1. Introduction
Ovarian cancer (OC) is among the deadliest cancers affecting women worldwide. Although OC accounts for only 3.7% of all female malignancies, it causes 4.7% of cancer-related deaths, and therefore is the most lethal female reproductive cancer [
1]. Approximately 310 000 new OC cases and 200 000 deaths were reported globally in 2020, with the incidence continuing to rise [
2]. Because OC often progresses with few symptoms, nearly two-thirds of patients are diagnosed at advanced or metastatic stages, leading to a five-year survival rate below 30% [
3]. Chemotherapy, particularly with platinum and paclitaxel, remains the primary treatment and aims to damage DNA in OC cells (OCCs). However, its effectiveness is limited by severe side effects, including bone marrow suppression, liver and kidney damage, and chemotherapy resistance, with a poor prognosis and high recurrence rate [
4]. Recently, poly (ADP-ribose) polymerase (PARP) inhibitor treatment has extended survival and recurrence intervals through synthetic lethality in certain patients with OC, particularly breast cancer susceptibility gene 1/2 mutation carriers [
5]. However, as these mutations are present in only 30% of OC patients, most patients lack targeted treatment options [
6], underscoring the urgent need for new therapeutic strategies and reliable therapeutic targets.
Autophagy, a cell survival mechanism invoked under stress conditions, such as starvation, is gaining attention in cancer research. During autophagy, cellular components, including proteins and organelles, are enclosed in autophagosomes that fuse with lysosomes. The cellular components are then broken down and recycled in a process known as the autophagy–lysosome pathway (ALP). While autophagy generally supports cell survival, excessive autophagy leads to autophagic cell death (ACD), a form of programmed cell death characterized by autophagosome accumulation [
7]. Previous studies demonstrated that excessive autophagy can overcome apoptosis resistance in tumor cells to directly trigger cancer cell death after drug resistance emerges [
8]. Autophagy initiation requires autophagy-related gene (
ATG)-encoded proteins, which facilitates autophagosome formation [
9]. Upon activation, microtubule-associated protein 1 light chain 3 (LC3)-I binds to phosphatidylethanolamine on the autophagosome membrane aided by the E1 enzyme ATG7, E2 enzyme ATG3, and E3 ligase ATG12–ATG5–ATG16L, to form LC3-II, a marker of autophagosomes [
10]. However, OCCs have been found to have a notable absence of ATG, including LC3. The use of autophagy selective therapeutics showed that LC3-II accumulation correlated with limited OC progression [
11]. For example, isoliquiritigenin inhibited OCC growth by increasing LC3-II expression, which was reversed by autophagy inhibitors [
12]. Disruptions in the autophagic flux, particularly in the autophagosome–lysosome fusion, can also cause autophagosomes to accumulate. Proteins such as lysosome-associated membrane protein 2 (LAMP2), syntaxin 17 (STX17), and ras-related protein Rab-7 (Rab7) are critical for fusion [
13], and lysosomal dysfunction is a key pathological feature associated with OC progression [
14]. Therefore, blocking autophagosome–lysosome fusion can lead to mitochondrial damage and apoptosis in OCCs [
15].
Sortilin 1 (SORT1), encoded by the
SORT1 gene, is primarily located on the endoplasmic reticulum and lysosomal membranes, where it is involved in protein transport, particularly to lysosomes [
16]. Studies have demonstrated that SORT1 mediates OC progression by enhancing cell proliferation, migration, and invasion [
17]. High SORT1 expression has been detected in over 75% of clinical ovarian tumor samples, and a targeted therapy for SORT1-positive OC patients has entered phase I clinical trials [
18]. Additionally, SORT1 was found to mediate lysosomal degradation of its substrates through autophagy, influencing lysosomal function and the ALP [
19]. Previous studies showed that SORT1 promoted gastric cancer progression by regulating autophagy [
20]. However, its specific role in OC progression and treatment remains unclear.
Herbal medicines have gained attention for their ability to enhance patient quality of life with few side effects, particularly as anti-tumor adjuvants [
21]. Bioactive compounds, such as paclitaxel, camptothecin, and vinblastine, remain a crucial source of anti-tumor drugs.
Salvia miltiorrhiza, a traditional Chinese medicine, contains lipid-soluble tanshinones and water-soluble salvianolic acids that have significant medicinal value. Tanshinones were shown to have significant anti-tumor activity. Notably, tanshinone IIA was shown to activate autophagy in glioblastoma multiforme and myofibroblasts [
22,
23]. Previous studies from our group revealed that dihydrotanshinone I (DHT) inhibited lung metastasis of orthotopic breast cancer in nude mice [
24]. However, the specific mechanisms by which DHT exerted anti-tumor effects in OC remain poorly understood.
Our previous study indicated that DHT induced mitochondrial damage in OCCs by increasing intracellular reactive oxygen species (ROS) levels [
25]. Interestingly, the antioxidant
N-acetyl-
L-cysteine could not fully reverse DHT-induced apoptosis, suggesting that DHT may inhibit OC growth through mechanisms in addition to activating oxidative stress. In this study, we identified potential targets of DHT in OC inhibition through proteomic analysis. We then explored the mechanisms by which DHT inhibited OC growth using an orthotopic OC model, transmission electron microscopy (TEM), co-immunoprecipitation (Co-IP), a cellular thermal shift assay (CETSA), and other techniques. Our findings suggested that targeting SORT1 may be effective against OC. By promoting SORT1 ubiquitination and degradation, DHT disrupted the autophagic flux in OCCs, ultimately leading to ACD.
2. Materials and methods
2.1. Reagents and antibodies
The reagents used in this study were purchased from Chinese suppliers: DHT (≥ 98.0%, PubChem CID: 11425923) was purchased from Yuanye Biotechnology (China). Rapamycin (RAPA, 99.06%, PubChem CID: 5284616), Torin 1 (99.07%, PubChem CID: 49836027), 3-methyladenine (3-MA, 99.97%, PubChem CID: 135398661), hydroxychloroquine (HCQ, 99.93%, PubChem CID: 3652), wortmannin (WORT, 99.97%, PubChem CID: 312145), carbobenzoxyl-L-leucyl-L-leucyl-L-leucinal (MG132, 99.69%, PubChem CID: 462382), quinolyl-valyl-O-methylaspartyl-[-2,6-difluorophenoxy]-methyl ketone (Q-VD-OPh, 99.82%, PubChem CID: 24794416), and benzyloxycarbonyl-Val–Ala–Asp-fluoromethyl ketone (Z-VAD-FMK, 99.79%, PubChem CID: 5497174) were purchased from Selleck Chemicals (USA). Dulbecco’s modified Eagle’s medium (DMEM), McCoy’s 5A medium, Roswell Park Memorial Institute (RPMI) 1640 medium, fetal bovine serum, a penicillin/streptomycin solution, and 0.25% trypsin-ethylenediaminetetraacetic acid (EDTA) were purchased from Gibco (USA). Adenovirus expressing mCherry–GFP–LC3B fusion protein recognizing CD46, Mito-Tracker Green, and Lyso-Tracker Red were purchased from Beyotime Biotechnology (China). Lipofectamine 2000 was purchased from Thermo Fisher (USA). The following antibodies were purchased from Proteintech (China) and used in this study: LC3 (14600-1-AP), LAMP2 (66301-1-Ig), SORT1 (12369-1-AP), ATG3 (11262-2-AP), ATG5 (10181-2-AP), ATG7 (10088-2-AP), ATG16L1 (67943-1-Ig), phosphatase and tensin homolog deleted on chromosome ten (PTEN, 60300-1-Ig), Beclin-1 (11306-1-AP), B-cell lymphoma 2 (BCL-2, 60178-1-Ig), STX17 (17815-1-AP), Rab7 (55469-1-AP), LAMP1 (67300-1-Ig), β-actin (66009-1-Ig), an immunoglobulin G control (IgG, 30000-0-AP), horseradish peroxidase (HRP)-conjugated goat anti-rabbit (SA00001-2), HRP-conjugated goat anti-mouse (SA00001-1), CoraLite 488 goat anti-rabbit (RGAR002), and CoraLite 594 goat anti-mouse (RGAM004). rProtein A/G agarose resin was purchased from Yeasen Biotechnology (China).
