Functional mechanism of hsa-miR-128-3p in epithelial-mesenchymal transition of pancreatic cancer cells via ZEB1 regulation

Ethics statement

The experiment was conducted with the approval of the Ethics Committee of Qilu Hospital of Shandong University.

Cell culture

Normal human pancreatic duct epithelial cell line (HPDE6c7) and human PC cell lines (AsPC-1, BxPC-3, CFPAC-1, and PANC-1) with different degrees of malignancy were procured from the cell bank of Chinese Academy of Science (Shanghai, China). All cells were cultured in Dulbecco’s modified Eagle’s medium (DMEM) supplemented with 10% fetal bovine serum (FBS) and 1% penicillin-streptomycin (Procell, Wuhan, Hubei, China) at 37 °C with 5% CO₂ in a humidified incubator (Thermo Forma 3111; Thermo Fisher Scientific, Waltham, MA, USA). The medium was refreshed every two days. The cells were detached with 0.25% trypsin and passaged every 3 days. After 3 passages, the morphology of cells at the logarithmic growth phase was observed under the light microscope (CX23; Olympus, Tokyo, Japan) before further experimentation.

Cell grouping and transfection

PANC-1 and AsPC-1 cells at the logarithmic growth phase were seeded in 6-well plates. Once obtaining 60–70% confluence, the cells were co-transfected with miR-128-3p mimic or miR-128-3p inhibitor, oe-ZEB1 lentivirus, or si-ZEB1 using Lipofectamine 2000 (YT1317; Yita, Beijing, China). miR-128-3p mimic, miR-128-3p inhibitor, oe-ZEB1 lentivirus, si-ZEB1, and their controls were synthesized by Biomics Biotech (Nantong, Jiangsu, China) and used as per the provided instructions. PANC-1 cells were divided into the following groups according to different transfections: the blank group, the miR-mimic NC group, the miR-mimic group, the si-NC group, the si-ZEB1 group, the miR-mimic + oe-NC group, and the miR-mimic + oe-ZEB1 group. AsPC-1 cells were assigned into the following groups according to different transfections: the blank group, the miR-inhi NC group, the miR-inhi group, the oe-NC group, the oe-ZEB1 group, the miR-inhi + si-NC group, and the miR-inhi + si-ZEB1 group. The best transfection condition and concentrations of transfection reagents were confirmed in preparatory experiments to guarantee transfection efficiency. RT-qPCR, Western blot, and further experimentation was conducted after 48-h transfection.

Transwell assay

Cells were harvested after 48-h transfection and made into cell suspension at the concentration of at 1 × 10⁵ cells/mL. Then, 200 µL cell suspension was slowly seeded in the apical chamber coated or not coated by Matrigel (60 mg/well, BD Bioscience, Franklin Lakes, NJ, USA) and 700 µL medium containing 10% FBS was supplemented to the basolateral chamber. Both chambers were placed in a 37 °C incubator with 5% CO₂ and saturated humidity. The non-invading and non-migrating cells in the apical chamber were removed using cotton swabs. Following 48-h incubation, cells were fixed with 4% paraformaldehyde and stained for 10 min with 0.1% crystal violet. The number of cells passing through the membrane pores was counted. Three random visual fields of each well were selected for observation. The invasion and migration rates of PC cells in different groups were determined.

Dual-luciferase assay

The binding sites of miR-128-3p and ZEB1 were predicted on bioinformatics online software TargetScan (http://www.targetscan.org). The Psicheck reporter gene detection plasmids (GenePharma, Shanghai, China) of ZEB1-WT (wild-type) and ZEB1-MUT (mutant) were constructed and co-delivered into PANC-1 cells with miR-128-3p mimic, and into AsPC-1 cells with miR-128-3p inhibitor in strict conformity with the instructions. Following 48-h transfection, relative fluorescence intensities of firefly luciferase and ranilla luciferase were detected.

