Downregulation of SRSF3 promotes ADD3 exon 14 skipping and drives the progression of endometriosis

Endometriosis (EMs) is a common and complex gynecological disease with an unclear pathogenesis. Splicing factors play critical roles in the progression of various diseases. However, the expression pattern and biological function of serine/arginine-rich splicing factor 3 (SRSF3) in EMs remain incompletely understood. This study aimed to investigate whether the splicing factor SRSF3 participates in the pathogenesis of endometriosis by regulating the alternative splicing of the Adducin 3 (ADD3) gene. Ectopic and eutopic endometrial tissues were collected from patients with EMs, with normal endometrial tissues from non-EMs women serving as controls. The expression levels of SRSF3 and ADD3 were detected by qRT-PCR and Western blot. Overexpression vectors for SRSF3 and two splice isoforms of ADD3 (ADD3(+ 14) and ADD3(− 14)) were constructed and transfected into ectopic endometrial stromal cells (ESCs). Subsequently, the effects on cell proliferation, apoptosis, migration, and invasion were assessed using CCK-8 assay, flow cytometry, wound healing assay, and Transwell assay. The binding of SRSF3 to the precursor mRNA of ADD3 was validated by RNA immunoprecipitation (RIP) assay. The results showed that SRSF3 expression was downregulated while ADD3 expression was upregulated in ectopic endometrial tissues. Overexpression of SRSF3 inhibited the splicing of ADD3 exon 14, leading to reduced expression of both isoforms. Functionally, overexpression of ADD3(− 14) enhanced the proliferation, migration, and invasion abilities of ESCs, whereas overexpression of SRSF3 exerted the opposite effects. RIP-qPCR confirmed that SRSF3 could bind to the exon 14 region of the ADD3 precursor mRNA. In conclusion, the downregulation of SRSF3 induces the skipping of ADD3 exon 14, generating the ADD3(− 14) isoform, thereby promoting the progression of endometriosis. These findings provide new insights into the pathogenesis of this disease and suggest a potential therapeutic target. Similar content being viewed by others

Endometriosis (EMs) is a prevalent benign gynecological disorder, affecting nearly 10% of women of reproductive age worldwide [1]. Despite extensive investigation, its pathogenesis remains incompletely defined. Evidence indicates that the disorder arises from the interplay of genetic predisposition, hormonal regulation, immune-mediated inflammation, and alterations in the local microenvironment [2, 3]. Recent research further indicates that alternative splicing events contribute significantly to the initiation and progression of diverse pathological conditions, offering a promising framework for elucidating the underlying mechanisms of EMs. In the preliminary exploration of this study, we conducted multi-omics analyses. First, ectopic and eutopic endometrial tissues were collected from patients with ovarian endometriotic cysts, with normal endometrial tissues from non-EMs women serving as controls. Integrated analysis using TMT-based quantitative proteomics and phosphoproteomics revealed that SRSF3 was significantly downregulated in ectopic lesions, whereas ADD3 was significantly upregulated. KEGG pathway enrichment analysis demonstrated that the "spliceosome" pathway was consistently enriched in comparisons between ectopic and eutopic endometrium, as well as between ectopic and normal endometrium, with SRSF3 identified as a core gene within this pathway. Subsequently, transcriptome sequencing combined with alternative splicing analysis further suggested that ADD3 undergoes significant alternative splicing events in ectopic tissues. These multi-omics data collectively indicate that SRSF3 and ADD3 are candidate regulators of alternative splicing in EMs, providing direction for subsequent functional studies. SRSF3, the smallest member of the serine/arginine-rich (SR) protein family, functions as a key multifunctional splicing factor. New functions of SRSF3 have been continuously discovered recently, especially its oncogenic function. SRSF3 plays essential roles in many cellular processes by regulating almost all aspects of RNA biogenesis and processing of many target genes, and thus, contributes to tumorigenesis when overexpressed or dysregulated [4]. As a member of the actin-binding protein family, ADD3 functions as a structural organizer of the cytoskeleton, contributing to intercellular junctional regulation and membrane stability [5, 6]. Previous studies have revealed that dysregulated expression and aberrant splicing of ADD3 in tumors can influence cell proliferation, migration, and invasion [7]. Given that the core pathological feature of endometriosis similarly involves the acquisition of aberrant migratory and invasive capabilities by endometrial stromal cells, we hypothesized that ADD3, a key cytoskeletal regulator, may play a significant role in the pathogenesis of EMs. However, its specific function and regulatory mechanisms in EMs, particularly whether it is subject to alternative splicing regulation, remain unknown. Notably, our preliminary multi-omics data precisely revealed significant alternative splicing events of the ADD3 gene in EMs lesions, providing direct clues for subsequent investigations. The serine/arginine (SR) splicing factor family exerts essential regulatory functions in mRNA splicing, including activation, repression, export, stability, and translation. In this study, endometrial tissue and peripheral blood samples were analyzed to determine the expression of SRSF3 and ADD3 in EMs. Subsequent experiments confirmed the existence of an ADD3 isoform and demonstrated the specific binding affinity of SRSF3 to this variant. Using primary endometriotic stromal cells (ESCs), the impact of SRSF3-mediated regulation of ADD3 exon skipping on key pathological traits of ovarian EMs—including proliferation, adhesion, migration, and invasion—was systematically examined, along with an initial assessment of the underlying mechanism. To date, this investigation provides the first evidence that SRSF3 modulates ADD3 alternative splicing in EMs and implicates this regulatory axis in disease pathogenesis.

