The human gastric cancer cell lines BGC-823, MGC-803, HGC-27, and MKN45 were obtained from the Cell Bank of the Chinese Academy of Sciences (Shanghai, China). NCI-N87, SGC-7901 and GES-1 were obtained from the Cell Bank of the Chinese Academy of Sciences, Beijing, China. The cells were maintained in the corresponding medium supplemented with 10% fetal bovine serum (FBS) at 37°C with 5% CO₂.

To prepare samples for immunoblotting, protein was extracted from cells with Radioimmunoprecipitation lysis buffer containing protease inhibitor to protect protein degradation. Cells were washed with phosphate buffer saline (PBS), mixed with lysis buffer and then centrifugation was applied at 12,000×g for 10 minutes at 4°C. The supernatant was collected and the protein concentration was measured using a Bicinchoninic Acid (BCA) Protein Assay Kit. Following, denatured at 100°C for 10 min, samples were subjected to SDS-PAGE, and then protein was transferred to 0.22 μm nitrocellulose membranes. After blocking with PBS containing 5% skim milk for 2h at room temperature, the membranes were probed with specific primary antibodies diluted in PBST at 4°C overnight, followed by treatment with a horseradish-peroxidase-conjugated secondary antibody at room temperature for 2h. Subsequently, the membranes were washed with PBST, and exposed to an enhanced chemiluminescence kit (Thermo Fisher Scientific, Rockford, IL). The antibodies used for this study were against USP7 (Abcam No. 10893), MDM2 (Cell Signaling Technology No. 86934S), p53 (Cell Signaling Technology No. 9282T), and p21 (Cell Signaling Technology No. 2946T), GAPDH (Goodhere No. AB-M-M 001), Nanog (Abcam No,21624), E-cadherin (Cell Signaling Technology No. 3195S), ZO1 (Cell Signaling Technology No. 8193S), Vimentin (Cell Signaling Technology No. 5741S).

Gastric carcinoma tissues were acquired from the Henan Provincial People’s Hospital. Prior to tissue collection, the clinical protocol was approved by the Ethics Review Board of the Henan Provincial People’s Hospital. Written informed consent was obtained from all patients. Every specimen was handled anonymously, according to ethical and legal standards. Deparaffinization of the slides containing histological samples were performed by xylene and different concentrations of ethanol and washed with PBS three times. Then the slides were incubated with 3% hydrogen peroxidase (H₂O₂) for 20 min at room temperature to block endogenous peroxidase, then rinsed in PBS, and blocked with 5% BSA. Anti-USP7 antibody was applied to the sections at 4°C overnight. After washing three times with PBS, the slides were incubated with the secondary antibody for 2h at 37°C, subsequently, the slides were stained with DAB and hematoxylin. Finally, the slides were dehydrated and mounted. The slides were scanned and analyzed using Aperio AT2 (Leica).

Cells were seeded in 96-well plate at approximately 30% confluency for 24h before shRNA transfection. After transfection with Lipofectamine 3000, the medium was replaced with fresh medium. Three days later, cells were harvested and used for the indicated experiments. The following USP7 shRNA sequence was used: 5’- AAGGACCCGCAAAA-3’ (15).

1×10³ transfected cells of BGC-823 and MKN45 cells were seeded in each well of 6-well plate in triplicate. 1×10³ MKN45 cells were incubated in 6-well plate with or without Almac4 (0, 2.5, 5, 10μM) for 24h. After that, the medium was replaced with fresh medium, and the cells were cultured for an additional 10 days. Formed colonies (>50 cells/colony) were fixed and stained with methanol containing1% crystal violet (Beyotime Biotechnology, Haimen, China) and colony number was counted using Image J software (National Institutes of Health). Three independent experiments were performed to collect data, and the results were presented as the mean ± SD.

