3.1. In silico analysis of the differential expression of YAP1 and p63 in ADC and SCC
The analysis of the microarray data suggested that the expression of Tp63 in SCC was substantially greater than that in ADC. According to the results, the expression of Tp63 was significantly greater in SCC than in ADC. Furthermore, the expression level of YAP1 was marginally lower in the SCC than in the ADC (Fig. 1A and B). The results of gene expression studies in Caski and HeLa cell lines confirmed the in silico results, showing higher and lower expression of Tp63 and YAP1, respectively, in Caski cells than in HeLa cells (Fig. 1C). miR-141-3p target prediction was then conducted using the mirDIP database. The results indicate that, within the set of crucial regulators involved in EMT, both ZEB1 and YAP1 are highly significant targets of miR-141-3p. The network depicted in Fig. 1D illustrates the close associations between miR-141-3p, YAP1, p63, and the regulators of EMT. These findings strongly indicate that aberrant expression patterns of miR-141-3p, YAP1, and p63 can impact EMT.
3.2. Dual role of ΔNp63 in regulating proliferation, migration, and invasion in ADC HeLa and SCC CaSki cells
The differential expression of the TP63 and YAP1 genes in cervical SCC and ADC suggested that p63 and YAP1 play opposing roles in the progression of these two types of cancer (see Fig. 1). The qRT‒PCR results showed that the expression of the TAp63 isoform was undetectable in HeLa and CaSki cells. In addition, several studies have reported that ΔNp63 is the main isoform of p63 that acts as both an oncogene and a tumor suppressor, especially during EMT progression [10, 11, 25-27]; therefore, we examined the effect of ΔNp63 downregulation on the proliferative, migratory, and invasive capacities of HeLa and CaSki cells. ΔNp63 silencing by sh-RNA was performed in both cell lines. Cells were transfected with either ΔNp63 sh-RNA or empty vector as a negative control. qRT‒PCR revealed an 8-fold decrease in ΔNp63 expression compared to that in control cells (Fig. 2A). Next, MTT and colony formation assays were performed to evaluate the impact of ΔNp63 knockdown on cell growth. The results indicated no significant difference between sh-ΔNp63-treated and control HeLa cells; however, an increase in colony size was observed. In contrast, in CaSki cells, ΔNp63 knockdown resulted in a decrease in cell proliferation with no difference in colony size between the control and sh-ΔNp63 groups (Fig. 2B and C).
Transwell assays demonstrated that the migratory and invasive abilities of CaSki cells were dramatically greater in the sh-ΔNp63-transfected group than in the control group. In contrast to CaSki cells, ΔNp63 downregulation decreased the migration and invasion of HeLa cells (Fig. 2D and E). Previous studies have also highlighted the contradictory tumor-suppressing or oncogenic roles of p63 in different cancer types [13, 25, 27]. Our results confirmed the dual role of ΔNp63 in various cervical cancer cell lines with distinct origins.
3.3 Pro- and antimetastatic effects of miR-141-3p in cervical cancer cell lines
To elucidate the function of miR-141-3p in the progression of ADC and SCC, we examined the proliferation, migration, and invasion abilities of HeLa and CaSki cells after transfection with miR-141-3p mimics and inhibitors. First, the transfection of miR-141-3p mimics and inhibitors was successfully confirmed by qRT‒PCR (Fig. 3A). According to the MTT results, the proliferation rate of HeLa cells after transfection with the miR-141-3p mimic at 48, 72, and 96 h was dramatically greater than that of the control group (Fig. 3. B and C). In addition, an increase in the size and number of colonies was observed after upregulation of miR-141-3p in these cells. As depicted in Fig. 3D and E, the migratory and invasive properties of HeLa cells were dramatically increased after miR-141-3p mimic transfection. However, the results of these experiments in CaSki cells were opposite to those in HeLa cells, demonstrating that the upregulation of miR-141 led to a decrease in the proliferation, colony formation (Fig. 3C), migration, and invasion of CaSki cells (Fig. 3D and E) compared to those in the control group. Accordingly, downregulation of miR-141-3p using miR-141-3p inhibitors suppressed the proliferation (Fig. 3B and C), migration, and invasion of HeLa cells (Fig. 3D and E); however, the proliferation (Fig. 3B and C), migration, and invasion of CaSki cells were significantly greater than those in the control group (Fig. 3D and E). These results confirmed the contradictory roles of miR-141-3p in suppressing and/or promoting EMT in cervical cancer cell lines with distinct origins, as previously demonstrated in some studies [21, 28, 29].
3.4 miR-141-3p upregulation reverses the effects of ΔNp63 knockdown on migration and invasion
Previously, we showed that decreased expression of ∆Np63 led to an increase in the migration and invasion ability of CaSki cells, while the opposite effects were observed in HeLa cells, showing that decreased expression of ∆Np63 reduced the migration and invasion abilities of HeLa cells compared to those in the control group (see Fig. 2). To investigate the role of miR-141-3p as a downstream effector of ∆Np63 and whether increased expression of miR-141-3p can reverse the effects of ∆Np63 knockdown, cotransfection experiments in which sh-∆Np63 and the miR-141-3p mimic were simultaneously transfected into cells were performed. Interestingly, the increased level of miR-141-3p compensated for the decreased expression of ∆Np63 and resulted in decreased proliferation (Fig. 4B and C), migration, and invasion (Fig. 4D and E) in CaSki cells. The same experiment in HeLa cells demonstrated an increase in proliferation, migration, and invasion in these cells (Fig 4). These observations confirmed the role of miR-141-3p as an intermediate player in the function of ∆Np63, as maintaining miR-141-3p at high levels neutralizes the effect of ∆Np63, indicating that without intermediating miR-141-3p, ∆Np63 cannot exert its tumor suppressor or oncogenic effects.