Study participants and sample collection Fifteen patients with histopathologically confirmed ovarian endometriotic cysts who underwent laparoscopic surgery at the Fifth Affiliated Hospital of Xinjiang Medical University between October 1, 2024, and March 31, 2025, were included. Their ages ranged from 34 to 50 years (mean, 36.9 years), with menstrual cycles of 27–35 days, and none had received hormone therapy during the preceding 3 months. Based on the American Society of Reproductive Medicine (ASRM) staging system, ectopic endometrium, eutopic endometrium, and peripheral blood were collected intraoperatively. Approximately 1 g of eutopic endometrium was scraped, together with four representative ectopic lesions (each approximately peanut-sized) and one to two samples of eutopic tissue. Portions of both ectopic and eutopic endometrium were processed for stromal cell isolation and culture. The control cohort comprised 15 patients undergoing surgery for benign ovarian cysts or subserosal uterine fibroids during the same period, with EMs excluded by intraoperative and postoperative pathology. Peripheral blood samples were obtained from participants aged 23–45 years (mean, 31.4 years), all of whom had regular cycles of 27–35 days. Tissue samples from all participants were collected during the proliferative phase, confirmed by preoperative history and histological evaluation. Ethical approval was granted by the Ethics Committee of the Fifth Affiliated Hospital of Xinjiang Medical University (approval number: XYDWFYLSk-2025-09), and the study was conducted in accordance with the Declaration of Helsinki. Written informed consent was obtained from all participants prior to enrollment. Multi-omics combined analysis In this study, transcriptomic alternative splicing analysis, as well as integrated phosphoproteomic and proteomic analyses, were performed on tissues collected via laparoscopic or hysteroscopic surgery. The samples included ectopic endometrium (PR group) and eutopic endometrium (PR_N group) from patients with ovarian endometriosis, and normal endometrial tissue (NR group) from non-endometriosis controls, all of which were flash-frozen for preservation. Specifically, protein samples were digested with trypsin, labeled with TMT reagents (Thermo Fisher Scientific, USA), and subjected to high-performance liquid chromatography fractionation and phosphopeptide enrichment, followed by detection using a Q Exactive HF-X mass spectrometer. Total RNA was extracted using TRIzol (Ambion, 15596026), and after library construction, sequencing was performed on the Illumina NovaSeq 6000 platform. In the subsequent bioinformatics analysis, raw mass spectrometry data were processed with Proteome Discoverer 2.2 software and searched against the MASCOT 2.6 server. Proteins and phosphopeptides were filtered with a false discovery rate FDR < 0.01. Quantitative data were log2-transformed and normalized by median centering. Batch effects were corrected using the ComBat algorithm in R 4.1 software. Differential expression analysis for proteins and phosphoproteins was conducted using the limma package (thresholds: FDR 1.2). Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway enrichment analyses were performed using the clusterProfiler package (FDR < 0.05). For RNA-seq data, alternative splicing events were identified using rMATS software (version 4.1.0), with events showing FDR < 0.05 defined as statistically significant. Isolation, culture, and characterization of ESCs ESCs were isolated from eutopic and ectopic endometrium collected during surgery in patients with EMs. Resected tissues were rinsed with PBS to remove residual blood, transferred into Ham’s F12/DMEM medium (Thermo Fisher, 11330032) containing antibiotic–antimycotic solution (Thermo Fisher, 15240062; penicillin 10,000 U/mL, streptomycin 10,000 μg/mL, amphotericin B 25 μg/mL; 1:100 dilution), stored at 4 °C, and delivered to the laboratory within 2 h. Samples were minced into fragments smaller than 1 mm3 and digested with a solution comprising HEPES (25 mmol/mL, Thermo Fisher, 15630106), collagenase (1 mg/mL, Sigma, C0130), and deoxyribonuclease (0.1 mg/mL, Sigma, D5319) in a shaking incubator at 37 °C for 1 h. Following digestion and filtration, cells were centrifuged at 1000 g for 5 min and seeded in Ham’s F12/DMEM supplemented with 10% FBS (Excell Bio, FND500) and 1% tertiary antibody, then maintained at 37 °C in a humidified incubator with 5% CO2. To ensure cellular stability and minimize dedifferentiation, ESCs at passages 3–5 were selected for downstream experiments. Immunocytochemical characterization was performed using antibodies against vimentin (Abcam, ab92547) and PCK (Abcam, ab52625), followed by DAB and hematoxylin staining for microscopic evaluation. The cultured cells exhibited strong vimentin positivity and absence of PCK expression, with