2.2. Cell culture
Human OCC lines ES2, OVCAR3, A2780, HO8910PM, SKOV3, and HEY were purchased from Pricella Biotechnology (China). Normal human ovarian epithelial cells, IOSE80, were purchased from WheLab (China). Cells were cultured in DMEM, RPMI 1640, or McCoy’s 5A medium containing 10% fetal bovine serum and 1% penicillin/streptomycin. Cultures were maintained at 37 °C in a humidified atmosphere with 5% CO2. Cells were used for the subsequent experiments were passaged four times. Unless otherwise indicated, cells were grown to 80% confluence followed by overnight starvation in serum-free medium before treatments.
2.3. Animals
Female BALB/c nude mice (4–6 weeks old) were purchased from Shanghai Slack Laboratory Animal Co., Ltd. (SCXK (HU) 2022-0004, China). Mice were housed under specific-pathogen-free conditions with a 12 h light-dark cycle and fed a standard chow diet. All animal experiments were conducted in compliance with the National Institutes of Health (NIH) Guide for the Care and Use of Laboratory Animals (No. 8023, revised 1978). All animal protocols and procedures were approved by the Animal Experimental Research Center of Zhejiang Chinese Medicine University, China (SYXK (ZHE) 2021-0012, ethics approval No. IACUC-20230904-05). Mice were euthanized by CO2 asphyxiation at the indicated time points.
2.4. Cell viability and colony formation assay
Cell viability was assessed using a 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) assay. Briefly, cells were seeded in 96-well plates at 1 × 104 cells∙well–1, and incubated with various compounds at the indicated concentrations for 24 h. Subsequently, 10 µL of MTT reagent was added to each well, and the plates were incubated at 37 °C for 4 h. Formazan crystals were dissolved by adding 100 µL of dimethyl sulfoxide (DMSO) to each well, and absorbance was measured at 490 nm using a microplate reader (Cytation 1; BioTek, USA). For the colony formation assay, OCCs with a SORT1 knockdown short hairpin RNA (shRNA, shSORT1) or scrambled shRNA (negative control) were seeded in six-well plates and treated with or without DHT the following day. After 12 days, cells were fixed with 4% paraformaldehyde for 15 min, washed with phosphate-buffered saline (PBS) and then stained with 0.1% crystal violet for 15 min. Excess stain was removed with distilled water, and the number of cell colonies was counted.
2.5. Proteomic analysis
Ovarian tumor samples with or without DHT intervention were stored at −80 °C (snap-frozen). Samples were sent to Jingjie PTM BioLab Co., Ltd. (China) for proteomic analysis. Four-dimensional label-free quantitative proteomics was performed on a Bruker timsTOF Pro/Pro2 with parallel accumulation-serial fragmentation technology for analysis. Protein identification was performed using Maxquant software (v.1.6.15.0). Differentially expressed proteins (DEPs) were identified based on the criteria p < 0.05 and |fold change| > 1.5.
2.6. Transmission electron microscopy
Cells treated with different compounds were collected and fixed in 2.5% glutaraldehyde at 4 °C for 24 h. The cells were then incubated with a 1% osmium acid and 1.5% K3[Fe(CN)3] solution at 4 °C for 2 h. The cells were then fixed, washed with ddH2O, and stained with 2% uranyl acetate at 37 °C for 2 h. After staining, the cells were dehydrated in graded concentrations of ethanol (70%, 80%, 90%, and 95%), embedded in epoxy resin for 72 h, and sectioned. Images were captured using a transmission electron microscope (H-7650; HITACHI, Japan).
2.7. mCherry–GFP–LC3 puncta assay
OCCs were seeded onto a glass cell culture dish and transfected with mCherry–GFP–LC3B adenovirus (multiplicity of infection (MOI) = 20) when the cell density reached 50%. After 24 h of transfection, the medium was replaced with medium containing various compounds, and the cells were incubated for another 24 h. The autophagic flux was then observed using a confocal microscope (LSM880; Zeiss, Germany). During autophagosome–lysosome fusion, the acidic lysosomal environment quenched GFP fluorescence, while mCherry fluorescence remained stable due to its resistance to acidic conditions. Therefore, by monitoring mCherry–GFP–LC3B expression, the autophagic flux was tracked. When autophagy occurred, mCherry–GFP–LC3B aggregated on the autophagosome membrane and appeared as yellow spots (LC3 puncta, mCherry+GFP+). When the autophagosomes fused with lysosomes, they appeared as red spots (mCherry+GFP−) due to partial quenching of GFP fluorescence.
2.8. Immunofluorescence
OCCs were seeded onto glass cell culture dishes and treated with various compounds. After treatment, cells were washed with PBS and fixed with paraformaldehyde for 15 min. The cells were then permeabilized with 0.4% Triton X-100 and blocked with 5% goat serum for 30 min, and incubated overnight at 4 °C with primary antibodies. The next day, the cells were incubated with secondary antibodies at 37 °C for 1 h. After staining the nuclei with 4′,6-diamidino-2-phenylindole (DAPI) for 10 min, the stained cells were visualized using a confocal microscope. For Mito-Tracker and Lyso-Tracker staining, the treated cells were washed with PBS and incubated in medium containing Mito-Tracker Green (100 nmol∙L–1) and Lyso-Tracker Red (75 nmol∙L–1) for 45 min. The stained cells were visualized using a confocal microscope to assess mitochondrial and lysosomal colocalization.
2.9. Real-time quantitative polymerase chain reaction (qPCR)
Total RNA was extracted from ES2 cells using TRIzol reagent (Vazyme Biotech Co., Ltd., China). Reverse transcription was performed using the HiScript IV All-in-One Ultra RT SuperMix for qPCR (Vazyme Biotech Co.,Ltd.) following the manufacturer’s protocol. qPCR was performed using Taq Pro Universal SYBR qPCR master mix (Vazyme Biotech Co., Ltd.) on an Applied Biosystems 7500 real-time PCR system (Thermo Fisher). The gene expression was determined using the 2−△△CT method. Glyceraldehyde-3-phosphate dehydrogenase (GAPDH) expression was used as an internal standard. Primer sequences are provided in Table S1 in Appendix A.
