USP18 promotes cell proliferation and suppressed apoptosis in cervical cancer cells via activating AKT signaling pathway

All cell lines used in the present study were purchased from the cell bank of the Shanghai Biology Institute (Shanghai, China), including normal cervical epithelial cells HcerEpic and the human cervical cancer cell lines, including Hela, C-33A, Caski, and SiHa. Cells were cultured in RPMI-1640 medium (SH30809.01B, Hyclone, USA) containing 10% foetal bovine serum (16000–044, Gibco, USA) and 1% penicillin-streptomycin (P1400–100, Solarbio, China) and maintained in a 37 °C incubator with 5% CO₂. The AKT inhibitor LY294002 (25 μmol/L; S1105, Selleck, USA) was dissolved in dimethyl sulfoxide (DMSO, D2650, Sigma, USA) and used to treat the cells.

Briefly, the full-length USP18 (NM_017414.4) coding region sequence (CDS) was inserted into the lentiviral-mediator vector (pLVX-Puro). Then, the recombinant vector was transfected into human Hela cells using Lipofectamine 2000, following the manufacturer’s protocols (Cat: 11668027, ThermoFisher, USA) (oeUSP18). A mock vector was transfected as corresponding control (oeNC).

Cell proliferation was determined by using the Cell Counting Kit-8 (CCK-8) (CP002, SAB, USA) following the manufacturer’s instructions. The OD450 value was quantified using a microplate reader (DNM-9602, Pulangxin, China). Three replications were analysed for each time point.

Briefly, the cell (Hela, Caski, and SiHa) were stained using an Annexin V-fluorescein isothiocyanate (FITC) apoptosis detection kit (Beyotime, China) according to the manufacturer’s instructions, at 48 h after transfection. Then, the proportion of apoptotic cells were determined using flow cytometer (BD, USA). Three replicates were necessary for each samples.

The total RNA from cell samples was extracted using the TRIzol Reagent (1596–026, Invitrogen, USA). Then, the cDNA synthesis kit (Fermentas, Canada) was used to reverse transcribe the RNA into complementary DNA (cDNA) according to the manufacturer’s instructions. GAPDH expression was functioned as internal reference and used to normalise gene expression. Gene expressions were determined using the 2^-ΔΔCt method [11]. Three biological replicates were included for each analysis. The primers that used in this research were listed as follows: USP18 F 5′ TCTGGAG GGCAGTATGAG 3′, USP18 R 5′ TGGTAGTTAGGATTTCCGTAG 3′; and GAPDH F 5′ GGATTGTCTGGCAGTAGCC 3′, GAPDH R 5’ATTGT GAAAGGCAGGGAG 3′.

Total protein was extracted using RIPA lysis buffer (JRDUN, Shanghai, China). A BCA protein assay kit (PICPI23223, Thermo Fisher, USA) was used to measure total protein concentrations. Equal amounts of proteins adjusted to 25 μg were separated by 10% SDS-PAGE and subsequently transferred onto PVDF nitrocellulose membranes (HATF00010, Millipore, USA) for 12 h. After that, the membranes were then probed with primary antibodies at 4 °C overnight, followed by the appropriate HRP-conjugated goat anti-rabbit IgG (A0208, Beyotime, China) at 37 °C for 60 min. Protein signals were detected using a chemiluminescence system (5200, Tanon, China). GAPDH served as an endogenous reference. The protein expression was quantified as Gene _{grey value}/GAPDH _{grey value}. Each analysis was performed in triplicate. The primary antibodies that used the current study were listed as follows: USP18 (AB168478, Abcam, UK), cleaved caspase-3 (AB32042, Abcam, UK), AKT (#4691, CST, Danvers, USA), p-AKT (#4060, CST, Danvers, USA), Ki-67 (ab92742, Abcam, UK), Cyclin D1 (ab16663, Abcam, UK), Cleaved PARP (ab32064, Abcam, UK), Bax (ab32503, Abcam, UK), β-catenin (ab32572, Abcam, UK) and GAPDH (#5174, CST, Danvers, USA). Primary antibodies were detected using HRP-conjugated anti-rabbit IgG (A0208, Beyotime, Shanghai, China) or anti-mouse IgG (A0216, Beyotime, Shanghai, China) secondary antibodies.