purity exceeding 96%, consistent with ESC identity. Lentiviral vector construction and cell transfection Lentiviral vectors for SRSF3 and ADD3 were synthesized by Genechem (GeneChem, Shanghai, China), including LV-SRSF3, LV-ADD3 (− 14), LV-ADD3 (+ 14), and the negative control LV-NC. Ectopic ESCs were seeded at 5 × 104 cells/mL in 6-well plates. At ~ 60% confluency, cells were infected with lentivirus (MOI = 100) in the presence of HiTransGP (GeneChem, Shanghai, China). After 12 h, the medium was replaced with Ham’s F12/DMEM supplemented with 10% FBS (Excell Bio, FND500) and 1% antibiotics (triple antibodies, Thermo Fisher Scientific, 15240062), followed by incubation at 37 °C in 5% CO2. Infection efficiency was evaluated by GFP fluorescence microscopy at 72 h post-transfection. Cells were subsequently harvested for qRT-PCR to quantify SRSF3 and ADD3 expression and then subjected to downstream functional assays. qRT-PCR Total RNA was isolated using TRIzol (Ambion, Catalog No. 15596026). Quantification was performed with a nucleic acid–protein analyzer, and purity was verified by a 260/280 ratio of 1.8–2.1. RNA integrity was confirmed by agarose gel electrophoresis. Reverse transcription was carried out with 5X All-In-One RT MasterMix (abm, Catalog No. G492). qRT-PCR was conducted in a 20 μL system containing 10 μL EvaGreen Express 2 × qPCR MasterMix, specific primers, cDNA, and RNase-free water. The amplification protocol consisted of an initial denaturation at 95 °C for 3 min, followed by 40 cycles of 95 °C for 15 s and 60 °C for 1 min. Primers for ADD3 (− 14) and ADD3 (+ 14) were designed across the unique exon junction of each isoform to ensure precise discrimination and quantification of the two splice variants. Primer sequences for SRSF3, ADD3, and GAPDH are provided in Supplementary Table S1. Relative expression of ADD3 (− 14), ADD3 (+ 14), and SRSF3 was analyzed using the 2-ΔΔCt method, with GAPDH as the internal reference. Western blotting Cells were rinsed with PBS and lysed in RIPA buffer (Boster, AR0105) supplemented with protease and phosphatase inhibitors. Lysis was maintained on ice for 60 min, followed by homogenization and centrifugation at 12,000 rpm for 15 min at 4 °C. Supernatants were collected for protein quantification using a BCA assay (Quanshijin, DQ111-01). Equivalent protein amounts were mixed with 5 × SDS-PAGE loading buffer containing β-mercaptoethanol (Amresco, 0482–100 mL) and denatured at 100 °C for 5 min. After centrifugation, the resulting supernatants were separated by SDS-PAGE on a 12% gel and transferred onto PVDF membranes (0.45 μm, Millipore, IPVH00010). Membranes were blocked with 5% skim milk in TBST for 1 h at room temperature and incubated overnight at 4 °C with primary antibodies against β-actin (1:1000, Sino Biological, 100166-MM10), ADD3 (1:300, Proteintech, 17585-1-AP), and SRSF3 (1:400, Affinity, DF3230). After TBST washes, HRP-conjugated secondary antibodies (goat anti-mouse IgG H&L, 1:10,000, Abcam, ab205719; goat anti-rabbit IgG H&L, 1:5000, Abcam, ab205718) were applied for 1 h at room temperature. Detection was performed using SuperSignal™ West Pico PLUS (Thermo Fisher, 34580), and protein bands were visualized with a ChemiScope 3000 chemiluminescence imaging system. Relative expression levels were normalized to β-actin as the internal reference. RIP-PCR HEK293/EMs cells (1 × 10⁷/group) were cross-linked with 0.3% formaldehyde at room temperature for 8 min, quenched with glycine, rinsed with pre-chilled PBS, and lysed on ice for 15 min in RIP lysis buffer (Merck Millipore, 17–701) supplemented with protease and RNase inhibitors. Lysates were centrifuged at 12,000 rpm, and the clarified supernatant was pre-incubated with Protein A/G magnetic beads (Thermo Fisher Scientific) to eliminate nonspecific protein binding. Subsequently, 5 μg of anti-SRSF3 antibody (Abcam, ab154195) or isotype IgG control (Thermo Fisher Scientific, 39000) was conjugated to the beads for 30 min at room temperature, followed by overnight incubation with lysates at 4 °C under rotation. Immune complexes were sequentially washed five times with high-salt buffer (500 mM NaCl) and then with low-salt buffer (150 mM NaCl). Cross-links were removed by proteinase K digestion, and RNA was purified using the Zymo RNA Clean & Concentrator-5 Kit (Zymo Research, R1016). qRT-PCR was performed to quantify enrichment of pre-mRNA containing the ADD3 exon 14–intron 14 junction, while RT-PCR using primers spanning exons 13–15 confirmed expression of the ADD3Δex14 isoform. The experiment included an Input group, an IgG isotype control group, and an empty vector blank control group, each consisting of three independent biological replicates (using different batches of cell lysates). Standard RIP-qPCR normalization was performed using the Percent Input Method. The calculation formula was 2^(Ct Input—Ct