2.10. Western blotting
The OCCs were lysed using radioimmunoprecipitation assay (RIPA) buffer (Beyotime Biotechnology) containing 1 mmol∙L
–1 phenylmethylsulfonyl fluoride (PMSF; Solarbio, China) for 20 min on ice. The lysate was centrifuged at 14 000
g and 4 °C for 10 min, and the supernatant was collected. The supernatant protein concentration was measured using a bicinchoninic acid (BCA) protein assay kit (Yuanye Biotechnology). Western blotting was performed following a standard protocol, as previously described [
26].
2.11. Co-IP assay
Cells were lysed with Pierce IP lysis buffer (Thermo Fisher) supplemented with 1 mmol∙L–1 PMSF. The supernatant was incubated with anti-SORT1 antibody or IgG control at 4 °C overnight. The next day, the samples were mixed with rProtein A/G agarose resin and incubated at room temperature for 2 h. After incubation, the resin was washed with IP lysis buffer and collected by centrifugation at 1500g for 10 min. The immunoprecipitated protein was subjected to sodium dodecyl-sulfate polyacrylamide gel electrophoresis (SDS-PAGE) with the indicated primary antibodies.
2.12. Cellular thermal shift assay
The interaction between DHT and target proteins in ES2 cells was analyzed by CETSA as previously described [
27]. Briefly, ES2 cells were incubated with or without DHT (350 μmol∙L
–1) for 2 h. Then, the cells were washed with PBS, collected, and subjected to CETSA. Pairs of samples consisting of a control sample and a DHT-treated sample were heated for 5 min at different temperatures (43, 46, 49, 52, 55, 58, 61, 64, 67, and 70 °C). After heating, the samples were flash-frozen in liquid nitrogen for 30 s. This procedure was performed three times. The cell lysate was centrifuged at 14 000
g and 4 °C for 10 min, and the supernatant was collected. The protein stability at various temperatures was assessed via Western blotting. A dose-dependent CETSA was also performed, in which cells were treated with various concentrations of DHT (0, 50, 150, 200, 250, 300, 350, 400, 450, and 500 μmol∙L
–1) for 2 h. Cells were then washed with PBS and collected. Samples were heated at 64 °C for 5 min, and the remaining steps were performed as described earlier. Western blot was used to analyze the degradation of specific proteins at different DHT concentrations.
2.13. Lentiviral transduction
A lentivirus carrying the luciferase gene with a puromycin selection marker was provided by GeneChem (China). SORT1-silenced OCCs were generated by lentiviral infection. HEK293T cells were co-transfected with lentiviral packaging plasmids (pCMV-Gag-Pol and pCMV-VSVG) and either shSORT1 plasmids (neomycin selection) or control shRNA plasmids to produce lentiviral particles (oligo sequences shown in Table S2 in Appendix A). After 48–72 h of transfection, the cell supernatant containing the lentiviruses was collected and filtered through a 0.45 μm filter. OCCs were infected with the lentiviruses in the presence of polybrene. After 48 h, cells were selected using puromycin (1 μg∙mL–1) or G418 (1 mg∙mL–1).
2.14. RNA interference
A small interfering RNA (siRNA) targeting ATG16L1 was synthesized by Haixing Biosciences (China). The siRNA sequences are provided in Table S3 in Appendix A. siRNA transfection into OCCs was performed using Lipofectamine 2000, following the manufacturer’s instructions. The OCCs were used 60 h post-transfection for experiments.
2.15. Subcutaneous xenograft model
A subcutaneous xenograft model was established as described previously [
25]. Briefly, ES2 cells (10
6 cells∙mice
–1,
n = 7) and ES2-sh
SORT1 or ES2-shNC cells (10
6 cells∙mice
–1,
n = 8) were harvested, resuspended in serum-free medium, and inoculated in the right armpit of nude mice. For the proteomic study evaluating DHT intervention in OC, DHT was dissolved in a solvent mixture of 35% DMSO, 50% PEG400, and 15% saline. Two days before the mice were inoculated with tumor cells, they were randomly divided into three groups: tumor control (TC), DHT (20 mg∙kg
–1, once a day on a 2-day-on/1-day-off schedule, intraperitoneally (i.p.)), and carboplatin (CBP, 12.5 mg∙kg
–1, twice a week, i.p.). The TC group was administered saline (i.p.). At the end of the experiment (18 days post-injection), mice were sacrificed under anesthesia. To study the effect of SORT1 on OC growth, mice were randomly divided into three groups: ES2-shNC, ES2-sh
SORT1, and ES2-sh
SORT1 + DHT (20 mg∙kg
–1, once a day on a 2-day-on/1-day-off schedule, i.p.). The ES2-shNC and ES2-sh
SORT1 groups received saline (i.p.). At the end of the experiment (21 days post-injection), the mice were sacrificed under anesthesia. Tumor volumes were measured using calipers and calculated as follows: tumor volume = 0.5 × length × width
2. Tumors were quickly dissected and either frozen at −80 °C or fixed in 4% paraformaldehyde for pathological analysis.
2.16. Orthotopic ovarian cancer model
An orthotopic OC model was established according to a previous study [
28]. Female mice were anesthetized by inhalation of isoflurane (5% in oxygen) in an induction chamber, and anesthesia was maintained with 2.5%–3.0% isoflurane during the procedure using a nosecone. Mice were subcutaneously injected with carprofen (5 mg∙kg
–1) before surgery. A small dorsomedial incision was made above the ovarian fat pad, followed by a secondary incision through the peritoneal wall. The ovarian fat pad was externalized and stabilized with a bull clip, and microforceps were used to locate the oviduct. ES2-luciferase cells (1.5 × 10
6, 15 μL) were injected under the left ovarian bursa. The peritoneal wall was sutured closed using 6/0 sutures, bupivacaine was applied topically, and the skin incision was closed with surgical staples. Six days after tumor cell inoculation, mice were randomly divided into four groups: TC, DHT (20 mg∙kg
–1,
quaque omni die (qod), i.p.), RAPA (6 mg∙kg
–1, qod, i.p.), and DHT + RAPA ((20 + 6) mg∙kg
–1, qod, i.p.). DHT and RAPA were dissolved in a mixed solvent containing 35% DMSO, 50% PEG400, and 15% saline. The TC group received saline (i.p.). At the end of the experiment (25 days post-injection), mice were sacrificed under anesthesia. Tumor progression and metastases were monitored using bioluminescence imaging on days 6, 12, 18, and 24. Mice were injected i.p. with
D-luciferin (150 mg∙kg
–1 body weight) after anesthesia and were imaged 5–30 min later using an
in vivo imaging system (IVIS-200; PerkinElmer, USA). Imaging data were analyzed with Living Image software (Caliper LifeSciences, USA). The tumors were weighed, photographed, and processed for subsequent analyses.