Reverse transcription-quantitative polymerase chain reaction (RT-qPCR)

Cells at the logarithmic growth phase were harvested and total RNA was extracted using RNAsimple Total RNA kits (R401-01; Vazyme, Nanjing, Jiangsu, China) and reverse-transcribed into cDNA using Taqman^® MicroRNA Reverse Transcription kit (Invitrogen, Carlsbad, CA, USA). PCR was performed on the ABI Prism 7300 system (ABI, Foster City, CA, USA) under the following conditions: pre-denaturation at 95 °C for 10 min and 40 cycles of 90 °C for 10 s, 60 °C for 20 s, and 72 °C for 30 s. U6 or β-actin served as internal controls. Relative gene expressions were calculated with the 2^−ΔΔCT method. The amplified primer sequences were synthesized by Sangon Biotech (Shanghai, China) and illustrated in Table 1.

Western blot

Cells at the logarithmic growth phase were collected and lysed with radio immunoprecipitation assay lysis buffer (SS0501; SORFA, Beijing, China) to extract the total protein. The protein concentration was quantified using BCA kits (PC0020-500, Acmec, Shanghai, China). The protein samples were denatured by boiling, separated using sodium dodecyl sulfate-polyacrylamide gel electrophoresis, electrophoretically transferred onto the polyvinylidene fluoride membranes, and blocked with 5% skim milk for 1 h. Next, the membranes were incubated with rabbit anti-human primary antibodies anti-E-cadherin (1:10000, ab40772; Abcam, Cambridge, MA, USA), anti-N-cadherin (1:1000, ab207608; Abcam), anti-ZEB1 (1:1000, ab203829; Abcam), and β-actin (1:1000, ab8227, Abcam) at 4 °C overnight. After washing, the membranes were probed with HRP-labeled goat anti-rabbit secondary antibody IgG H&L (1:300, ab7090; Abcam) for 1 h. The membranes were visualized with enhanced chemiluminescence (ECL) and photographed. The gray value of the protein bands was analyzed using the Image-Pro 6.0 software with β-actin as the internal control.

Cell fluorescence staining

Cell climbing sheets were prepared, placed in the culture plate, washed thrice with PBS, fixed in 4% paraformaldehyde for 15 min, and cleared with 0.1% Triton X-100 for 10 min. After three PBS washes, the sheets were blocked with 10% goat serum for 30 min and incubated with primary antibodies E-cadherin (1:10000, ab40772), N-cadherin (1:200, ab18203) at 4 °C overnight. Following three PBS washes, the sheets were incubated with secondary antibody Alexa Fluor 488 (1:1000, ab150077) and counterstained with Alexa Fluor 594 (1:200, ab195889) at room temperature for 2 h under conditions devoid of light. After washing with PBS three times, DAPI was used for nuclear counterstaining. The sheets were sealed with an anti-fluorescence quenching sealing liquid (P0131-5mL; Beyotime, Shanghai, China) after three PBS washes. Images were captured under a fluorescence microscope (CKX53, Olympus).

Statistical analysis

Data were processed and plotted using SPSS 21.0 (IBM Corp., Armonk, NY, USA) and GraphPad Prism8.0 (GraphPad Software Inc., San Diego, CA, USA) software. Data were described as mean ± standard deviation. Comparisons among multi-groups were analyzed using one-way analysis of variance (ANOVA) following by Tukey’s multiple comparisons test. The p value was obtained using two-sided tests. Differences were regarded as statistical significance with p < 0.05, and ns meant no significant difference.

hsa-miR-128-3p was weakly expressed while ZEB1 was highly expressed in different PC cells

