DHX32 is upregulated in HCC cells and predicts poor survival in HCC patients
The expression levels of DHX32 in human HCC cells and human immortalized normal liver LO2 cells were assessed by RT-PCR and Western blot assays. We found that the expression levels of DHX32 mRNA and protein were significantly higher in five HCC cell lines, including HepG2, Hep3B, Huh7, SNU-181, and SNU-387 cell, than those in human normal immortalized liver cell LO2 (Figure 1, A and B). In addition, we also found that high level of DHX32 expression had a negative correlation with overall survival in patients with HCC using Kaplan-Meier Plotter (Figure 1C). Together, these data indicate that DHX32 might serve as a contributor to HCC progression.
Ectopic expression of DHX32 induces EMT and enhances the migration, invasion, and proliferation of HCC cells
To explore whether ectopic expression of DHX32 promote HCC progression, the migration, invasion, and proliferation of HCC cells were tested. Huh7 cells after the indicated transfections were performed with RT-PCR assay, Western blot assay, wound-healing assay, Transwell invasion assay, and EdU cell proliferation assay. We confirmed that the expression levels of DHX32 mRNA and protein were upregulated in Huh7 cells transfected with DHX32 lentiviral particles compared with cells transfected with vector lentiviral particles (Figure 2, A and B). Epithelial mesenchymal transition is one of the key regulators of invasion and metastasis in HCC [14]. We found that the overexpression of DHX32 induced EMT in HCC cells, leading to the upregulation in the expression of mesenchymal markers N-cadherin and vimentin and the reduction in the expression of epithelial marker E-cadherin (Figure 2, B and C). Then, the effect of DHX32 on HCC cell migration and invasion was examined. Wound-healing assay revealed that DHX32 overexpression increased the migration capacity of Huh7 cells and facilitated the closure of wound width of cell monolayers (Figure 2D). We also observed that the overexpression of DHX32 significantly increased the number of invasive HCC cells (Figure 2E). Moreover, we also detected the effect of DHX32 on HCC cell proliferation. EdU cell proliferation assay showed that ectopic expression of DHX32 promoted HCC cell proliferation, as evidenced by much increase in the number of EdU-positive Huh7 cells (Figure 2F). Taken together, our results suggest that the overexpression of DHX32 induces EMT and promotes the mobility and growth of HCC cells.
The knockdown of DHX32 reverses EMT and inhibits the migration, invasion, and proliferation of HCC cells
Next, we further investigated the effect of DHX32 knockdown on HCC progression. The expression of DHX32 in Huh7 cells were stably silenced with its specific shRNA lentiviral particles. RT-PCR and Western blot assays showed that the expression of DHX32 was dramatically downregulated in Huh7 cells infected with DHX32 shRNA lentiviral particles (Figure 3, A and B). Then, whether inhibition of DHX32 suppressed EMT in HCC cells was determined. We found that DHX32 shRNA dramatically increased E-cadherin mRNA and protein expression levels, while decreased the expression of N-cadherin and vimentin in Huh7 cells, which indicated that DHX32 shRNA inhibited EMT (Figure 3, B and C). Wound-healing assay showed that silencing DHX32 inhibited the closure of wound width of Huh7 cell monolayers (Figure 3D). Transwell invasion assay revealed that the knockdown of DHX32 remarkably reduced the invasive capacity of HCC cells (Figure 3E). In addition, we found that DHX32 shRNA significantly decreased the number of EdU-positive Huh7 cells compared with cells transfected with control shRNA (Figure 3F). Thus, these data suggest that silencing DHX32 suppresses EMT and inhibits the migration, invasion, and proliferation of HCC cells.
DHX32 regulates the activation of β-catenin pathway in HCC cells
We next investigated the mechanisms of DHX32-medaited EMT and aggressiveness in HCC cells. Since various signaling pathways are involved in EMT and tumor progression, we traced β-catenin pathway. We found that the overexpression of DHX32 increased CTNNB1 mRNA expression in Huh7 cells (Figure 4A). Then, we performed Western blot assay to determine whether DHX32 increased the expression β-catenin in nucleus, which is responsible for the activation of β-catenin pathway. We found that the expression of β-catenin in nucleus was much higher in Huh7 cells after transfection with DHX32 lentiviral particle than vector (Figure 4B). RT-PCR assay revealed that ectopic expression of DHX32 upregulated the mRNA expression of CCND1, COX2, and MMP7 that were the target genes of β-catenin pathway (Figure 4C). DHX32 overexpression also reduced the expression of WIF1, a negative regulator of Wnt/β-catenin pathway (Figure 4C). Moreover, we further determined whether DHX32 knockdown inactivated β-catenin pathway in Huh7 cells. We found that silencing DHX32 downregulated the expression of CTNNB1 mRNA and decreased the expression of β-catenin in nucleus of Huh7 cells (Figure 4, D and E). In addition, RT-PCR assay further confirmed that the knockdown of DHX32 downregulated the mRNA expression of CCND1, COX2, and MMP7, while upregulated the expression of WIF1 mRNA in HCC cells (Figure 4F). Together, these findings suggest that DHX32 could activate β-catenin pathway in HCC cells via promoting the expression of β-catenin in nucleus.
β-catenin siRNA abrogates DHX32-induced EMT, migration, invasion, and proliferation in HCC cells
Then, we further determined the role of β-catenin in DHX32-mediated mobility and growth in HCC cells. Huh7 cells with stably overexpression of DHX32 were transfected with β-catenin siRNA or control siRNA. RT-PCR and Western blot assays revealed that DHX32 lentiviral particle significantly upregulated the mRNA and protein expression of β-catenin in Huh7 cells, which was downregulated in cells after co-transfection with DHX32 lentiviral particle plus β-catenin siRNA (Figure 5, A and B). Then, we detected whether DHX32-induced EMT in HCC cells can be reversed by β-catenin siRNA. We found that β-catenin siRNA reversed DHX32-induced EMT in HCC cells, and led to the increase in E-cadherin expression and the downregulation of N-cadherin and vimentin expression (Figure 5, C and D). Expectedly, we found that β-catenin siRNA decreased the ability of DHX32 to increase the proliferation of Huh7 cells and resulted in a significant decrease in the number of EdU-positive cells (Figure 5E). Wound-healing assay revealed that the closure of wound width was attenuated in cells after co-transfection with DHX32 lentiviral particle plus β-catenin siRNA, when compared to cells after transfection with DHX32 lentiviral particle (Figure 5F). In addition, we also found that β-catenin siRNA reduced the activity of DHX32 to promote the invasion capacity of HCC cells and decreased the number of invasive cells (Figure 5G). Taken together, these results demonstrate that DHX32-mediated HCC progression is regulated by β-catenin pathway.
Inhibition of DHX32 suppresses HCC tumor growth
Next, we tested the effect of DHX32 on HCC tumor growth. Huh7 cells after the indicated transfections were subcutaneously injected into the flank of male Balb/c Nude mice and tumor sizes were measured every five day for 4 weeks. Compared to Huh7 cells transfected with control shRNA, silencing DHX32 impaired tumor growth (Figure 6A) and decreased the weight of Huh7 xenograft tumors (Figure 6B). Furthermore, we found that ectopic expression of DHX32 promoted HCC tumor growth in comparison with vector-transfected groups (Figure 6, C and D). Therefore, our in vivo study indicates that suppression of DHX32 blocks HCC tumor growth.