IP) × 100% × dilution factor correction factor (a fivefold dilution was used in this study). The Ct values for Input samples ranged from 20 to 25, and the enrichment signal in the IP samples was significantly higher than that in the negative control IgG. In this study, a positive result was defined as an enrichment ratio (IP/IgG or IP/Input) greater than 2 with a P-value less than 0.05. Data were analyzed using GraphPad Prism 10.0 software with a t-test (P < 0.01). Cell function assay CCK-8 assay for cell proliferation Ectopic ESCs were seeded in 96-well plates at a density of 5 × 104 cells/mL, with 100 μL of complete medium added per well, followed by incubation at 37 °C in 5% CO2 for 24 h. After cell adherence, groups were established as follows: blank control (Control), empty vector (OE-NC), ADD3(+ 14) overexpression (OE-ADD3(+ 14)), ADD3(− 14) overexpression (OE-ADD3(− 14)), and SRSF3 overexpression (OE-SRSF3). Each group contained five replicate wells. Following 72 h of treatment, the medium was replaced with 100 μL of fresh medium containing 10% CCK-8 solution. After 1 h incubation under the same conditions, absorbance at 450 nm was determined using a multifunctional microplate reader to quantify proliferative activity [8]. Flow cytometry assay for apoptosis ESCs were seeded into 6-well plates at 5 × 104 cells/mL and cultured at 37 °C in 5% CO2 until attachment. Cells were then exposed for 72 h under the indicated conditions (Control, OE-NC, OE-ADD3(+ 14), OE-ADD3(− 14), and OE-SRSF3). After treatment, both medium and PBS wash were collected, and cells were digested to obtain a single-cell suspension. The suspension was centrifuged at 1000 rpm for 5 min, the supernatant removed, and the pellet washed twice with chilled PBS. Cells were resuspended in 500 μL of 1 × Binding Buffer, passed through a filter, and stained with 5 μL Annexin V-PE and 10 μL 7-AAD (BD, 559763) [9]. After gentle mixing, the samples were incubated at 4 °C in the dark for 10 min. Apoptotic cell percentages were quantified within 30 min using flow cytometry (BD FACSAria II). Wound healing assay for cell migration ESCs were suspended at 5 × 105 cells/mL, and 500 μL of the suspension was seeded into each well of a 24-well plate. Cultures were maintained at 37 °C in 5% CO2 until a confluent monolayer was established. A vertical scratch was introduced using a sterile pipette tip, followed by two washes with PBS to remove detached cells. Serum-free medium was added, and cells were subjected to treatments according to the designated groups (Control, OE-NC, OE-ADD3(+ 14), OE-ADD3(− 14), OE-SRSF3). Images of the wound area were captured at 0 and 24 h, and ImageJ software was employed to quantify the remaining scratch area. The migration rate was calculated to evaluate the repair capacity of the cells [10]. Transwell assay for cell invasion A 100 μL volume of diluted Matrigel (Corning, 356234, 1 mg/mL) was applied to the upper chamber of a Transwell insert (Corning, 3422) and incubated at 37 °C for 5 h to form a uniform matrix. Following treatment, ESCs were washed with serum-free medium, adjusted to a density of 2 × 105 cells/mL, and seeded at 100 μL per upper chamber. The lower chambers were filled with 600 μL of complete medium containing 10% FBS. After incubation at 37 °C in 5% CO2 for 72 h, chambers were rinsed twice with PBS, fixed with 4% paraformaldehyde for 20 min, washed again with PBS, and stained with 0.1% crystal violet (Solabol, G1062) for 5 min. Non-invading cells on the upper surface were removed using a moistened cotton swab, followed by two PBS washes and air-drying. Cells that had migrated to the lower surface were visualized and recorded microscopically to assess invasive capacity [11]. Statistical analysis Data obtained in this study were analyzed using SPSS 19.0 software. Quantitative data were presented as mean ± standard deviation. All in vitro experiments were performed with at least three independent biological replicates, and the quantitative data were first assessed for normality using the Shapiro–Wilk test. For data following a normal distribution, comparisons between two groups were conducted using the independent samples t-test (unpaired two-tailed t-test), and comparisons among multiple groups were performed using one-way analysis of variance (ANOVA). If the ANOVA result was significant, further pairwise comparisons were conducted using Tukey’s post hoc test. For data not following a normal distribution, the nonparametric Mann–Whitney U test (for two groups) or the Kruskal–Wallis test (for multiple groups) was employed. To control for false positives due to multiple comparisons, P-values from all post hoc tests were corrected for the false discovery rate (FDR) using the Benjamini–Hochberg method. A P-value < 0.05 or an adjusted P-value < 0.05 was defined as indicating a statistically significant difference.

