ER stress-induced LINC00173 promotes the apoptosis of ovarian granulosa cells by regulating the HRK/PI3K/AKT pathway in polycystic ovary syndrome.

This study was approved by the Clinical Ethics Review Board of the Fourth Hospital of Shijiazhuang (ethical approval number: 20220112), and written informed consent was obtained from all patients. Of the total 100 patients who undergoing in vitro fertilization (IVF) at the Department of Reproductive Medicine Center, the Fourth Hospital of Shijiazhuang between January 2023 and December 2023. 50 patients were diagnosed with PCOS based on the 2003 Rotterdam criteria, and the remaining 50 patients served as controls. The control group was patients with normal ovulatory cycles but infertility due to male factors or tubal/cervical factor. Patients with endocrine diseases (e.g., Cushing’s syndrome, hyperprolactinemia, diabetes, thyroid dysfunction, androgen secreting tumor), endometriosis, chromosomal abnormalities, immune diseases, chemotherapy or radiation therapy, acute infections and other reproductive disorders (e.g., recurrent miscarriage and recurrent implantation failure) were excluded from this study. The patient’s general information is listed in Table  1 . Additionally, this study was conducted in accordance with the 1964 Declaration of Helsinki. All methods were performed in accordance with the relevant guidelines and regulatory “ Methods ” section. Table 1 Comparison of clinical baseline characteristics between healthy control and PCOS group. variables Healthy control group (n = 50) PCOS group (n = 50) Z/t values P values Age (years) a 29.98 ± 2.83 28.80 ± 3.80 1.762 0.081 Infertility years (years) b 3.00 (2.00, 4.00) 3.00 (1.75, 4.15) − 0.094 0.925 Body mass index (kg/m 2 ) b 22.50 (20.63, 25.18) 26.85 (23.40, 29.48) − 4.337 0.000* Basal FSH (IU/L) b 5.17 (4.36, 6.35) 4.53 (3.95, 5.61) − 2.110 0.035* Basal LH (IU/L) b 3.40 (1.45, 3.88) 3.97 (2.40, 6.82) − 2.172 0.030* Basal E2 (pg/mL) b 31.00 (22.75, 42.25) 37.00 (29.00, 48.25) − 1.997 0.046* Basal PRL (μg/L) b 13.81 (10.50, 19.42) 13.32 (9.49, 17.97) − 1.213 0.225 Basal T (nmol/L) b 0.98 (0.81, 1.27) 1.48 (1.13, 1.85) − 4.943 0.000* Antral follicle count (n) b 17.00 (12.00, 22.25) 25.00 (18.75, 37.00) − 3.713 0.000* Note: Values are described as a mean ± SD or b median (25th quartile,75th quartile). * P < 0.05. FSH follicle-stimulating hormone, E2 estradiol, LH luteinizing hormone, PRL prolactin, T testosterone. Comparison of clinical baseline characteristics between healthy control and PCOS group. Note: Values are described as a mean ± SD or b median (25th quartile,75th quartile). * P < 0.05. FSH follicle-stimulating hormone, E2 estradiol, LH luteinizing hormone, PRL prolactin, T testosterone. The GSE95728 and GSE137684 datasets were used to analyze differentially expressed lincRNAs in the GCs of healthy women and patients with PCOS. The datasets were retrieved from the Gene Expression Omnibus (GEO) database. The GSE95728 dataset contains 7 PCOS patients and 7 healthy women, while the GSE137684 dataset contains 8 PCOS patients and 4 healthy women. Differentially expressed lincRNAs in the GSE95728 and GSE137684 datasets were analyzed by using the Limma package. Heatmaps were generated using Z scores to analyze the differentially expressed lincRNAs in the GSE95728 and GSE137684 datasets. 100 cases of follicular fluids were obtained from 50 patients with PCOS and 50 healthy women on the day of oocyte retrieval, as described previously 43 . All identifiable follicles (≥ 14 mm) were subjected to aspiration. After the retrieval of oocytes, any unclear FF or samples that showed visible blood contamination were excluded. Clear FF samples, devoid of blood impurities, were gathered in a 50 mL sterile tube, correctly labeled, and immediately sent to the laboratory for centrifugation at 2000× g for 10 min at room temperature. The collected cell pellets were then resuspended in 5 mL of phosphate-buffered saline (PBS). Subsequently, the cell suspension was added to 5 mL of sample density separation solution (TBD Technology Co., Ltd., Tianjin, China) and centrifuged at 1500× g for 15 min. The intermediate white membrane layer was collected and washed twice with PBS. Finally, freshly prepared granulosa cells were immediately used for experiments. The KGN cell line, sourced from Shanghai Zhong Qiao Xin Zhou Biotechnology in Shanghai, China, was used in this study. Prior to the study, the cells were subjected to STR profiling for verification. The cells were cultured in DMEM/F-12 (Thermo Fisher, Carlsbad, CA, USA) supplemented with 10% fetal bovine serum (FBS, Thermo Fisher) and maintained at 37 °C in a humidified atmosphere with 5% CO 2 . For the experiments, KGN cells were pretreated with 100 nM thapsigargin (Tg) (Calbiochem, Burlington, MA, USA) for 6 h. Subsequent experiments were conducted after harvesting the cells. A plasmid for LINC00173 overexpression (pcDNA3.1-LINC00173) was constructed by GenScript (Nanjing GenScript Co., Ltd., Nanjing, China). siRNAs targeting human LINC00173 (siLINC00173), HRK (siHRK), ERN1 (siERN1), ATF6 (siATF6), and EIF2AK3 (siEIF2AK3) and negative control siRNA (siNC) were purchased from Gene Pharma (GenePharma