Hypoxia-Induced miR-675-5p Supports β-Catenin Nuclear Localization by Regulating GSK3-β Activity in Colorectal Cancer Cell Lines

Abstract

The reduction of oxygen partial pressure in growing tumors triggers numerous survival strategies driven by the transcription factor complex HIF1 (Hypoxia Inducible Factor-1). Recent evidence revealed that HIF1 promotes rapid and effective phenotypic changes through the induction of non-coding RNAs, whose contribution has not yet been fully described. Here we investigated the role of the hypoxia-induced, long non-coding RNA H19 (lncH19) and its intragenic miRNA (miR-675-5p) into HIF1-Wnt crosstalk. During hypoxic stimulation, colorectal cancer cell lines up-regulated the levels of both the lncH19 and its intragenic miR-675-5p. Loss of expression experiments revealed that miR-675-5p inhibition, in hypoxic cells, hampered β-catenin nuclear localization and its transcriptional activity, while lncH19 silencing did not induce the same effects. Interestingly, our data revealed that miRNA inhibition in hypoxic cells restored the activity of Glycogen Synthase Kinase 3β (GSK-3β) reducing the amount of P-Ser9 kinase, thus unveiling a role of the miR-675-5p in controlling GSK-3β activity. Bioinformatics analyses highlighted the serine/threonine-protein phosphatases PPP2CA, responsible for GSK-3β activation, among the miR-675-5p targets, thus indicating the molecular mediator through which miR-675-5p may control β-catenin nuclear localization. In conclusion, here we demonstrated that the inhibition of the hypoxia-induced non-coding RNA miR-675-5p hampered the nuclear localization of β-catenin by regulating GSK-3β activity, thus proposing the miR-675-5p as a new therapeutic target for the treatment of colorectal cancer.

Keywords: colorectal cancer; hypoxia; long non-coding H19; miR-675; β-catenin.

Figure 1

CRC cell lines up-regulated miR-675-5p and lncH19 after 18 h of hypoxic stimulation. ELISA assay for HIF-1α performed with nuclear extracts of SW620 (A) and HCT116 (B) after 18 h of hypoxic stimulation, positive control was furnished by the kit and it is the nuclear extract of HeLa cells pretreated with LiCl. Data, expressed as absorbance (ABS) values at 450 nm, are the mean ± SD of three independent experiments. Real-time PCR for VEGF, SAIL1, miR-675-5p and lncH19 in SW620 (C,E) and HCT116 (D,F) stimulated for 18 h in the hypoxic chamber. MiR-675-5p data were normalized for RNU6 (RNA, U6 Small Nuclear 1) while VEGF, SNAIL1, and lncH19 levels were normalized for β-actin, ΔΔCt is expressed as fold of increase (FOI) with respect to normoxic conditions. Data are expressed as the mean ± SD of three independent experiments and p-values are indicated in the graph.

Figure 2

MiR-675-5p regulation (A) Kaplan–Meier curve for overall survival in the cohort of 160 rectal cancer patients. The plot was drawn by using the online Kaplan–Meier plotter tool. Patients were divided into low and high expression groups based on an upper-tertile cut off value of 20. (B) Real-time PCR for miR-675-5p in SW620 (on the left) and HCT116 (on the right) transfected with AntimiR-675-5p, siH19 or relative scrambled negative control and subjected to 18 h of hypoxic stimulation. (C) Real-time PCR for lncH19 in SW620 (on the left) and HCT116 (on the right) transfected with AntimiR-675-5p, siH19 or relative scrambled negative control and subjected to 18 h of hypoxic stimulation. MiR-675-5p data were normalized for RNU6 (RNA, U6 Small Nuclear) while lncH19 levels were normalized for β-actin, ΔΔCt is expressed as fold of increase (FOI) with respect to scrambled negative control. Data are expressed as the mean ± SD of three independent experiments and p-values are indicated in the graph.