2.17. Hematoxylin and eosin (H&E) staining and immunohistochemistry (IHC) analysis
Tumor tissue was fixed in 10% paraformaldehyde overnight, then dehydrated, embedded in paraffin, sectioned, and stained with H&E according to a standard H&E procedure. To prepare tissue for IHC analysis, tissue slides were incubated with 0.1 mol∙L–1 citrate buffer for 30 min, followed by 3% H2O2 treatment for 15 min to quench the endogenous peroxidase activity. After blocking with 10% goat serum for 1 h at room temperature, the slides were incubated with primary antibodies overnight at 4 °C. The tissue slides were then incubated with secondary antibodies for 1 h at 25 °C, followed by 3,3'-diaminobenzidine (DAB) staining and hematoxylin counterstaining. All images were visualized using a microscope (Zeiss AXIO SCOPE A1). Protein expression was analyzed using Image Pro Plus software (Media Cybernetics, USA).
2.18. Statistical analysis
All data are represented as mean ± standard deviation (SD) from at least three independent experiments. Statistical analyses were performed using IBM SPSS Statistics 22.0 software (SPSS Inc., USA) and GraphPad Prism 10.0 (GraphPad Software Inc., USA). Student’s t test was used to analyze differences between two groups. One-way analysis of variance (ANOVA) was used to compare multiple sets of quantitative data. p < 0.05 was considered statistically significant.
3. Results
3.1. DHT inhibition of OC involves the ALP
Two OCC lines, ES2 and OVCAR3, were used to investigate the pharmacological action of DHT on OC. After DHT treatment, colony formation assays showed a marked reduction in clonogenic survival in both cell lines (
Figs. 1(a)–(c)). In prior studies that screened anti-tumor compounds in
Salvia miltiorrhiza, we found that DHT significantly inhibited the growth of OC xenografts (Figs. S1(a) and (b) in Appendix A) and induced mitochondrial depolarization and apoptosis in OCCs [
25]. To further determine the role of DHT in OC, we performed a proteomic analysis on OC xenografts treated with DHT. We identified 373 DEPs (proteomic information is shown in Table S4 in Appendix A) and found distinct clustering between the DHT-treated and control groups using projection latent structure discriminant analysis (PLS-DA) (
Fig. 1(d)). A volcano plot revealed that 203 proteins were downregulated and 170 proteins were upregulated in the DHT treatment group (Fig. S1(c) in Appendix A). A bioinformatics analysis of the DEPs identified changes in key pathways, with the lysosome (hsa04142) and phagosome (hsa04145) pathways prominently enriched among the top ten Kyoto Encyclopedia of Genes and Genomes (KEGG) pathways ranked by
p-value (
Fig. 1(e)). The results of a gene set enrichment analysis (GSEA) confirmed that following DHT administration, the lysosome pathway was significantly inhibited (
Fig. 1(f)). Fatty acid metabolism activation is also recognized as a characteristic feature of autophagy [
29]. The GSEA results also indicated that the fatty acid metabolism pathway was activated (
Fig. 1(g)). Consistently, a gene ontology (GO) biological process analysis of the upregulated proteins identified significant activation of pathways involving tricarboxylic acid (TCA) metabolic processes, lipid catabolism, and lipid oxidation (
Fig. 1(h)). These findings collectively indicated that abnormalities of the ALP may be involved in DHT-mediated OC growth inhibition. A heatmap showed that 15 significantly downregulated proteins were enriched in the lysosome pathway (
Fig. 1(i)). We analyzed the DEPs in the phagosome and lysosome pathways using a protein–protein interaction network to further identify key targets causing autophagic flux abnormalities in OCCs (Figs. S1(d) and (e) in Appendix A). Among these, SORT1 and LAMP2 attracted our attention, and particularly SORT1, as a Human Protein Atlas (HPA) analysis showed high SORT1 expression in over half of OC patient tumors (
Fig. 1(j), Fig. S1(f) in Appendix A). The Cancer Genome Atlas (TCGA) also confirmed elevated
SORT1 gene expression in OC patients (
Fig. 1(k)), with high correlations between
SORT1 and
LAMP2 expression in OC patients but not in the normal population (
Figs. 1(l) and
(m)). Consistent with the publicly available data, a multivariate correlation analysis of the proteomic data indicated a significant positive correlation between the DEPs in the lysosome and phagosome pathways (Fig. S1(g) in Appendix A). Furthermore, a Kaplan–Meier survival analysis showed that patients with high
SORT1 expression had significantly lower overall survival (OS) and progression-free survival (PFS) (
Figs. 1(n) and
(o)). Together, these findings underscore a strong association between SORT1 expression and the impact of DHT on the autophagy pathway in OCCs.
3.2. DHT induces autophagosome accumulation in OCCs by disrupting the autophagic flux
To examine the effect of DHT on the autophagic flux in OCCs, we evaluated the autophagosome formation and referred to previous studies identifying double-membraned autophagosomes via TEM [
30]. ES2 cells treated with 1.8, 3.5, and 7.0 μmol∙L
–1 of DHT had an increased number of autophagosomes (
Fig. 2(a)). A similar trend was observed in A2780 cells (Fig. S2(a) in Appendix A). As expected, the accumulation of autophagosomes was accompanied by a marked increase in the autophagosome marker LC3. Our results showed that DHT treatment promoted the LC3-I to lipidated LC3-II conversion in OCCs (
Fig. 2(b)). Autophagosome-specific monodansylcadaverine (MDC) staining further confirmed this accumulation, which aligned with the TEM findings (Fig. S2(b) in Appendix A). The entire process of autophagy involves autophagosome formation, autolysosome development, and degradation. We transfected OCCs with an mCherry–GFP–LC3B adenovirus to monitor the complete autophagic flux process. ES2 cells treated with DHT displayed yellow puncta (indicating mCherry+GFP+ labeling of autophagosomes), whereas cells treated with the autophagy activator RAPA showed red puncta (indicating mCherry+GFP− labeling of autolysosomes). The phosphatidylinositol 3-kinase (PI3K) inhibitor 3-MA prominently counteracted mCherry+GFP+ signaling in DHT-treated ES2 cells (
Figs. 2(c) and
(d)), suggesting that DHT impaired the autophagic flux in OCCs. Similar results were observed in A2780 cells (Figs. S2(c) and (d) in Appendix A). A further analysis using Mito-Tracker and Lyso-Tracker staining revealed fragmented mitochondria in ES2 cells treated with DHT, contrasting with the normal tubular mitochondrial network in control cells. Treatment with 3-MA notably reversed DHT-induced mitochondrial autophagy in ES2 cells (
Fig. 2(e)). In comparison, RAPA treatment showed enhanced colocalized Mito-Tracker and Lyso-Tracker signals. However, DHT significantly reduced Lyso-Tracker signals and disrupted colocalization of damaged mitochondria and lysosomes in ES2 cells compared with both RAPA-treated and control cells (
Figs. 2(e) and
(f)). Similar findings were validated in A2780 cells (Figs. S2(e) and (f) in Appendix A). Additionally, immunoblotting showed that DHT reduced LAMP2 expression in a dose-dependent manner (
Fig. 2(g)). Together, these results indicated that DHT promoted autophagy initiation in OCCs, but blocked the autophagic flux by impairing autolysosome biogenesis. The disrupted mitochondrial and lysosomal colocalization further supported our hypothesis that DHT both altered the mitochondrial structure and inhibited the lysosomal degradation of damaged mitochondria.
