{"paper_id":"c95be6c2-9e06-43c0-a9fd-adc2ef880c78","body_text":"Clin. Exp. Obstet. Gynecol. 2026; 53(8): 48963\nhttps://doi.org/10.31083/CEOG48963\nCopyright: © 2026 The Author(s). Published by IMR Press.\nThis is an open access article under the CC BY 4.0 license .\nPublisher’s Note: IMR Press stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.\nOriginal Research\nSPARC Promotes the Proliferation and Migration of Endometriotic\nStromal Cells via the PI3K/AKT and MAPK/ERK Signaling Pathways\nXiaoqing Li1, Zunpeng Wang1, Y an He2, Shuo Wang1, Peng Liu 3, Jianfeng Wu4,*,\nShan Y u1,*\n1Department of Pathology, Second Affiliated Hospital of Harbin Medical University, 150086 Harbin, Heilongjiang, China\n2Health Record Management, Second Affiliated Hospital of Harbin Medical University, 150086 Harbin, Heilongjiang, China\n3Laboratory of Medical Genetics, Harbin Medical University, 150080 Harbin, Heilongjiang, China\n4State Key Laboratory of Holistic Integrative Management of Gastrointestinal Cancers and Department of Pathology, Xijing Hospital and School of\nBasic Medicine, Fourth Military Medical University (Air Force Medical University), 710000 Xi’an, Shaanxi, China\n*Correspondence: kobewoo1989@163.com (Jianfeng Wu); yushan@hrbmu.edu.cn (Shan Y u)\nAcademic Editor: V alerio Gaetano V ellone\nSubmitted: 9 December 2025 Revised: 10 April 2026 Accepted: 21 April 2026 Published: 19 August 2026\nAbstract\nBackground: Endometriosis is a major contributor to infertility, and an effective curative treatment remains unavailable. SP ARChas\nbeen identified as a differentially expressed gene in endometriosis between peritoneal and ovarian lesions. In this study, our aim was\nto comprehensively evaluate the effects of SP ARCon the proliferation, migration, apoptosis, necrosis, and adhesion of endometriotic\nstromal cells (ESCs) in vitro and to explore the associated molecular signaling pathways. Methods: Using ESCs transfected with\nsi-SP ARCfor gene silencing and those with SP ARCoverexpression; we evaluated cell proliferation through the Cell Counting Kit-8\n(CCK-8) assay. Migratory capacity was examined using the wound-healing and Transwell assays. Cell apoptosis, necrosis, and adhesion\ncapabilities were evaluated using the corresponding experimental methods. Immunohistochemistry was utilized to confirm SP ARCex-\npression in endometriotic lesions, and Western blot was conducted to assess protein expression. Results: The CCK‑8 assay demonstrated\nthat si‑SPARC inhibited ESCs proliferation, whereas SPARC overexpression promoted it. Transwell and wound-healing assays showed\nthat si‑SPARC attenuated the migration ability of ESCs, whereas SPARC overexpression was associated with enhanced cell migration.\nBased on qualitative Hoechst/propidium iodide (PI) staining, SPARC did not show a significant effect on apoptosis or necrosis. More-\nover, the study did not conclusively establish that SPARC regulates ESC adhesion to extracellular matrix (ECM) components, such as\ncollagen I, fibronectin, poly‑L‑lysine, and laminin. Immunohistochemistry results further showed that SPARC expression was higher\nin ectopic endometrium compared with orthotopic endometrium. Finally, we observed that SPARC was involved in the regulation of\nthe mitogen-activated protein kinases/extracellular regulated protein kinases (MAPK/ERK) and phosphatidylinositol 3-kinase/protein\nkinase B (PI3K/AKT) signaling pathways. Conclusions: Our in vitro findings suggest that SPARC may promote ESC proliferation and\nmigration, potentially via activation of the MAPK/ERK and PI3K/AKT signaling pathways. Qualitative Hoechst/PI staining showed no\napparent effect of SPARC on apoptosis or necrosis, although subtle effects cannot be excluded. The effects of SPARC on ESCs' adhesion\nto the ECM were inconsistent, possibly reflecting differences between cell models.\nKeywords: endometriosis; SPARC; endometriotic stromal cells; PI3K; cell signaling\n1. Introduction\nEndometriosis is a prevalent gynecological disorder\ncharacterized by the aberrant proliferation of endometrial\ntissue at extra-uterine locations, including the ovary, fal-\nlopian tube, bladder, and even the intestinal tract. This ab-\nnormal proliferation of endometrial tissue gives rise to a\nrange of symptoms, such as menstrual irregularities, pelvic\npain, and dyspareunia (pain during sexual intercourse). En-\ndometriosis also significantly increases the risk of infertility\n[1]. The invasion and metastasis of endometriotic stromal\ncells (ESCs) underlie the pathological development of ec-\ntopic endometrium. On the other hand, ESC proliferation\nand death form the physiological basis for the cyclical ex-\npansion and shedding of the endometrium. Consequently,\nstrategies aimed at halting the growth, invasion, and metas-\ntasis of ESCs represent a promising therapeutic approach\nfor endometriosis [2].\nThe gene for Secreted Protein Acidic and Rich in Cys-\nteine ( SP ARC) is located on human chromosome 5q33.1\nand encodes a multifunctional glycoprotein that functions\nwithin the extracellular matrix (ECM) [ 3]. SPARC binds\nECM components to regulate cell adhesion, proliferation,\nmigration, and growth factor signaling, thereby playing a\nkey role in tissue remodeling and damage [ 4]. Mature\nSPARC protein consists of three domains: the N-terminal\n(NT) domain has a low-affinity calcium-binding site, the\nF-spondin (FS) domain contains internal disulfide bonds\nand glycosylation sites, and the extracellular (EC) domain\ncontains collagen-binding motifs and anti-proliferative pep-\ntides that specifically inhibit endothelial cell growth [ 5].\n\nRecent evidence has demonstrated that SPARC is differ-\nentially expressed in the peritoneal and ovarian lesions of\nendometriosis [ 6]. However, it is not yet known whether\nSPARC can modulate the progression of endometriosis.