Co., Ltd., Shanghai, China). The original pcDNA3.1 plasmid or negative control siRNA was used as a transfection control for gene overexpression or knockdown, respectively. KGN cells were transiently transfected using Lipofectamine 2000 (Thermo Fisher) according to the manufacturer’s protocol for 48 h. For plasmid/siRNA cotransfection, 2 µg of plasmid and 100 pmoL of siRNA were added to each well. All siRNA sequences are listed in Table S1 . Total RNA from granulosa cells and KGN cells was extracted using TRIzol reagent (Thermo Fisher). cDNA was synthesized from the RNA using the Prime-Script RT Reagent Kit (Roche Diagnostics, Indianapolis, IN, USA). The relative RNA expression of LINC00173 and apoptosis markers was measured using SYBR Green Mix (Yi Sheng Co., Ltd., Shanghai, China). The PCR conditions were as follows: 95 °C for 3 min, 95 °C for 5 s, 95 °C for 30 s, and 60 °C for 30 s for a total of 40 cycles. The relative expression of the target genes was normalized to that of β-actin and calculated using the 2 −ΔΔCT method. The primer sequences are listed in Table S2 . Granulosa cells and KGN cells were lysed using RIPA lysis buffer (Solarbio, Beijing, China) supplemented with protease and phosphatase inhibitors. The protein concentrations in the lysed cells were quantified using a Nanodrop 2000. Adequate amounts of protein were electrophoresed on a 10% sodium dodecyl sulfate‒polyacrylamide gel and transferred to a polyvinylidene fluoride (PVDF) membrane (Millipore, Bedford, MA, USA). The membrane was blocked with 5% nonfat skim milk for 1 h at room temperature and then incubated overnight at 4 °C with primary antibodies. Subsequently, the membrane was washed three times with Tris-buffered saline supplemented with 0.1% Tween 20 (TBS-T) and incubated with the corresponding secondary antibodies for 1 h at room temperature. Finally, the membrane was washed three times with TBS-T, and proteins were detected with a chemiluminescence imaging system. Detailed information on the primary antibodies is provided in Table S3. The LINC00173 promoter reporter plasmid, which contains ATF4 binding sites, was constructed by Nanjing GenScript Biotech Co., Ltd. (Nanjing, China). Renilia luciferase activity was used as an internal control. The LINC00173 promoter reporter plasmid and pRL-TK vector were cotransfected into KGN cells at a ratio of 10:1 to 20:1. After 48 h, the cells were stimulated with 100 nM Tg for an additional 6 h. Subsequently, the cells were lysed, and dual luciferase activity was determined by a dual luciferase assay kit (Promega, Madison, WI, USA) according to the manufacturer’s instructions. KGN cells were treated with Tg (100 nM) or DMSO for 6 h. Subsequently, crosslinking was performed with 1% formaldehyde at 37 °C for 10 min. Afterwards, 0.125 M glycine was added to each dish, and the mixture was incubated for 5 min at room temperature to remove unreacted formaldehyde. The cells were then lysed, sonicated, and centrifuged to eliminate insoluble cell debris. The supernatants were immunoprecipitated with an anti-ATF4 antibody, rabbit IgG (negative control), or anti-RNA polymerase II (positive control) using a ChIP kit (Millipore) following the manufacturer’s instructions. DNA was purified using a PCR purification kit (Thermo Fisher). Finally, the promoter region of LINC00173 containing the putative ATF4 binding site was amplified through PCR using purified DNA as a template. The primer sequences can be found in Table S4. The differentially expressed genes in LINC00173-overexpressing KGN cells were identified using RNA-sequencing (RNA-Seq) analysis. Total RNA was extracted and sent to Sangon Technology Co., Ltd., for sequencing (Shanghai, China). FastQC (version 0.11.2) was used for evaluating the quality of the sequenced data. DESeq2 (version 1.12.4) was used to determine differentially expressed genes between two samples. Genes were considered significantly differentially expressed if the q value was ≤ 0.001 and |fold change| was ≥ 2. The cell apoptosis assay was conducted following the standard protocol provided by the manufacturer’s instructions for the PE Annexin V Apoptosis Detection Kit (BD Biosciences, San Jose, CA, USA). Briefly, KGN cells were seeded at a density of 1 × 10 6 cells per well in a 6-well plate. The cells were then collected and diluted to a density of 1 × 10 5 cells using 1× binding buffer. Subsequently, the cells were stained with 5 µL of PE-conjugated Annexin V and 5 µL of 7-AAD for 15 min at room temperature in the dark. Finally, the apoptosis assay was performed using a flow cytometer (BD FACS Melody, USA), and the data were analyzed using Flow Jo software (version 10.8.1, USA). All the statistical analyses were performed using GraphPad Prism 7.0 (GraphPad Prism, San Diego, CA, USA). The data are presented as the mean ± standard deviation (SD) of at least three independent replicates. Significant differences between two groups were determined using an independent-samples t test (unpaired two-tailed t test). Statistical differences among groups were assessed using one-way analysis of variance (ANOVA) with Tukey’s multiple comparison test. P values less than 0.05 were considered to indicate statistical significance (two-tailed).

