MiR-34b promotes oxidative stress and induces cellular senescence through TWIST1 in human cervical cancer

Abstract

Purpose: The aim of this research was to elucidate the role of miR-34b in cervical cancer progression and the underlying mechanism behind the miR-34b-mediated tumor suppression. The study revealed the role of miR-34b as a senescence inducer and serves as a potential therapeutic target in developing combination therapy with senotherapeutics.

Methods: MiR-34b was ectopically expressed in cervical cancer cell lines using a tetracycline inducible system and its effects on cell viability, apoptosis, senescence, DNA damage and oxidative stress were studied using MTT assay, acridine orange/ ethidium bromide staining, senescence associated β-galactosidase assay, gamma H2AX foci staining assay, western blotting and specific dyes for the detection of total and individual ROS species.

Results: Ectopic expression of miR-34b promoted cellular senescence but no significant induction of apoptosis was observed in cervical cancer cell lines. MiR-34b promoted increase in oxidative stress through increase in total and individual ROS species and contributed to increase in cellular senescence. Mechanistically, miR-34b mediates its action by targeting TWIST1 as evidenced by the similar actions of TWIST1 shRNA in cervical cancer cell lines. Furthermore, our study revealed TWIST1 is one of the most significant targets of miR-34b targetome and identified RITA as a novel senolytic agent for use in combination therapy with miR-34b.

Conclusion: MiR-34b promotes cellular senescence and oxidative stress by targeting TWIST1, a known oncogene and EMT regulator. This study delved into the mechanism of miR-34b-mediated tumor suppression and provided novel insights for development of miR-34b based therapeutics for cervical cancer.

Keywords: Cervical cancer; MiR-34b; Oxidative stress; Senescence; TWIST1.

Graphical abstract

Fig. 1

Higher expression of miR-34b correlates with patient survival in CESC and UCEC, and stably induced ectopic miR-34b expression decreases cell viability without apoptosis induction a) Survival analysis of miR-34b across 6 different cancers b) Expression of miR-34b on doxycycline induction at different time points relative to RNU6 quantified by qRT-PCR (n = 3) c) Representative image of GFP expressing cells on induction of miR-34b expression by doxycycline in cervical cancer cell lines d) Relative basal expression levels of miR-34b in different cervical cancer cell lines (n = 3) e) Percentage change in cell viability on ectopic expression of miR-34b in different cell lines (n = 3) f) Representative BF and AO/EB staining merged(GFP/TRITC filters) images of control and miR-34b expressing cells g) Percentages of live, necrotic and apoptotic populations quantified from acridine orange/ ethidium bromide immunofluorescence staining in control and miR-34b expressing cell lines h) Immunoblotting images of total and cleaved caspase3 and PARP in positive control for apoptosis, control and miR-34b expressing cells with β-actin as loading control (numbers indicate molecular weight (M.Wt) in kilodalton or kDa). Values are depicted as mean± SD and p-values are depicted as * p ≤ 0.05 ** p ≤ 0.01 *** p ≤ 0.001.

Fig. 2

MiR-34b promotes cellular senescence in cervical cancer cells a) Bright-field images of control and miR-34b expressing cells after day 5 and day 7 post maximal induction of miR-34b showing positive SA-β gal staining b) Quantification of total positively stained cells in each group (n = 4) c) Bright field, DAPI stained nucleus and FITC stained γH2AX foci representative images of control and miR-34b expressing cells at 2 h and 24 h post maximal induction of miR-34b d) Quantification of corrected total cell fluorescence (CTCF) of γH2AX foci (n = 4) e) Immunoblotting images of p53, p21 and lamin A/C in control and miR-34b expressing cells with β-actin as loading control. Values are depicted as mean± SD, and p-values are depicted as ** p ≤ 0.01 *** p ≤ 0.001.

Fig. 3

MiR-34b drives oxidative stress a) Percentage change in total ROS detected by DCFDA in control and miR-34b expressing cells at 0 and 24 h post maximal induction b) Percentage change in superoxide radicals detected by DHE in control and miR-34b expressing cells at 0 and 24 h post maximal induction c) Percentage change in hydroxyl and peroxynitrite radicals detected by HPF in control and miR-34b expressing cells at 0 and 24 h post maximal induction d) Representative hyPer Red Mito protein detector for H₂O₂ in control and miR-34b expressing cells post maximal induction e) Quantification of corrected total cell fluorescence (CTCF) of hyPer Red detector in control and miR-34b expressing cells at 0 and 24 h post maximal induction (n = 4). Values are depicted as mean± SD, and p-values are depicted as * p ≤ 0.05 ** p ≤ 0.01 *** p ≤ 0.001.