This assay was performed according to a previous reference [12]. In brief, The tissue sections were fixed in methanol (4%) for 30 min. Then, endogenous peroxidase activity was blocked by incubating with H₂O₂ (3%) for 10 min. The tissue sections were then incubated with the USP18 primary antibody (ab115618, Abcam, UK) at room temperature for 1 h, followed by the HRP-labelled secondary antibody for 30 min. Then, the sections were stained with DAB and re-stained with haematoxylin for 3 min. An upright microscope (ECLIPSE Ni, NIKON, Japan) was utilised to obtain images, which were analysed using the microscope image analysis system (DS-Ri2, NIKON, Japan) at a magnification of 200 × .

All in vivo experiments were performed according to the Institute’s guidelines for animal experiments and approved by the independent ethics committee of Shanghai First Maternity and Infant Hospital, Tongji University School of Medicine, Shanghai, China.

All animals were treated in accordance with the Institutional Animal Care and Use Committee. An equal number of siNC or siUSP18 transfected Caski cells (n = 5 × 10⁶) were injected subcutaneously into the right flank of 4–6-week-old nude mice (n = 5 for each group; Shanghai Laboratory Animal Company, China). The length and width of the tumours were determined every 3 days for 33 days after cell injections. The volumes of the tumours were calculated according to the formula as follows: length × (width2/2). Then, both siNC and siTRIM37-injected mice were sacrificed by cervical dislocation, and then the tumour tissues were surgically removed and fixed in 4% formalin for further analysis.

TUNEL assays were performed with sections using an ApopTag kit (Intergene) according to the supplier’s instructions. Three replicates were analysed for each sample.

To examine the relationship between USP18 and cervical cancer, we collected data from the UALCAN (http://​ualcan.​path.​uab.​edu/​cgi-bin/​TCGAExResultNew2​.​pl?​genenam=​USP18&​ctype=​CESC) database. As presented in Fig. 1a, the level of USP18 was substantially higher in cervical squamous cell carcinoma (CESC) samples than that in para-cancer samples. Additionally, 30 pairs of cervical cancer and matched adjacent para-cancer tissues were used to examine the mRNA level of USP18 further. Our results also indicated that USP18 was overexpressed in human cervical cancer tissues (Fig. 1b).

Moreover, GSEA indicated that USP18 was enriched in the PI3K/AKT signalling pathway (Fig. 1c). Next, Western blot analysis was used to examine the protein levels of USP18 and phosphorylated AKT (p-AKT) in four pairs of human cervical cancer and matched adjacent para-cancer tissues. As shown in Fig. 1d, both USP18 and p-AKT were upregulated significantly in human cervical cancer tissues. To further determine the relationship between USP18 and AKT, IHC staining assays were performed to determine the protein contents of USP18 and p-AKT in human cervical cancer and adjacent para-cancer tissues (n = 30). Correlation analysis indicated that USP18 was correlated with p-AKT expression in human cervical cancer tissues (Fig. 1e).

Next, we compared the relative mRNA and protein levels of USP18 between normal human cervical epithelial (HcerEpic) and cervical cancer cells, including Hela, C-33A, Caski, and SiHa. Both the relative mRNA and protein levels of USP18 were upregulated significantly in cervical cancer cells (except for Hela), especially Caski and SiHa cells (Fig. 2a & b). Therefore, USP18 expression increased in Caski and SiHa cells.

For silencing, three siRNAs targeting different regions of the human USP18 gene (siUSP18–1, siUSP18–1, and siUSP18–3) and siNC were synthesised and transfected into Caski and SiHa cells. Untreated cells functioned as a blank control (Blank). Both the relative mRNA and protein levels of USP18 were reduced significantly in Caski and SiHa cells transfected with siUSP18–1, siUSP18–2, and siUSP18–3 (Fig. 2c & d). Also, HcerEpic cells were transfected with an overexpressing USP18 plasmid (oeUSP18). A mock plasmid functioned as the negative control (oeNC). As shown in Fig. 2e & f, oeUSP18 remarkably promoted the ectopic expression of USP18 in HcerEpic cells.