The gene expression datasets GSE95728 and GSE137684 were downloaded from the GEO database to analyze differentially expressed lincRNAs in GCs from the PCOS and normal groups. In the GSE95728 dataset, a total of 1501 differentially expressed RNAs were identified. Similarly, in the GSE137684 dataset, 470 significantly differentially expressed genes were detected (Fig.  1 a,b). By comparing the GSE95728 and GSE137684 datasets, it was found that 25 differentially expressed RNAs were co-regulated genes. Among these, only LINC00173 and MCF2L-AS1 were identified as differentially expressed lincRNAs (Fig.  1 b). Notably, both LINC00173 and MCF2L-AS1 exhibited high expression levels in the PCOS group in both the GSE95728 and GSE137684 datasets ( P  < 0.05) (Fig.  1 c,d). To further investigate this phenomenon, we performed RT‒qPCR analysis of GCs extracted from the follicular fluids of 50 PCOS patients and 50 normal controls. The results demonstrated that only LINC00173 was highly expressed in the GCs of patients with PCOS ( P   0.05) (Fig.  1 e,f). These findings suggest that LINC00173 may play a role in the development of PCOS. Fig. 1 LINC00173 is highly expressed in granulosa cells from PCOS patients. ( a ) Heatmaps were generated by analyzing the differentially expressed RNAs in GCs from both normal controls and patients with PCOS ( GSE95728 and GSE137684 ). ( b ) Venn diagram showing the overlap of differentially expressed RNAs between the two datasets. ( c , d ) The expression levels of LINC00173 and MCF2L-AS1 in normal controls and patients with PCOS ( GSE95728 : normal, n = 7; PCOS, n = 7), ( GSE137684 : normal, n = 4; PCOS, n = 8). ( e , f ) The expression levels of LINC00173 and MCF2L-AS1 in GCs from both normal controls and patients with PCOS were detected by RT‒qPCR (normal, n = 50; PCOS, n = 50). Statistical analysis was conducted using an independent sample t test, and a significance level of P  < 0.05 was used. GEO refers to the Gene Expression Omnibus, a database managed by the National Center for Biotechnology Information. LINC00173 is highly expressed in granulosa cells from PCOS patients. ( a ) Heatmaps were generated by analyzing the differentially expressed RNAs in GCs from both normal controls and patients with PCOS ( GSE95728 and GSE137684 ). ( b ) Venn diagram showing the overlap of differentially expressed RNAs between the two datasets. ( c , d ) The expression levels of LINC00173 and MCF2L-AS1 in normal controls and patients with PCOS ( GSE95728 : normal, n = 7; PCOS, n = 7), ( GSE137684 : normal, n = 4; PCOS, n = 8). ( e , f ) The expression levels of LINC00173 and MCF2L-AS1 in GCs from both normal controls and patients with PCOS were detected by RT‒qPCR (normal, n = 50; PCOS, n = 50). Statistical analysis was conducted using an independent sample t test, and a significance level of P  < 0.05 was used. GEO refers to the Gene Expression Omnibus, a database managed by the National Center for Biotechnology Information. To evaluate the role of LINC00173 in apoptosis in PCOS, KGN cells were transiently transfected with pcDNA3.1-LINC00173 or siRNA targeting LINC00173 to successfully increase or decrease the expression level of LINC00173, respectively ( P  < 0.05) (Fig.  2 a,b). Cell apoptosis assays demonstrated that overexpression of LINC00173 distinctly promoted the apoptotic capabilities of KGN cells, whereas knockdown of LINC00173 elicited the opposite effect ( P  < 0.05) (Fig.  2 c,d). Furthermore, RT‒qPCR and western blotting revealed that LINC00173 overexpression significantly increased the expression of the pro-apoptotic factor Bax, while LINC00173 knockdown decreased its expression ( P  < 0.05). Conversely, overexpression of LINC00173 decreased the expression level of the antiapoptotic factor Bcl2, while knockdown of LINC00173 increased its expression ( P  < 0.05) (Fig.  2 e–h). These findings indicate that LINC00173 plays a crucial role in enhancing GC apoptosis. Fig. 2 LINC00173 promotes GC apoptosis. ( a , b ) The expression of LINC00173 in KGN cells transfected with pcLINC00173, siLINC00173, vector, or siNC was detected by RT‒qPCR. ( c , d ) The apoptotic rate of KGN cells transfected with pcLINC00173, siLINC00173, vector, or siNC was measured by flow cytometry. ( e–h ) The mRNA and protein levels of Bax and Bcl2 in KGN cells transfected with pcLINC00173, siLINC00173, vector, or siNC were evaluated by RT‒qPCR ( e , f ) and western blotting ( g , h ), and protein expression was quantified by grayscale analysis, original blots/gels are presented in Supplementary Figure S1. The data are presented as the mean ± SD, n = 3. Statistical analysis was performed by independent-sample t test (* P  < 0.05, ** P  < 0.01, *** P   0.05). LINC00173 promotes GC apoptosis. ( a , b ) The expression of LINC00173 in KGN cells transfected with pcLINC00173, siLINC00173, vector, or siNC was detected by RT‒qPCR. ( c , d ) The apoptotic rate of KGN cells transfected with pcLINC00173, siLINC00173, vector, or siNC was measured by flow cytometry. ( e–h ) The mRNA and protein levels of Bax and Bcl2 in KGN cells transfected with pcLINC00173, siLINC00173, vector, or siNC were evaluated by RT‒qPCR ( e , f ) and western blotting ( g , h ), and protein expression was quantified by grayscale analysis, original blots/gels are presented in Supplementary Figure S1. The data are presented as the mean ± SD, n = 3. Statistical analysis was performed by independent-sample t test (* P  < 0.05, ** P  < 0.01, *** P   0.05). During the progression of PCOS, GCs in the follicular microenvironment are exposed to various stress conditions, potentially leading to ER stress. Previous studies have demonstrated that prolonged and severe ER stress can induce apoptosis 25 , 26 . Therefore, we hypothesized that there might be a connection between ER stress and apoptosis in GCs. As depicted in Fig.  3 a–c, the mRNA expression levels of the ER stress markers XBP1 , EIF2AK3 , and CHOP were significantly greater in the GCs of PCOS patients than in those of control patients ( P  < 0.0001), indicating that the GCs of PCOS patients experience ER stress. Furthermore, we found that the expressions of XBP1 , EIF2AK3 , CHOP and LINC00173 were significantly upregulated in KGN cells treated with the ER stress inducer Tg ( P  < 0.05) (Fig.  3 d,e). These findings suggest that LINC00173 may act as a target gene of ER stress to induce GC apoptosis. To identify the specific pathway of the UPR responsible for the expression of LINC00173, we individually knocked down the three ER transmembrane sensors in KGN cells. As shown in Fig.  3 f–h, knockdown of EIF2AK3, but not of ERN1 or ATF6, impaired the expression of LINC00173 induced by Tg. These findings suggest that ER stress induces LINC00173 expression through the PERK pathway. Fig. 3 ER stress is activated in granulosa cells of PCOS patients, and ER stress induces LINC00173 expression in KGN cells. ( a–c ) The expression of ER stress markers (XBP1, EIF2AK3, and CHOP) in GCs from control (n = 50) and PCOS (n = 50) patients was detected by RT‒qPCR. ( d , e ) Cells were pretreated with DMSO or 100 nM Tg for 6 h, and the mRNA levels of ER stress markers (XBP1, EIF2AK3, and CHOP) and LINC00173 were detected by RT‒qPCR. ( f–h ) KGN cells were transfected with siNC, siERN1, siATF6 or siEIF2AK3 and then treated with DMSO or Tg for 6 h. The expression of LINC00173 and ERN1, ATF6 or EIF2AK3 was detected by RT‒qPCR. The data are presented as means ± SD, n = 3. Statistical analysis was performed by one-way ANOVA ( f–h ) (* P  < 0.05, ** P  < 0.01, *** P   0.05). ER stress is activated in granulosa cells of PCOS patients, and ER stress induces LINC00173 expression in KGN cells. ( a–c ) The expression of ER stress markers (XBP1, EIF2AK3, and CHOP) in GCs from control (n = 50) and PCOS (n = 50) patients was detected by RT‒qPCR. ( d , e ) Cells were pretreated with DMSO or 100 nM Tg for 6 h, and the mRNA levels of ER stress markers (XBP1, EIF2AK3, and CHOP) and LINC00173 were detected by RT‒qPCR. ( f–h ) KGN cells were transfected with siNC, siERN1, siATF6 or siEIF2AK3 and then treated with DMSO or Tg for 6 h. The expression of LINC00173 and ERN1, ATF6 or EIF2AK3 was detected by RT‒qPCR. The data are presented as means ± SD, n = 3. Statistical analysis was performed by one-way ANOVA ( f–h ) (* P  < 0.05, ** P  < 0.01, *** P   0.05). Considering that ATF4 is a multifunctional transcription factor, we further analyzed the proximal promoter of LINC00173 for a typical ATF4-binding element. As expected, we identified a potential ATF4-binding element in the proximal promoter of LINC00173 (Fig.  4 a). RT‒qPCR assays confirmed that ATF4 knockdown significantly inhibited the ER stress-induced expression of LINC00173 ( P  < 0.05) (Fig.  4 b). To further validate the binding of ATF4 to the promoter of LINC00173, KGN cells were treated with or without Tg and then subjected to ChIP assays using an anti-ATF4 antibody. The results showed that the binding capacity of ATF4 to the LINC00173 promoter was markedly enhanced in KGN cells treated with Tg (Fig.  4 c). Furthermore, luciferase reporter assays showed that the promoter luciferase activity of LINC00173 was significantly enhanced in KGN cells treated with Tg only in the presence of the ATF4 binding site ( P  < 0.05) (Fig.  4 d). These findings provide evidence that LINC00173 expression is induced by the PERK pathway in an ATF4-dependent manner in response to ER stress. Fig. 4 ATF4 transcriptionally activates LINC00173 expression. ( a ) Schematics showing the putative ATF4-binding sites on the LINC00173 promoter region. ( b ) Cells were transfected with siATF4 or siNC and then treated with Tg for 6 h. LINC00173 expression levels were measured by RT‒qPCR. ( c ) A ChIP assay was performed to demonstrate that ATF4 specifically immunoprecipitated with the LINC00173 promoter region. KGN cells were pretreated with or without Tg for 6 h, followed by immunoprecipitation of the cell lysates using an anti-ATF4 antibody. An anti-RNA polymerase II antibody was used as a positive