Figure 3

Effects induced on β-catenin by miR-675-5p inhibition or lncH19 silencing. Representative images and densitometric analyses of the Western blots for β-catenin on total extract proteins from SW620 (A) and HCT116 (B) transfected with AntimiR-675-5p, siH19 or scrambled negative control and subjected to 18 h of hypoxic stimulation. Data are expressed as the mean ± SD of three independent experiments. (C) Immunofluorescence for β-catenin on SW620 (upper panels) and HCT116 (lower panels) in the different culture conditions. Normoxic cells transfected with scrambled negative control, hypoxic cells transfected with AntimiR-675-5p, siH19 or relative scrambled negative control. β-catenin in green, Hoechst stained nuclei in blue. The blue scale bar is 50 µm (in SW620) while the white one is 20 µm (in HCT116).

Figure 3

Effects induced on β-catenin by miR-675-5p inhibition or lncH19 silencing. Representative images and densitometric analyses of the Western blots for β-catenin on total extract proteins from SW620 (A) and HCT116 (B) transfected with AntimiR-675-5p, siH19 or scrambled negative control and subjected to 18 h of hypoxic stimulation. Data are expressed as the mean ± SD of three independent experiments. (C) Immunofluorescence for β-catenin on SW620 (upper panels) and HCT116 (lower panels) in the different culture conditions. Normoxic cells transfected with scrambled negative control, hypoxic cells transfected with AntimiR-675-5p, siH19 or relative scrambled negative control. β-catenin in green, Hoechst stained nuclei in blue. The blue scale bar is 50 µm (in SW620) while the white one is 20 µm (in HCT116).

Figure 4

Effects induced on β-catenin activity by miR-675-5p inhibition or lncH19 silencing in hypoxic cells. (A) Dual Glo luciferase assay on SW620 (left graph) and HCT116 (right graph) transfected with TOP- Flash or FOP-Flash plasmid in normoxic condition or co-transfected with AntimiR-675-5p, siH19 or scrambled negative control, and subjected to 18 h of hypoxic stimulation. Normoxic cells transfected with TOP-Flash or FOP-Flash plasmid and treated with LiCl were used as positive control of the assay. Data are expressed as the mean ± SD of three independent experiments. The p-values of the differences between TOP vectors are indicated in the graph, while the * indicates the p-values of the difference between TOP and FOP for each condition. * < 0.05, ** < 0.005, *** < 0.0005. (B) Real-time PCR for c-MYC and Cyclin D1 in SW620 (left panels) and HCT116 (right panels) transfected with AntimiR-675-5p, siH19 or relative scrambled negative control and subjected to 18 h of hypoxic stimulation. Gene expression data were normalized for β-actin, ΔΔct is expressed as fold of increase (FOI) with respect to the expression in the scrambled negative control transfected cells. Data are expressed as the mean ± SD of three independent experiments and p-values are indicated in the graph.

Figure 5

MiR-675-5p inhibition favors GSK3β activation in hypoxic cells. Total GSK3β assay in SW620 (A) and HCT116 (B) transfected with AntimiR-675-5p, siH19 or relative negative control and subjected to 18 h of hypoxic stimulation, or transfected with the scrambled negative control and maintained in normoxic condition. Data are expressed as absorbance (ABS) values at 450 nm, and are represented as the mean ± SD of three independent experiments. Total/P-Ser9-GSK3β ELISA assay in SW620 (C) and HCT116 (D) treated as described above. Data of P-Ser9-GSK3β were normalized for total GSK3β and expressed as fold of increase (FOI) with respect to the P-Ser9 GSK3β levels in hypoxic cells transfected with scrambled negative control. Data are expressed as the mean ± SD of three independent experiments and p-values indicated in the graph.

Figure 6

Effect of miR-675-5p on the phosphatases’ mRNA level. Real-time PCR for PP2CA, PP2R2B and PP2R1A in SW620 (A) and HCT116 (B) under normoxic condition or subjected to 18 h of hypoxic stimulation. Gene expression data were normalized for β-actin, ΔΔct is expressed as fold of increase (FOI) with respect to the expression in normoxic cells. Real-time PCR PP2CA, PP2R2B and PP2R1A in SW620 (C) and HCT116 (D) transfected with miR-675-5p mimic or scrambled negative control in normoxic conditions and subjected to 18 h of hypoxic stimulation. Gene expression data were normalized for β-actin, ΔΔct is expressed as fold of increase (FOI) with respect to the expression in scrambled negative control transfected cells. Data are expressed as the mean ± SD of three independent experiments and p-values are indicated in the graph.