3.5. ΔNp63 targets YAP1 via miR-141-3p
p63 functions as a regulator of a wide variety of microRNAs in different types of cancers [27, 30, 31]. An integrative data mining approach identified p63 as a key regulator of microRNAs with different expression profiles in ovarian carcinomas, including miR-141-3p [19]. By using three distinct miRNA target prediction databases (TargetScan, miRWalk, and MIRDIP), YAP1 was identified as a potential target of miR-141-3p. To verify the functional relationships among ΔNp63, miR-141-3p, and YAP1, we performed a series of transfection experiments with sh-ΔNp63, the miR-141-3p inhibitor, the miR-141 mimic, and the sh-ΔNp63/miR-141 mimic in HeLa and CaSki cells. sh-NC, inh-NC (inhibitor-negative control), and mi-NC (mimic-negative control) were used as negative controls. The transfection efficiency was validated by qRT‒PCR after 48 h. First, knockdown of ΔNp63 by sh-RNA in both CaSki and HeLa cells resulted in a significant decrease in miR-141-3p expression compared to that in empty vector-transfected cells (Fig. 5A). Interestingly, in CaSki cells, the YAP1 expression level was obviously elevated in response to sh-ΔNp63 treatment (Fig. 5A). In contrast, YAP1 was slightly downregulated in HeLa cells in response to ΔNp63 knockdown (Fig. 5B). Subsequently, transfection of CaSki cells with the miR-141-3p inhibitor or mimic resulted in increased or decreased YAP1 expression, respectively, at both the mRNA and protein levels (Fig. 5 A and C). Interestingly, miR-141-3p inhibitor and mimic transfection had opposite effects on HeLa cells, as the expression of YAP1 decreased slightly in response to miR-141-3p knockdown and increased upon miR-141-3p overexpression (Fig. 5B and C). Finally, to investigate whether the regulatory effect of ΔNp63 on YAP1 expression is mediated by miR-141-3p, we cotransfected these two cell lines with sh-ΔNp63/miR-141-mimic. As presented in Fig. 5A and B, the up-/downregulation of YAP1 due to ΔNp63 knockdown in CaSki and HeLa cells was reversed by miR-141-3p overexpression. Thus, these results suggest that miR-141-3p acts as an intermediate regulator between YAP1 and ΔNp63. These experimental results were also confirmed by in silico studies, as mentioned previously (see Fig. 1D). Notably, while the expression of YAP1 is highly regulated by this axis (ΔNp63-miR-141-3p-YAP1) in CaSki cells, the expression of YAP1 is not merely regulated by ΔNp63-miR-141-3p in HeLa cells. Thus, other upstream effectors other than p63-miR-141-3p might regulate the expression of YAP1 in HeLa cells. The results also demonstrated a positive correlation between p63-miR-141-3p and YAP1 expression in HeLa cells, whereas a negative correlation was observed in CaSki cells.
3.6. Regulation of EMT progression by the ∆Np63-miR-141-3p-YAP1 axis in different cervical cancer cell lines
To further investigate the influence of this regulatory axis on the progression of EMT, we analyzed the expression of epithelial and mesenchymal markers by western blotting. As depicted in Fig. 6A, reduced levels of E-cadherin (an epithelial marker) and increased levels of vimentin (a mesenchymal marker) were observed in sh-∆Np63- and miR-141-3p inhibitor-transfected CaSki cells. In contrast, we observed the opposite results in the miR-141-3p mimic- and sh-∆Np63-miR-141-3p mimic-transfected groups of these cells (Fig. 6A). These findings reinforced the findings of the migration and invasion experiments in CaSki cells. Morphological changes in CaSki cells 3 days after sh-∆Np63 transfection revealed an elongated spindle shape and weakened cell–cell adhesion (Fig. 6B), which was consistent with the loss of E-cadherin. As predicted, cotransfection of cells with the sh-∆Np63/miR-141-3p mimic prevented the acquisition of mesenchymal characteristics (Fig. 6B). In more aggressive ADC HeLa cells, the expression patterns of EMT markers (E-cadherin and vimentin) differed from those in CaSki cells. Downregulation of ∆Np63 and miR-141-3p, which are positive regulators of YAP1 in HeLa cells, led to increased levels of E-cadherin and decreased levels of vimentin. Conversely, transfection of the miR-141-3p mimic and cotransfection of the sh-∆Np63/1miR-141-3p mimic resulted in decreases in E-cadherin and vimentin at the protein level (Fig. 6A). From a morphological perspective, CaSki cells exhibited significant morphological changes after transfection with sh-∆Np63 and miR-141-3p inhibitors and after cotransfection with the sh-∆Np63/miR-141-3p mimic (Fig. 6B), while no significant morphological changes were observed in HeLa cells after transfection, perhaps because these cells are more aggressive and have a natural spindle shape.