\nIn this study, we systematically evaluated the impact of\nSPARC on the proliferation, migration, apoptosis, and ad-\nhesion of ESCs in an in vitro environment. Additionally,\nwe investigated the associated signaling pathways to gain a\ndeeper understanding of the underlying molecular mecha-\nnisms.\nNotably, SPARC is overexpressed in aggressive sub-\nclones of gynecological malignancies, where it consistently\npromotes cell proliferation, invasion, and metastasis. In\novarian cancer, its silencing suppresses these processes\nand induces apoptosis [ 4]. Similarly, high SPARC expres-\nsion in cervical squamous cell carcinoma (CESC) drives\ntumor progression by enhancing proliferation, migration,\nand epithelial–mesenchymal transition (EMT) [ 7]. Fur-\nthermore, SPARC contributes to tumor-stroma crosstalk.\nFor example, in endometrial cancer, it collaborates with fi-\nbronectin to activate fibroblasts, thereby accelerating can-\ncer cell invasion [ 8]. Collectively, SPARC plays a multi-\nfaceted pro-tumorigenic role across gynecological cancers.\n2. Materials and Methods\n2.1 Cell Culture\nhEM15A cell line was obtained from American Type\nCulture Collection (A TCC). We tested for mycoplasma con-\ntamination and used the short tandem repeat (STR) test to\nverify cell origin. Cell morphology was confirmed by a\npathologist before the experiments. Cells were cultured\nin Roswell Park Memorial Institute (RPMI) 1640 medium\n(Biosharp, BL303A, Hefei, AnHui, China) mixed with 10%\nfetal bovine serum (FBS) (Biosharp, BL201A, Hefei, An-\nhui, China), 100 U/mL penicillin, and 100 U/mL strepto-\nmycin (Phygene, PH1513, Fuzhou, Fujian, China) in a hu-\nmidified incubator at 37 °C with 5% CO 2.\nPrimary ESCs were a gift from Prof. Zongfeng\nZhang, Gynecology Laboratory, Second Affiliated Hospi-\ntal of Harbin Medical University. The primary cells were\nvalidated by flow cytometry analysis with CD10 antibody\n(detailed results are provided in the Supplementary Ma-\nterials File ). Flow cytometry data were analyzed using a\nhierarchical gating strategy. First, the cell population was\ninitially identified on a forward scatter area (FSC-A) versus\nside scatter area (SSC-A) plot to exclude debris and non-\ncellular events. Next, single cells were selected by plot-\nting FSC-A against FSC-W (or FSC-H) to exclude dou-\nblets and cell aggregates. Subsequently, live cells were\ngated based on the exclusion of a viability dye (e.g., DAPI-\nnegative or PI-negative). Finally, the target cell populations\nwere defined based on specific fluorescence markers, with\nthresholds established using unstained and fluorescence-\nminus-one (FMO) controls. Furthermore, all primary cells\ntested negative for mycoplasma contamination. All cultures\n(passages 3–5) were grown in Dulbecco’s Modified Ea-\ngle Medium (DMEM) (Biosharp, BL301A, Hefei, Anhui,\nChina) supplemented with 15% FBS (Biosharp, BL201A,\nHefei, Anhui, China), 100 U/mL penicillin, and 100 U/mL\nstreptomycin (Phygene, PH1513, Fuzhou, Fujian, China) in\na cell incubator under standard conditions.\n2.2 Transfection of Cells\nThe SPARC overexpression (SPARC-OE) plasmid,\nfeaturing the complete coding sequence inserted into the\npcDNA3.1+ vector, was synthesized by FengHuiShengWu\n(Changsha, Hunan, China). This construct, and the empty\nvector control, were subsequently purified from glyc-\nerol stocks using a mini-plasmid extraction kit (Solarbio,\nD1100, Beijing, China). The SPARC-OE plasmid was\ntransfected into hEM15A cells and primary ESCs by Lipo-\nfectamine 8000 (Beyotime, C0533, Shanghai, China). An\nempty plasmid was transfected as an internal control.\nSmall interfering RNAs (siRNAs) were designed and\nproduced by Sangon Biotech (Shanghai, China) (Table 1).\nLipofectamine 8000 was used to transfect each of these siR-\nNAs into the two cell lines independently, followed by incu-\nbation at 37 °C for 24 h. The knockdown potency of siRNA\nwas assessed using quantitative reverse transcription poly-\nmerase chain reaction (qRT-PCR). In order to ensure their\neffectiveness, all siRNAs were transfected into the cells si-\nmultaneously in the subsequent in vitro cell experiments.\n2.3 Wound Healing Assay\nThe hEM15A siNC cells, hEM15A si-SPARC cells,\nhEM15A V ector cells, and hEM15A SPARC-OE cells (5 ×\n105 cells/well) were grown to a density of approximately\n90% in 6-well plates. The same procedure was performed\nfor primary ESCs. A p1000 pipette tip was used to pro-\nduce a single scratch in the middle of the plate. Photomicro-\ngraphs were captured at 0 h and 24 h using an EVOS light\nmicroscope (ThermoFisher, Waltham, MA, USA). Cell mi-\ngration fronts were subsequently quantified using ImageJ\n(v1.54h, LOCI, University of Wisconsin, Madison, WI,\nUSA).\n2.4 Migration Assay\nThe hEM15A siNC cells, hEM15A si-SPARC cells,\nhEM15A V ector cells, or hEM15A SPARC-OE cells (7\n× 10 4 cells/well) were seeded into the upper chamber of\na 24-well Transwell plate (Corning Inc., Corning, NY ,\nUSA) in serum-free medium. The same procedure was\nused for primary ESCs. The medium in the lower cham-\nber was supplemented with 10% FBS. Following incuba-\ntion for 24 h, non-migrated cells on the upper surface were\nremoved with a cotton swab. Migrated cells were fixed\nwith methanol, stained with 0.5% crystal violet (Phygene,\nPH1277, Fuzhou, Fujian, China), and imaged. An EVOS\nlight microscope (ThermoFisher, MA) was used to take pic-\n2\n\n\nTable 1. si-SPARC and control sequences .