Increased expression of TRIM44 in the EOC tissues and cell lines
We first examined the mRNA and protein levels of TRIM44 in 5 fresh EOC samples and 3 normal ovarian tissue samples. The results showed that the mRNA and protein expression of TRIM44 was significantly upregulated in the EOC samples compared to the normal noncancerous tissue samples (Figure 1A, B and C; P<0.05). We then determined the TRIM44 expression in five human EOC cell lines: CAOV3, OVCAR3, A2780, HO-8910PM and SKOV3. Western blot analyses revealed that TRIM44 expression was the highest in the A2780 cells (Figure 1D and E; P<0.05). Taken together, these data suggest that TRIM44 may serve as a vital biomarker of EOC.
Lentivirus-mediated RNAi significantly inhibited TRIM44 expression in SKOV3 and A2780 cells
The individual shRNA oligos targeting TRIM44 were validated. The Western blot results indicated that lentiviral vectors with the TRIM44 shRNA sequence can effectively decrease the expression of TRIM44 in SKOV3 and A2780 cells (Figure 2A and B).
Knockdown of TRIM44 suppressed the SKOV3 and A2780 cell proliferation in vitro
To detect whether TRIM44 influenced the malignant properties of EOC cells, we first performed cell proliferation analysis by CCK-8 assays. We found that knockdown of TRIM44 resulted in a significant reduction in cell proliferation (Figure 2C and D). As shown in Figure 2, the growth curve of the shTRIM44-treated cells started to drop from the third day compared with that of the nontreated and shNC-treated cells. The decline reached 54.14% and 76.93% for the shTRIM44-treated SKOV3 and A2780 cells, respectively, on the fifth day compared with the shNC-treated cells (P<0.001).
Knockdown of TRIM44 suppressed the invasion and migration of the SKOV3 and A2780 cells in vitro
In the Transwell invasion assay, we observed that invasion was inhibited in the lenti-shTRIM44-transfected cells. The migration experiment showed that knockdown of TRIM44 reduced the migration of the lenti-shTRIM44 virus-transfected cells (Figure 2E, F and H). The migration of the SKOV3 cells transfected with TRIM44 shRNA was significantly inhibited compared with that of the control cells (0.115 ± 0.002 versus 0.258 ± 0.002, P< 0.001). The migration of the A2780 cells transfected with TRIM44 shRNA was significantly inhibited (0.030 ± 0.003 versus 0.086 ± 0.006, P< 0.001). The invasion experiment showed that knockdown of TRIM44 reduced the invasion of the lenti-shTRIM44 virus-transfected cells (Figure 2E, F and G). The invasion of the SKOV3 cells transfected with TRIM44 shRNA was significantly inhibited compared with that of the control cells (0.100 ±0.004 versus 0.289 ± 0.005, P< 0.001). The invasion of the A2780 cells transfected with TRIM44 shRNA was significantly inhibited (0.308 ±0.007 versus 0.497 ± 0.004, P< 0.001). The above results indicate that lentiviral vectors with the TRIM44 shRNA sequence can effectively inhibit the migration and invasion induced by TRIM44.
Knockdown of TRIM44 induced the apoptosis of SKOV3 and A2780 cells in vitro
We investigated potential cellular apoptosis events during cell death caused by TRIM44 knockdown in the ovarian cancer cell lines SKOV3 and A2780. The results showed that apoptosis increased in both the SKOV3 and A2780 cell lines after transfection with lenti-shTRIM44 (Figure 2I and J). Depletion of TRIM44 caused drastic morphological changes and a substantial decrease in cell number. Taken together, these results clearly indicated that TRIM44 depletion led to apoptotic cell death in the SKOV3 and A2780 cancer cells.
Knockdown of TRIM44 significantly inhibited the colony-forming ability of SKOV3 and A2780 cells
The long-term effect of TRIM44 silencing on the colony-forming ability of SKOV3 and A2780 cells was determined by colony formation assays. The size of the independent colonies was much smaller in the shTRIM44-treated cells than in the nontreated and shNC-treated cells. Moreover, the number of colonies formed by the SKOV3 and A2780 cells was significantly (P<0.05) decreased following TRIM44 knockdown (Figure 2K and L). The data indicated that TRIM44 knockdown also significantly inhibited colony formation in the SKOV3 and A2780 ovarian cancer cells.