3.3. DHT inhibits orthotopic ovarian tumor growth by inducing ACD
Given that DHT caused autophagosomes to accumulate in OCCs, we applied autophagy inhibitors and activators to further investigate the effects of DHT. Our results showed that 3-MA and WORT did not significantly inhibit OCC growth at low concentrations, whereas RAPA and Torin 1 effectively suppressed growth (Figs. S3(a)–(c) in Appendix A). Notably, cotreatment with DHT and HCQ (a late-stage autophagy inhibitor) induced more cell death than treatment with DHT alone (Fig. S3(d) in Appendix A). However, 3-MA and WORT significantly reversed the inhibitory effect of DHT on OCC proliferation (Figs. S3(e) and (f) in Appendix A). Additionally, the apoptosis inhibitors Q-VD-OPh and Z-VAD-FMK failed to protect against DHT-induced cell death in ES2 cells (Fig. S4 in Appendix A), indicating that DHT inhibited OCC growth in an apoptosis-independent manner. These findings suggested that DHT may cause ACD by inducing autophagosome accumulation in OCCs. Therefore, we investigated whether DHT triggered ACD in orthotopic ovarian tumors. We assessed the effects in ES2 orthotopic tumors using the following treatments: TC, DHT (20 mg∙kg
–1, qod, i.p.), RAPA (6 mg∙kg
–1, qod, i.p.), and DHT and RAPA (combined) at the same dosages (
Fig. 3(a)). Both DHT and RAPA significantly slowed tumor growth (
Fig. 3(b)), reduced tumor nodule formation (
Fig. 3(c)), and decreased the tumor weight (
Fig. 3(d)). Notably, DHT and RAPA combination therapy did not enhance the anti-tumor effect beyond that of either monotherapy, suggesting that DHT may inhibit OC growth through autophagy activation alone. DHT was well tolerated, with minor weight loss that was attributed to the reduced tumor burden (
Fig. 3(e)). Live imaging based on ES2-luciferase cells further supported the above conclusions (
Figs. 3(f) and
(g)). A histological analysis using H&E and IHC staining (
Fig. 3(h)) showed that both DHT and RAPA treatments caused cells to exhibit abnormal morphology, including nuclear condensation, vacuole formation, and cell dispersion. Additionally, DHT significantly decreased SORT1 expression in OC to levels close to those in normal ovarian tissue. To confirm that DHT inhibited OC growth by inducing ACD, we examined ALP protein expression in xenograft tumors. Both DHT and RAPA reduced the expression of the proliferation marker Ki67 in OCCs. As expected, DHT treatment resulted in stronger ATG5, ATG16L1, LC3, and PTEN staining of tumor tissue sections than did TC treatment (Fig. S5 in Appendix A). DHT also significantly reduced the expression of LAMP2, STX17, and Rab7 in tumor tissue, but RAPA did not (Fig. S6 in Appendix A). This difference may explain why DHT exhibited stronger anti-tumor effects than RAPA (
Fig. 3(d)), supporting our earlier findings that DHT disrupted the autophagic flux in OCCs. In summary, these results confirmed that excessive autophagy activation inhibited OC growth and that DHT exerted its anti-cancer effects by promoting ACD.
3.4. SORT1 knockdown induces ACD in OCCs
Building on our findings that SORT1 may be involved in DHT-induced autophagic flux disruption, we investigated whether
SORT1 deficiency alone caused autophagosomes to accumulate in OCCs. First, we assessed SORT1 expression across six types of OCCs. Compared with normal human ovarian epithelial cells (IOSE80), SORT1 was highly expressed in most of the OCCs evaluated, particularly ES2 and A2780 cells (
Figs. 4(a) and
(b)). To verify the role of SORT1 in OCC autophagy, we constructed a
SORT1 knockdown cell line by transfecting OCCs with a sh
SORT1 plasmid packaged in a lentivirus. Both sh
SORT1-1 and sh
SORT1-2 constructs significantly reduced SORT1 expression in ES2 and A2780 cells (Fig. S7 in Appendix A). MDC staining indicated a marked increase in labeled autophagosomes in ES2/A2780-sh
SORT1 cells compared with control cells (
Fig. 4(c)). A colony formation assay demonstrated that SORT1 depletion significantly inhibited cell growth in ES2 and A2780 cells (
Fig. 4(d), Fig. S8 in Appendix A). Notably,
SORT1 knockdown did not alter the expression of ATGs, including
ATG3,
ATG5,
ATG7,
ATG16L1,
BECN1, and
LC3B (
Fig. 4(e)). However, TEM imaging of ES2-sh
SORT1 cells revealed a substantial increase in the number of autophagosomes (
Fig. 4(f)). This strongly suggested that SORT1, which is overexpressed in OCCs, inhibited autophagosome formation. Treatment with 3-MA reduced the number of autophagosomes in ES2-sh
SORT1 cells, whereas HCQ did not significantly affect on autophagosome accumulation caused by
SORT1 deficiency (
Fig. 4(f)). Building on our previous discovery of the DHT-induced disruption of autophagic flux and the inhibitory effect of SORT1, we further investigated the potential underlying causes of autophagosome accumulation. Our findings revealed that after DHT treatment, the endogenous LC3 levels in OCCs significantly increased. However, compared with the LC3 and LAMP1 (a lysosome marker) colocalization induced by RAPA treatment, no significant colocalization (orange puncta) of LC3 and LAMP1 was observed in the DHT treatment group (
Fig. 4(g)). This suggested that early autophagy activation combined with the dysfunctional autophagosome–lysosome fusion mediated the accumulation of autophagosomes caused by SORT1 deficiency. Collectively, these results confirmed that DHT-induced ACD in OCCs was mediated by SORT1.