\nsi-RNA name Sequence (5′ to 3′) Base count (nt) with 3'-TT\nhSparc-209-a AAAUUCUCCUACUUCCACC 21\nhSparc-209-s GGUGGAAGUAGGAGAAUUU 21\nhSparc-928-a UUCUGCUUGAUGCCGAAGC 21\nhSparc-928-s GCUUCGGCAUCAAGCAGAA 21\nhSparc-715-a UAGUUCUUCUCGAAGUCCC 21\nhSparc-715-s GGGACUUCGAGAAGAACUA 21\nNegative control UUCUCCGAACGUGUCACGU 21\nACGUGACACGUUCGGAGAA 21\nPositive control (GAPDH) GUAUGACAACAGCCUCAAG 21\nCUUGAGGCUGUUGUCAUAC 21\nGAPDH, glyceraldehyde 3-phosphate dehydrogenase.\ntures at 40× magnification after drying. Stained areas were\nquantified using ImageJ.\n2.5 Cell Counting Kit-8 (CCK-8) Assay\nThe hEM15A siNC cells, hEM15A si-SPARC cells,\nhEM15A V ector cells, and hEM15A SPARC-OE cells (1 ×\n103 cells/well) were seeded into 96-well plates with growth\nmedium. The same procedure was performed for primary\nESCs. CCK-8 (Beyotime, China) (10%) was added to each\nwell on 24 h, 48 h, and 72 h. After 37 ℃-30 min incu-\nbation in cell incubator, the 450 nm absorbance was then\nmeasured.\n2.6 Cell Adhesion Assay\nA precoated 24-well plate was added by 300 μL\nof PBS containing collagen I (100 μg/mL), fibronectin\n(20 μg/mL), poly-L-lysine (100 μg/mL), or laminin (100\nμg/mL) to each well and incubating the plate for 1 h. Sub-\nsequently, 300 μL of blocking buffer (0.5% bovine serum\nalbumin [BSA] in medium) was added to each well. Af-\nter 60 min, hEM15A siNC cells, hEM15A si-SPARC cells,\nhEM15A V ector cells, or hEM15A SPARC-OE cells (8 ×\n105 cells/well) in serum-free medium were added for 90\nmin. Then, the culture medium was removed, and the plate\nwas left with the remaining cells were fixed with 100 μL\nmethanol for 10 min and subsequently stained with 0.5%\ncrystal violet (Phygene, China) for 10 min. Finally, the\nplate was washed with PBS and dried at room temperature.\nThe same procedure was performed for the primary ESCs.\nPhotomicrographs were taken of each well, and the cells\nwere counted.\n2.7 Apoptosis and Necrosis Assay\nA 6-well plate was used to grow hEM15A siNC\ncells, hEM15A si-SPARC cells, hEM15A V ector cells, and\nhEM15A SPARC-OE cells (1 × 10 3 cells/well). Follow-\ning 24 h incubation, 1 mL of cell staining buffer, 5 μL\nHoechst 33342, and 5 μL propidium iodide (PI) (Beyotime,\nC1052, Shanghai, China) were added and mixed, and the\nplate was placed on ice for 30 min. The staining solu-\ntion was then removed, and the cells were washed with\nPBS. The same procedure was performed for the primary\nESCs. The cells were viewed under a fluorescent micro-\nscope after anti-fluorescent mounting solution was added.\nHoechst 33342 stained the nuclei of all cells, while PI la-\nbeled cells with compromised membrane integrity. Apop-\ntotic cells were identified by characteristic nuclear morpho-\nlogical changes, including condensation and fragmentation,\nas visualized under fluorescence microscopy. Representa-\ntive images were selected to reflect the overall observations\nacross experimental groups.\n2.8 qRT-PCR\nRNA was extracted from both treated hEM15A cells\nand primary ESCs using Trizol reagent (Life Sciences,\nShanghai, China), as recommended by the manufacturer.\nThe concentration of extracted RNA was measured using\na NanoDrop spectrophotometer (Thermo Fisher Scientific,\nUSA), then converted into cDNA using the HiScript II Q RT\nSuperMix kit (V azyme, Nanjing, Jiangsu, China; R122-01).\nThe program was set as follows: incubation at 37 °C for 2\nmin, 55 °C for 15 min, and 85 °C for 5 min. The resulting\ncDNA was stored at –20 °C for subsequent analysis.\nThe AceQ-quantitative polymerase chain reaction\n(qPCR) SYBR Green Master Mix (V azyme, Q111-02) and\na LightCycler 480 II real-time PCR system (Roche, Basel,\nSwitzerland) were utilized for qPCR. The primer sequences\n(5′ to 3′) for SPARC were TGAGGTA TCTGTGGGAGC-\nTAA T (forward) and CCTTGCCGTGTTTGCAGTG (re-\nverse). Pre-denaturation was performed at 95 °C for 10 min,\nfollowed by 40 cycles of denaturation at 95 °C for 10 sec,\nand annealing and extension at 60 °C for 30 sec. The ex-\npression levels of target genes were normalized to that of\nglyceraldehyde 3-phosphate dehydrogenase (GAPDH) us-\ning the 2 –ΔΔCt method.\n2.9 Immunohistochemistry\nA total of 32 cases of paraffin-embedded tissue were\ncollected from patients who underwent surgery for en-\ndometriosis between 01/06/2023 and 01/06/2024 in the De-\n3\n\npartment of Obstetrics and Gynecology, Second Affiliated\nHospital of Harbin Medical University. These comprised\n20 cases of paired ovarian endometriosis and in situ en-\ndometrium, and 12 cases of abdominal wall endometriosis.\nThe samples were analyzed on 05/02/2025. Prior to im-\nmunostaining, the slides containing tumor tissue sections\nwere deparaffinized, rehydrated, and then immersed in a\n3% H 2O2/PBS solution for 15 min to deactivate endoge-\nnous peroxidase activity. Sections were then autoclaved\nin a buffered Tris-EDTA solution for two min at 121 °C\nto expose the antigen. After blocking with serum for 1\nh, the sections were incubated overnight at 4 ℃ with anti-\nSPARC antibody (diluted 1:800, catalog number AF8043,\nBeyotime, China), followed by incubation with a secondary\nantibody for 1 h. Freshly prepared 3,3'-diaminobenzidine\n(DAB, Maxim Biotechnologies, DAB-0031, Fuzhou, Fu-\njian, China) solution was then added to each slide. Finally,\nthe slides were mounted with coverslips after undergoing\nhematoxylin counterstaining.\nEach slide was categorized into four classes accord-\ning to the intensity of cell staining: no positive staining\n(negative: 0 points), light yellow staining (weak positive: 1\npoint), brown/yellow staining (positive: 2 points), tan stain-\ning (strong positive: 3 points). The percentage of positive\ncells was also evaluated and categorized: ≤25% (1 point),\n26%–50% (2 points), 51%–75% (3 points), and >75% (4\npoints). The final score for each sample was calculated by\nmultiplying the two scores for intensity and percentage of\ncell staining.