control, original gels are presented in Supplementary Fig. S2. ( d ) The activity of the wild-type or mutant LINC00173 promoter was assessed in KGN cells by a dual-luciferase reporter assay. All the data are presented as means ± SD, n = 3. Statistical analysis was performed by independent-sample t tests (* P  < 0.05, ** P  < 0.01, *** P   0.05). ATF4 transcriptionally activates LINC00173 expression. ( a ) Schematics showing the putative ATF4-binding sites on the LINC00173 promoter region. ( b ) Cells were transfected with siATF4 or siNC and then treated with Tg for 6 h. LINC00173 expression levels were measured by RT‒qPCR. ( c ) A ChIP assay was performed to demonstrate that ATF4 specifically immunoprecipitated with the LINC00173 promoter region. KGN cells were pretreated with or without Tg for 6 h, followed by immunoprecipitation of the cell lysates using an anti-ATF4 antibody. An anti-RNA polymerase II antibody was used as a positive control, original gels are presented in Supplementary Fig. S2. ( d ) The activity of the wild-type or mutant LINC00173 promoter was assessed in KGN cells by a dual-luciferase reporter assay. All the data are presented as means ± SD, n = 3. Statistical analysis was performed by independent-sample t tests (* P  < 0.05, ** P  < 0.01, *** P   0.05). We further investigated whether LINC00173 mediated ER stress-regulated apoptosis in KGN cells. To achieve this goal, siRNAs targeting LINC00173 were transiently transfected into KGN cells prior to Tg treatment. Treatment of KGN cells with Tg increased the expression of the proapoptotic factor Bax and decreased the expression of the antiapoptotic factor Bcl2, as detected by RT‒qPCR ( P  < 0.05) (Fig.  5 a,b) and western blotting ( P  < 0.05) (Fig.  5 c). Conversely, knockdown of LINC00173 elicited the opposite effects in KGN cells ( P  < 0.05) (Fig.  5 a–c). Importantly, knockdown of LINC00173 significantly suppressed the Tg-triggered upregulation of Bax at both the mRNA and protein levels, while Bcl2 exhibited the opposite trend to that of Bax ( P  < 0.05) (Fig.  5 a–c). Moreover, the Tg-induced apoptosis of KGN cells was also strongly decreased by LINC00173 knockdown ( P  < 0.05) (Fig.  5 d). These findings suggest that LINC00173 mediates the ER stress-induced apoptosis of GCs in PCOS. Fig. 5 LINC00173 mediates ER stress-induced apoptosis. ( a–c ) The mRNA and protein expression levels of Bax and Bcl2 in KGN cells were analysed by RT‒qPCR (a, b) and western blotting (c) after transfection with siNC or siLINC00173 for 48 h and then treatment with Tg (100 nM, 6 h). The protein expression was quantified by grayscale analysis. Original blots/gels are presented in Supplementary Figure S3. ( d ) The apoptotic rate of KGN cells transfected with siNC or siLINC00173 for 48 h and then treated with Tg (100 nM, 6 h) was determined by flow cytometry. All the data are presented as mean ± SD, n = 3. Statistical analysis was performed by one-way ANOVA (* P  < 0.05, ** P  < 0.01, *** P   0.05). LINC00173 mediates ER stress-induced apoptosis. ( a–c ) The mRNA and protein expression levels of Bax and Bcl2 in KGN cells were analysed by RT‒qPCR (a, b) and western blotting (c) after transfection with siNC or siLINC00173 for 48 h and then treatment with Tg (100 nM, 6 h). The protein expression was quantified by grayscale analysis. Original blots/gels are presented in Supplementary Figure S3. ( d ) The apoptotic rate of KGN cells transfected with siNC or siLINC00173 for 48 h and then treated with Tg (100 nM, 6 h) was determined by flow cytometry. All the data are presented as mean ± SD, n = 3. Statistical analysis was performed by one-way ANOVA (* P  < 0.05, ** P  < 0.01, *** P   0.05). To further investigate the potential gene involved in LINC00173 function, we performed RNA sequencing (RNA-seq) assays in KGN cells. The results showed that HRK was significantly upregulated in LINC00173-overexpressing KGN cells (Fig.  6 a). RT‒qPCR and western blotting analysis confirmed that HRK expression in KGN cells was significantly increased by LINC00173 overexpression, while the opposite effect was observed in LINC00173-knockdown KGN cells ( P  < 0.05) (Fig.  6 b,c). Pathway enrichment analysis of the differentially expressed genes identified by RNA sequencing revealed significant alterations in the PI3K/AKT signaling pathway (Fig.  6 d). Previous studies have also demonstrated that the PI3K/AKT pathway can inhibit cell apoptosis 27 . Hence, we selected the HRK and PI3K/AKT pathways for further investigation. Subsequently, western blotting analysis showed that the phosphorylation of PI3K (p-PI3K), AKT (p-AKT), and mTOR (p-mTOR) was significantly decreased in LINC00173-overexpressing KGN cells ( P  < 0.05) (Fig.  6 e). Conversely, knockdown of LINC00173 in KGN cells had the opposite effect ( P  < 0.05) (Fig.  6 f). Importantly, knockdown of HRK reversed the phosphorylation of PI3K, AKT, and mTOR induced by LINC00173 overexpression ( P  < 0.05) (Fig.  6 g). These findings suggest that LINC00173 suppresses the