3.5. SORT1 depletion stimulates ACD via ATG family proteins in OCCs
Given that
SORT1 knockdown induced autophagosomes to accumulate in OCCs but did not significantly affect autophagosome-related gene expression, we hypothesized that SORT1 regulated ACD through protein–protein interactions. The known autophagy regulators, including ATG3, ATG7, and the ATG5–ATG12–ATG16L1 complex, are essential for LC3 lipidation and autophagosome maturation, in which the phagophore closes to form an autophagosome [
31]. Western blotting revealed that DHT treatment upregulated ATG5 and ATG16L1 and downregulated SORT1 in ES2 and A2780 cells (
Figs. 5(a)–(d)). To rule out other factors affecting ATG5 and ATG16L1 levels, an examination of the PI3K–protein kinase B (AKT)–mechanistic target of rapamycin (mTOR) pathway showed no significant changes in PI3K, AKT, p-AKT, mTOR, or p-mTOR levels after DHT treatment (Figs. S9(a) and (b) in Appendix A). Additionally, a qPCR analysis confirmed that DHT did not alter the mRNA levels of
SORT1,
ATG5, or
ATG16L1 (Figs. S9(c) and (d) in Appendix A), indicating that the reduction in SORT1 was not due to transcriptional changes. SORT1 degradation is mainly mediated by the ubiquitin–proteasome pathway; therefore, to explore the mechanism by which DHT downregulated SORT1, we assessed SORT1 ubiquitination [
32]. Co-IP assays showed both increased SORT1 ubiquitination (upward shifted ubiquitinated bands were observed) and decreased SORT1 protein levels following DHT treatment (
Fig. 5(e)). Excitingly, a Co-IP analysis further revealed interactions between SORT1 and both ATG5 and ATG16L1. DHT suppressed SORT1 expression, which reduced the levels of ATG5 and ATG16L1 bound to SORT1, while the unbound levels of ATG5 and ATG16L1 significantly increased (
Fig. 5(f)). These results suggested that DHT indirectly decreased binding between ATG and SORT1 by inhibiting SORT1 expression and that SORT1 negatively regulated endogenous ATG5 and ATG16L1 through direct interactions. Unexpectedly, co-treatment with the proteasome inhibitor MG132 did not prevent the effects of DHT on SORT1 or restore LC3-II levels (
Fig. 5(g), Fig. S10 in Appendix A), suggesting that alternative pathways may also regulate SORT1 degradation. Nonetheless, the above evidence sufficiently supported that DHT-mediated SORT1 depletion promoted autophagosome formation.
To confirm the role of SORT1 in autophagy, we examined ALP proteins in ES2 cells after
SORT1 knockdown or DHT treatment (
Fig. 5(h), Figs. S11(a) and (b) in Appendix A). The results showed elevated levels of ATG5, ATG16L1, and LC3-II in ES2-sh
SORT1 cells. Notably, PTEN levels were unaffected by
SORT1 knockdown. However, DHT treatment increased PTEN expression. Additionally,
SORT1 knockdown increased Beclin1 and decreased BCL-2 expression, thus affecting the Beclin1/BCL-2 complex, which regulates autophagy levels. Interestingly, we found that LAMP2 and STX17, which are both essential for autophagosome–lysosome fusion, were significantly downregulated in ES2-sh
SORT1 cells, whereas Rab7 levels remained unchanged. This suggested that SORT1 participated in the DHT-induced autophagic flux blockade. Furthermore,
ATG16L1 knockdown partially reversed the increased LC3-II/I ratio in ES2-sh
SORT1 cells (
Fig. 5(i), Fig. S11(c) in Appendix A). Additionally, 3-MA and WORT treatments promoted ES2-sh
SORT1 cell proliferation, while HCQ attenuated cell proliferation (
Fig. 5(j)). This indicated that excessive autophagosome accumulation, rather than other effects of
SORT1 knockdown, led to cell death in OCCs. DHT and RAPA had little inhibitory effect on ES2-sh
SORT1 cells, further supporting this conclusion (
Fig. 5(k)). Indeed, knocking down
ATG16L1 to suppress sh
SORT1-induced autophagy alleviated sh
SORT1-induced cytotoxicity in OCCs (
Fig. 5(l), Fig. S12 in Appendix A). A similar tendency was observed in the wound healing ability of ES2-shNC and ES2-sh
SORT1 cells (Fig. S13 in Appendix A).
ATG16L1 knockdown partially reversed the inhibited migration of OCCs induced by sh
SORT1. Notably, DHT did not further inhibit ES2-sh
SORT1 cell migration. These findings strongly suggested that SORT1 promoted autophagosome accumulation and induced ACD in OCCs through interactions with
ATG-encoded proteins.
3.6. DHT interacts with SORT1 through its carbonyl group
Because DHT did not affect
SORT1 mRNA expression, we hypothesized a direct interaction between DHT and SORT1. To test this, we conducted a CETSA, an assay based on the principle that ligand binding enhances the target protein thermal stability [
33]. OCCs treated with increasing doses of DHT exhibited increased SORT1 levels after heat treatment at 63 °C, with significantly reduced SORT1 degradation at a concentration of 350 μmol∙L
–1 (
Figs. 6(a) and
(b)). To explore the binding mode between DHT and SORT1, we used molecular docking to predict the interaction site. The specific analysis method is described in detail in a previous study [
24]. An Autodock (Scripps Research Institute, USA) analysis indicated that DHT bound to SORT1 in a pocket containing the amino acid residues TYR189 and MET187, with a binding energy of −6.15 kcal∙mol
–1. A LigPlot+ (EMBL-EBI, UK) analysis further revealed that the carbonyl group of DHT interacted with TYR189 of SORT1 (
Fig. 6(c)). This interaction was further validated by temperature-dependent CETSA (
Fig. 6(d)). The SORT1 band of ES2 and A2780 cells treated with DMSO nearly disappeared between 61 and 70 °C, whereas it remained stable in cells treated with 350 μmol∙L
–1 DHT (
Figs. 6(e) and
(f)). The thermal melting curves showed a distinct shift in SORT1 stability in DHT-treated ES2 and A2780 cells, while β-actin stability remained unchanged (
Figs. 6(g) and
(h)). These results suggested that DHT may directly bind to SORT1. However, the specific mechanism by which DHT bound to SORT1 and inhibited its expression warrants further investigation.
3.7. SORT1 depletion inhibits OC growth in a xenograft model by disrupting the autophagic flux
The results indicated that DHT affected key proteins in the ALP, including ATG5 and ATG16L1, by inhibiting SORT1 expression, which ultimately disrupted the autophagic flux and induced ACD. To evaluate the therapeutic potential of SORT1 inhibition in OC, we established a xenograft model by subcutaneously injecting mice with ES2-sh
SORT1 cells.
SORT1 knockdown significantly reduced the size and weight of the OC xenografts, and DHT had no additional inhibitory effect on ES2-sh
SORT1 xenograft growth (
Figs. 7(a)–(c)). H&E staining of ES2-shNC tumor sections showed densely packed cells with a regular morphology, large size, and high nucleoplasm ratios. In contrast, ES2-sh
SORT1 tumor sections had more loosely arranged cells with abnormal morphology (
Fig. 7(d)). An IHC analysis further revealed a marked decrease in Ki67-positive cells in tumors derived from
SORT1-depleted ES2 cells compared with the control xenografts. Consistent with our
in vitro findings, ES2-sh
SORT1 xenografts had higher ATG5 and ATG16L1 levels (
Figs. 7(d) and
(e)) and lower LAMP2 and STX17 levels compared with ES2-shNC xenografts (
Figs. 7(f) and
(g)). DHT treatment did not alter levels of these proteins in ES2-sh
SORT1 xenografts, suggesting that DHT-regulated autophagy of OCCs is dependent on SORT1. In summary, these results demonstrated that SORT1 depletion disrupted the autophagic flux, thereby inducing ACD and inhibiting OC growth
in vivo.