\n2.10 Western Blot\nCells were lysed using radioimmunoprecipitation as-\nsay (RIPA) solution (BioSharp, BL504, Hefei, Anhui,\nChina). The extracted protein was mixed with 5× loading\nbuffer (BioSharp, Hefei, Anhui, China), denatured by boil-\ning, and then separated via SDS-PAGE (Epizyme, China).\nThe protein was transferred onto a polyvinylidene fluoride\n(PVDF) membrane (0.45 μm), blocked with 5% nonfat dry\nmilk for 1 h, and then incubated overnight at 4 °C with\nprimary antibodies targeting PI3 Kinase p85α (Beyotime,\nChina, AF7742; 1:1000), protein kinase B (AKT1/2/3)\n(Beyotime, China, AF1789; 1:1000), Phospho-AKT1\n(Thr308) (Beyotime, China, AF5734; 1:1000), extracellu-\nlar regulated protein kinases 1 (ERK1) (Beyotime, China,\nAF1315; 1:1000), Phospho-Erk1 (Thr202/Tyr204)/Erk2\n(Thr185/Tyr187) (Beyotime, China, AF1891; 1:1000), and\nSPARC (Beyotime, China, AF8043; 1:1000). GAPDH (zs-\nbio, China, TA-08; 1:2000) served as the internal control.\nThe next day, the membranes were washed three times\nwith Tris-buffered saline with Tween 20 (TBST) (Solar-\nbio, T1081, Beijing, China) and incubated with the fol-\nlowing secondary antibodies for 1 h at room temperature:\nhorseradish peroxidase-conjugated goat anti-rabbit/mouse\nIgG (H+L) (Beyotime, China, A0208/A0216; 1:1000). Af-\nter adding ECL reagent (Beyotime, China), the membranes\nwere visualized using a ChemiDoc imaging system (Bio-\nRad, Hercules, CA, USA). Relative protein expression lev-\nels were quantified using ImageJ software based on the gray\nvalues of the protein bands.\n2.11 Analysis of SP ARC in the Turku Endometriosis\nDatabase\nThe Turku Endometriosis Database [ 9] was used to\ncompare SPARC gene-related information between patient\nand control groups using the link https://endometdb.utu.fi\n/gene_analysis/. After entering SPARC, the data was ana-\nlyzed and a boxplot was chosen.\n2.12 Acquisition of Microarray Data Information\nThe Gene Expression Omnibus (GEO) database ( http\ns://ncbi.nlm.nih.gov/geo/) was used to select the GSE5108\ndataset based on the GPL2895 platform. This includes 22\nsamples from endometriosis patients, comprising 11 pairs\nof ectopic and eutopic endometrium [ 10].\n2.13 Identification of DEGs and GO/KEGG Enrichment\nAnalysis\nGEO2R ( https://ncbi.nlm.nih.gov/geo/geo2r/) was\nused to identify differentially expressed genes (DEGs) in\nthe GEO dataset. Ectopic endometrial tissues were used\nas the experimental group, and eutopic endometrium as\nthe control group. The GEO2R online analysis tool was\nused to analyze and screen out DEGs from the GSE5108\ndataset. The screening criteria were p < 0.05 and |logFC|\n≥1.5, with logFC ≥1.5 defined as up-regulated genes, and\nlogFC ≤–1.5 defined as down-regulated genes. p < 0.05\nwas used as the criterion for significance in both the Gene\nOntology (GO) and Kyoto Encyclopedia of Genes and\nGenomes (KEGG) enrichment analyses.\n2.14 Statistical Analysis\nData were analyzed using Prism software 9.4.1\n(GraphPad, San Diego, CA, USA). Two-tailed unpaired\nStudent’s t-test was employed for comparisons between two\ngroups, and one-way analysis of variance (ANOV A) for\ncomparisons across three or more groups. To ensure ade-\nquate statistical power, the sample size for each experiment\nwas determined based on previous studies in the field and\non our preliminary data. All quantitative data are presented\nas the mean ± standard deviation (SD) from at least three\nindependent experiments. Differences were considered sta-\ntistically significant at *p < 0.05, ** p < 0.01, *** p < 0.001\nand **** p < 0.0001.\n3. Results\n3.1 SP ARC Expression Is Higher in Ectopic Than in\nEutopic Endometrium in the Turku Endometriosis\nDatabase and the GEO Database\nCompared to controls, endometriosis patients showed\nhigher levels of SPARC expression in their endometrium\n4\n\n\nand peritoneum. Additionally, analysis of the Turku En-\ndometriosis Database revealed that the ovary and deep tis-\nsues expressed higher SPARC levels than the endometrium\n(Fig. 1A). According to this database, SPARC expression\ndiffers in the endometrium, peritoneum, deep tissue, and\novary, as well as in the various stages of endometriosis.\nThe expression of SPARC in the endometrium of stage I–IV\npatients was higher than that of the control group. More-\nover, SPARC expression in the peritoneum, deep tissues\nand ovary was different at different disease stages com-\npared with the control group (Fig. 1B). SPARC was a\ndifferentially expressed gene in the endometriosis dataset\nGSE5108 from the GEO database, and showed significant\nupregulation (Fig. 1C). This finding led us to hypothesize\nthat SPARC might play a functional role in the pathogen-\nesis of endometriosis. KEGG enrichment analysis identi-\nfied that SPARC expression was associated with PI3K/AKT\nand mitogen-activated protein kinases (MAPK) signaling\npathways (Fig. 1D). This bioinformatic prediction directly\nguided our subsequent mechanistic investigation, which\naimed to validate these specific pathways. GO biological\nprocess enrichment analysis focused mainly on cell migra-\ntion, activation, adhesion, and immunity (Fig. 1E). Conse-\nquently, these GO terms framed the scope of our in vitro\nfunctional assays, which were designed to specifically test\nthe effects of SPARC expression on cell proliferation, mi-\ngration, and adhesion.\n3.2 SP ARC Regulates ESC Proliferation and Migration\nFollowing transfection with si-SPARC-209, si-\nSPARC-928, and si-SPARC-715 for 24 h, the two cell\nlines showed a significant decrease in SPARC RNA\nexpression, as observed by qRT-PCR. Similarly, SPARC\nRNA levels were significantly increased after transfection\nwith SPARC-OE (Fig. 2A). Western blot results revealed\nlower SPARC protein levels after interference for 24 h, and\nincreased levels after SPARC overexpression for 24 h (Fig.