PI3K/AKT/mTOR signaling pathway through upregulation of HRK. Fig. 6 LINC00173 suppresses the PI3K/AKT pathway by upregulating HRK expression. ( a ) Heatmaps were drawn according to the analysis of the differentially expressed mRNAs in KGN cells with normal expression or overexpression of LINC00173 using RNA-seq. ( b , c ) HRK mRNA and protein levels in KGN cells transfected with pcLINC00173, siLINC00173, vector or siNC were detected by RT‒qPCR ( b ) and western blotting ( c ), original blots/gels are presented in Supplementary Fig. S4. ( d ) A bubble plot was used to represent significantly enriched pathways based on KEGG pathway analysis of differentially expressed genes between the control group and the LINC00173 high-expression group. ( e , f ) The protein levels of PI3K, p-PI3K, AKT, p-AKT, mTOR and p-mTOR were measured in KGN cells transfected with pcLINC00173, siLINC00173, vector or siNC for 48 h by western blotting, and protein expression was quantified by grayscale analysis. Original blots/gels are presented in Supplementary Figure S5. The data are presented as mean ± SD. * P  < 0.05 versus vector or siNC. ( g ) The protein levels of PI3K, p-PI3K, AKT, p-AKT, mTOR and p-mTOR were detected in KGN cells cotransfected with vector + siNC, pcLINC00173 + siNC, siHRK + vector or pcLINC00173 + siHRK by western blotting, and the protein expression was quantified by grayscale analysis. Original blots/gels are presented in Supplementary Figure S6. All the data are presented as mean ± SD, n = 3. Statistical analysis was performed by one-way ANOVA ( g ) (* P  < 0.05, ** P  < 0.01, *** P   0.05). LINC00173 suppresses the PI3K/AKT pathway by upregulating HRK expression. ( a ) Heatmaps were drawn according to the analysis of the differentially expressed mRNAs in KGN cells with normal expression or overexpression of LINC00173 using RNA-seq. ( b , c ) HRK mRNA and protein levels in KGN cells transfected with pcLINC00173, siLINC00173, vector or siNC were detected by RT‒qPCR ( b ) and western blotting ( c ), original blots/gels are presented in Supplementary Fig. S4. ( d ) A bubble plot was used to represent significantly enriched pathways based on KEGG pathway analysis of differentially expressed genes between the control group and the LINC00173 high-expression group. ( e , f ) The protein levels of PI3K, p-PI3K, AKT, p-AKT, mTOR and p-mTOR were measured in KGN cells transfected with pcLINC00173, siLINC00173, vector or siNC for 48 h by western blotting, and protein expression was quantified by grayscale analysis. Original blots/gels are presented in Supplementary Figure S5. The data are presented as mean ± SD. * P  < 0.05 versus vector or siNC. ( g ) The protein levels of PI3K, p-PI3K, AKT, p-AKT, mTOR and p-mTOR were detected in KGN cells cotransfected with vector + siNC, pcLINC00173 + siNC, siHRK + vector or pcLINC00173 + siHRK by western blotting, and the protein expression was quantified by grayscale analysis. Original blots/gels are presented in Supplementary Figure S6. All the data are presented as mean ± SD, n = 3. Statistical analysis was performed by one-way ANOVA ( g ) (* P  < 0.05, ** P  < 0.01, *** P   0.05). We further investigated whether HRK mediates the LINC00173-mediated regulation of KGN cell apoptosis. The results showed that overexpression of LINC00173 increased the expression of Bax, and this effect was reversed by HRK knockdown ( P  < 0.05) (Fig.  7 a–c). Additionally, Bcl2 was reduced by LINC00173 overexpression, which was reversed by HRK knockdown ( P  < 0.05). Consistent with these findings, knockdown of HRK reversed the promoting effect of LINC00173 overexpression on KGN cell apoptosis ( P  < 0.05) (Fig.  7 d). These findings suggest that HRK is required for LINC00173-regulated apoptosis of KGN cells in PCOS. Fig. 7 Knockdown of HRK suppresses LINC00173-induced GC apoptosis. ( a , b ) The protein levels of Bax and Bcl2 in KGN cells cotransfected with vector + siNC, pcLINC00173 + siNC, siHRK + vector or pcLINC00173 + siHRK were detected by western blotting. Original blots/gels are presented in Supplementary Figure S7. ( c , d ) The apoptotic rate of KGN cells cotransfected with vector + siNC, pcLINC00173 + siNC, siHRK + vector or pcLINC00173 + siHRK was detected by flow cytometry. All the data are expressed as mean ± SD with (n = 3). Statistical analysis was performed by one-way ANOVA (* P  < 0.05, ** P  < 0.01, *** P   0.05). Knockdown of HRK suppresses LINC00173-induced GC apoptosis. ( a , b ) The protein levels of Bax and Bcl2 in KGN cells cotransfected with vector + siNC, pcLINC00173 + siNC, siHRK + vector or pcLINC00173 + siHRK were detected by western blotting. Original blots/gels are presented in Supplementary Figure S7. ( c , d ) The apoptotic rate of KGN cells cotransfected with vector + siNC, pcLINC00173 + siNC, siHRK + vector or pcLINC00173 + siHRK was detected by flow cytometry. All the data are expressed as mean ± SD with (n = 3). Statistical analysis was performed by one-way ANOVA (* P  < 0.05, ** P  < 0.01, *** P   0.05).

