The RNA-seq gene expression and clinical information of breast cancer samples were obtained from the TCGA-BRCA dataset via UCSC Xena (https://xenabrowser.net/) (accessed on 02 March 2025) [12]. Based on receptor status determined by immunohistochemistry, 113 normal breast tissue samples, and 121 TNBC samples were selected for differential expression analysis. The analysis was conducted using the R package “limma” (version 3.62.1) in R software (version 4.2.2) [13]. The list of deubiquitinases was obtained from the DUBase database (https://ehubio.ehu.eus/dubase/) (accessed on 02 March 2025) and provided in the supplementary materials. Deubiquitinases with significant differential expression were identified based on the criteria of |log₂FC| > 0.585 and adj p-value < 0.05. Additionally, the GSE21653 dataset was analyzed for differential expression using the GEO2R online analysis tool (https://www.ncbi.nlm.nih.gov/geo/info/geo2r.html) (accessed on 02 March 2025) available on the GEO platform (https://www.ncbi.nlm.nih.gov/geo/) (accessed on 02 March 2025) [14]. Sangerbox (http://www.sangerbox.com/) (accessed on 02 March 2025) was utilized as a comprehensive bioinformatics analysis platform [15]. The pan-cancer dataset (TCGA TARGET GTEx, PANCAN, N = 19,131, G = 60,499) was downloaded from UCSC Xena (https://xenabrowser.net/) (accessed on 02 March 2025), which provides uniformly processed and standardized expression data. The expression levels of ENSG00000089723 (OTUB2) were extracted from this dataset and analyzed across different cancer types. The TIMER (Tumor Immune Estimation Resource, http://timer.cistrome.org/) (accessed on 02 March 2025) and GEPIA2 (http://gepia2.cancer-pku.cn) were utilized to further validate the differential expression of the gene across pan-cancer datasets. [16,17]. In this study, box plots generated by GEPIA2 and UALCAN (http://ualcan.path.uab.edu) (accessed on 02 March 2025) were used to visualize OTUB2 expression in TNBC [18]. Survival analysis was conducted using the Kaplan-Meier plotter (https://kmplot.com/analysis/) (accessed on 02 March 2025) database to evaluate the effect of OTUB2 on recurrence-free survival (RFS) in breast cancer and triple-negative breast cancer. The protein-protein interaction (PPI) network was constructed using the STRING database (version 12.0, https://www.string-db.org) (accessed on 02 March 2025) with a confidence score threshold of >0.7 [19]. The resulting work was then downloaded and further analyzed in Cytoscape software (version 3.6.0). UbiBrowser (http://ubibrowser.ncpsb.org/ubibrowser/) (accessed on 02 March 2025) is a web-based application that utilizes a naïve Bayesian computational framework to reliably predict interactions between human E3 ubiquitin ligase and their corresponding substrates [20].

The gene expression profiles of 121 TNBC specimens from the TCGA-BRCA dataset were analyzed to explore gene expression differences based on OTUB2 expression levels. Gene set enrichment analysis (GSEA) was performed using GSEA software (version 4.0.1) with the curated C2 gene set collection from the Molecular Signatures Database to identify pathways associated with high or low OTUB2 expression. Results were reported as normalized enrichment scores, with statistical significance defined by a false discovery rate (FDR) < 0.25 and p < 0.05 [21].

Human TNBC cell lines (MDA-MB-231, BT-549, Hs578T, and MDA-MB-468) and HEK293T cells (human embryonic kidney cell line) were cultured in Dulbecco’s Modified Eagle’s Medium (DMEM) (Thermo Fisher Scientific, C11995500BT, Lenexa, KS, USA) supplemented with 10% fetal bovine serum (FBS) (Excell Bio, FSP500, Suzhou, China) and antibiotics (100 U/mL penicillin and 100 μg/mL streptomycin) (Beyotime, C0222, Shanghai, China). MDA-MB-157 cells were cultured in Leibovitz’s L-15 medium (Thermo Fisher Scientific, 11415064, USA) supplemented with 10% FBS and antibiotics. Human normal mammary epithelial cells (MCF10A) were cultured in DMEM/F12 medium (Thermo Fisher Scientific, C11330500BT, USA) supplemented with 10% FBS and antibiotics [22,23]. All cells were incubated at 37°C in a humidified incubator with 5% CO₂ atmosphere. Mycoplasma contamination was regularly monitored and tested.

