MiR-612 regulates invadopodia of hepatocellular carcinoma by HADHA-mediated lipid reprogramming

Abstract

Background: MicroRNA-612 (miR-612) has been proven to suppress EMT, stemness, and tumor metastasis of hepatocellular carcinoma (HCC) via PI3K/AKT2 and Sp1/Nanog signaling. However, its biological roles on HCC progression are far from elucidated.

Methods: We found direct downstream target of miR-612, hadha by RNA immunoprecipitation and sequencing. To explore its biological characteristic, potential molecular mechanism, and clinical relevance in HCC patients, we performed several in-vitro and in-vivo models, as well as human tissue chip.

Results: Ectopic expression of miR-612 could partially reverse the level of HADHA, then suppress function of pseudopods, and diminish metastatic and invasive potential of HCC by lipid reprogramming. In detail, miR-612 might reduce invadopodia formation via HADHA-mediated cell membrane cholesterol alteration and accompanied with the inhibition of Wnt/β-catenin regulated EMT occurrence. Our results showed that the maximum oxygen consumption rates (OCR) of HCCLM3^miR-612-OE and HCCLM3^hadha-KD cells were decreased nearly by 40% and 60% of their counterparts (p < 0.05). The levels of acetyl CoA were significantly decreased, about 1/3 (p > 0.05) or 1/2 (p < 0.05) of their controls, in exogenous miR-612 or hadha-shRNA transfected HCCLM3 cell lines. Besides, overexpression of hadha cell lines had a high expression level of total cholesterol, especially 27-hydroxycholesterol (p < 0.005). SREBP2 protein expression level as well as its downstream targets, HMGCS1, HMGCR, MVD, SQLE were all deregulated by HADHA. Meanwhile, the ATP levels were reduced to 1/2 and 1/4 in HCCLM3^miR-612-OE (p < 0.05) and HCCLM3^hadha-KD (p < 0.01) respectively. Moreover, patients with low miR-612 levels and high HADHA levels had a poor prognosis with shorter overall survival.

Conclusion: miR-612 can suppress the formation of invadopodia, EMT, and HCC metastasis and by HADHA-mediated lipid programming, which may provide a new insight of miR-612 on tumor metastasis and progression.

Keywords: Hepatocellular carcinoma; Invadopodia; Metastasis; miR-612; β-Oxidation.

Fig. 1

MiR-612 and hadha, its new target, in HCC. a, b Representative images with high and low expressed level of miR-612 in HCC and their paired adjacent tissues in immunohistochemistry. Bars: (left) magnification × 100, (right) magnification × 400. c, d Kaplan–Meier analysis of PFS and OS in HCC patients using SPSS 22.0. e Volcano map of the differential expressed genes (Ordinate: statistical significance of changes in expression; abscissa: multiple changes in the differential genes. Orange represents upregulated genes and downregulated ones in green). f Repeated downregulated genes in two independent RNA-seqs. g KEGG pathway analysis (Ordinate: the KEGG signal path; abscissa: p value). h Gene profiles related to lipid metabolism, including hadha. i Schematic diagram of the dual luciferase reporter vector of miRNA target. j Luciferase activity in 293T cells. H2294:pMIR-REPORT hadha WT 3′-UTR; H2336:pMIR-REPORT hadha mut 3′-UTR. Statistical analysis by Student’s t test. (***p < 0.001). Data are mean ± SEM of three independent experiments

Fig. 2

Morphology and mobility of HCC regulated by miR-612 and hadha. a, b Morphological changes of HepG2 and HCCLM3 infected with indicated lentivirus in microscope. Scale bars, 200 μm. c, d Morphological changes of HepG2 and HCCLM3 infected with indicated lentivirus in electron microscopy. e, g Wound healing of HepG2 infected with miR-612-KD or hadha-OE lentivirus. Scale bars, 200 μm. f, h Wound healing of HCCLM3 infected with miR-612-OE and hadha-KD lentivirus. Scale bars, 200 μm. (ns: not significant; *p < 0.05; ***p < 0.001). Data are mean ± SEM of three independent experiments

Fig. 3

MiR-612 suppresses invasion and migration of HCC by targeting hadha. a, b Cell migration abilities and statistic results of HepG2^miR-612-KD and HCCLM3^miR-612-OE cells. Scale bars, 200 μm. c, d Cell invasion abilities and statistic results of HepG2^miR-612-KD and HCCLM3^miR-612-OE cells. Scale bars, 200 μm. e, f Cell migration abilities and statistic results of HepG2^hadha-OE and HCCLM3^hadha-KD cells. Scale bars, 200 μm. g, h Cell invasion abilities and statistic results of HepG2^hadha-OE and HCCLM3^hadha-KD cells. Cell migration abilities and statistic results of HepG2^miR-612-KD cells after hadha was knocked down. Scale bars, 200 μm. i, j Cell migration abilities and statistic results of HCCLM3 ^miR-612-OE cells after HADHA was rescued. Scale bars, 50 μm. k, l Cells (*p < 0.05; **p < 0.01). Data are mean ± SEM of three independent experiments. Scale bars, 50 μm