\n2B). Transwell assay showed that SPARC-OE increased\nthe migration of hEM15A cells and primary ESCs under\nstarvation conditions, whereas si-SPARC attenuated the\nmigration of these cell lines (Fig. 2C). The wound healing\nexperiment revealed that cells with increased SPARC\nexpression exhibited improved scratch closure at 24 h,\nwhereas cells in the si-SPARC group displayed poorer\nhealing. The blue vertical line represents the fixed initial\nscratch boundary and the yellow line indicates the cell\nedge at 0h and 24h in Fig. 2D. The CCK-8 assay for\ncell proliferation in hEM15A cells and primary ESCs\nrevealed that cell proliferation was greater in the SPARC\noverexpression group and lower in the si-SPARC group\n(Fig. 2E).\n3.3 SP ARC Did Not Regulate Apoptosis and Necrosis of\nESCs and Did Not Affect Cell Adhesion to the\nExtracellular Matrix\nObservation of both cell lines by fluorescence mi-\ncroscopy showed that si-SPARC and SPARC-OE did not\nalter apoptosis (as revealed by Hoechst staining) or necrosis\n(as revealed by PI staining) (Fig. 3A). The effect of SPARC\non cell adhesion to ECM proteins was evaluated in both\nhEM15A cells and primary ESCs. In primary ESCs, neither\nSPARC-OE nor si-SPARC significantly altered cell adhe-\nsion to Collagen I, fibronectin, poly-L-lysine, or laminin.\nIn contrast, in hEM15A cells, both SPARC-OE and si-\nSPARC led to a marked reduction in adhesion specifically\nto Collagen I, whereas si-SPARC increased the adhesion of\nhEM15A cells to poly-L-lysine but decreased their adhe-\nsion to laminin (Fig. 3B). Taken together, these results sug-\ngest that SPARC does not substantially influence the adhe-\nsive capacity of endometrial stromal cells. The decreased\nadhesion to Collagen I observed exclusively in hEM15A\ncells is likely cell line–specific and does not reflect a general\nrole of SPARC in regulating ECM adhesion in this context.\n3.4 SP ARC Is Associated With Activation of PI3K/AKT and\nMAPK/ERK in Endometriotic Stromal Cells\nWestern blot assay confirmed that si-SPARC de-\ncreased the phosphorylation of ERK and AKT in ESCs,\nwhereas overexpression of SPARC increased the phospho-\nrylation of ERK and AKT (Fig. 4A). Immunohistochemi-\ncal results showed a significantly higher staining intensity\nfor SPARC in endometriotic tissues (including stroma and\nglands) of the ovary and abdominal wall compared with\neutopic endometrium (including stroma and glands) (Fig.\n4B).\n4. Discussion\nEndometriosis is one of the main causes of infertil-\nity, with complicated pathophysiologic mechanisms and\nunclear etiopathogenesis [ 2]. Despite its classification as\na benign condition, endometriosis exhibits striking paral-\nlels to malignant neoplasms, characterized by aberrant cell\nproliferation, tissue invasion, and metastatic-like dissem-\nination [ 11,12]. A widely accepted theory is that during\nmenstruation, viable endometrial fragments undergo retro-\ngrade transport via the fallopian tubes into the peritoneal\ncavity [13,14]. Here, the cells implant and adhere to peri-\ntoneal surfaces, initiating a cascade of events that includes\nangiogenic remodeling, immune evasion, and progressive\nstromal infiltration, ultimately forming endometriotic le-\nsions [15]. The aberrant proliferation and invasive capacity\nof ESCs underlie the pathological progression of ectopic\nendometrial lesions, while their regulated turnover main-\ntains the endometrium’s cyclical remodeling [ 16]. The tar-\ngeting of ESC hyperactivity—including proliferation, inva-\nsion, and immune evasion—represents a promising thera-\npeutic avenue, as demonstrated by a recent study that inhib-\n5\n\nFig. 1. SPARC expression in endometriosis in the Turku Endometriosis Database and GEO database . (A) SPARC expression in\ndifferent endometriotic sites was shown by the Turku Endometriosis Database. (B) Turku Endometriosis Database revealed the SPARC\nexpression in different stages of endometriosis. (C) SPARC was up-regulated in endometriosis in GSE5108. (D) The Kyoto Encyclopedia\nof Genes and Genomes (KEGG) enrichment bubble plot clearly showed the involvement of signaling pathways. (E) Bubble plot of Gene\nOntology (GO) biological process analysis. SPARC, Secreted Protein Acidic and Rich in Cysteine; GEO, Gene Expression Omnibus;\nECM, extracellular matrix.\n6\n\n\nFig. 2. SPARC promotes endometriotic stromal cells proliferation and migration . (A) The RNA levels of SPARC were altered by\ninterference and overexpression in hEM15A and primary endometriotic stromal cells (ESCs) cell lines, according to Quantitative Real-\ntime polymerase chain reaction. (B) Western blot results demonstrated that the protein level of SPARC was decreased and increased after\ninterference and overexpression for 24 h, respectively. (C) Transwell assay showed that SPARC overexpression enhanced the migration\nof hEM15A cells and primary ESCs under starvation conditions, while the si-SPARC group attenuated the migration of the above two\ncell lines. Scale bar = 50 μm. (D) The results of the wound healing experiment showed that the scratch healing was better when the\nexpression level of SPARC was increased at 24 h, while the healing of the si-SPARC group was worse at 24 h. The blue vertical line\nrepresents the fixed initial scratch boundary and the yellow line indicates the cell edge at 0 h and 24 h. Scale bar = 50 μm. (E) Cell\nCounting Kit-8 (CCK-8) time line plot of cell proliferation in both hEM15A cells and the primary ESCs: the cell proliferation ability was\nthe strongest in the SPARC overexpression group and the weakest in the si-SPARC group. Data are representative of three independent\nexperiments (n = 3). * indicates p < 0.05, ** indicates p < 0.01, *** indicates p < 0.001, **** indicates p < 0.0001, ns = not significant.