Multi-omics analysis identified SRSF3 and ADD3 as candidate factors associated with alternative splicing in EMs Integrated proteomic and phosphoproteomic profiling revealed marked downregulation of SRSF3 and upregulation of ADD3 within protein-coding genes containing differentially expressed peptides. KEGG enrichment indicated significant involvement of the spliceosome pathway in both “ectopic endometrium (PR) vs. normal endometrium (NR)” (Fig. 1A) and “ectopic endometrium (PR) vs. eutopic endometrium (PR_N)” (Fig. 1B), with SRSF3 positioned within this pathway and suppressed. Complementary transcriptomic analysis of alternative splicing demonstrated pronounced exon-skipping events in ADD3 across both comparisons. Collectively, the results point to SRSF3 and ADD3 as candidate regulators of alternative splicing in EMs, warranting further functional validation in subsequent studies. In endometriosis, SRSF3 is downregulated, whereas ADD3 and its splicing isoform ADD3(− 14) are upregulated The ADD3 gene undergoes exon skipping at exon 14, generating two splicing isoforms: ADD3(+ 14), which retains this exon, and ADD3(− 14), which excludes it. qRT- PCR analysis demonstrated that the expression level of ADD3(− 14) was significantly higher in ectopic endometrium (PR) than in eutopic endometrium (PR_N) and normal endometrium (NR) (Fig. 2A). Validation by RT-PCR and electrophoresis consistently detected a specific 135 bp band corresponding to ADD3(− 14) in both tissue and blood samples, whereas the 100 bp band representing ADD3(+ 14) showed a weak signal, visually confirming the presence of both splice variants and corroborating the quantitative qRT-PCR results (Fig. S1). Furthermore, in peripheral blood from patients with endometriosis, SRSF3 mRNA expression was significantly lower in the PR group compared with the NR group, while total ADD3 mRNA expression was markedly elevated (Fig. 2B); a consistent expression trend for SRSF3 and ADD3 mRNA was also observed in ectopic endometrial tissues (Fig. 2C). At the protein level, Western blot analysis confirmed that SRSF3 protein abundance was reduced in PR tissues relative to PR_N and NR tissues, whereas the expression of both ADD3(+ 14) and ADD3(− 14) proteins was significantly higher in the PR group than in the NR group (Fig. 2D, E). Notably, although the mRNA signal of ADD3(+ 14) was faint, its protein product was clearly detectable, suggesting that mRNA abundance does not fully determine protein output and implying substantial post-transcriptional regulation of this isoform. For instance, the ADD3(+ 14) mRNA may harbor specific upstream open reading frames (uORFs) that enable efficient translation, or other RNA-binding proteins might bind to sequences unique to this isoform, thereby enhancing its mRNA stability and promoting translation. This pattern of downregulated SRSF3 and upregulated ADD3 (including both isoforms) was similarly reproduced in peripheral blood samples (Fig. 2F, G). Integrating the findings from both tissue and blood analyses, we confirmed that endometriosis is characterized by decreased expression of SRSF3 and increased expression of ADD3 (encompassing both splicing isoforms). Overexpression of SRSF3 suppresses the expression of ADD3 splicing isoforms To investigate the functional role of SRSF3 in endometriosis, we established an SRSF3 overexpression model in primary endometriotic stromal cells using a lentiviral vector system (MOI = 100, with polybrene-assisted transduction via HiTransG P to maximize infection efficiency) (Fig. 3A). qRT-PCR validation confirmed that SRSF3 mRNA levels were significantly elevated in the SRSF3-overexpressing group compared with the control (Fig. 3B). Similarly, successfully constructed ADD3(− 14) and ADD3(+ 14) overexpression models showed significant upregulation of the respective splicing isoforms (Fig. 3C, D), indicating the effective establishment of key gene overexpression systems. Subsequently, we examined the effect of SRSF3 overexpression on ADD3. Both qRT-PCR and Western blot analyses revealed that SRSF3 overexpression led to significant downregulation of the two ADD3 splicing isoforms, ADD3(+ 14) and ADD3(− 14), at both the mRNA and protein levels (Fig. 3E–I). In summary, we successfully generated overexpression models and demonstrated that SRSF3 overexpression effectively suppresses the expression of both ADD3 splicing isoforms. SRSF3 binds to ADD3 pre-mRNA To preliminarily verify the direct interaction between SRSF3 and ADD3 pre-mRNA, RIP-PCR analysis was performed in the easily transfectable HEK293/EMs cell model. Results showed that both ADD3(+ 14) and ADD3(− 14) were detected within the SRSF3 immunoprecipitate, with a markedly stronger signal observed for ADD3(− 14). No specific bands were observed in the IgG control (Fig. 4A). Quantitative evaluation demonstrated a significantly greater proportion of SRSF3 bound to ADD3(− 14) compared with ADD3(+ 14) (P < 0.001), while GAPDH exhibited no detectable interaction (Fig. 4B). These data indicate that SRSF3 directly binds to ADD3 pre-mRNA. ADD3(− 14) splicing isoform enhanced proliferation, migration, and invasion of ESCs The functional impact of the ADD3(− 14) splicing isoform on ESCs was examined by evaluating proliferation, migration, invasion, and apoptosis across experimental groups. CCK-8 assays demonstrated the highest survival rate in the OE-ADD3(− 14) group and the lowest in the OE-SRSF3 group (P < 0.001) (Fig. 5G). Wound healing analysis indicated that the OE-ADD3(− 14) group exhibited the most rapid closure at 24 h, whereas the OE-SRSF3 group displayed the slowest (P < 0.001) (Fig. 5A, B). Transwell assays showed significantly greater invasiveness in the OE-ADD3(− 14) group, while the OE-SRSF3 group presented the lowest number of penetrating cells (P < 0.001) (Fig. 5C, D). Flow cytometry further revealed minimal apoptosis in the OE-ADD3(− 14) group, contrasted with the maximal rate observed in the OE-SRSF3 group (Fig. 5E, F). Collectively, the ADD3(− 14) isoform was shown to augment proliferative, migratory, and invasive capacities of ESCs, in direct opposition to the effects of SRSF3 overexpression.