The equal number of the BGC-823 and MKN45 cells (1×103 cells per well) were seeded in a 96-well plate after transfection in triplicate. One thousand MKN45 cells were seeded in 96-well plate with or without Almac4 (0, 6.25, 12.5, 25, 50μM). At 1, 2, 3, 4, 5, 6, and 7 days, each well was added with 20μl of MTT [3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide] (obtained from Solarbio, Beijing, China) solution (5mg/ml) and then incubated at 37°C for 4h. Then 200 μL of dimethyl sulfoxide (DMSO) was added to dissolve formed formazan crystals, and the absorbance was taken at 570 nm detected on a microplate reader (Perkin Elmer).

The cells were harvested, centrifuged, and fixed with ice-cold 70% ethanol at 4°C overnight. Then cells were rinsed with PBS before staining with Propidium Iodide (PI) and suspended in staining buffer (50 μg/ml PI, 1% Triton X-100 and 0.5 mg/ml RNase in PBS) at 37°C for 30 min in the dark. Finally, the cells were analyzed using a BD Calibur flow cytometer (BD, USA). The results were analyzed using FlowJo V7.6 software (TreeStar, USA). DNA distribution at G0/G1, G2/M, or S phase was analyzed using FlowJo software and presented as histogram.

All animal experiments were conducted in accordance with the guidelines of the Institutional Animal Care and Use Committee of the Zhengzhou University. Female BALB/c nude mice of age range from four- to five-week-old with weighing 18-21g, were purchased from Hunan Slack Scene of the Laboratory Animal Company, LTD (Hunan, China). Optimum food and water were supplied to each animal and the nursing was done under sterilized conditions. The right scapular region of mice was selected and the cancer cells (5×10⁶) were injected. The weight of mice and the tumors were measured every 3 days. On the 18th day, the tumors were obtained from mice and weight was measured. The size of tumor was measured using vernier caliper.

Data was presented as the mean ± Standard Deviation. The significance of the difference between different groups was calculated using analysis of variance (ANOVA) and Student’s t-test. Student’s t-test was used to compare the difference between the two groups using SPSS 17.0. Results were regarded statistically significant at 0.01<p<0.05. p<0.01 was regarded highly significant.

First, UALCAN was used to evaluate the expression level of USP7 in GC tissues (16). Figure 1A showed that USP7 was overexpressed in GC tissues compared to that in normal gastric tissues. To further confirm this finding, the expression of USP7 in 287 human GC tissues with adjacent normal tissues was examined in our in-house tissue library, and the results showed that the expression of USP7 in tumor tissues was higher than that in adjacent normal tissues ( Figures 1B–E ), which is consistent with the finding from UALCAN. After confirmation the upregulation of USP7 in GC tissues, the clinical relevancy of USP7 with GC progression was investigated. As shown in Table 1 , USP7 expression was correlated with the stage of differentiation (p = 0.011), but not with age, gender and metastasis in GC patients. In addition, Kaplan-Meier analysis revealed that patients with high USP7 expression had significantly worse overall survival than those with low USP7 expression ( Figure 1F ). Overall, these data indicated that USP7 was overexpressed in GC and that USP7 overexpression may contribute to the poor survival outcome of GC patients.

As USP7 was confirmed to be overexpressed in GC, role of USP7 in the development and progression of GC remains unexplored. Hence, the expression of USP7 in seven gastric cancer cell lines was investigated. Results in Figure 2A showed that six human GC cell lines express high amount of USP7 comparing with the normal gastric epithelium cell line GES-1. Among these cell lines, MKN45 and BGC-823 were chosen to explore the function of endogenous USP7 in cell proliferation due to their high expression of USP7. Once USP7 was knocked down ( Figure 2B ), the colony formation efficiency of MKN45 and BGC-823 cells was inhibited compared to that in control cells ( Figure 2C ). What is more, proliferation of MKN45 and BGC-823 cells was also inhibited when USP7 was abrogated ( Figure 2D ). In the wound healing assay, knockdown USP7 decreased healing of wounds in MKN45 cells compared with the control group ( Supplementary Figure S2A ). Another study also found that the USP7 inhibitor slowed the closure of the would relative to the control group in MGC803 cells (17). These data indicated that knockdown of USP7 reduced colony formation ability and suppressed the proliferative ability on MKN45 and BGC-823 cells.

