Human urine stem cells protect against cyclophosphamide-induced premature ovarian failure by inhibiting SLC1A4-mediated outflux of intracellular serine in ovarian granulosa cells

Abstract

Background: Cyclophosphamide (CTX) is the first-line medication for the treatment of breast cancer, although it potentially leads to severe ovarian dysfunction and even premature ovarian failure (POF). However, the mechanism of CTX-induced POF remains unclear. Mesenchymal stem cell-based therapy has been wildly used for treating numerous diseases. Therefore, our study aims to elucidate the underlying mechanism of CTX-induced POF and to explore the therapeutic effect of human urine stem cells (hUSCs) in POF.

Methods: CTX-induced POF or ovarian granulosa cell (GCs) apoptosis were treated with hUSCs and their exosomes in vitro and in vivo. Morphological, histological, and functional alternations were examined using multiple approaches. The effector molecules of hUSC-derived exosomes (hUSC-Exo) were determined by differential expression analysis in the ovaries. The target genes of miRNA were accessed by transcriptome sequencing in GCs, and the underlying mechanisms were further elucidated.

Results: hUSCs remarkably inhibited CTX-induced apoptosis and promoted the proliferation of GCs, respectively. In addition, we observed that miR-27b-3p was highly expressed in hUSC-Exo and markedly suppressed CTX-induced GC apoptosis by specifically inhibiting the expression of SLC1A4, a serine transporter, in ovarian GCs, which, in turn, elevated the concentration of the intracellular serine by inhibiting the outflux of cellular serine. More importantly, the knockdown of SLC1A4 or simple supplementation of serine suppressed CTX-induced apoptosis of GCs. Finally, we demonstrated that CTX-induced apoptosis of ovarian GCs was essential for POF by reducing the intracellular serine concentration via elevating the expression of SLC1A4, whereas hUSCs protected against CTX-induced POF via miR-27b-3p/SLC1A4/serine axis-mediated activation of the PI3K/AKT/mTOR signaling pathway.

Conclusions: Our study suggests that hUSC-based cell therapy or simple supplementation of serine may provide an efficient therapeutic approach for the prevention and treatment of CTX-induced POF clinically.

Keywords: Cyclophosphamide; Human urine-derived stem cells; Ovarian granulosa cells; Premature ovarian failure; SLC1A4; miR-27b-3p.

Fig. 1

Characteristics analysis of hUSCs. A Morphological images of hUSCs at different culture passages (scale bar = 100 μm and 50 μm). B The expressions of CD29, CD90, CD105, CD73, HLA-ABC, CD34, CD45, and HLA-DR and the co-stimulatory molecules CD80, CD86, and CD40 were examined by flow cytometry analysis. C The expressions of Nanog, Oct4, CD105, CD29, CD34, and CD133 were analyzed by RT-PCR analysis. D The expressions of the embryonic stem cell surface markers Oct4, SSEA-4, and Nanog in hUSCs were identified by immunofluorescence staining. E The hUSCs were differentiated into mature adipocytes and osteoblasts, with osteogenic or lipogenic differentiation media for the corresponding times. Osteoblasts and adipocytes differentiated by hUSCs were identified by staining with alizarin red and oil red O, respectively. F In vitro tumorigenicity of hUSCs was conducted in soft agar. Numerous colonies appeared in the soft agar in the 4T1 tumor cells (1 × 10³/well) but not in the hUSCs, after an inoculation of 30 days. G, H In vivo tumorigenicity of hUSCs was performed in NOD-SCID mice with injection of 1 × 10⁶ cells/100 μl. Visible tumor was observed in the 4T1 cell group at the breast pads in week 2 after inoculation, and the tumor presented in the image was at week 8 after inoculation, whereas there was no tumor formation in the hUSCs group after 20 weeks of inoculation

Fig. 2

Homing of hUSCs in mouse ovaries. A Lentiviruses expressing GFP and luciferase genes were transfected into hUSCs. The transfection efficiency was detected by fluorescence microscopy. B hUSCs expressing GFP genes were observed under fluorescence microscopy. C A schematic diagram of the animal experimental procedure for preparing the CTX-induced POF model, as well as the cell transplantation. D The distribution of GFP⁺hUSCs was observed by in vivo imaging. The distribution of GFP⁺hUSCs in various organs of mice including heart, liver, spleen, lungs, kidneys, brain and ovaries was observed by in vivo imaging at the time of the sample harvest. E The hUSCs were stained with MAB1281, and the distribution of hUSCs in the CTX modeling ovaries was observed by imaging under fluorescence microscopy