Then, we examined the proliferation rate of Caski and SiHa cells transfected with siUSP18–1 or siUSP18–2, using CCK-8 assays. As presented in Fig. 3a & b, both siUSP18–1 and siUSP18–2 significantly abolished the proliferation of Caski and SiHa cells. Moreover, the apoptosis of cells transfected with siUSP18–1 and siUSP18–2 was substantially higher than that in siNC-transfected Caski and SiHa cells (Fig. 3c).

Ki-67 and Cyclin D1 are well-known cell proliferation markers [13]. As presented in Fig. 3d, the proteins levels of Ki-67 and Cyclin D1 decreased significantly in Caski cells transfected with siUSP18–1 or siUSP18–2. Moreover, we also measured the levels of pro- or anti-apoptotic biomarkers in Caski cells, as indicated. Both siUSP18–1 and siUSP18–2 increased the expression of the pro-apoptosis factors cleaved PARP and Bax. Interestingly, the protein level of β-catenin, an AKT downstream factor, was also significantly decreased in USP18 siRNA-transfected cells. Importantly, similar results were obtained in SiHa cells (Fig. 3e). These results suggested that USP18 was a pro-proliferation and anti-apoptosis mediator in human cervical cancer cells.

Cleaved caspase-3 is a positive apoptotic factor [14]. In the current study, Western blotting was performed to quantify the protein contents of USP10, cleaved caspase-3, p-AKT and AKT in Caski and SiHa cells transfected with siUSP18–1 or siUSP18–2. Our results suggested that the protein content of cleaved caspase-3 increased significantly in siUSP18–1- or siUSP18–2-transfected cells. Interestingly, USP18 siRNAs suppressed the phosphorylation of AKT in Caski and SiHa cells (Fig. 3f & g).

To further confirm the function of USP18 in Caski cells, a rescue assay was established by transfecting oeUSP18 into USP18-silenced Caski cells. Our results indicated that oeUSP18 rescued the function of USP18 in siUSP18-transfected cells, further demonstrating USP18’s role as an oncogene in human cervical cancer cells (Figure S1).

The PI3/AKT inhibitor LY294002 was used to block AKT’s endogenous activity to assess the relationship between USP18 and AKT in cervical cancer cells. As shown in Fig. 4a, USP18 overexpression improved HcerEpic cell proliferation, but the inhibitor LY294002 disrupted this function. Moreover, oeUSP18 significantly reduced the apoptosis of HcerEpic cells, whereas LY294002 reversed this suppression (Fig. 4b).

Moreover, oeUSP18 promoted the protein levels of Ki-67 and Cyclin D1, while inhibited the levels of cleaved PARP and Bax. Further, the protein level of cleaved caspase-3 was suppressed in HcerEpic cells transfected with oeUSP18. However, LY294002 remarkably significantly disrupted oeUSP18 fucntion in HcerEpic cells. Importantly, AKT phosphorylation was higher in oeUSP18-transfected cells but significantly suppressed by LY294002 (Fig. 4c).

An equal number of Caski cells, transfected with siNC or siUSP18, was hypodermically injected into nude mice (n = 5 for each group) to investigate USP18’s role in regulating tumorigenicity in vivo. Tumour formation was assessed every 3 days for 33 days (started on day 12). Although both the injected cell types were capable of developing tumours, it was evident that USP18-siRNA cells suppressed the tumour growth rate (Fig. 5a & b). The volume and weight of USP18-siRNA tumours were reduced significantly compared with those of siNC tumours. Moreover, the TUNEL staining assay results indicated that the apoptosis of cells in siUSP18 tumours was substantially higher than that in siNC tumours (Fig. 5c).

Further, the protein content of USP18 was downregulated in siUSP18 tumours. Additionally, the phosphorylation levels of AKT and β-catenin also decreased in siUSP18 tumours (Fig. 5d). Together, these results suggested that USP18 silencing inhibited the tumorigenicity of human cervical cancer cells in vivo.