Cell Cycle and Apoptosis Analysis Kit (C1052) were purchased from Beyotime Biotech Inc, Annexin V-APC/PI Apoptosis Kit (E-CK-A217) were purchased from Elabscience Biotechnology Co.,Ltd., Click-iT EdU Cell Proliferation Kit with Alexa Fluor 488 (BL915A) were purchased from Labgic Technology Co., Ltd. Antibodies CDK4 (ab199728), CDK6 (ab241554), Cyclin D1 (ab134175), Cyclin D2 (ab308258), Cyclin D3 (ab289546), Rb (ab181616), p-Rb (ab184796), E2F1 (ab317385) were purchased from Abcam. Antibodies EZH2 (5246T) were purchased from Cell Signaling Technology. Antibodies OLFML2A (DF5039) was purchased from Affinity Biosciences Research Center Co., Ltd., Antibodies β-Actin (ET1702-67) were purchased from Hangzhou HUABIO Biotechnology Co., Ltd., TRIzol reagent (AG21101) was used for total RNA extraction and was obtained from Accurate Biology. Rabbit IgG Nanoselector Magnetic beads (025-101-003) were purchased from Chengdu Critical Point Biotechnology Co., Ltd. CCK8 (BS350B) kit were purchased from Labgic Technology Co., Ltd. GSK126 (S7061) were purchased from Selleckchem.

The MDA-MB-231 human breast cancer cell line utilized in this study was procured from Guangzhou Yuanjing Biotechnology Co., Ltd. The cells were cultured in high-glucose Dulbecco’s Modified Eagle Medium (DMEM) obtained from Jinyuan, Shanghai, which was supplemented with 10% fetal bovine serum and 1% penicillin/streptomycin to facilitate cell proliferation. The cell cultures were maintained under standard conditions at 37°C in a humidified atmosphere with 5% CO₂.

The MDA-MB-231 triple-negative breast cancer cell line used in this study was obtained from Guangzhou Yuanjing Biotechnology Co., Ltd. To ensure phenotypic specificity, we generated a panel of isogenic cell models via lentiviral transduction using a pre-engineered MDA-MB-231-Cas9 stable cell line (Guangzhou Yuanjing Biotechnology Co., Ltd. Catalog No. YC-D005-Cas9-H) as the parental line. This Cas9-expressing line carries a hygromycin resistance marker based on the YCas-LV002 backbone (non-fluorescent). Using this platform, we established three genetically defined models: (1) an Empty Vector Control (Control-OE), created by transducing cells with the lentiviral backbone vector without the OLFML2A insert to control for transduction and selection artifacts; (2) an OLFML2A Knockout (OLFML2A-KO) model, generated by infecting cells with a lentiviral sgRNA vector (based on the Source Bioscience backbone YKO-001) targeting OLFML2A; and (3) an OLFML2A Overexpression (OLFML2A-OE) model, produced by transducing cells with a lentiviral vector containing the full-length OLFML2A coding sequence.

Flow cytometric analysis was performed to assess the cell cycle distribution of MDA-MB-231 cells. The cells were harvested, washed with phosphate-buffered saline (PBS), and subsequently fixed in 70% ethanol at 4°C overnight. Prior to analysis, the fixed cells were stained in the dark at 37°C for 30 minutes with a solution containing propidium iodide (PI) and RNase A. Fluorescence emission was collected after excitation at 488 nm, and the proportion of cells in each phase of the cell cycle was determined based on DNA content.

The apoptotic rate of MDA-MB-231 cells was assessed via flow cytometry utilizing dual staining with Annexin V-APC and propidium iodide. In summary, harvested cells underwent washing with phosphate-buffered saline and were subsequently resuspended in 500 μL of 1× binding buffer. The cell suspension was then stained with 5 μL of both Annexin V-APC and PI, followed by a 15-minute incubation at room temperature under dark conditions. Flow cytometric analysis was conducted immediately following the incubation period.