EMs is a benign gynecological disorder characterized by invasive growth and ectopic implantation of endometrial cells. Its initiation and progression are driven by aberrant proliferation, migration, invasion, and immune dysregulation. Advances in multi-omics technologies have enabled deeper insights into its molecular complexity [12, 13]. Integrated analyses of proteomics, phosphoproteomics, and transcriptomics demonstrated significant enrichment of the spliceosome pathway in ectopic EMs tissues, with SRSF3 and ADD3 identified as potential regulatory nodes, implicating alternative splicing in disease pathogenesis. In this study, SRSF3 is downregulated in endometriosis, while ADD3 expression is elevated; further validation revealed that the downregulation of SRSF3 promotes increased skipping of exon 14 in ADD3 pre-mRNA, thereby generating the pro-endometriotic splice isoform ADD3(− 14). Mechanistically, RNA immunoprecipitation assays demonstrated that SRSF3 directly binds to the exon 14 region of ADD3 pre-mRNA, and its overexpression effectively suppresses the production of ADD3(− 14). Functionally, cellular experiments indicated that ADD3(− 14) significantly enhances the proliferation, migration, and invasion of endometrial stromal cells, whereas overexpression of SRSF3 reverses these disease-promoting phenotypes. Collectively, spanning from multi-omics screening to functional validation, this study elucidates for the first time a novel post-transcriptional regulatory pathway in which downregulation of SRSF3 induces exon 14 skipping of ADD3, thereby driving the progression of endometriosis. Adducin is a widely distributed cytoskeletal protein comprising three isoforms, α, β, and γ. Among them, γ-adducin (ADD3) is broadly expressed across tissues and maps to the functional tumor suppressor locus at chromosome 10q25.1–25.2 [14, 15], where it has been implicated in regulating cancer cell migration, proliferation, and angiogenesis [16]. The human ADD3 gene consists of 15 exons, with exon 14 (96 nt) undergoing alternative splicing. Structurally, ADD3 contains an N-terminal globular head, a central neck region, and a C-terminal tail. The terminal portion of the tail harbors a MARCKS-related lysine-rich domain. Exon 14 encodes a 32-amino acid sequence positioned between the neck and MARCKS-related domains, and its selective inclusion produces a protein variant distinct from the shorter isoform. This retained region introduces additional serine/threonine phosphorylation sites and expands protein interaction networks, ultimately influencing proliferative and migratory behavior through modulation of spectrin–actin cytoskeletal organization [17]. Previous research has demonstrated that ADD3 exhibits considerable heterogeneity across different pathological contexts. Navarro et al. [18] reported that adducin may function either as an oncogene or as a tumor suppressor in malignancies, with its precise role remaining unresolved. Kiang et al. [6] observed a marked reduction of ADD3 expression in glioblastoma multiforme (GBM), whereas no comparable decrease was detected in gliomas of lower grade, indicating that diminished ADD3 expression represents a potential driver of malignant progression. In vivo analyses further confirmed that loss of ADD3 accelerated GBM growth and angiogenesis. In contrast, another investigation [19] identified elevated ADD3 expression in aging adipose-derived mesenchymal stem cells (ASCs); silencing ADD3 enhanced proliferation and migration through suppression of p21 and p53, thereby delaying senescence. Similarly, Wang et al. [17] demonstrated that increased inclusion of ADD3 exon 14 intron in lung cancer cells enhanced both proliferation and migration. The present study revealed significant upregulation of ADD3(− 14) in EMs, promoting pathological cellular activity in a manner consistent with its role in invasive phenotypes of specific diseases. Notably, this effect contrasts with its function in GBM, ASCs, and other tissues, indicating that tissue-specific outcomes are tightly governed by splicing regulation. Future investigations integrating single-cell omics with functional assays are expected to clarify the independent regulatory mechanisms of ADD3 isoforms within distinct tissue networks. Serine/arginine-rich splicing factors (SRSFs) constitute a family of 12 RNA-binding proteins. Among them, SRSF3 is the smallest member, characterized by one or two N-terminal RNA recognition motifs (RRMs) and a C-terminal serine/arginine-rich (RS) domain, and functions as a central regulator of precursor messenger RNA (pre-mRNA) splicing [20, 21]. Beyond its role in alternative splicing, SRSF3 influences diverse cellular processes by modulating RNA biogenesis and the processing of multiple downstream targets. Aberrant expression disrupts splicing homeostasis, leading to impaired cell growth, apoptosis, or conversely, enhanced proliferation and tumor development [4, 22]. Elevated SRSF3 expression has been documented in numerous malignancies, where it promotes tumor cell proliferation, migration, and invasion [23]. In colorectal cancer (CRC), SRSF3 expression is significantly increased, positively correlates with serum response factor (SRF), and modulates angiogenesis-associated genes. Mechanistically, SRSF3 binds to the CAUC motif within exon 6 of SRF, promoting intron exclusion, thereby altering SRF splicing patterns and augmenting VEGF secretion. Knockout of SRSF3 diminishes VEGF release and suppresses the migration, invasion, and angiogenic