MiR-146a expression is elevated and FLAP expression is reduced in hepatocellular carcinoma
To investigate whether miR-146a was involved in HCC, we first examined its expression in 16 human HCC tissues and para-carcinoma tissues. MiR-146a expression was higher in HCC tissues than in para-carcinoma tissues (Fig. 1a). Importantly, there were no differences in miR-146a expression between HCC tissues and para-carcinoma tissues in six cases (37.5%). We also found that FLAP expression was lower in HCC tissues than in para-carcinoma tissues (Fig. 1b, c). Next, we compared miR-146a and FLAP mRNA and protein expression in human normal liver cell line, HL-7702 cells (our experimental control), and three liver cancer cell lines: HepG2, Hep3B, and PLC cells. Consistent with the tissue results, we found that miR-146a expression was higher in HepG2 cells than in HL-7702 cells (Fig. 1d). Additionally, FLAP protein and mRNA were reduced in HepG2, Hep3B, and PLC cells than those in control cells (Fig. 1e).
MiR-146a-5p directly targets FLAP mRNA
The presence of binding sites between the 3'–UTR of FLAP and miR-146a-5p was predicted by TargetScan software (Fig. 2a). Furthermore, qRT-PCR and WB detection showed that the expression level of FLAP was downregulated in HCC tissues and HepG2 cells (Fig. 1b, c, e), which was contrary to the expression trend of miR-146a-5p in HCC tissues and HepG2 cells (Fig. 1a, d). To further verify the structural similarities between FLAP and miR-146a-5p, we constructed a 3'–UTR wild-type (WT) and mutated (MUT) plasmid containing FLAP for dual luciferase reporter gene assay (Fig. 2b). The results of the dual luciferase reporter gene assay showed that, compared with normal control (NC), mimic, miR-146a-5p mimic could significantly down-regulate the luciferase activity of h-ALOX5AP-3'–UTR-WT. Furthermore, miR-146a-5p mimic had no significant effect on the luciferase activity of h-ALOX5AP-3'–UTR-MUT (Fig. 2c). Together, these results demonstrate that miR-146a-5p can directly target FLAP mRNA.
MiR-146a promoted HepG2 cell proliferation, cycle progression, migration, and invasion
Given that miR-146a expression was elevated in HCC, we speculated whether miR-146a promotes HCC cell proliferation. To investigate this, HepG2 cells were stably transfected with control (LV-vector) or Lentivirus packaged miR-146a overexpression vector (LV-miR-146a). QRT-PCR (Fig. 3a) confirmed that the LV-miR-146a substantially elevated miR-146a expression. To investigate the effect of increased miR-146a on FLAP expression, we found that the LV-miR-146a substantially reduced FLAP protein and mRNA expression (Fig. 3b). We re-plated the cells and followed proliferation for 48 h and found that LV-miR-146a significantly increased the proliferation of HepG2 cells (Fig. 3c).
Next, the expression of the cell cycle regulatory proteins were examined: CDK4 (G1 phase), CDK2 (G1 phase), Cyclin E1 (S-phase transition), Cyclin B1 (transition from G2 to M), Cyclin D1 (throughout the cell cycle), p21 (cell cycle inhibitor)[23]. Expression of Cyclins D1, E1, B1, CDK2 and CDK4 was efficiently increased in cells with overexpressed miR-146a, whereas the expression of p21 was reduced (Fig. 3c). These results suggested that high miR-146a expression promotes HepG2 cell proliferation by inducing G2/M cell cycle arrest.
Next, we evaluated the effect of LV-miR-146a on cell apoptosis. Flow cytometry analysis revealed that, LV-miR-146a significantly decreased the apoptosis of HepG2 cells (Fig. 3d).
Metastasis, the main cause of most cancer-related deaths, occurs after cells undergo the epithelial-to-mesenchymal transformation (EMT)[24,25]. To determine whether miR-146a was required for tumor cell migration, we performed wound-healing scratch assays in cells in which miR-146a was overexpressed. At 48 h post-scratch, HepG2 cells expressing high miR-146a were significantly more migratory than LV-vector (Fig. 3e). To evaluate whether miR-146a was required for tumor cell invasion, we performed Matrigel invasion assay in cells with overexpressed miR-146a. Compared to LV-vector, HepG2 cells expressing high miR-146a were significantly more capable to invade through Matrigel (Fig. 3f). Next, we examined the expression of adhesion markers: E-cadherin and vimentin associated with EMT. We found that cells in which miR-146a was overexpressed had lower E-cadherin expression and higher vimentin expression than in LV-vector (Fig. 3f). Together, these results suggest that miR-146a promotes migration and invasion behaviors of HepG2 cells.