According to the ENCORI database (https://starbase.sysu.edu.cn/), low expression of miR-128-3p is an indicator of poor prognosis of PC patients (Fig. 1A). It’s been reported that EMT is a pivotal step in the invasion and metastasis of PC cells (Li et al., 2020; Rhim et al., 2012; Zhou et al., 2017). ZEB1 is a key transcriptional factor in EMT activation (Banerjee et al., 2021; Garinet et al., 2021). There is evidence that miR-128-3p inhibits metastasis and EMT in esophageal squamous-cell cancer via targeting ZEB1 (Zhao et al., 2018). Therefore, it is not unreasonable to assume a similar functional mechanism in PC. Four PC cell lines (PANC-1, CFPAC-1, BxPC-3, and AsPC-1) with different degrees of malignancy and normal PC cell line HPDE6c7 were selected to investigate the involvement of hsa-miR-128-3p and ZEB1 in EMT of PC cells. AsPC-1 cells and BxPC-3 cells were round- or ellipse-shaped with tight intercellular junctions and obvious cell polarity, while CFPAC-1 cells and PANC-1 cells showed a fusiform shape with loose intercellular junctions and loss of cell polarity (Fig. 1B). Western blot demonstrated increased E-cadherin (epithelial marker) and decreased N-cadherin (mesenchymal marker) in AsPC-1 and BxPC-3 cells, while the opposite was found in CFPAC-1 and PANC-1 cells (Fig. 1C, all p < 0.05). These results suggested that AsPC-1 cells and BxPC-3 cells were PC cell lines with epithelial phenotypes while CFPAC-1 cells and PANC-1 cells were with mesenchymal phenotypes.

RT-qPCR and Western blot showed decreased miR-128-3p expression (Fig. 1D), and increased ZEB1 expression in AsPC-1, BxPC-3, CFPAC-1 and PANC-1 cells relative to that in normal HPDE6c7 cells (Figs. 1E–1F, all p < 0.05). Among the PC cell lines, miR-128-3p expression was increased while ZEB1 expression was decreased in AsPC-1 cells with epithelial phenotypes relative to those in PANC-1 cells with mesenchymal phenotypes (all p < 0.05). Transwell assay showed stronger invasive and migrative abilities in PANC-1 cells relative to those in AsPC-1 cells (Fig. 1G, all p < 0.05). The above results demonstrated that hsa-miR-128-3p was poorly expressed while ZEB1 was highly expressed in PC cell lines and might be related to the EMT and malignant biological behaviors of PC cells.

hsa-miR-128-3p overexpression partially inhibited the EMT, invasion, and migration of PC cells

To further validate the involvement of hsa-miR-128-3p in the EMT of PC cells, PANC-1 cells with relatively less expressed miR-128-3p and mesenchymal phenotypes and AsPC-1 cells with relatively more expressed miR-128-3p and epithelial phenotypes were selected for further experimentation. miR-128-3p mimic was delivered into PANC-1 cells to overexpress miR-128-3p and miR-128-3p inhibitor was delivered into AsPC-1 cells to downregulate miR-128-3p. The result showed upregulated expression of miR-128-3p in PANC-1 cells and downregulated expression of miR-128-3p in AsPC-1 cells after transfection (Fig. 2A, all p < 0.001), indicating successful transfection. PANC-1 cells transformed from mesenchymal phenotypes to epithelial phenotypes after miR-128-3p overexpression, while AsPC-1 cells transformed from epithelial phenotypes to mesenchymal phenotypes after miR-128-3p downregulation (Fig. 2B). Furthermore, Western blot showed that E-cadherin was increased and N-cadherin was decreased in PANC-1 cells after transfection with miR-128-3p mimic, while the opposite trend was found in AsPC-1 cells after transfection with miR-128-3p inhibitor (Figs. 2C–2D, all p < 0.001). The consistent tendency was found in cell fluorescence staining (Fig. 2E). Briefly, hsa-miR-128-3p overexpression inhibited the EMT of PC cells while downregulation of hsa-miR-128-3p facilitated the EMT of PC cells.

A close association between EMT and invasion and migration of PC cells has been documented (Sun, Chen & Dong, 2021; Xu et al., 2021). To further understand whether miR-128-3p affects invasion and migration of PC cells by manipulating the EMT process, we examined the effect of miR-128-3p on PC cell invasion and migration by Transwell assay, which showed decreased invasion and migration of PANC-1 cells transfected with miR-128-3p mimic, and increased invasion and migration of AsPC-1 cells transfected with miR-128-3p inhibitor (Fig. 2F, all p < 0.05). These results suggested that hsa-miR-128-3p overexpression repressed PC cell migration and invasion, whereas down-regulation of hsa-miR-128-3p enhanced PC cell migration and invasion.