\nited IDO1-mediated immune tolerance and NF-κB-driven\ninflammatory signaling [ 1]. Since the current therapeutic\nschedule only provides pain relief, the identification of reli-\nable molecular targets and biomarkers is crucial [17,18,19].\nSPARC was reported to be involved in regulating the pro-\nliferation and migration of gynecologic neoplasms, includ-\n7\n\nFig. 3. SPARC does not regulate apoptosis and necrosis of ectopic endometrial stromal cells and does not affect cell adhesion to\nthe extracellular matrix . (A) After staining the cells and observing them under a fluorescence microscope, the results indicated that\nsi-SPARC and SPARC overexpression (SPARC-OE) did not alter apoptosis (stained with Hoechst) and necrosis (stained with PI) in both\ncell lines. Apoptotic cells were distinguished by Hoechst staining based on nuclear condensation and fragmentation. Scale bar = 10 μm.\n(B) Cell-to-matrix adherence test showed that: in the hEM15A cells, si-SPARC and SPARC-OE altered the ability of cells to adhere to\ndifferent extracellular matrices. However, in the primary ESCs, altering SPARC expression did not significantly change cell adhesion\nability. Scale bar = 50 μm. Data are representative of three independent experiments (n = 3). * indicates p < 0.05, ** indicates p < 0.01,\nns = not significant.\ning endometrial carcinoma and ovarian cancer [ 4,20]. It\nwas also shown to be a differentially expressed gene in\nendometriosis between peritoneal and ovarian lesions [ 6].\nHowever, little is known about its functional significance\n8\n\n\nFig. 4. SPARC is associated with activation of PI3K/AKT and MAPK/ERK in endometriotic stromal cells . (A) Western blot assay\nconfirmed that si-SPARC attenuated the phosphorylation of ERK and AKT, and overexpression of SPARC enhanced the phosphorylation\nof ERK and AKT. (B) Immunohistochemical results showed that SPARC staining intensity was significantly higher in endometriotic\ntissues (including stroma and glands) of the ovary and abdominal wall than in eutopic endometrium (including stroma and glands). Scale\nbars represent 100 μm in the upper image, scale bars represent 10 μm in the enlarged views. * indicates p < 0.05, ** indicates p < 0.01,\n*** indicates p < 0.001, **** indicates p < 0.0001, ns = not significant. PI3K/AKT, phosphatidylinositol 3-kinase/protein kinase B;\nMAPK/ERK, mitogen-activated protein kinases/extracellular regulated protein kinases.\nin endometriosis. Our study addresses this gap by eluci-\ndating the functional role of SPARC and its associated sig-\nnaling pathways in ESCs. We found that dysregulation of\nSPARC contributes to the loss of cellular homeostasis in\nendometriosis.\nThe role of SPARC in regulating ESC adhesion to\nthe ECM appears to be context-dependent, with our find-\nings reflecting this complexity. As a canonical matricel-\nlular protein, SPARC is known to fine-tune cell-ECM in-\nteractions rather than simply promoting or inhibiting adhe-\nsion [21,22]. SPARC can bind to ECM components, mod-\nulate integrin clustering and signaling, and facilitate focal\nadhesion disassembly via its FS and EC domains to en-\nable cell motility [ 23,24,25]. The discrepancies we ob-\nserved between immortalized hEM15A cells and primary\nESCs in the adhesion assays likely stem from fundamental\ndifferences in their ECM receptor repertoire and signaling\nnetworks. The immortalized line, with a potentially more\nhomogenized signaling environment, may have a greater\ndependency on SPARC for modulating adhesion. Con-\nsistent with this, both SPARC overexpression and knock-\ndown reduced hEM15A adhesion to Collagen I—a para-\ndoxical result that can be explained by a biphasic, optimal-\nconcentration-dependent model: an appropriate SPARC\nlevel is required for normal focal adhesion dynamics, and\neither excess or deficiency impairs adhesion. This context-\ndependent variability highlights an important limitation in\nthe generalizability of our adhesion-related findings, as the\n9\n\nobserved effects may not extend broadly across different\nexperimental systems or physiological conditions.\nOur data confirm that SPARC is highly expressed\nin clinical endometriotic tissues (ovarian and abdominal\nwall) and is associated with activation of the PI3K/AKT\nand MAPK/ERK pathways. This aligns with findings in\nother diseases. For instance, Deng et al. [ 26] reported that\nSPARC promotes cell proliferation, migration, invasion,\nand EMT in cholangiocarcinoma via PI3K/AKT activation.\nThus, SPARC-mediated activation of core oncogenic path-\nways such as PI3K/AKT appears to be a conserved mecha-\nnism that drives invasive behavior, not only in cancers but\nalso in the progression of endometriosis. From a trans-\nlational perspective, these findings suggest that elevated\nSPARC expression could potentially serve as a biomarker;\nhowever, this remains hypothetical at this stage. Its detec-\ntion might hypothetically indicate the presence of lesions\nor could potentially identify patients with a higher risk of\nmultifocal or occult disease at other sites, but such appli-\ncations would require rigorous validation in large, well-\ncharacterized patient cohorts. Furthermore, the assessment\nof SPARC levels in ectopic lesions could, in principle, in-\nform prognosis and therapeutic strategy, although this pos-\nsibility is speculative and awaits further investigation. This\ncross-disease commonality strengthens the therapeutic ra-\ntionale for targeting SPARC and its downstream signaling.