Silencing of miR-146a reduced HepG2 cell proliferation, cycle progression, migration, and invasion
We stably transfected HepG2 cells with control (inhibitor NC) or Lentivirus packaged miR-146a knockdown vector (miR-146a inhibitor), to further prove that miR-146a silencing reduced HepG2 cell proliferation. To investigate this, qRT-PCR (Fig. 4a) confirmed that the miR-146a inhibitor substantially knocked down miR-146a expression. Consistent with the results in Figure 3b, we found that the miR-146a inhibitor substantially increased FLAP protein and mRNA expression (Fig. 4b). We re-plated the cells and followed proliferation for 48 h and found that miR-146a inhibitor significantly reduced the proliferation of HepG2 cells (Fig. 4c).
Next, we examined the expression of the cell cycle regulatory proteins: Cyclin D1, Cyclin E1, Cyclin B1, p21, CDK4, and CDK2. Expression of Cyclins D1, E1, B1, CDK2, and CDK4 was efficiently decreased in cells in which miR-146a was knocked down, whereas the expression of p21 was elevated (Fig. 4c). These results suggested that miR-146a silencing reduced HepG2 cell proliferation by inducing G2/M cell cycle arrest.
Next, we evaluated the effect of miR-146a inhibitor on cell apoptosis. Flow cytometry analysis revealed that, miR-146a inhibitor significantly increased the apoptosis of HepG2 cells (Fig. 4d).
To determine whether miR-146a silencing reduced HepG2 cell migration, we performed wound-healing scratch assays in cells with miR-146a knockdown. At 48 h post-scratch, HepG2 cells with miR-146a silencing were significantly less migratory than inhibitor NC (Fig. 4e). To evaluate whether miR-146a silencing reduced HepG2 cell invasion, we performed Matrigel invasion assay in cells in which miR-146a was knocked down. Compared to inhibitor NC, HepG2 cells with miR-146a silencing were significantly less capable to invade through Matrigel (Fig. 4f). Next, we examined the expression of adhesion markers: E-cadherin and vimentin associated with EMT. We found that cells in which miR-146a was knocked down had higher E-cadherin expression and lower vimentin expression than inhibitor NC (Fig. 4f). These results suggest that miR-146a silencing reduced HepG2 cells migration and invasion.
High FLAP expression inhibited HepG2 cell proliferation, cycle progression, migration, and invasion
Given that LV-miR-146a substantially reduced FLAP protein and mRNA expression, and miR-146a promotes HepG2 cell proliferation, it was speculated whether FLAP restrained HepG2 cell proliferation. To investigate this, HepG2 cells were stably transfected with control (LV-vector), or Lentivirus packaged FLAP overexpression vector (LV-FLAP). To investigate this, WB and qRT-PCR (Fig. 5a) confirmed that the LV-FLAP substantially elevated FLAP protein and mRNA expression. We re-plated the cells and followed proliferation for 48 h and found that the LV-FLAP significantly reduced the proliferation of HepG2 cells (Fig. 5b).
Next, we examined the expression of cell cycle regulatory proteins: Cyclin D1, Cyclin E1, Cyclin B1, p21, CDK4, and CDK2. Expression of Cyclins D1, E1, B1, CDK2, and CDK4 was efficiently decreased in cells in which FLAP was overexpressed, whereas the expression of p21 was elevated (Fig. 5b). These results suggested that a high FLAP expression reduced HepG2 cell proliferation by inducing G2/M cell cycle arrest.
Next, we evaluated the effect of LV-FLAP on cell apoptosis. Flow cytometry analysis revealed that LV-FLAP significantly increased the apoptosis of HepG2 cells (Fig. 5c).