LINC00173 is a novel lncRNA that has been identified as a driver gene in various diseases 28 . Although LINC00173 was recently reported to regulate PCOS progression via functioning as a competing endogenous RNA for miR-124-3p to facilitate JAG1 expression, the functional significance of LINC00173 in PCOS is far from clear 15 . In the present study, we observed that LINC00173 is highly expressed and plays a critical role in inducing apoptosis in the GCs of PCOS patients. Furthermore, LINC00173, which was transactivated by ATF4 in GCs under ER stress, could inhibit the activation of the PI3K/AKT pathway through upregulating the expression of HRK. Our study elucidated the upstream and downstream regulatory mechanisms of LINC00173 in the progression of PCOS, thereby providing a rationale for targeting LINC00173 as a novel therapeutic strategy. Dysregulation of LINC00173 has been detected in various diseases. Yang et al. reported that overexpression of LINC00173 repressed cell proliferation and induced apoptosis in non-small-cell lung cancer via the AGER/NF-kB pathway 29 . Similarly, Zhang et al. demonstrated that overexpression of LINC00173 inhibited the proliferation of cervical cancer cells by regulating FBXW7 expression through its interaction with miR-182-5p 30 . Additionally, Li et al. elaborated that LINC00173 was significantly upregulated and promoted apoptosis by downregulating sphingosine kinase 1 protein expression in pancreatic cancer 31 . Consistently, we also found that LINC00173 was upregulated in the GCs of PCOS patients, and the expression of LINC00173 was necessary and sufficient for inducing apoptosis in GCs. However, the key upstream regulator of LINC00173 and the specific regulatory mechanism by which LINC00173 contributes to the development of PCOS remain to be further investigated. ER stress could be triggered by local hyperandrogenism, inflammation, or oxidative stress within the follicular microenvironment of patients with PCOS 32 . The activation of the UPR pathway has been reported to contribute the pathophysiology of PCOS by disrupting the functions of GCs in various ways 18 , 26 . ER stress triggered the expression of pro-apoptotic factors, particularly CHOP and death receptor 5 (DR5), which were elevated in GCs of patients with PCOS 33 . In the present study, we confirmed that ER stress was indeed activated in the GCs of PCOS patients. Furthermore, the ER stress sensor ATF4 positively regulated the expression of LINC00173 by activating its transcription in the GCs of PCOS patients. Functional experiments demonstrated that silencing of LINC00173 blocked the enhanced apoptotic ability of GCs induced by ER stress, suggesting that LINC00173 was a key target of ER stress in GCs of PCOS. We further performed RNA-seq analysis in LINC00173-overexpressing KGN cells and control KGN cells to explore the potential downstream mechanisms involved. Our results revealed that HRK expression was significantly upregulated in LINC00173-overexpressing KGN cells. HRK, a BH-3 only member of the BCL2 family, promoted apoptosis by binding to and antagonizing the antiapoptotic action of BCL2- and BCL2-like proteins 34 . In this study, we found that knockdown of HRK inhibited apoptosis of KGN cell. Furthermore, pathway enrichment analysis of the differentially expressed genes identified by RNA sequencing revealed significant alterations in the PI3K/AKT signaling pathway. Finally, we confirmed that LINC00173 promoted apoptosis of GCs by targeting HRK-mediated activation of the PI3K/AKT pathway. Several key signaling pathways, especially responded to estrogen, have been reported to be closely involved in the pathogenesis of PCOS, including the PI3K/AKT, NF-κB, Keap1/Nrf2, and AMPK pathways, among others 19 , 35 – 38 . The PI3K/AKT pathway plays an essential role in cell survival and glucose homeostasis, the NF-κB pathway mediates inflammatory responses, the Keap1/Nrf2 pathway functions as a primary antioxidative defense, and the AMPK pathway regulates energy metabolism 39 – 41 . The dysregulation of these pathways jointly contributes to the pathological states involved in PCOS. Targeting these aberrant pathways in PCOS can alleviate inflammation, insulin resistance, androgen excess, and ovarian fibrosis. In the current study, we found that LINC00173 hindered the phosphorylation of PI3K, AKT, and mTOR, leading to the apoptosis of GC in PCOS patients. We speculate that LINC00173 may also influence the function of other signaling pathways indirectly due to the intriguing interactions that occur between each signaling cascade. This inference should be validated by further experiments. There are also still some limitations in this study. The exact molecular mechanism by which LINC00173 regulates HRK expression remains unconfirmed. Considering that the regulatory effects of lncRNAs on gene expression and protein functions primarily depend on their localization and interactions with DNA, RNA, and proteins, additionally experiments are required to investigate where and how LINC00173 controls HRK expression. Furthermore, the potential relationship between LINC00173 and glucose metabolism has also not been fully explored in our current research 42 . These aspects will be addressed in our future investigations. Taken together, our study reveals the upregulation and pro-apoptotic role of LINC00173 in the GCs of PCOS patients. The transcription of LINC00173 is activated by ATF4 in response to ER stress, while LINC00173 inhibits the PI3K/AKT pathway by upregulating the expression of HRK, thereby inducing GC apoptosis in PCOS patients. Our data reveals that LINC00173 promotes GC apoptosis in PCOS patients in response to ER stress through the HRK/PI3K/AKT pathway. These findings suggest that LINC00173 holds promise as a novel biomarker and potential therapeutic target for PCOS.