Cell proliferation was assessed utilizing a 5-ethynyl-2′-deoxyuridine (EdU) labeling kit, in accordance with the manufacturer’s instructions. Following cell plating, the EdU incorporation procedure was performed, which encompassed fixation and permeabilization steps. Fluorescence images were subsequently acquired using an EVOS M7000 imaging system (Thermo USA). For the purpose of analysis, the proliferation rate was quantified as the ratio of EdU-positive nuclei to the total number of nuclei counted.

Total RNA was isolated utilizing the TRIzol reagent (Accurate Biology, China) following the protocol provided by the manufacturer. Subsequently, the isolated RNA was reverse-transcribed into complementary DNA (cDNA), and quantitative reverse transcription PCR (RT-qPCR) was conducted using SYBR Green (Accurate Biology, China) to assess the transcript levels of CDK4, CDK6, Cyclin D1, Cyclin D2, Cyclin D3, RB, E2F1, and EZH2. The specific sequences of primers used for amplification are presented in Table 1.

Tissue samples were homogenized in ice-cold lysis buffer with protease and phosphatase inhibitors. After centrifugation at 4°C, supernatants were collected for protein quantification. Samples were boiled, resolved on SDS-PAGE gels, and transferred to PVDF membranes. These were blocked for 30 minutes, then probed with primary antibodies overnight at 4°C. After TBST washes, secondary antibodies were applied for 1 hour at room temperature. Immunoreactive bands were visualized using enhanced chemiluminescence and quantified with a digital imaging system.

The study investigated the localization and expression of CDK4, Cyclin D1, p-RB, RB, E2F1, EZH2 and OLFML2A in MDA-MB-231 cells using immunofluorescence techniques. The cultured cells underwent fixation with 4% paraformaldehyde, permeabilization with 0.1% Triton X-100, and blocking with 5% bovine serum albumin (BSA). Subsequently, the fixed samples were incubated with specific primary antibodies at 4°C, and the nuclei were counterstained with DAPI. Imaging was performed using an inverted fluorescence microscope (EVOS M7000, Thermo Fisher Scientific, USA).

For the CCK−8 assay, logarithmically growing cells were seeded in 96−well plates and allowed to adhere for 24h. After replacing the medium with fresh medium containing a concentration gradient of GSK126, cells were incubated for an additional 24h. The medium was then aspirated and replaced with 100 μL of fresh medium containing 10% CCK−8 reagent. Following gentle mixing, the plates were incubated for 2 h, and absorbance was read at 450 nm on a microplate reader.

Following cell lysis, a portion of the lysate was reserved as the Input control. The remaining sample was incubated with either an anti-OLFML2A antibody or a control IgG, together with Protein A/G magnetic beads, at 4°C for 2 to 4 hours. The beads were subsequently isolated by centrifugation and washed three to four times with ice-cold lysis buffer. Bound proteins were eluted by boiling the beads for 10 minutes in 2× SDS loading buffer, followed by separation via SDS-PAGE and analysis through Western blotting.

Following an acclimation period of approximately one week, the mice were randomly allocated into three experimental groups. Cells from the KO, C-OE, and OE lines, each labeled with EGFP, were cultured until reaching the logarithmic growth phase. Subsequently, each BALB/c-nu mouse was administered a subcutaneous injection of 100 μL containing 1×10⁷ cells into the axillary region. Tumor progression and potential metastasis were systematically monitored after the establishment of the xenografts.

After dissection, the mouse tumor samples were fixed in 4% paraformaldehyde and subsequently processed for paraffin embedding. Histological analysis was conducted on tissue sections stained with hematoxylin and eosin (H&E), utilizing a Pannoramic MIDI light microscope for examination.

Cellular proteomic alterations were examined utilizing label-free quantitative proteomics facilitated by data-independent acquisition (DIA) mass spectrometry. Following cell harvesting, lysis was performed, and proteins were alkylated with iodoacetamide (IAM) in the absence of light for one hour at ambient temperature (22 ± 2°C). Subsequently, proteins underwent digestion, desalting, and analysis via liquid chromatography–tandem mass spectrometry (LC-MS/MS) operating in DIA mode. The resulting mass spectrometry data were processed and analyzed through bioinformatic methodologies.