To determine whether high FLAP expression reduced HepG2 cell migration, we performed wound-healing scratch assays in cells in which FLAP was overexpressed. At 48 h post-scratch, HepG2 cells with high FLAP expression were significantly less migratory than inhibitor NC (Fig. 5d). To evaluate whether high FLAP expression reduced HepG2 cell invasion, we performed Matrigel invasion assay in cells in which FLAP was overexpressed. Compared to LV-vector, HepG2 cells with high FLAP expression were significantly less capable to invade through Matrigel (Fig. 5e). Next, we examined the expression of adhesion markers: E-cadherin and vimentin associated with EMT. We found that cells in which FLAP was elevated had a higher E-cadherin expression and lower vimentin expression than LV-vector (Fig. 5f). Together, these results suggest that a high FLAP expression reduces HepG2 cells migration and invasion.
MiR-146a promoted HepG2 cell proliferation, cycle progression, migration, and invasion by targeting FLAP mRNA
To further confirm that miR-146a promoted HepG2 cell proliferation, cycle progression, migration, and invasion by targeting FLAP mRNA, FLAP-NC (control) or FLAP-targeted small interfering RNA (siRNA) were transiently transfected to HepG2 cells, and which were stably transfected with miR-146a inhibitor. We called them “HepG2 cells with inhibitor+FLAP-NC” and “HepG2 cells with inhibitor+si-FLAP”, respectively. QRT-PCR confirmed that FLAP-targeted small interfering RNA-2 (si-FLAP-2) had the best interference effect compared with that of the FLAP-NC group (Fig. 6a). Si-FLAP-2 was used in subsequent experiments. WB confirmed that the si-FLAP-2 substantially reduced FLAP protein expression in HepG2 cells, which were stably transfected with miR-146a inhibitor (Fig. 6b). We re-plated the cells and followed proliferation for 48 h and found that the inhibitory effect of miR-146a inhibitor on proliferation of HepG2 cells was neutralized by si-FLAP-2 (Fig. 6c).
Next, we examined the expression of the cell cycle regulatory proteins: Cyclin D1, Cyclin E1, Cyclin B1, p21, CDK4, and CDK2. Expression of Cyclins D1, E1, B1, CDK2, and CDK4 was efficiently increased in HepG2 cells with inhibitor+si-FLAP-2, whereas the expression of p21 was reduced (Fig. 6c). These results suggested that the inhibitory effect of miR-146a inhibitor on proliferation of HepG2 cells was neutralized by si-FLAP-2.
To determine whether miR-146a inhibited HepG2 cell apoptosis by targeting FLAP mRNA, we performed apoptosis assays in HepG2 cells with inhibitor+FLAP-NC and HepG2 cells with inhibitor+si-FLAP-2. Flow cytometry analysis revealed that the promoting of miR-146a inhibitor on apoptosis of HepG2 cells was neutralized by si-FLAP-2 (Fig. 6c).
To determine whether miR-146a promoted HepG2 cell migration by targeting FLAP mRNA, we performed wound-healing scratch assays in HepG2 cells with inhibitor+FLAP-NC and HepG2 cells with inhibitor+si-FLAP-2. At 48 h post-scratch, HepG2 cells with inhibitor+si-FLAP-2 were significantly more migratory than HepG2 cells with inhibitor+FLAP-NC, suggesting that the inhibitory effect of miR-146a inhibitor on migration of HepG2 cells was neutralized by si-FLAP-2 (Fig. 6e). To evaluate whether miR-146a promoted HepG2 cell invasion by targeting FLAP mRNA, we performed Matrigel invasion assay in cells in HepG2 cells with inhibitor+FLAP-NC and HepG2 cells with inhibitor+si-FLAP-2. Compared to HepG2 cells with inhibitor+FLAP-NC, HepG2 cells with inhibitor+si-FLAP-2 were significantly more capable to invade through Matrigel (Fig. 6f). Next, we examined the expression of adhesion markers: E-cadherin and vimentin associated with EMT. We found that HepG2 cells with inhibitor+si-FLAP-2 had lower E-cadherin expression and higher vimentin expression than HepG2 cells with inhibitor+FLAP-NC, suggesting that the inhibitory effect of miR-146a inhibitor on the invasion of HepG2 cells was neutralized by si-FLAP-2 (Fig. 6f). Together, these results suggest that miR-146a promoted HepG2 cell migration and invasion by targeting FLAP mRNA.