activity of human umbilical vein endothelial cells (HUVECs). Silencing SRF abrogates angiogenesis, whereas SRF overexpression reverses the inhibitory effects of SRSF3 depletion. Collectively, these findings indicate that SRSF3 regulates angiogenesis in CRC through modulation of SRF splicing. Although SRSF3 is frequently upregulated in diverse malignancies, reduced expression has also been linked to tumorigenic processes. In renal cell carcinoma (RCC), SRSF3 acts as a splicing regulator whose suppression in tumor tissues correlates with poor overall survival, whereas its overexpression restrains malignant progression [24]. In hepatocellular carcinoma, SRSF3 loss enhances proliferation of liver cancer cells through altered splicing of the insulin receptor (INSR) [25, 26]. SRSF3 expression is further modulated by autoregulatory splicing, as retention of exon 4 introduces a premature termination codon that induces mRNA decay or truncated protein formation, ultimately lowering protein abundance [27]. Proteasome-mediated degradation in hepatocytes represents an additional pathway contributing to SRSF3 downregulation [28]. Within EMs, SRSF3 displayed tumor-suppressive characteristics, paralleling its reported functions in renal and hepatic malignancies. Our study found that SRSF3 is similarly downregulated in endometriosis and exerts a disease-suppressive role, further supporting its complex function as a tissue-specific regulator and suggesting that its decreased expression may be a key event in the pathogenesis of EMs. Although this study is the first to reveal the significant role of the SRSF3–ADD3(− 14) axis in endometriosis, several limitations remain, which also point to valuable directions for future research. Primarily, as the conclusions are mainly drawn from clinical tissue samples and in vitro cell experiments, the functionality of this regulatory axis within the complex in vivo pathophysiological environment urgently requires further validation of its causality through animal models or loss-of-function experiments of SRSF3 in primary cells. Secondly, the understanding of the upstream inducers and downstream effector pathways of this axis is incomplete, constituting several important future research topics. For instance, the hypoxic microenvironment characteristic of EMs has been proven to play a crucial role in its pathogenesis [29], and hypoxia can influence SRSF3 expression and induce autophagy [30]. Therefore, whether hypoxia within endometriotic lesions is a key upstream factor leading to the sustained low expression of SRSF3 warrants in-depth investigation. Furthermore, SRSF3 has been reported in other systems to affect cell survival and inflammatory status by modulating signaling pathways such as PI3K/Akt/NF-κB [31, 32], raising a core scientific question: in endometriosis, do the downregulation of SRSF3 and upregulation of ADD3(− 14) influence disease progression through these classical signaling pathways or via broader remodeling of the immune microenvironment? The answer to this question awaits future validation through specifically designed experiments, such as simulating hypoxia in primary ESCs, detecting changes in inflammatory factor expression profiles, or employing single-cell multi-omics technologies for high-resolution analysis. Concurrently, future studies need to investigate whether the SRSF3–ADD3 axis is regulated by hormones in larger cohorts encompassing different menstrual cycle stages and to further validate the robustness and clinical translational value of this axis by quantifying the splicing efficiency of ADD3(− 14) (e.g., using Percent Spliced In, PSI values) and correlating it with clinicopathological features.

In summary, this study elucidates for the first time that in endometriosis, the downregulation of the splicing factor SRSF3 promotes exon 14 skipping of the ADD3 gene, leading to the generation of the pro-disease splice isoform ADD3(− 14), which ultimately enhances the proliferative, migratory, and invasive capacities of endometrial stromal cells. This discovery provides novel insights into the pathogenesis of endometriosis at the post-transcriptional regulatory level and suggests that the SRSF3–ADD3(− 14) axis represents a potential therapeutic target for this disease. Data availability The datasets used and/or analyzed in the current study are available from the corresponding author upon reasonable request.

Frisendahl C, et al. miR-193b-5p and miR-374b-5p are aberrantly expressed in endometriosis and suppress endometrial cell migration in vitro. Biomolecules. 2024. https://doi.org/10.3390/biom14111400. Smolarz B, Szyłło K, Romanowicz H. Endometriosis: epidemiology, classification, pathogenesis, treatment and genetics (review of literature). Int J Mol Sci. 2021. https://doi.org/10.3390/ijms221910554. Moazen S, Arjmand MH. The role of local Angiotensin II/Angiotensin Type 1 Receptor in endometriosis: a potential target for new treatment approaches. Curr Mol Pharmacol. 2024;17:e18761429315431. Jia R, Zheng ZM. Oncogenic SRSF3 in health and diseases. Int J Biol Sci. 2023;19(10):3057–76. Liu CM, et al. Adducin family proteins possess different nuclear export potentials. J Biomed Sci. 2017;24(1):30. Kiang KM, et al. Loss of cytoskeleton protein ADD3 promotes tumor growth and angiogenesis in glioblastoma multiforme. Cancer Lett. 2020;474:118–26. Suo Z, et al. Posttranscriptional inhibition of gamma-adducin promotes the proliferation and migration of osteosarcoma cells. Tumori. 