The OTUB2 siRNA (siOTUB2) constructs were synthesized by GenePharma (Suzhou, China), and the transfection control sequence (siNC) was purchased from GenePharma (A06001). The OTUB2 (NM_023112.4) and TRAF6 (NM_145803.3) coding regions were tagged with MYC and FLAG, respectively, and cloned into empty loading plasmids to obtain the overexpression plasmids pSin-EF2-puro-OTUB2-MYC and pSin-EF2-TRAF6-FLAG. The shRNA sequences targeting OTUB2 were synthesized and inserted into the vector pLKO.1-RFP to obtain the pLKO.1-shOTUB2#1/2 plasmids. Stable cell lines with OTUB2 overexpression or knockdown were generated via infection with retrovirus packaged from HEK293T cells. The viral supernatants were infected into target cells for 48 h, and then puromycin (1 μg/mL) was used to select cell lines for 1 week [24]. The target sequences of siRNA and shRNA are listed in Table 1.

MDA-MB-231 and BT-549 cells were transfected with the indicated siRNAs or plasmids using Lipofectamine 3000 (Invitrogen, L3000015, Lenexa, KS, USA) or Neofect (Neofect Biotechnologies, TF20121201, Beijing, China) according to the manufacturer’s instruction. The transfection efficiency was evaluated by quantitative RT-qPCR and western blotting after 24–48 h of transfection.

Total RNA was extracted by RNA Quick Purification kit (ESscience, RN001, Shanghai, China), and cDNA was synthesized using the NoScript Reverse Transcription System (Promega, A5001, Beijing, China). Quantitative PCR was performed with ChamQ SYBR qPCR Master Mix (Vazyme, Q311-02, Nanjing, China) on a LightCycler 480 II System (Roche Molecular Diagnostics, 5015243001, USA). The RT-qPCR cycling conditions were as follows: initial denaturation at 95°C for 30 s, followed by 40 cycles of denaturation at 95°C for 10 s, and annealing at 60°C for 30 s. A melting curve analysis was conducted using the instrument’s default settings. GAPDH was used as an internal control, and relative expression was calculated by the 2^−ΔΔC_T method [25]. The primer sequences are shown in Table 1.

MDA-MB-231 and BT-549 cells were plated into 96-well plates at a density of 800 cells per well. On the indicated day (days 0, 1, 2, 3, and 4), 10 μL/well of Cell Counting Kit-8 (CCK-8) reagent (TargetMol, C0005, Shanghai, China) was added to the 96-well plates [26]. After incubation at 37°C for 2 h, the absorbance of each well at 450 nm was detected on a spectrophotometer (EPOCH2, BioTek Instruments, Inc., Winooski, VT, USA).

Single-cell suspensions (800–1000 cells per well) were seeded in 6-well plates. Two weeks later, the colonies were rinsed with PBS, fixed with methanol, and stained with crystal violet (Beyotime, C0121, Shanghai, China). The number of colonies was then counted.

MDA-MB-231 and BT-549 cells were seeded in 12-well plates and cultured to near confluence. A wound was created in the cell monolayer using a sterile 10 μL pipette tip. Images of the wound area were captured at specific time points (0 and 48 h) using an inverted microscope (Nikon ECLIPS Ti2-U, Nikon Corporation, Tokyo, Japan).