Fig. 3

hUSCs and hUSC-CM transplantation improved CTX-induced ovarian dysfunction via inhibiting apoptosis of ovarian granulosa cells A Representative ovaries of each group of mice at the time of sample harvesting. B The weight changes of each group of mice were examined every 5 days, starting from the day before the CTX modeling. C The weights of mouse ovaries in different groups were measured and calculated for the ovarian coefficient. The percentage of the ovarian coefficient was calculated as: (ovarian tissue weight/body weight (g)) × 100. D Measurement and comparison of serum E2 between groups using ELISA. E Measurement and comparison of serum FSH between groups using ELISA. F Representative photomicrographs of H&E staining sections of mouse ovaries in the normal, CTX, CTX + hUSCs, and CTX + hUSC-CM groups. The black arrows represent primordial follicles, blue arrows represent primary follicles, gray arrows represent secondary follicles and red arrows represent atretic follicles. G, H Three non-adjacent H&E staining sections were randomly selected for counting the number of primordial follicles and primary follicles (G), and secondary follicles and atresia follicles (H) in mouse ovaries from each group. I, J Western blot assay for FHSR (I) and Bcl2 (J) expressions in ovarian tissues of different groups. K Detection of apoptosis and proliferation of ovarian GCs in each group using the TUNEL method and PCNA immunofluorescence staining. Significance was measured using a one-way ANOVA. *P < 0.05, **P < 0.01, ***P < 0.001

Fig. 4

hUSCs inhibited the apoptosis and promoted the proliferation of ovarian granulosa cells induced by CTX through paracrine effects. A, B Microscopic observation of the apoptosis of GCs (A) and the apoptotic ratios of GCs (B) in the normal group, CTX group, CTX + hUSCs group, and CTX + hUSC-CM group were measured by microscope and flow cytometry. C In vitro GC injury models were established with CTX (5 mg/mL, 24 h) in GCs. Then, the GCs were treated with hUSCs, hUSC-CM, and hUSC-Exo for an additional 24 h. D A representative image of hUSC-Exo was taken under the electron microscope. E The particle size of hUSC-Exo was measured by particle size analysis. F The expressions of CD81, TSG101, CD63, and CD9 were determined by Western blot in hUSC-Exo and hUSCs. G Live–dead staining of cells was performed with immunofluorescence analysis to identify death and apoptosis of GCs and the expression of PCNA

Fig. 5

hUSC inhibited CTX-induced GC apoptosis through miRNA-221-3p and miR-27b-3p in hUSC-Exo. A Volcano plot of differential expression miRNAs in the hUSC-Exo and DFL-Exo groups, n = 3. B Heatmap of the differentially expressed miRNAs between hUSC-Exo and DFL-Exo (red: increased expression; blue: decreased expression). C High expression of miR-27b-3p and miR-221-3p in hUSC-Exo was confirmed using qPCR. D miR-221-3p and miR-27b-3p expression of GCs was elevated after transient overexpression of miR-221-3p and miR-27b-3p. E The apoptotic ratios of GCs in the normal + NC group, CTX + NC group, CTX + miR-221-3p mimic group, and CTX + miR-27b-3p mimic group were detected by flow cytometry. F Live–dead staining to identify apoptosis of GCs was conducted using immunofluorescence staining *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001

Fig. 6

The related genes and signal pathways that have an antiapoptotic role associated with hUSC-Exo were determined in CTX-induced GC-injury models. A, B Venn diagram statistics for up- or downregulated differential genes between the CTX group, CTX + hUSC-Exo group, and CTX + hUSC-Exo group, respectively. C, D Heat map of differential genes was obtained from Venn plot statistics using differential expression ploidy. E Reactome analysis of the differential genes. F GO enrichment analysis of the differential genes. G The expressions of genes causing Th17 cell differentiation were determined using GSEA analysis in the different groups. H KEGG analysis of the differential genes revealed that most of them were enriched in the AKT signaling pathway

Fig. 7

miR-27b-3p inhibited the outflux of intracellular serine by selectively suppressing the expressions of SLC1A4 in CTX-induced POF. A The expressions of Slc1a4, Tmem86a, Pdia4, Ankrd1, and Heg1 were detected in GCs by qRT-PCR. B Predicted target sites of Slc1a4 3′UTR (red letters show the predicted pairing of the target region and miRNA). C Schematic diagram of SLC1A4 and serine transmembrane transport. D, E Detection of knockdown efficiency of Slc1a4 on GCs after siRNA transfection by qPCR and Western blotting. F The relative intensity of Slc1a4 in Western blotting. G Live–dead staining was used to detect death in the normal + NC group, CTX + NC group, CTX + siRNA group, and CTX + serine group. H The apoptotic ratios of GCs in each group were detected by flow cytometry. I The apoptosis of GCs was quantitatively detected by flow cytometry *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001

Fig. 8

hUSC-Exo inhibited CTX-induced apoptosis of GCs by activating the PI3K/AKT/mTOR signal pathway. A, B The expressions of the proteins related to the PI3K/Akt/mTOR signaling pathway, such as P-mTOR, P-PI3K ,and P-AKT, were detected by Western blot analysis in GCs in the treatment of hUSC-Exo. C, D The expressions of the proteins related to the PI3K/Akt/mTOR signaling pathway were determined by Western blot analysis in GCs in the treatment of hUSC-Exo with or without LY294002, an inhibitor of the PI3K/AKT/mTOR signaling pathway. E The apoptotic ratios of GCs in the normal group, CTX group, CTX + hUSC-Exo group, and CTX + hUSC-Exo + LY294002 group were detected by flow cytometry. F The quantitative results were obtained from the flow cytometry (E) *P < 0.05, **P < 0.01