Fig. 4

Higher expression levels of TWIST1 in cervical cancer correlate with poor survival and TWIST1 is a target of miR-34b modulating its mRNA and protein levels a) Survival analysis of TWIST1 in cervical cancer predicted by Kaplan Meier plotter using data from Gene Expression Analysis Profiling Interactive Analysis (GEPIA) b) Relative expression (as relative units, RU) of TWIST1 levels in normal and cervical cancer, samples retrieved from GSE9750 c) Schematic representation of the alignment of miR-34b and its binding site present in the 3′ UTR of TWIST1 mRNA d) Relative luciferase activity of 3′UTR dual luciferase assay in cervical cancer cell lines. Control or miR-34b expressing constructs were co-transfected with wild-type (WT) or mutant (MUT) 3′ UTR miRNA binding sites containing constructs. 48 h post-transfection, cells were lysed, and the renilla and firefly luciferase activities were measured in control transfection groups (Untransfected, control vector, Mutant TWIST1 3′ UTR miRNA binding site containing vector) and experimental groups (control/miR-34b expression vector and WT/MUT TWIST1 3′ UTR miRNA binding site containing vector) (n = 3) e) Expression of TWIST1 at different time points post induction of miR-34b using dox (n = 3) normalized to β-actin by qRT-PCR f) Protein levels of TWIST1 post maximal induction of miR-34b in SiHa, HeLa and C33A with β-actin as loading control (n = 3). Values are depicted as mean± SD and p-values are depicted as * p ≤ 0.05 ** p ≤ 0.01 *** p ≤ 0.001.

Fig. 5

TWIST1 shRNA mimics miR-34b effect on cervical cancer cells a) mRNA levels of TWIST1 on shRNA mediated downregulation relative to β-actin quantified by qRT-PCR (n = 3) b) Protein levels of TWIST1 on shRNA mediated downregulation using β-actin as a loading control (n = 3) c) Bright field images of control and sh-TWIST1 expressing cells after day 5 and day 7 post seeding showing positive SA-β gal staining d) Quantification of total positively SA-β gal stained cells in each group (n = 4) e) Bright field, DAPI stained nucleus and FITC stained γH2AX foci representative images of control and sh-TWIST1 expressing cells at 2 h and 24 h post seeding f) Quantification of corrected total cell fluorescence (CTCF) of γH2AX foci (n = 4) g) Percentage change in total ROS detected by DCFDA in control and sh-TWIST1 expressing cells h) Percentage change in superoxide radicals detected by DHE in control and sh-TWIST1 expressing cells i) Percentage change in hydroxyl and peroxynitrite radicals detected by HPF in control and sh-TWIST1 expressing cells j) Representative images for hyPer Red Mito protein detector for H₂O₂ in control and sh-TWIST1 expressing cells k) Quantification of corrected total cell fluorescence (CTCF) of γH2AX foci (n = 4) l) Immunoblotting images of p53, p21 and lamin A/C in control and miR-34b expressing cells with β-actin as loading control. Values are depicted as mean± SD and p-values are depicted as * p ≤ 0.05 ** p ≤ 0.01 *** p ≤ 0.001.

Fig. 6

Effects of miR-34b and sh-TWIST1 introduction on total ROS, SA- β galactosidase activity and senescence markers p53, p21 and lamin A/C in SiHa cells a) Percentage change of total ROS in miR-34b and/or sh-TWIST1 expressing cells (n = 3) b) Quantification of total positively SA-β gal-stained cells in each group (n = 4) c) Protein levels of p53, p21, Lamin A/C in each group using β-actin as loading control (n = 3). Values are depicted as mean± SD, and p-values are depicted as ** p ≤ 0.01 *** p ≤ 0.001.

Fig. 7

RITA, a p53 activator, acts as senolytic agent for miR-34b induced senescent cells. Percentage change in cell viability due to RITA treatment in the presence and absence of miR-34b represented by IC50 values (n = 3).