\nThe potential utility of SPARC as a target is further high-\nlighted by recent translational work. One example is the\nuse of a targeted nano-delivery system (BSA@Mif NPs)\ndesigned to selectively target SPARC-overexpressing M2\nmacrophages in ectopic lesions. This strategy enhances\ndrug accumulation in endometriotic tissue and modulates\nthe local immune microenvironment, showing marked effi-\ncacy in a mouse model [ 27]. Nonetheless, the clinical ap-\nplicability of targeting SPARC remains to be established,\nand any such strategies would require extensive preclinical\nand clinical evaluation.\nLimitations\nThe negative apoptosis/necrosis findings in this study\nwere limited by the sensitivity of the qualitative assay.\nMoreover, the results of the adhesion experiments likely re-\nflect the complex role of SPARC as a matricellular regula-\ntor that primarily drives cell motility, with statistical power\nbeing a potential factor. The in vitro experiments were per-\nformed without formal power calculations and without cor-\nrection for multiple comparisons. Further quantitative stud-\nies are required to clarify these issues.\nThe assessment of apoptosis and necrosis relied solely\non qualitative fluorescence microscopy. Future studies em-\nploying quantitative methods such as flow cytometry with\nAnnexin V/PI staining are required to definitively rule out\nany subtle effects of SPARC on cell death pathways. While\nour data show a significant correlation between SPARC\nexpression levels and the phosphorylation status of AKT\nand ERK, the present study cannot fully distinguish be-\ntween whether these changes represent direct, causative ac-\ntivation of these pathways, or whether they represent sec-\nondary, correlative adaptations. Additional experiments\nare needed to establish a direct causal link, such as co-\nimmunoprecipitation to probe for physical interactions be-\ntween SPARC-associated complexes and key components\nof the MAPK/ERK and PI3K/AKT molecular pathways,\nand kinase assays to directly assess the impact of SPARC\non AKT/ERK phosphorylation in vitro . Until such direct\nevidence is obtained, our conclusions regarding pathway\nactivation should be interpreted as indicative of a strong\ncorrelative relationship that is functionally significant in\nour model systems. Furthermore, the immunohistochemi-\ncal analysis in this study was limited by the relatively small\nsample size and single-center design, which may affect the\ngeneralizability of the findings. Future prospective cohort\nstudies with larger, multi-center samples are warranted to\nfurther validate the expression pattern and clinical signif-\nicance of SPARC across different subtypes and stages of\nendometriosis.\n5. Conclusions\nOur in vitro results suggest that SPARC may promote\nESC proliferation and migration, potentially via activation\nof the MAPK/ERK and PI3K/AKT pathways. Qualitative\nHoechst/PI staining showed no obvious effect of SPARC\non apoptosis or necrosis, though subtle effects cannot be\nexcluded. The effect of SPARC on ESC adhesion to the\nECM was inconsistent, possibly reflecting differences be-\ntween cell models. These findings offer preliminary in-\nsights into endometriosis pathogenesis and raise the possi-\nbility of SPARC as a diagnostic or therapeutic target, which\nrequires validation in larger cohorts and mechanistic stud-\nies.\nAvailability of Data and Materials\nThe data that support the findings of this study are\navailable from the corresponding authors upon reasonable\nrequest.\nAuthor Contributions\nXQL performed the literature search, in vitro experi-\nments and drafted the manuscript. YH performed the sta-\ntistical analysis. ZPW, SW and PL performed the IHC and\nrevised the manuscript. SY and JFW designed the concept\nof the study and revised the whole manuscript. All authors\ncontributed to editorial changes in the manuscript. All au-\nthors read and approved the final manuscript. All authors\nhave participated sufficiently in the work and agreed to be\naccountable for all aspects of the work.\n10\n\n\nEthics Approval and Consent to Participate\nThis study has been approved by the Second Affiliated\nHospital of Harbin Medical University, and the approval\nnumber is KY2024-277. The study was carried out in ac-\ncordance with the guidelines of the Declaration of Helsinki.\nThis was a retrospective study utilizing archived paraffin-\nembedded tissue blocks. The use and analysis of all the tis-\nsue samples were carried out strictly in accordance with the\napproved protocols after obtaining ethical approval. Writ-\nten informed consent was obtained from all patients whose\narchived tissue blocks were used in this study.\nAcknowledgment\nWe thank all who helped us in writing the manuscript\nand Professor Zhang for his primary ESCs as a gift to sup-\nport our project.\nFunding\nThis study was supported by the Research Project of\nChina Primary Health Care Foundation (2023001).\nConflicts of Interest\nThe authors declare no conflicts of interest.\nSupplementary Material\nSupplementary material associated with this article\ncan be found, in the online version, at https://doi.org/10.\n31083/CEOG48963.\nReferences\n[1] Hung SW, Zhang R, Tan Z, Chung JPW, Zhang T, Wang CC.\nPharmaceuticals targeting signaling pathways of endometriosis\nas potential new medical treatment: A review. Medicinal Re-\nsearch Reviews. 2021; 41: 2489–2564. https://doi.org/10.1002/\nmed.21802\n[2] Horne AW, Missmer SA. Pathophysiology, diagnosis, and man-\nagement of endometriosis. BMJ (Clinical Research Ed.). 2022;\n379: e070750. https://doi.org/10.1136/bmj-2022-070750\n[3] Said N. Roles of SPARC in urothelial carcinogenesis, progres-\nsion and metastasis. Oncotarget. 2016; 7: 67574–67585. https:\n//doi.org/10.18632/oncotarget.11590\n[4] Chen J, Wang M, Xi B, Xue J, He D, Zhang J, et al. SPARC is a\nkey regulator of proliferation, apoptosis and invasion in human\novarian cancer. PloS One. 2012; 7: e42413. https://doi.org/10.