Knockdown of TRIM44 inhibited xenotransplanted tumor growth in nude mice in vivo
To evaluate the effect of TRIM44 on ovarian carcinogenesis in vivo, we established xenograft tumor models in nude mice with A2780 cells transfected with lenti-shTRIM44 treatment. Xenograft tumors were observed in the nude mice at the injection site, and the tumors were harvested 30 days after injection. The tumor volumes were measured once a week. As shown in Figure3, the tumor volume in the lenti-shTRIM44-transfected cell treatment group was significantly decreased compared to that in the control group (Figure 4, P < 0.05). The above results indicated that lentiviral vectors with the TRIM44 shRNA sequence can inhibit xenotransplanted tumor growth, indicating that downregulation of TRIM44 may be beneficial for the treatment of ovarian cancer.
Global gene expression analysis following TRIM44 knockdown revealed modulation of key pathways in epithelial ovarian cancer cells
To elucidate the molecular mechanisms by which TRIM44 contributes to the carcinogenesis of lung cancer cells, we carried out gene chip analysis (gene expression profiling, using the Human Gene Expression Array PathArrayTM platform) comparing the gene expression of SKOV3 NS cells versus SKOV3 shTRIM44 cells. Differentially expressed genes with at least a 1.2-fold change were identified. The common differentially expressed genes were analyzed by Ingenuity Pathway Analysis (IPA).We identified 1172 genes with significant expression changes (>1.2-fold with adjusted P< 0.05) (data not shown in this paper). The bioinformatics analysis showed that the expression of 436 genes was significantly increased (P<0.05) and that of 736 genes was significantly decreased (P<0.05). The most markedly altered genes were AKT1S1 (AKT1 substrate 1, decreased1.96-fold), USP14 (ubiquitin specific peptidase 14, decreased1.85-fold), STK38L (serine/threonine kinase 38 like, decreased1.83-fold), GLIPR1 (GLI pathogenesis-related 1, increased1.84-fold), SEC23A (Sec23 homolog A, COPII coat complex component, decreased 1.76-fold), and SPC24 (NDC80 kinetochore complex component, decreased1.72-fold). Many of these TRIM44-regulated differentially expressed genes in EOC cells are involved in regulating cell proliferation, wound healing, the cell cycle, DNA damage, and cellular assembly and organization. In addition, functional analysis of the genes using IPA revealed that differentially expressed genes were mainly enriched in the superpathway of cholesterol biosynthesis (ACAT2, FDPS, IDI1, LBR, LSS and MVD, etc.), the superpathway of serine and glycine biosynthesis I (PHGDH, PSAT1, PSPH, and SHMT2), cyclins and cell cycle regulation (CCNA1, CCNA2, CCNB2, CCND3, CCNE2, CDK1, E2F2,E2F8, etc.), estrogen-mediated S-phase entry (CCNA1, CCNA2, CCNE2, CDK1,E2F2, etc.) the superpathway of geranylger anyldiphosphate biosynthesis I signaling (ACAT2, FDPS, IDI1, MVD, and MVK), the NER pathway (ACAT2, FDPS, IDI1, MVD, and MVK), cell cycle control of chromosomal replication (CDC6, CDK1, CDK10, LIG1, and MCM3,etc.),and oxidative phosphorylation (COX7A1, NDUFAB1, and NDUFB11, etc.). Taken together, these data indicate that TRIM44 may be a key regulator of cell proliferation and migration, therefore controlling cancer progression. Most of these pathways are implicated in tumorigenesis or chemotherapy resistance (all P <0.05; Figure 5).
TRIM44 knockdown suppressed the FOXM1-EZH2 signaling pathway in epithelial ovarian cancer cells
The potential gene network interacting with TRIM44 was also analyzed by IPA and showed that FOXM1-EZH2 signaling was the downstream signaling pathway regulated by TRIM44 (Figure6). Subsequent Western blot assays showed that silencing TRIM44 dramatically downregulated the expression of several key proteins in the FOXM1-EZH2 signaling pathway, including FOXM1, EZH2, CCNE2, CCND3 and BIRC5 (Figure 6, P < 0.05). These data further suggest that TRIM44 promotes human epithelial ovarian cancer progression via the FOXM1-EZH2 signaling pathway.