Polycystic ovary syndrome (PCOS) is a prevalent endocrine disorder and metabolic abnormality 1 , 2 . Approximately 10–13% of women worldwide suffer from PCOS 3 . Although there have been advancements in managing PCOS symptoms, abnormal ovarian folliculogenesis continues to be a major challenge in improving clinical pregnancy outcomes for PCOS patients 4 , 5 . Granulosa cells (GCs) are an important component of the follicular microenvironment and play a critical role in follicle development, oocyte growth and maturation 6 , 7 . A growing body of evidence suggests that the occurrence and progression of PCOS are linked to GC apoptosis 8 . However, the precise underlying mechanism of this association in the stressful follicular microenvironment remains unclear. Long noncoding RNAs (lncRNAs) are a class of noncoding RNA molecules that exceed 200 nucleotides in length and lack protein-coding capabilities 9 . LncRNAs are usually involved in the development of human diseases through various mechanisms, including chromatin modification, interaction with microRNAs, and interaction with proteins 10 . These mechanisms regulate cell development, differentiation, proliferation, and apoptosis 11 . Previous studies have revealed that certain lncRNAs, such as lnc-MAP3K13-7:1, lnc-CCNL1-3:1, and NEAT1, play a critical role in PCOS by regulating the functions of GCs 12 – 14 . However, the clinical potential of these lncRNAs as diagnostic or therapeutic targets in PCOS remains to be verified through large-cohort validation and more intensive experiments. LINC00173 was recently identified as a driver of GC apoptosis by sponging miR-124-3p to facilitate JAG1 expression 15 . However, the key upstream regulator of LINC00173 and the specific regulatory mechanism by which LINC00173 contributes to the development of PCOS remain to be further elucidated. Endoplasmic reticulum (ER) stress is a cellular stress response that occurs when unfolded or misfolded proteins accumulate in the ER lumen 16 . Persistent ER stress triggers a protective cascade known as the unfolded protein response (UPR) through three ER transmembrane sensors—inositol-requiring protein 1α (IRE1α, encoded by ERN1 ), activating transcription factor 6 (ATF6), and PKR-like ER kinase (PERK, encoded by EIF2AK3 )—and activates the following downstream signaling pathways: IRE1α/XBP1, ATF6/CHOP, and PERK/ATF4 17 . In the follicular microenvironment, various factors, such as excessive androgens, inflammation, and oxidative stress, disrupt ER homeostasis and trigger ER stress in ovarian GCs, leading to GC apoptosis 18 . However, the mechanism by which GCs undergo apoptosis in response to ER stress requires further investigation. PCOS is an intricately complex disorder, with multiple signaling pathways involved in its pathophysiological processes. Among these pathways, the phosphoinositide-3 kinase/protein kinase B (PI3K/AKT) pathway is one of the most extensively studied signaling pathways in PCOS and the dysregulation of the PI3K/AKT pathway significantly contributes to the pathogenesis of PCOS, particularly in relation to insulin resistance, elevated androgen levels, and follicular dysplasia 19 . This pathway plays a crucial role in regulating the growth and apoptosis of GCs during follicular development 20 – 22 . Recent studies have reported that the activation of the PI3K/AKT pathway induced by certain drugs can reduce the apoptosis of GCs and consequently improve PCOS symptoms 23 , 24 . However, the relationship between ER stress and the PI3K/AKT pathway in GCs of the PCOS ovary remains poorly understood. In this study, we demonstrated that ER stress induced the transcription of LINC00173 in an ATF4-dependent manner. LINC00173 further induced the expression of Harakiri (HRK), a gene encoding a proapoptotic protein, which is essential for inhibiting the PI3K/AKT pathway in GCs and inducing GC apoptosis in PCOS. The findings from our research will enhance our understanding of the development of PCOS and may aid in identifying potential therapeutic targets for its treatment.

Below is the link to the electronic supplementary material. Supplementary Material 1 Supplementary Material 1 Supplementary Material 2 Supplementary Material 2