The data are expressed as mean ± standard deviation (SD). All statistical analyses were conducted using GraphPad Prism software (version 9.0). For the purpose of group comparisons, a one-way analysis of variance (ANOVA) was utilized. A p-value of less than 0.05 was deemed statistically significant, with specific significance levels indicated as *p < 0.05 and **p < 0.01.

To explore the expression pattern of OLFML2A across various cancer types, we interrogated the Human Protein Atlas database (https://www.proteinatlas.org/). Immunohistochemistry results using the antibody HPA021180 (Figure 1A) revealed that OLFML2A exhibits a cancer type-specific expression profile, with notably high cytoplasmic staining in several solid malignancies. Elevated protein levels were consistently observed in colorectal, breast, liver, and pancreatic carcinomas. In contrast, minimal to no OLFML2A expression was detected in cancers such as lymphoma and glioma. This distinct expression pattern suggests a potential tissue- or lineage-specific role for OLFML2A in human oncogenesis. We then evaluated the correlation between OLFML2A expression and the prognosis of TNBC patients using the Kaplan-Meier database (Figure 1B). The results showed that high expression of OLFML2A was potentially associated with an increased risk of recurrence in TNBC patients, although this trend did not reach statistical significance (p > 0.05). Importantly, high levels of OLFML2A expression were linked to worse overall survival, with the high-expression group having a significantly lower chance of survival compared to the low-expression group (P < 0.05). These findings suggest that OLFML2A expression levels could be a potential prognostic indicator for TNBC.

To elucidate the biological function of OLFML2A in triple-negative breast cancer cells, we engineered MDA-MB-231 cell line variants with either knockout (KO) or overexpression (OE) of OLFML2A. Flow cytometry was employed to analyze cell cycle distribution and apoptosis. The findings revealed that OLFML2A depletion resulted in a marked accumulation of cells in the G1 phase and an elevated rate of apoptosis. Conversely, OLFML2A overexpression facilitated cell cycle progression and diminished apoptotic activity (Figures 2A, B). These results were corroborated by the EdU incorporation assay, which demonstrated that OLFML2A knockout reduced proliferative capacity, whereas OLFML2A overexpression enhanced it relative to control cells (Figure 2C). Collectively, these data suggest that OLFML2A plays a critical role in promoting cell proliferation and inhibiting apoptosis in MDA-MB-231 cells, underscoring its potential significance in sustaining tumor cell viability.

To elucidate the molecular mechanisms by which OLFML2A induces G1 phase arrest in MDA-MB-231 cells, we conducted an analysis of the expression levels of pivotal regulators involved in the G1/S phase transition. Employing RT-qPCR and western blotting techniques, we evaluated the expression of CDK4, CDK6, Cyclins D1–D3, RB, and E2F1 in both OLFML2A-knockout and control cell lines. The knockout of OLFML2A resulted in a significant reduction in the mRNA and protein levels of CDK4, CDK6, Cyclins D1–D3, and E2F1. Importantly, although the total RB protein expression remained constant, there was a notable decrease in RB phosphorylation (Figures 3A, B). These observations were further substantiated by immunofluorescence staining, which visually confirmed the downregulation of these cell-cycle regulators following OLFML2A depletion (Figures 4A, B). Taken together, these findings suggest that OLFML2A facilitates G1/S phase progression by modulating the expression and phosphorylation of critical cell-cycle proteins.

To elucidate the role of OLFML2A in the regulation of the tumor cell cycle in vivo, we developed a xenograft model of triple-negative breast cancer in mice. All experimental groups exhibited palpable tumors, albeit with distinct growth patterns. Compared to the control group (C-OE), the knockout of OLFML2A resulted in significantly slower tumor growth and reduced final tumor volume, whereas overexpression of OLFML2A led to accelerated tumor progression and increased tumor sizes (Figures 5A–D). To investigate the underlying molecular mechanisms, we conducted RT-qPCR and western blot analyses on the tumor tissues. The results indicated that OLFML2A modulates the expression of key regulators of the G1/S transition, including CDK4, CDK6, Cyclin D isoforms, and E2F1, corroborating our previous in vitro findings (Figures 5E, F). Based on these data, we propose that the depletion of OLFML2A induces G1 phase arrest by inhibiting the CDK4/6–Cyclin D complex, leading to reduced phosphorylation of RB and subsequent suppression of E2F1-mediated transcriptional activation. This pathway likely contributes to the observed inhibition of proliferation and increased apoptosis in breast cancer cells.

