Fig. 5E). These results revealed that ER stress was involved in the process of C28-induced autophagy of FGSCs.

Discussion
Presently, infertility is a severe problem worldwide. Successfully isolating FGSCs from postnatal mammalian ovaries and culturing them in vitro may facilitate treatment of female infertility [43,44]. Many studies have provided convincing evidence that small molecular compounds affect the physiological functions of mammalian cells, such as proliferation, differentiation, autophagy, and apoptosis [13,14,29,45]. In this study, we found that C28 induced significant autophagy, and H3K27ac and ER stress might play roles in C28-induced autophagy of FGSCs in vitro.
Autophagy is a self-digestion process in cells, which plays an essential role in cell development and the cellular response to stress [46]. However, excessive autophagy triggers apoptosis-mediated cell death (also a known type of programmed cell death)[47].
Our results showed that the expression of autophagic marker LC3B-II was significantly increased and that of SQSTM1 was significantly reduced in FGSCs after treatment with C28 for 3h, which indicated that C28 induced FGSC autophagy.
To explore the underlying mechanism of C28-induced autophagy in FGSCs, we performed genome wide profiling of H3K27ac and RNA-Seq analysis in the C28-treated group. . In our study, the autophagy induced by C28 in FGSCs might be mitochondrial autophagy. Autophagy is an important evolutionary conserved process in eukaryotes, which degrades intracellular materials. In this process, lysosomes are vesicles produced by Golgi. In the present study, the low levels of H3K27ac at genes enriched for Golgi vesicle transport in C28-treated FGSCs might lead to cell death. ROS play an important role in cellular metabolism, such as apoptosis [53,54] and autophag [55]. Autophagy reduces the cytotoxic damage of ROS to protect cells by clearing organelles and proteins [55]. Our ChIP-Seq results showed that exposure of FGSCs to C28 may induce the cellular response to ROS, leading to autophagy.
The results of RNA-Seq showed that ER stress-related genes (Trib3, DDIT3, and ATF4 ) were highly expressed in C28-treated FGSCs and qPCR validated this result. According to previous reports, DDIT3 is an ER stress-related gene that induces ER stress to inhibit the mTOR pathway by regulating TRIB3 and ATG5 [56][57][58]. It has been revealed that the ATF4-DDIT3/CHOP-TRIB3-AKT1-mTOR pathway induces autophagy [57,59]. Additionally, the ATG12-ATG5-ATG16 complex mediates covalent binding of LC3B-I and phosphatidy lethanolamine to form lipid-soluble LC3B-II, thereby participating in extension of the autophagic bilayer membrane and the formation of autophagosomes [60,61]. Our study showed that C28 upregulated the expression of ER stress-related genes Trib3, DDIT3, and ATF4 in FGSCs, and the protein expression of LC3B-II was also upregulated in FGSCs treated with C28. These results suggested that C28 activates ER stress by upregulating ER stress-related genes in FGSCs, inhibiting the mTOR pathway to induce autophagy and decreasing of the number of FGSCs.

Conclusions
We found that C28 might induce autophagy by changing epigenetics and upregulating ER stress in FGSCs (Fig. 6), leading to a decrease in the number of FGSCs. This study provides evidence to support the novel finding that C28-induced autophagy is related to epigenetic regulatory mechanisms and ER stress. Our study also provides the framework for clinical application of C28.

Animals
Five-day-old C57BL/6 mice were purchased from SLAC Laboratory Animal Co. (Shanghai, China). All procedures involving animals were approved by the Institutional Animal Care and Use Committee of Shanghai and conducted in accordance with the National Research Council Guide for the Care and Use of Laboratory Animals.

C28 was synthesized by Professor Huchen Zhou from the Schools of Pharmacy and
Medicine, Shanghai Jiao Tong University. The chemical structure of C28 is shown in Figure   1A.

FGSC culture in vitro
The mouse FGSC line used in this study was characterized and established previously [40].

EdU assay
A Cell-Light EdU Apollo ® 567 in vitro imaging kit (RiboBio, Guangzhou, China) was used to assay cell proliferation, according to the manufacturer's protocol. The cells were incubated with 50μM EdU for 2h at 37°C after C28 treatment for 24h. Then, the cells were fixed in 4% paraformaldehyde for 20min at room temperature and washed with 2mg/ml glycine for 5min on a shaker. Permeabilization was conducted by incubation in 0.5% Triton X-100 for 1h. Then, 1×Apollo was added, followed by incubation for 30min on a shaker.
Subsequently, cells were washed three times with PBS. Then, 1×Hoechst 33342 was used to stain nuclei. Images were obtained under a DM2500 microscope (Leica, Germany) and analyzed by Image J.

Apoptosis analysis
Apoptotic cells were quantified using an annexin V-FITC/PI detection kit (Invitrogen, Carlsbad, CA) and flow cytometry. Briefly, FGSCs were treated with 0.5μM C28 for 3 and 24h, washed with PBS, collected, and resuspended in 1×Binding Buffer. The cells were then stained for 15min at room temperature with annexin V and propidium iodide, and analyzed by flow cytometry (BD FACSCalibur, BD Biosciences) to quantify apoptosis.

Reverse transcription-polymerase chain reaction and quantitative real-time polymerase chain reaction
FGSCs were cultured in vitro and treated with 0.5µM C28 for 24 and 48h. Total RNA was extracted from ovarian tissues of neonatal mice and FGSCs treated with C28 using TRIzol reagent (Life Technologies, CA), according to the manufacturer's instructions. RNA quantity and concentration were assessed using a NanoDrop 2000 spectrophotometer (Thermo Scientific, Wilmington, DE, USA). HiScript II Q RT SuperMix for qPCR kit (Vazyme Biotech) was used to obtain cDNA. Approximately 1µg RNA was used to synthesize cDNA in a 20µl reaction volume. Reverse transcription-polymerase chain reaction (RT-PCR) analysis was performed with Taq DNA polymerase. The RT-PCR products were separated on 1% agarose gels and imaged with a bioimaging system (Alpha Innotech). RT-PCR products were confirmed by sequencing.

Quantitative real-time polymerase chain reaction (qPCR) was conducted with SYBR Premix
Ex Taq (Takara, Shanghai, China) and an ABI 7500 Real-Time PCR System (Applied Biosystems) to measure expression levels of Trib3, DDIT3, and ATF4. The qPCR reaction volume of 20μl included 10μl SYBR Green PCR Master Mix (Roche, Basel, Switzerland), 1μl cDNA template, 2μl primer mixture, and 7μl water. qPCR conditions were as follows: 95°C for 30s 40 cycles of 95°C for 15s and 60°C for 60s, and then then 95°C for 15s, 60°C for 60s, and 95°C for 15s. All assays were repeated three times. Fold changes in expression were calculated using the 2 − ΔΔCt method. ΔΔCt = ΔCt experimental group -ΔCt control group, and ΔCt = Ct target gene -Ct Gapdh. Primers are listed in Additional file 1: Table   S1.

Western blotting
FGSCs were cultured in vitro and treated with 0.5µM C28 for 3 and 6h. Cells were lysed with RIPA buffer (Shanghai Yeasen Biotechnology Co., Ltd) containing a protease inhibitor cocktail. Protein concentrations were measured using the bicinchoninic acid (BCA) assay.
The western blotting procedure was as follows. A total of 20µg protein was separated by 15% w/v sodium dodecyl sulfate polyacrylamide gel electrophoresis and transferred onto a PVDF membrane. Then, the membranes were blocked with 5% dry nonfat milk in Tris-      Table S1.docx figure S2.tiff