4. Discussion
OC is a prevalent female reproductive system malignancy characterized by a high mortality rate and vague early symptoms, earning it the moniker “the silent killer.” Most OC cases are diagnosed at advanced stages, often with metastasis, limiting the effectiveness of surgical treatments [
34]. Increasingly more efficient autophagy-based therapeutic agents and protocols for treating cancer have been proposed. Certain nanomedicines that induce autophagosome formation have shown exciting anti-tumor effects [
35]. Therefore, inducing ACD by disrupting the autophagic flux is a promising treatment strategy for OC. In this study, we investigated whether the ALP influenced DHT inhibition of OC. Our primary finding showed that DHT disrupted the autophagic flux in OCCs by inhibiting SORT1 expression, causing autophagosomes to accumulate with subsequent ACD. Notably, our study highlighted the previously unreported role of SORT1 in OCC autophagy, in which it interacted with the autophagy-related proteins ATG5 and ATG16L1.
Proteomic and bioinformatic analyses revealed that DHT altered the expression of proteins that were significantly enriched in the phagosome and lysosome pathways. GSEA and GO analyses further demonstrated that DHT inhibited the lysosome pathway in OC while overactivating energy metabolism pathways, such as fatty acid metabolism and the TCA cycle. Several factors may explain this finding. First, increased lipid metabolism and TCA cycle activity in cancer cells are often linked to increased autophagy [
36,
37]. Second, most cancer cells exhibit the Warburg effect, in which an overreliance on the TCA cycle may deplete intermediate metabolites (such as nucleotides) in the glycolytic pathway, thus limiting DNA and protein synthesis [
38]. Third, increased fatty acid metabolism and lipid oxidation cause ROS levels to rise. Our previous studies indicated that DHT induced oxidative stress in OC [
25], potentially by regulating autophagy. Thus, we hypothesized that DHT disrupted the autophagic flux in OCCs. Specifically, DHT may activate autophagy in OCCs while simultaneously disrupting the fusion of autophagosomes and lysosomes. Next, we focused on key DEPs within the ALP. The four proteins with the highest expression changes in the lysosome pathway were β-glucuronidase,
N-acetylgalactosamine-6-sulfatase, cathepsin C, and lysosomal α-mannosidase. These proteins are lysosomal enzymes that primarily degrade cysteine, glycosaminoglycans, and certain oligosaccharides within cells and do not directly influence autophagy. SORT1 and LAMP2 ranked fifth and sixth, respectively, in terms of expression changes. SORT1 is primarily located on the endoplasmic reticulum, lysosomes, and cell membrane, where it facilitates intracellular protein transport [
16]. It was discovered in 2018 that SORT1 mediated lysosomal degradation of its substrate proteins through autophagy [
19], and subsequent studies have reported its regulation of autophagy [
39]. However, the specific mechanism by which SORT1 regulates autophagy in cancer cells remains unclear. A recent study suggested that SORT1, regulated by circular RNA (hsa_circ_0110389), may promote gastric cancer progression through abnormal autophagy [
20]. Our study showed that although SORT1 was not a central protein in the protein–protein interaction network, public databases such as HPA and TCGA indicate high SORT1 expression in OC patients at both the gene and protein levels, suggesting its significance in OC progression. Notably, the expression of SORT1 and LAMP2 was shown to be strongly and positively correlated in OC patients but not in healthy individuals. LAMP2, a highly glycosylated protein on the lysosomal membrane, stabilizes the lysosomal membrane and promotes autophagosome–lysosome fusion.
LAMP2 deficiencies can reduce this fusion, with inefficient autophagic cargo clearance. These findings implied that SORT1 is essential for regulating autophagic flux and is likely involved in DHT inhibition of OC.
To further explore the role of SORT1 in autophagy, we tested whether DHT affected OC through the ALP. Although the initial research on autophagy suggested it had a dual role in cancer, it is now widely recognized to suppress tumor initiation [
40]. Excessive autophagy activation can lead to ACD, also called type II cell death, which differs from apoptosis because it involves extensive degradation of cellular components via autophagosomes. Our data now show for the first time that DHT promotes autophagosome generation in OCCs. However, the autophagosomes accumulate due to both activation of the autophagy initiation pathways and inhibition of autophagosome–lysosome fusion. DHT treatment increased levels of LC3-II, which led to high levels of LC3 accumulation in autophagosomes, and not the autolysosomes, as demonstrated with RAPA treatment. This suggested that DHT promoted autophagosome formation while simultaneously blocking the autophagic flux in OCCs. Additionally, the colocalization of mitochondria and lysosomes supported our hypothesis that DHT disrupted the mitochondrial structure and prevented lysosomes from binding to damaged mitochondria. This caused damaged mitochondria to accumulate in cells, which may explain the DHT-induced ACD in OCCs. Notably, similar mechanisms have been reported in which cancer cells become more sensitive to chemotherapeutic agents [
41]. Interestingly, 3-MA and WORT, which inhibit autophagy initiation, reversed the inhibitory effect of DHT on OCCs. In contrast, HCQ enhanced the inhibitory effect of DHT on OCCs. This suggested that DHT induced ACD in OCCs. To further confirm this, we used an orthotopic OC model. A previous study reported that OCCs typically exhibit low autophagy levels, with LC3 expression significantly lower in OC than in benign or borderline ovarian tumors, which is further reduced in International Federation of Gynecology and Obstetrics (FIGO) stages III and IV compared with stages I and II [
42]. Tanshinone IIA was shown to promote autophagy and inhibit glioblastoma multiforme growth by regulating mechanistic target of rapamycin complex 1 [
22]. Additionally, high mitofusin 2 expression was found to trigger AMP-activated protein kinase, promote autophagy, and inhibit OC progression by downregulating the p-mTOR and phosphorylated extracellular signal-regulated kinase axis [
43]. This suggested that activating autophagy may provide a therapeutic approach for OC. Consistently, RAPA significantly increased ATG5, ATG16L1, and LC3 expression in our model and inhibited OC growth. DHT similarly mimicked these effects in OC but more effectively suppressed tumors than RAPA alone, potentially because it inhibited autolysosome formation by reducing LAMP2, STX17, and Rab7 expression. Furthermore, through a colocalization analysis of LC3 and LAMP1, we found that while DHT increased LC3 levels, it did not enhance the colocalization of LC3 and lysosomes. This phenomenon may be the primary means by which DHT inhibited OC growth. This aligned with the results of recent studies showing that initiating autophagy while also blocking late-stage autophagy caused excessive autophagosome accumulation, which transformed autophagy into a destructive, ACD-inducing process [
44].