Silencing ZEB1 inhibited malignant biological behavior of PC cells

To explore the function of ZEB1 on EMT in PC, PANC-1 cells with mesenchymal phenotypes and higher ZEB1 expression were transfected with si-ZEB1, whereas AsPC-1 cells with epithelial phenotypes and lower ZEB1 expression were transfected with oe-ZEB1. RT-qPCR and Western blot were used to confirm the transfection efficiency (Figs. 3A–3B, all p < 0.05). Silencing ZEB1 promoted the transformation of PANC-1 cells from mesenchymal phenotype to epithelial phenotype, while ZEB1 overexpression facilitated the transformation of AsPC-1 cells from epithelial phenotype to mesenchymal phenotype (Fig. 3C). According to Western blot, E-cadherin was increased and N-cadherin was decreased in PANC-1 cells transfected with si-ZEB1, whereas E-cadherin was decreased and N-cadherin was increased in AsPC-1 cells transfected with oe-ZEB1 (Figs. 3D–3E, all p < 0.001). The same tendency was observed in cell fluorescence staining (Fig. 3F). These results together illustrated that silencing ZEB1 could suppress EMT of PC cells, while ZEB1 overexpression could facilitate EMT of PC cells.

To further investigate whether ZEB1 could affect PC cell invasion and migration via modulation of the EMT process, we detected the effect of ZEB1 on PC cell invasion and migration by Transwell assay, which exhibited decreased invasive and migrative abilities of PANC-1 cells transfected with si-ZEB1, and increased invasive and migrative abilities of AsPC-1 cells transfected with oe-ZEB1 (Fig. 3G, all p < 0.05). The above results indicated that silencing ZEB1 inhibited PC cell migration and invasion, whereas ZEB1 overexpression enhanced PC cell migration and invasion.

hsa-miR-128-3p targeted ZEB1

Based on preliminary experimental results and background information, we speculated that miR-128-3p could target ZEB1. To testify our hypothesis, the binding sites of miR-128-3p and ZEB1 3′ UTR were predicted on the TargetScan database (http://www.targetscan.org), and the binding sequence was CACUGUG (Fig. 4A). The target relationship of miR-128-3p and ZEB1 was validated by a dual-luciferase assay, and miR-128-3p mimic in PANC-1 cells repressed the relative fluorescence intensity of ZEB1-WT reporter gene (Fig. 4B, p < 0.01), while miR-128-3p inhibitor in AsPC-1 cells enhanced the fluorescence intensity of ZEB1-WT (p < 0.001), but no apparent difference was observed in that of ZEB1-MUT. RT-qPCR and Western blot showed that ZEB1 was downregulated in PANC-1 cells after transfection with miR-128-3p mimic (Fig. 4C, all p < 0.001), and upregulated in AsPC-1 cells after transfection with miR-128-3p inhibitor (Fig. 4D, all p < 0.001). These results elicited that miR-128-3p targeted ZEB1.

ZEB1 overexpression antagonized the inhibition of hsa-miR-128-3p overexpression on the EMT of PC cells

To identify the function of miR-128-3p on the EMT of PC cells via ZEB1, PANC-1 cells were transfected with oe-ZEB1 and miR-128-3p mimic, and AsPC-1 cells with si-ZEB1 and miR-128-3p inhibitor for 48 h. Western blot showed upregulated ZEB1 level in PANC-1 cells in the miR-mimic + oe-ZEB1 group (Fig. 5A, all p < 0.001), and downregulated ZEB1 level in AsPC-1 cells in the miR-inhi + si-ZEB1 group (Fig. 5B, all p < 0.001), indicating successful transfection of oe-ZEB1 or si-ZEB1. Furthermore, ZEB1 overexpression partially averted the inhibition of miR-128-3p overexpression on EMT and invasive and migrative abilities of PC cells. Conversely, silencing ZEB1 partially reversed the promotional effect of miR-128-3p downregulation on EMT and invasive and migrative abilities of PC cells (Figs. 5C–5G, all p < 0.001).