\n1371/journal.pone.0042413\n[5] Brekken RA, Sage EH. SPARC, a matricellular protein: at the\ncrossroads of cell-matrix communication. Matrix Biology: Jour-\nnal of the International Society for Matrix Biology. 2001; 19:\n816–827. https://doi.org/10.1016/s0945-053x(00)00133-5\n[6] Meola J, Rosa e Silva JC, Dentillo DB, da Silva W A, Jr, V eiga-\nCastelli LC, Bernardes LADS, et al. Differentially expressed\ngenes in eutopic and ectopic endometrium of women with en-\ndometriosis. Fertility and Sterility. 2010; 93: 1750–1773. https:\n//doi.org/10.1016/j.fertnstert.2008.12.058\n[7] Jia Q, Li M, Ma Y , Liu J, Jia Q, Zhang Q, et al. SPARC promotes\nmalignant progression and predicts poor prognosis in cervical\nsquamous cell carcinoma. Pathology, Research and Practice.\n2025; 272: 156096. https://doi.org/10.1016/j.prp.2025.156096\n[8] Y oshida S, Asanoma K, Y agi H, Onoyama I, Hori E, Matsumura\nY , et al. Fibronectin mediates activation of stromal fibroblasts\nby SPARC in endometrial cancer cells. BMC Cancer. 2021; 21:\n156. https://doi.org/10.1186/s12885-021-07875-9\n[9] Gabriel M, Fey V , Heinosalo T, Adhikari P , Rytkönen K, Ko-\nmulainen T, et al. A relational database to identify differen-\ntially expressed genes in the endometrium and endometriosis\nlesions. Scientific Data. 2020; 7: 284. https://doi.org/10.1038/\ns41597-020-00623-x\n[10] Edgar R, Domrachev M, Lash AE. Gene Expression Omnibus:\nNCBI gene expression and hybridization array data repository.\nNucleic Acids Research. 2002; 30: 207–210. https://doi.org/10.\n1093/nar/30.1.207\n[11] Ellis K, Wood R. Endometriosis and the Hallmarks of Cancer.\nReproductive Sciences (Thousand Oaks, Calif.). 2026; 33: 32–\n46. https://doi.org/10.1007/s43032-025-02024-0\n[12] Zhang M, Xu T, Tong D, Li S, Y u X, Liu B, et al. Research ad-\nvances in endometriosis-related signaling pathways: A review.\nBiomedicine & Pharmacotherapy = Biomedecine & Pharma-\ncotherapie. 2023; 164: 114909. https://doi.org/10.1016/j.biopha\n.2023.114909\n[13] Viganò P , Caprara F, Giola F, Di Stefano G, Somigliana E, V er-\ncellini P . Is retrograde menstruation a universal, recurrent, phys-\niological phenomenon? A systematic review of the evidence in\nhumans and non-human primates. Human Reproduction Open.\n2024; 2024: hoae045. https://doi.org/10.1093/hropen/hoae045\n[14] Hsu CF, Khine AA, Huang HS, Chu TY . The Double Engines\nand Single Checkpoint Theory of Endometriosis. Biomedicines.\n2022; 10: 1403. https://doi.org/10.3390/biomedicines10061403\n[15] Laganà AS, Garzon S, Götte M, Viganò P , Franchi M, Ghezzi\nF, et al. The Pathogenesis of Endometriosis: Molecular and Cell\nBiology Insights. International Journal of Molecular Sciences.\n2019; 20: 5615. https://doi.org/10.3390/ijms20225615\n[16] Bulun SE. Endometriosis and ovulatory menstruation: beyond\nthe Sampson principle. The Journal of Clinical Investigation.\n2025; 135: e188787. https://doi.org/10.1172/JCI188787\n[17] Billow M, Sridhar S, Mintz G. Endometrioma: Contemporary\nApproach to Diagnosis and Management. Seminars in Repro-\nductive Medicine. 2025; 43: 221–230. https://doi.org/10.1055/\ns-0045-1810430\n[18] Al Hussaini HAD, Alatawi ESE, Shabani JAJ, Edhrabooh\nMIH, Alhawaj SAA, Almahfoodh MS, et al. Management of\nEndometriosis-Related Pain: Comparing the Effectiveness of\nHormonal Therapy, Surgical Interventions, and Complementary\nTherapies. Cureus. 2024; 16: e75590. https://doi.org/10.7759/cu\nreus.75590\n[19] Perrone U, Evangelisti G, Laganà AS, Bogliolo S, Ceccaroni\nM, Izzotti A, et al. A review of phase II and III drugs for the\ntreatment and management of endometriosis. Expert Opinion on\nEmerging Drugs. 2023; 28: 333–351. https://doi.org/10.1080/\n14728214.2023.2296080\n[20] Y usuf N, Inagaki T, Kusunoki S, Okabe H, Y amada I, Mat-\nsumoto A, et al. SPARC was overexpressed in human endome-\ntrial cancer stem-like cells and promoted migration activity. Gy-\nnecologic Oncology. 2014; 134: 356–363. https://doi.org/10.\n1016/j.ygyno.2014.04.009\n[21] Nagaraju GP , Dontula R, El-Rayes BF, Lakka SS. Molecular\nmechanisms underlying the divergent roles of SPARC in hu-\nman carcinogenesis. Carcinogenesis. 2014; 35: 967–973. https:\n//doi.org/10.1093/carcin/bgu072\n[22] Bradshaw AD. The role of SPARC in extracellular matrix as-\nsembly. Journal of Cell Communication and Signaling. 2009; 3:\n239–246. https://doi.org/10.1007/s12079-009-0062-6\n[23] Murphy-Ullrich JE, Lane TF, Pallero MA, Sage EH. SPARC me-\ndiates focal adhesion disassembly in endothelial cells through a\nfollistatin-like region and the Ca(2+)-binding EF-hand. Journal\n11\n\nof Cellular Biochemistry. 1995; 57: 341–350. https://doi.org/10.\n1002/jcb.240570218\n[24] Barker TH, Baneyx G, Cardó-Vila M, Workman GA, Weaver M,\nMenon PM, et al. SPARC regulates extracellular matrix organi-\nzation through its modulation of integrin-linked kinase activity.\nThe Journal of Biological Chemistry. 2005; 280: 36483–36493.\nhttps://doi.org/10.1074/jbc.M504663200\n[25] Bhoopathi P , Gondi CS, Gujrati M, Dinh DH, Lakka SS. SPARC\nmediates Src-induced disruption of actin cytoskeleton via inac-\ntivation of small GTPases Rho-Rac-Cdc42. Cellular Signalling.\n2011; 23: 1978–1987. https://doi.org/10.1016/j.cellsig.2011.07.\n008\n[26] Deng S, Zhang L, Li J, Jin Y , Wang J. Activation of the PI3K-\nAKT signaling pathway by SPARC contributes to the malignant\nphenotype of cholangiocarcinoma cells. Tissue & Cell. 2022; 76:\n101756. https://doi.org/10.1016/j.tice.2022.101756\n[27] Zhang M, Y e Y , Chen Z, Wu X, Chen Y , Zhao P , et al. Tar-\ngeting delivery of mifepristone to endometrial dysfunctional\nmacrophages for endometriosis therapy. Acta Biomaterialia.\n2024; 189: 505–518. https://doi.org/10.1016/j.actbio.2024.09.\n037\n12","source_license":"CC0","license_restricted":false}