The migration assay was conducted using Transwell Chambers (LABSELECT, 14341, Hefei, China). MDA-MB-231 or BT-549 cells (1 × 10⁵ cells per well) were suspended in 200 μL serum-free medium and added into the upper chamber. The lower chamber was filled with 500 μL medium supplemented with 10% FBS. After incubation for 12 or 16 h, cells were fixed with methanol, stained with 0.1% crystal violet, and captured by an inverted microscope (NIKON Eclipse Ti2).

Cell lysis was performed on ice using RIPA buffer (Merck Millipore, R0278, USA) with a protease inhibitor cocktail (Beyotime, P1046, China). The lysate proteins were isolated using 7.5%–15% SDS-PAGE gels (Epizyme, PG111-3, Shanghai, China) and subsequently transferred to Polyvinylidene Fluoride (PVDF) membranes (Merck Millipore, IPVH00010, Burlington, MA, USA). Following blocking of the membranes with 5% nonfat milk, primary antibodies against various proteins, including OTUB2 (1:1000, Affinity, AF9147, Liyang, China), Tubulin (1:5000, Proteintech, 11224-1-AP, Wuhan, China), DYKDDDDK-Tag (1:2000, Proteintech, 66008-4-Ig), Myc-Tag (1:1000, CST, 2278S, USA), HA-Tag (1:1000, Sigma, H9658, Burbank, CA, USA), AKT (1:2000, Proteintech, 10176-2-AP), Phospho-AKT (1:1000, CST, 4060S), were applied. After washing 3 times with TBST (TBS: Biosharp, BL602A-25, Hefei, China; Tween-20: Solarbio, T8220, Beijing, China), the membranes were incubated with specific secondary antibodies (1:5000, Proteintech, SA00001-2-100UL, SA00001-1-100UL) at room temperature for 1 h. The target protein bands were visualized using an enhanced chemiluminescence substrate (APPLYGEN, P1050-500, Beijing, China) and imaged with the ChemiDic MP Imaging System (Bio-Rad, 12003154, Boulder, CO, USA) [24]. The relative protein expression was quantified using ImageJ (version 1.52a, NIH, Bethesda, MD, USA) software by measuring band intensities. To account for variations in sample loading, the grayscale values of the target protein were normalized to Tubulin. The normalized expression levels of the experimental groups were then compared to the control group, with the control set as the baseline, to determine the relative protein expression [27].

Cells were lysed in IP lysis buffer (Beyotime, P0013), scraped off from the petri dish by a cell scraper, and transferred to a tube for centrifugating and sonicating. The protein supernatant was incubated with anti-Flag magnetic beads (Selleck, B26102, Houston, TX, USA) overnight at 4°C. The beads were washed with IP wash buffer and boiled in 1× SDS loading buffer for WB analysis [28].

HEK293T cells were transfected for 24 h and then incubated with 10 μM MG132 (Selleck, S2619, USA), a proteasome inhibitor, for 6 h. Subsequently, cells were lysed in RIPA buffer containing 1% SDS on ice, and the lysates were heated at 100°C for 5 min to denature the proteins. The denatured lysates were then diluted to 0.1% SDS concentration with RIPA buffer and immunoprecipitated with Anti-Flag beads (Selleck, B26102, USA). The level of ubiquitination was detected using western blotting with an anti-HA-Tag antibody (1:1000, Sigma, H9658).

Immunohistochemistry was performed as previously described [28]. Six pairs of matched cancerous and adjacent normal tissue samples from patients with TNBC were obtained from Sun Yat-Sen University Cancer Center (G2023-213-01). In brief, the sections were sequentially deparaffinized and rehydrated with xylene and ethanol, and endogenous peroxidase activity was blocked. The sections were then subjected to antigen retrieval. Finally, the sections were incubated with anti-OTUB2 antibody (1:200, Affinity, AF9147) overnight at 4°C, followed by the addition of secondary antibody (ZSGB-BIO, PV-6000, Beijing, China) and incubation at room temperature for 25 min, DAB (ZSGB-BIO, ZLI-9018) development and hematoxylin (Beyotime, C0107) staining. The stained slides were scanned and imaged using a Jiangfeng automated slide scanning system (KF-PRO-020, China) evaluated by an experienced pathologist blinded to the clinical data.