Fig. 4

MiR-612 restrains HCC invadopodia by HADHA. a, c Expression of invasive pseudopodia-related proteins, Cortactin, F-actin, and Caveolin-1, in HepG2 cells with indicated treatments. a Scale bars, 5 μm. c Scale bars, 30 μm (b, d) Expression of invasive pseudopodia-related proteins, Cortactin, F-actin and Caveolin-1, in HCCLM3 cells with indicated treatments. b Scale bars, 50 μm. d Scale bars, 15 μm. e, f The number and statistic results of invasive pseudopods of HCCLM3 cells. Scale bars, 15 μm. g EMT and invadopodia biomarkers in HepG2 and HCCLM3 cells with indicated treatments analyzed by Western blots

Fig. 5

HADHA promotes metastasis of HCC in vivo. a Metastatic foci in lung were imaged by micro-spiral CT scans using 3D synthesis. b, c Metastatic foci in lung and their statistic results of indicated HCCLM3 orthotopic xenografts (HCCLM3^hadha-KD or HCCLM3^NC) in H-E staining. d, e Metastatic foci in liver and their statistic results of indicated HCCLM3 orthotopic xenografts in H-E staining. Bars: (left) magnification × 100, scale bars, 200 μm. (right) magnification × 200, scale bars, 100 μm. f, g The levels of Cortactin and Caveolin-1 in liver orthotopic xenografts using IHC analyses. Bars: (up) magnification× 100, scale bars, 200 μm. (down) magnification × 200. scale bars, 100 μm. (*p < 0.05; **p < 0.01)

Fig. 6

HADHA promotes β-oxidation of fatty acids in HCC. a OCR, the activities of β-oxidation, in HepG2 cells treated with 100 μM ETO, as well as in HepG2^miR-612-KD, HepG2^hadha-OE, and HepG2^NC cells. b OCR, the activities of β-oxidation, in HCCLM3 cells treated with 100 μM ETO, as well as in HCCLM3^miR-612-OE, HCCLM3^hadha-KD, and HCCLM3^NC cells. c The protein levels of Cortactin in HCCLM3 cells treated with 100 μM ETO or 100 mM linoleic acid, as well as in HCCLM3^miR-612-o, HCCLM3^miR-612-i, HCCLM3^hadha-o, and HCCLM3^hadha-i cells. d The intracellular levels of acetyl CoA in HCCLM3^miR-612-o, HCCLM3^hadha-i, and HCCLM3^NC cells. e The cellular levels of cholesterol in HepG2^miR-612-KD, HCCLM3^miR-612-OE, HepG2^hadha-OE, and HCCLM3^hadha-KD cells. f The protein levels of Cortactin, Caveolin-1, E-cadherin, and GAPDH in HCCLM3 cells treated with or without 1 mM MβCD. g Fluorescence intensity in HepG2^miR-612-i, HCCLM3^miR-612-o, HepG2^hadha-o, and HCCLM3^hadha-i cells detected by fluorescence spectrophotometer (Ex/Em = 360/460 nm). Fluorescence polarization is calculated by the formula (h) The levels of ATP in HepG2^miR-612-KD, HCCLM3^miR-612-OE, HepG2^hadha-OE, and HCCLM3^hadha-KD cells analyzed by spectrophotometer (λ = 636 nm). (NS: no significance; *p < 0.05; **p < 0.01; ***p < 0.001). Data are mean ± SEM of three independent experiments

Fig. 7

Clinical significance of HADHA in HCC patients. a The cholesterol and 27-hydroxycholesterol of HepG2 and HCCLM3 cells relative quantification based on liquid chromatography mass spectrometry (LC-MS)/MS system. (NS: no significance; **p < 0.01; ***p < 0.001). b Immunoblotting of proteins involved in cholesterol-biosynthesis pathway in HepG2^hadha-OE, HCCLM3^hadha-KD cell, and their negative control cells. (Cells treated with negative control lentivirus). c Negative correlation between miR-612 and hadha mRNA in 15 HCC and their adjacent normal tissues. d Representative images of negative and positive HADHA staining in tumor and peritumor tissue from one HCC patient. Bars: (left) magnification × 100, scale bars, 500 μm (right) magnification × 400, scale bars, 50 μm. e The average densities of HADHA in 134 HCC and their paired normal tissues. f, g Kaplan-Meier analyses of PFS and OS in HCC patients using SPSS 22.0. h Hypothesis diagram of lipid reprogramming in HCC cells modulated by miR-612/HADHA/cholesterol axis