To elucidate the functional role of OLFML2A in TNBC, we employed the STRING database (https://cn.string-db.org/) to construct a protein–protein interaction (PPI) network. This approach enabled a systematic analysis of potential interaction partners and helped clarify the biological pathways associated with OLFML2A in TNBC. Differentially expressed proteins (DEPs) were identified based on screening criteria of FDR ≤ 0.05, FC ≥ 2, and p ≤ 0.05. Comparison between the OE and KO groups revealed a total of 1,383 DEPs, of which 1,026 were up-regulated and 357 were down-regulated. From these, the top 45 proteins ranked by p-value were selected for PPI network analysis (Figure 6A). By integrating the proteomic results with literature evidence, we identified EZH2 as one of the key interactors of OLFML2A in TNBC, showing a positive correlation with OLFML2A expression. To elucidate the functional roles of the DEPs, we performed Gene Ontology (GO) enrichment analysis using GOatools (https://github.com/tanghaibao/GOatools) (Figure 6B). Enrichment significance was assessed based on Fisher’s exact test, and p-values were adjusted for multiple testing using FDR correction. GO terms with an adjusted p≤ 0.05 were considered significantly enriched. Key enriched biological processes included”immune response”and”immune system process,”while molecular functions such as “signaling receptor activity” and “molecular transducer activity”were highlighted, along with cellular components such as”myosin complex.”Subsequently, Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway enrichment analysis was performed on the DEPs (Figure 6C), revealing significant associations with pathways related to immune and infectious diseases, metabolism, cancer and other diseases, as well as cellular structure and signaling.

To validate the interaction between OLFML2A and EZH2, we performed Co-IP assays in MDA-MB-231 cells: KO, C-OE, and OE. Cell lysates were immunoprecipitated using an anti-OLFML2A antibody, followed by immunoblotting with an anti-EZH2 antibody. As shown in Figure 6D, EZH2 was readily co-precipitated with OLFML2A in both C-OE and OE cells, whereas the interaction signal was nearly undetectable in KO cells. These results indicate a specific and OLFML2A dependent association between OLFML2A and EZH2 in TNBC cells. Immunofluorescence staining further demonstrated that knockout of OLFML2A significantly diminished EZH2 intensity, whereas overexpression of OLFML2A augmented EZH2 fluorescence signals relative to controls (Figure 6E). In alignment with these findings, both mRNA and protein levels of EZH2 were downregulated upon OLFML2A depletion and upregulated following OLFML2A overexpression, as confirmed by RT-qPCR (Figure 6F) and western blot analysis (Figure 6G). Collectively, these results establish EZH2 as a downstream interacting partner of OLFML2A and suggest that its expression is positively regulated by OLFML2A in triple-negative breast cancer (TNBC) cells.

To investigate the functional role of EZH2 downstream of OLFML2A, we employed the EZH2-specific inhibitor GSK126. First, an appropriate working concentration of GSK126 was determined via CCK-8 assay (Figure 6H). Treatment with 3 μM GSK126 resulted in approximately 60% relative viability in OLFML2A-OE cells, indicating effective EZH2 inhibition while preserving basic cellular metabolic activity. At this concentration, EdU proliferation assays were performed (Figure 6I). The results demonstrated that GSK126 treatment significantly reduced the percentage of EdU−positive cells compared with untreated OE controls, suggesting that EZH2 inhibition effectively reverses the enhanced proliferative capacity driven by OLFML2A overexpression. Together, these data further support that EZH2 acts downstream of OLFML2A and mediates its pro−proliferative function.