Female BALB/c nude mice, aged 3–4 weeks, were obtained from Charles River Laboratories (Guangzhou, China), housed in a specific pathogen-free (SPF) facility under controlled conditions (22°C–24°C, 12/12-h light/dark cycle) with autoclaved food and water ad libitum. All animal procedures were approved by the Animal Ethics Committee of Sun Yat-Sen University Cancer Center (L025501202404001). The mice were randomly assigned to two groups. To assess tumor growth, 1 × 10⁷ shOTUB2#1 or control vector MDA-MB-231 cells were resuspended in 200 μL of PBS and injected into the mammary fat pads of the mice (n = 5 per group) [23]. Tumor volume was measured every three days starting from day 6 post-injection (Volume = (length × width × width)/2). After 18 days, the mice were sacrificed, and the tumors were isolated for weighing and photographing.

Data are presented as the mean ± standard deviation (SD) from three independent experiments performed in duplicate. Continuous variables were compared using Student’s t-test and two-way ANOVA followed Dunnett’s multiple comparisons test. Statistical analyses were performed using GraphPad Prism software (version 8.0.1, GraphPad Software, LLC), and p-value < 0.05 was considered statistically significant.

We first analyzed the expression of deubiquitinases (DUBs) in TNBC using the TCGA-BRCA dataset, which was downloaded from the UCSC Xena platform. From the dataset, we specifically selected TNBC samples based on clinical information. Then, we conducted a differential analysis on 91 DUBs retrieved from the DUBase database (https://ehubio.ehu.eus/dubase/) (accessed on 02 March 2025). The results showed that 10 DUBs were downregulated, while 21 were upregulated in TNBC. Among these, OTUB2, a DUB with limited previous research in TNBC, exhibited high expression, which prompted subsequent investigation (Fig. 1A). To further explore the significance of OTUB2 in cancer, we examined its expression across various cancer types (http://www.sangerbox.com/) (accessed on 02 March 2025). Compared with normal tissues, the expression of OTUB2 was generally upregulated in solid tumors (Fig. 1B). We further validated OTUB2 expression in pan-cancers using the UALCAN (https://ualcan.path.uab.edu/) (accessed on 02 March 2025) and TIMER2.0 (http://timer.cistrome.org/) (accessed on 02 March 2025) databases (Fig. S1A,B). Additionally, analysis of the GSE21653 dataset revealed that OTUB2 expression was significantly higher in TNBC tissues compared to normal tissues (Fig. 1C). The GEPIA (http://gepia2.cancer-pku.cn/) (accessed on 02 March 2025) and UALCAN databases also confirmed that the transcriptional level of OTUB2 was elevated in breast cancer tissues (Fig. S1C,D). Kaplan-Meier plots (https://kmplot.com/analysis/) (accessed on 02 March 2025) showed that patients with higher OTUB2 expression in breast cancer or TNBC had shorter relapse-free survival (RFS) (Fig. 1D), suggesting that OTUB2 may influence the prognosis of TNBC patients. Immunohistochemical (IHC) staining revealed that OTUB2 was predominantly localized in the cytoplasm and exhibited stronger staining in TNBC tissues compared to adjacent normal tissues (Fig. 1E). Moreover, qPCR analysis of mRNA extracted from TNBC cell lines showed that OTUB2 expression was significantly higher in TNBC cell lines compared to normal breast epithelial cells (Fig. 1F).