2022;108(6):600–8. Ning Y, et al. Exosome miR-30a-5p regulates glomerular endothelial cells’ EndMT and angiogenesis by modulating Notch1/VEGF signaling pathway. Curr Gene Ther. 2024;24(2):159–77. Hekmatshoar Y, et al. Methylsulfonylmethane induces caspase-dependent apoptosis in acute myeloid leukemia cell lines. Fundam Clin Pharmacol. 2024;38(6):1094–102. Liu C, et al. SHR6390 combined with Cabozantinib inhibits tumor progression in the hepatocellular carcinoma mouse model. Curr Gene Ther. 2024;24(5):453–64. Qiu J, et al. Pan-cancer analysis identifies AIMP2 as a potential biomarker for breast cancer. Curr Genomics. 2023;24(5):307–29. Shi J, et al. Kallikrein-related peptidase 4 promotes proliferation, migration, invasion, and pro-angiogenesis of endometrial stromal cells via regulation of Brain-Derived Neurotrophic Factor production in endometriosis. Am J Pathol. 2024;194(1):121–34. Costa V, et al. Uncovering the complexity of transcriptomes with RNA-Seq. J Biomed Biotechnol. 2010;2010:853916. Matsuoka Y, Li X, Bennett V. Adducin: structure, function and regulation. Cell Mol Life Sci. 2000;57(6):884–95. Kiang KM, Sun S, Leung GK. ADD3 deletion in glioblastoma predicts disease status and survival. Front Oncol. 2021;11:717793. Lechuga S, et al. Adducins inhibit lung cancer cell migration through mechanisms involving regulation of cell-matrix adhesion and cadherin-11 expression. Biochim Biophys Acta Mol Cell Res. 2019;1866(3):395–408. Wang JZ, et al. QKI-5 regulates the alternative splicing of cytoskeletal gene ADD3 in lung cancer. J Mol Cell Biol. 2021;13(5):347–60. Navarro L, et al. Identification of new genetic clusters in Glioblastoma Multiforme: EGFR status and ADD3 losses influence prognosis. Cells. 2020. https://doi.org/10.3390/cells9112429. Xiong H, et al. Knockdown of long noncoding RNA SAN rejuvenates aged adipose-derived stem cells via miR-143-3p/ADD3 axis. Stem Cell Res Ther. 2023;14(1):213. Zhang S, et al. Tumor-promoting effect and tumor immunity of SRSFs. Front Cell Dev Biol. 2025;13:1527309. Manley JL, Krainer AR. A rational nomenclature for serine/arginine-rich protein splicing factors (SR proteins). Genes Dev. 2010;24(11):1073–4. Jia R, et al. Oncogenic splicing factor SRSF3 regulates ILF3 alternative splicing to promote cancer cell proliferation and transformation. RNA. 2019;25(5):630–44. Chen Y, et al. SRSF3 promotes angiogenesis in colorectal cancer by splicing SRF. Front Oncol. 2022;12:810610. Zhang L, et al. SRSF3 suppresses RCC tumorigenesis and progression via regulating SP4 alternative splicing. Biochimica et Biophysica Acta (BBA). 2024;1871(8):119841. Sen S, et al. Deletion of serine/arginine-rich splicing factor 3 in hepatocytes predisposes to hepatocellular carcinoma in mice. Hepatology. 2015;61(1):171–83. Torres S, et al. Proteomic characterization of transcription and splicing factors associated with a metastatic phenotype in colorectal cancer. J Proteome Res. 2018;17(1):252–64. Guo J, Jia J, Jia R. PTBP1 and PTBP2 impaired autoregulation of SRSF3 in cancer cells. Sci Rep. 2015;5:14548. Kumar D, et al. Degradation of splicing factor SRSF3 contributes to progressive liver disease. J Clin Invest. 2019;129(10):4477–91. Hoffmann-Mlodzianowska M, et al. Endometriosis development in relation to hypoxia: a murine model study. Mol Med. 2024;30(1):195. Zhou L, Guo J, Jia R. Oncogene SRSF3 suppresses autophagy via inhibiting BECN1 expression. Biochem Biophys Res Commun. 2019;509(4):966–72. Gao P, et al. SRSF3 inhibits high glucose and palmitic acid-induced cardiomyocyte apoptosis through the PI3K/AKT and NF-kappaB pathways. J Biochem Mol Toxicol. 2025;39(7):e70395. Fu Y, et al. SRSF3 knockdown inhibits lipopolysaccharide-induced inflammatory response in macrophages. Curr Issues Mol Biol. 2024;46(6):6237–47.

Not applicable. Funding This study was supported by the Natural Science Foundation of Xinjiang Uygur Autonomous Region (No. 2024D01C160, Jinling Yi). Author information Authors and Affiliations Contributions Conception and design: Yanli Shen, Jinling Yi; Administrative support: Yanli Shen, Wenguang Feng; Provision of study materials: Nurimanguli Aishanjiang, Lin Ma, Mian Sun, Kejian Li, Meimei Zhang, Lizhu Chen; Collection and assembly of data: Nurimanguli Aishanjiang, Lin Ma, Mian Sun, Kejian Li, Meimei Zhang, Lizhu Chen; Data analysis and interpretation: Peishuang Sun, Yanli Shen; Manuscript writing: Yanli Shen, Peishuang Sun, Wenguang Feng; Final approval of manuscript: All authors. Corresponding author Ethics declarations Ethics approval and consent to participate This study was approved by the Ethics Committee of the Fifth Affiliated Hospital of Xinjiang Medical University (Approval No.: XYDWFYLSk-2025-09), and the study was conducted in accordance with the Declaration of Helsinki. Written informed consent was obtained from all participants prior to enrollment. Consent for publication Written informed consent was obtained from all participants prior to enrollment. Competing interests The authors declare no competing interests. Additional information Publisher's Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. Supplementary Information Rights and permissions Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/. About this article Cite this article Shen, Y., Feng, W., Sun, P. et al. Downregulation of SRSF3 promotes ADD3 exon 14 skipping and drives the progression of endometriosis. Eur J Med Res 31, 708 (2026). https://doi.org/10.1186/s40001-026-04281-4 Received: Accepted: Published: Version of record: DOI: https://doi.org/10.1186/s40001-026-04281-4