As reported, the TRAF domain and C-terminal (amino acids 801-1050) of USP7 interact with MDM2, which contributes to its stability through erasing its tagged ubiquitin, and protecting it from proteasome degradation (11). Meanwhile, MDM2 is the key E3 ligase of p53 and can write a poly-ubiquitin tag on p53 for subsequent 26S proteasome-dependent degradation (18). In contrast, the CDK inhibitor p21 was the first identified transcriptional target for p53. Once p53 is activated, p21 is transcriptionally upregulated via direct binding of p53 to the p21^CDKN1A promoter. Then p21 inhibits cyclin-dependent kinases (CDKs) activity, which phosphorylate the pRB-related proteins p107 and p130, and induces cell cycle arrest to inhibit cell viability (19). Thereafter, impact of USP7 on cell cycle was tested and results showed that abrogation of USP7 in MKN45 and BGC-823 cells promotes cell cycle arrest at G2/M phase ( Figure 2E ), while knockdown of USP7 has no effect on MKN45 cells apoptosis ( Supplementary Figure S1A ). To further characterize the molecular mechanisms which govern G2/M arrest as a result of USP7 abrogation, the expressions of p53 and p21 in MKN45 and BGC-823 cells were investigated, and the results showed that both proteins accumulated upon knockdown of USP7 ( Figure 2F ). After seeing that USP7 knockdown inhibits MKN45 cell migration, we investigated the status of mesenchymal marker proteins. The results showed that knockdown USP7 reduced the level of the mesenchymal marker Vimentin in MKN45 cells ( Supplementary Figure S2B ). A similar downregulation of Vimentin was observed when USP7 was knockdown in HCT116 cells (20). To characterize the accumulation mechanism of p53 upon USP7 abrogation, half-life of p53 was investigated and results indicated that abrogation of USP7 can stabilize p53 ( Figure 2G , Supplementary Figures S1B, C ). All these data collectively indicated that USP7 is involved in cell cycle progression by promoting the stability of p53.

Based on the results above, although USP7 was confirmed as a promoter in GC progression, whether USP7 is a drug target in GC still remains to be answered. Hence, USP7 inhibitor Almac4 was applied to verify the USP7 function on cell proliferation and colony formation, and results showed that Almac4 decreased the colony formation and proliferation activity of MKN45 cells in a dose-dependent manner ( Figures 3A, B ). In addition, cell cycle analysis uncovered that cell numbers accumulate in the G2/M phase in the presence of Almac4 in a dose dependent manner either ( Figure 3C ). Further pharmacological inactivation using Almac4 also induced the expression of p53 and p21 in MKN45 cells in a dose-dependent manner ( Figure 3D ). To further explore the accumulation mechanism of Almac4 on p53, half-life of p53 was tested and results suggested that Almac4 reduced endogenous p53 turnover in the presence of cycloheximide ( Figure 3E & Supplementary Figure S1D ). These data revealed that USP7 inhibitor suppresses GC cell proliferation and cell cycle progression.

Above results let us make a hypothesis that knockdown of USP7 can suppress GC tumor growth in vivo, the ability of USP7-knockdown cells to affect tumor growth was detected in xenograft model bearing Con and USP7 KD MKN45 cells. The results in Figure 4A showed that knockdown of USP7 in MKN45 cells significantly reduced tumor size compared to the control group. Besides, the expression levels of USP7 in tumors were tested and lower USP7 protein levels are detected in the USP7 knockdown group compared to control group in general ( Figure 4B ). At the same time, abrogation of USP7 decreased the tumor volume and weight in the USP7 KD group compared to the control group, and the difference between these two groups was significant ( Figures 4C, D ). Overall, these results suggested that the abrogation of USP7 can restrict the proliferation of tumor cells in vivo.