To further investigate the potential biological functions of OTUB2, we conducted gene set enrichment analysis (GSEA), which revealed that OTUB2 was significantly associated with processes such as proliferation and metastasis (Fig. 2A). We then established stable OTUB2 overexpression or knockdown cell lines in MDA-MB-231 and BT-549 cells, with efficiency confirmed by qPCR and Western blotting (Fig. 2B–E). To assess the effect of OTUB2 on the proliferation ability of TNBC cells, we transiently overexpressed or knocked down the expression of OTUB2 in MDA-MB-231 and BT-549 cells and performed cell viability assays and clonogenic assays. The results demonstrated that OTUB2 knockdown significantly inhibited TNBC cell growth, while overexpression promoted cell growth in a time-dependent manner (Fig. 3A,B). The clonogenic assay further confirmed these findings, showing a similar trend in long-term proliferation (Fig. 3C,D). Finally, we injected MDA-MB-231 cells with stable shOTUB2#1 into the fat pads of female nude mice. The subcutaneous xenograft model showed that OTUB2 knockdown suppressed tumor growth in vivo (Fig. 3E–G).

The results of the Transwell assay showed that the knockdown of OTUB2 significantly impaired the migratory and invasive abilities of TNBC cells (Fig. 4A,B), indicated by the reduced number of cells migrating through the membrane. In contrast, overexpression of OTUB2 enhanced the migratory abilities of MDA-MB-231 and BT-549 cells (Fig. 4C,D). These findings were further corroborated by the wound-healing assay, which assessed the ability of cells to close a scratch wound. In this assay, OTUB2 overexpression significantly accelerated the wound closure rate, indicating enhanced cell migration (Fig. 4G,H). Conversely, the knockdown of OTUB2 resulted in slower wound healing, reinforcing the Transwell results and confirming that OTUB2 depletion inhibits TNBC cell migration (Fig. 4E,F). Taken together, these complementary assays provide strong evidence that OTUB2 plays a critical role in enhancing the migratory and invasive abilities of TNBC cells, which are key processes in tumor migration.

To further investigate the mechanism by which OTUB2 regulates proliferation and migration, we analyzed its potential interacting partners using the STRING online tool. This analysis revealed a potential interaction between OTUB2 and TRAF6 (Fig. 5A), and subsequent studies confirmed that TRAF6 is a substrate of OTUB2. Numerous studies have confirmed that TRAF6 is closely associated with tumorigenesis and is a potential target in malignant tumors. For instance, POU5F1 promotes gastric cancer (GC) cell proliferation and migration by inhibiting TRAF6 ubiquitination and degradation, thereby activating the NF-κB pathway [29], and KDM48 stimulates TRAF6-mediated AKT activation and facilitates the progression of rectal cancer [30]. Therewith, we demonstrated the exogenous binding between OTUB2 and TRAF6 in HEK293T cells by Co-immunoprecipitation (Co-IP) assays (Fig. 5B). Furthermore, we observed that overexpression of OTUB2 increased the protein levels of TRAF6 in a dose-dependent manner, although it did not affect TRAF6 mRNA expression (Fig. 5C,D, Fig. S2A), indicating that OTUB2 regulates TRAF6 expression at the post-transcriptional level. OTUB2, as a member of the deubiquitinase family, promotes substrate stability by cleaving ubiquitin chains and exhibits ubiquitin hydrolase activity. Therefore, we examined the effects of OTUB2 on TRAF6 ubiquitination and found that OTUB2 overexpression reduced the polyubiquitination of TRAF6 (Fig. 5E).

Given that TRAF6 functions as an E3 ubiquitin ligase with a RING domain, we utilized the UbiBrowser 2.0 (http://ubibrowser.bio-it.cn/) (accessed on 02 March 2025) to identify its protein substrates. Among the predicted substrates, AKT, which had the highest score, drew our attention (Fig. 5F). TRAF6 has been identified as a direct E3 ligase for AKT, and its ubiquitination is a crucial step in oncogenic AKT activation [31]. Based on these findings, we hypothesized that OTUB2 activates AKT by deubiquitinating TRAF6, thereby exerting a pro-oncogenic effect. To test this hypothesis, we transiently knocked down OTUB2 expression, and Western blot analysis revealed a decrease in AKT phosphorylation. Conversely, overexpression of OTUB2 led to a corresponding increase in p-AKT levels (Fig. 5G, Fig. S2B–E).