DHT inhibited SORT1 expression in an orthotopic xenograft, which validated the proteomic results. Therefore, we explored the role of SORT1 in DHT-induced ACD. Previous studies established that various cancers, including breast, prostate, and ovarian, had high SORT1 expression [
45]. Our research confirmed elevated SORT1 levels in OCCs (OVCAR3, ES2, SKOV3, and A2780) compared with normal ovarian epithelial cells. We observed that the autophagosome formation inhibitor 3-MA significantly reduced autophagosome accumulation in
SORT1 knockdown cells, while the autophagic flux inhibitor HCQ had little effect on the accumulation. This discovery preliminarily confirmed that excessive autophagosome accumulation in OCCs caused by DHT was SORT1-dependent. SORT1 has been shown to increase tumor aggressiveness by promoting the mesenchymal transition, deregulating lipid metabolism, and conferring chemotherapy resistance [
[46],
[47],
[48]]. Here, we found that knocking down
SORT1 inhibited OCC growth. Additionally, 3-MA and WORT enhanced ES2-sh
SORT1 proliferation, while HCQ attenuated ES2-sh
SORT1 proliferation. Furthermore, DHT and RAPA showed only minimal inhibition of ES2-sh
SORT1, confirming that SORT1 depletion induced ACD in OCCs. However, the molecular mechanism by which SORT1 regulates autophagy requires further study. Various oncogenic mutations suppress autophagy by activating mTORC1 or inhibiting the Beclin1/class III PI3K complex [
49]. While the PI3K–AKT–mTOR pathway is well known for initiating autophagy, DHT did not affect PI3K–AKT–mTOR signaling, and SORT1 depletion did not alter the expression of autophagosome formation-related genes. We therefore hypothesized that SORT1 accelerated autophagosome generation by interacting with
ATG-encoded proteins.
The ATG12–ATG5–ATG16L1 complex facilitates the lipid conjugation of LC3/γ-aminobutyric acid receptor-associated protein, which is important during cargo recruitment and autophagosome maturation [
50]. DHT significantly increased ATG5 and ATG16L1 expression while inhibiting SORT1 expression. Notably,
ATG-encoded proteins sometimes become mutated or inactivated to evade tumor-suppressive autophagy as cancer progresses [
51]. In a dormant breast cancer model induced by doxorubicin, an autophagy deficiency mediated by
ATG5 knockdown led to metastatic recurrence earlier than in control cells [
52]. Excitingly, our study is the first to reveal that SORT1 promoted autophagosome generation by binding to ATG5 and ATG16L1, rather than to LC3. Knocking down
ATG16L1 further supported this finding, as si
ATG16L1 partially reversed the inhibited proliferation and migration of OCCs caused by SORT1 deficiency. One study demonstrated that activating ATG16L1 induced autophagy and inhibited the progression of clear cell renal cell carcinoma [
53], which may explain why SORT1 deficiency led to ACD in OCCs. Previous research showed that a primary function of SORT1 is mediating the transport of target proteins to lysosomes. For example, SORT1 was shown to redirect apoB100 from the secretion pathway to autophagosomes for degradation via lysosomes [
19]. This indicated that DHT inhibited the SORT1-dependent lysosomal degradation of ATG5 and ATG16L1, causing autophagosome accumulation. Furthermore, SORT1 depletion decreased the expression of proteins involved in autophagosome–lysosome fusion, including LAMP2, STX17, and Rab7. Our prior bioinformatics analysis also found a significant positive correlation between SORT1 and LAMP2 in OC patients. Furthermore, SORT1 and LAMP2 were reported to co-mediate extracellular vesicle secretion and cell adhesion, contributing to lenalidomide resistance in multiple myeloma [
54]. In summary, disrupting the autophagic flux and inducing ACD in OCCs by inhibiting SORT1 appears to be a potential mechanism by which DHT inhibits OC growth. However, further research is needed to clarify whether SORT1 is critical for inhibiting autophagosome–lysosome fusion.
SORT1 can be degraded in cells through both the classical ubiquitin proteasome pathway and lysosome pathway. In the present study, DHT intervention did not affect
SORT1 mRNA levels, leading us to speculate that DHT regulated SORT1 expression post-translationally. Further investigation confirmed that DHT promoted SORT1 ubiquitination, which explained its attenuation of SORT1 protein levels. Interestingly, one study showed that progranulin enhanced SORT1 ubiquitination, leading to its internalization via endocytosis and subsequent sorting into early endosomes for lysosomal degradation [
55]. This suggested that SORT1 ubiquitination may bypass the classical proteasome pathway, explaining why the proteasome inhibitor MG132 did not reverse the effect of DHT on SORT1. CETSA and molecular docking analyses provided insight into the mechanism by which DHT acted on SORT1. The CETSA results confirmed a direct interaction between DHT and SORT1, while molecular docking indicated that DHT bound to TYR189 in SORT1 via its carbonyl group at the C10 position. Previous studies showed that inhibiting SORT1 SER825 phosphorylation accelerated proteasome-dependent degradation of SORT1 [
56]. Similarly, DHT may promote SORT1 ubiquitination by occupying specific amino acid sites, although further research is required to confirm this mechanism. While the effect of SORT1 on the ALP has been partially studied [
57], its regulation of the ALP to influence tumor growth remains unclear. Finally, we demonstrated that SORT1 depletion increased ATG5 and ATG16L1 expression
in vivo, thereby inhibiting OC xenograft growth and supporting previous
in vitro findings. Additionally, the effect of DHT on
ATG-encoded protein expression and xenograft growth was lost in ES2-sh
SORT1 cells, further confirming that DHT-induced ACD in OCCs was dependent on SORT1.
5. Conclusions
Our study revealed that DHT regulated ACD in OCCs by disrupting the ALP, thus identifying SORT1 as a potential therapeutic target of OC treatment. Specifically, DHT promoted SORT1 ubiquitination-dependent degradation, reduced binding between
ATG-encoded proteins and SORT1, increased ATG5 and ATG16L1 expression, and impeded autophagosome–lysosome fusion, ultimately causing autophagosomes to accumulate in OCCs (
Fig. 8). Our findings offer new insights into anti-OC drug development through autophagy regulation.
CRediT authorship contribution statement
Chengtao Sun: Writing – original draft, Data curation, Conceptualization. Shengqian Deng: Project administration, Methodology, Data curation. Bing Han: Writing – review & editing, Supervision. Xiaoxiao Han: Visualization, Project administration. Yanan Yu: Visualization, Software. Man Li: Visualization, Software. Jiayi Lou: Data curation. Chengping Wen: Project administration. Jiong Wu: Resources, Methodology. Guoyin Kai: Writing – review & editing, Supervision, Funding acquisition.
Declaration of competing interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
Acknowledgments
This work was supported by the National Key Research and Development Program of China (2023YFC3503900); the National Natural Science Foundation of China (82305001); the Zhejiang Provincial Natural Science Foundation of China (LQ24H280011); the Science Research Fund of Administration of Traditional Chinese Medicine of Zhejiang Province (2023ZR014); the National Young Qihuang Scholars Training Program; the Research Project of Zhejiang Chinese Medical University (2022RCZXZK18, 2023JKZKTS17). We appreciate the technical support from the Pharmaceutical/Medical Research Center, Academy of Chinese Medical Sciences, Zhejiang Chinese Medical University. We thank LetPub for its linguistic assistance during the preparation of this manuscript.
Appendix A. Supplementary data
Supplementary data to this article can be found online at
https://doi.org/10.1016/j.eng.2025.06.020.
PDF
(9976KB)
