Long non-coding RNA MAPKAPK5-AS1/PLAGL2/HIF-1α signaling loop promotes hepatocellular carcinoma progression

Abstract

Background: Long non-coding RNAs (lncRNAs) are widely involved in human cancers' progression by regulating tumor cells' various malignant behaviors. MAPKAPK5-AS1 has been recognized as an oncogene in colorectal cancer. However, the biological role of MAPKAPK5-AS1 in hepatocellular carcinoma (HCC) has not been explored.

Methods: Quantitative real-time PCR was performed to detect the level of MAPKAPK5-AS1 in HCC tissues and cell lines. The effects of MAPKAPK5-AS1 on tumor growth and metastasis were assessed via in vitro experiments, including MTT, colony formation, EdU, flow cytometry, transwell assays, and nude mice models. The western blotting analysis was carried out to determine epithelial-mesenchymal transition (EMT) markers and AKT signaling. The interaction between MAPKAPK5-AS1, miR-154-5p, and PLAGL2 were explored by luciferase reporter assay and RNA immunoprecipitation. The regulatory effect of HIF-1α on MAPKAPK5-AS1 was evaluated by chromatin immunoprecipitation.

Results: MAPKAPK5-AS1 expression was significantly elevated in HCC, and its overexpression associated with malignant clinical features and reduced survival. Functionally, MAPKAPK5-AS1 knockdown repressed the proliferation, mobility, and EMT of HCC cells and induced apoptosis. Ectopic expression of MAPKAPK5-AS1 contributed to HCC cell proliferation and invasion in vitro. Furthermore, MAPKAPK5-AS1 silencing suppressed, while MAPKAPK5-AS1 overexpression enhanced HCC growth and lung metastasis in vivo. Mechanistically, MAPKAPK5-AS1 upregulated PLAG1 like zinc finger 2 (PLAGL2) expression by acting as an endogenous competing RNA (ceRNA) to sponge miR-154-5p, thereby activating EGFR/AKT signaling. Importantly, rescue experiments demonstrated that the miR-154-5p/PLAGL2 axis mediated the function of MAPKAPK5-AS1 in HCC cells. Interestingly, we found that hypoxia-inducible factor 1α (HIF-1α), a transcript factor, could directly bind to the promoter to activate MAPKAPK5-AS1 transcription. MAPKAPK5-AS1 regulated HIF-1α expression through PLAGL2 to form a hypoxia-mediated MAPKAPK5-AS1/PLAGL2/HIF-1α signaling loop in HCC.

Conclusions: Our results reveal a MAPKAPK5-AS1/PLAGL2/HIF-1α signaling loop in HCC progression and suggest that MAPKAPK5-AS1 could be a potential novel therapeutic target of HCC.

Keywords: Hepatocellular carcinoma progression; Hypoxia; MAPKAPK5-AS1; PLAGL2; miR-154-5p.

Fig. 1

MAPKAPK5-AS1 upregulation is associated with poor clinical features and prognosis in HCC. a TCGA dataset showed the high expression level of MAPKAPK5-AS1 in HCC tissues compared to normal tissues. b GSE45436 dataset further exhibited high expression level of MAPKAPK5-AS1 in HCC tissues. c The level of MAPKAPK5-AS1 in 97 pairs HCC tissues and adjacent non-tumor tissues. d Kaplan-Meier survival analysis showed that HCC patients with high MAPKAPK5-AS1 expression exhibited a worse overall survival. e-f TCGA data showed that high MAPKAPK5-AS1 expression indicated poor overall survival (OS) and disease-free survival (DFS)

Fig. 2

MAPKAPK5-AS1 promotes proliferation and inhibits apoptosis of HCC cells. a-b MTT assay was carried out to examined the effect of MAPKAPK5-AS1 knockdown or overexpression on HCC cells viability. c-d Colony formation confirmed the negative regulatory effect of MAPKAPK5-AS1 on tumor cells proliferation. e-f The influence of MAPKAPK5-AS1 on DNA synthesis activity of HCC cell was evaluated by EdU assay. g-h The effect of MAPKAPK5-AS1 on cell apoptosis was assessed by flow cytometry. *p < 0.05

Fig. 3

MAPKAPK5-AS1 enhances migration, invasion, and EMT process of HCC cells. a-b Transwell assay were performed to evaluate the effects of MAPKAPK5-AS1 on the tumor cells abilities of migration and invasion. c-d Expression of EMT markers was detected by western blotting to assess the effect of MAPKAPK5-AS1 on EMT process. *p < 0.05

Fig. 4

MAPKAPK5-AS1 promotes the growth and metastasis of HCC cell in vivo. a-b Xenograft tumor model showed the negative regulatory effect of MAPKAPK5-AS1 on tumor size, growth rate, and tumor weight. c-d Representative images of tail vein injection model and the number of lung metastatic nodes. e-f IHC results of Ki67 and EMT markers for xenograft tumors. *p < 0.05

Fig. 5

MAPKAPK5-AS1 acts as a sponge of miR-154-5p in HCC cells. a Subcellular fractionation assay was applied to determine MAPKAPK5-AS1 subcellular localization. b 15 putative targets of MAPKAPK5-AS1 was predicted by applying two bioinformatics prediction tools. c qRT-PCR showed that MAPKAPK5-AS1 negatively regulated miR-154-5p expression in HCC cells. d The predicted binding sites between miR-154-5p and MAPKAPK5-AS1 and mutant sequences of the potential miR-154-5p binding site in MAPKAPK5-AS1. Additionally, the result of luciferase reporter assay is shown here. (E) Anti-Ago2 RIP assay was performed in HCCLM3 with miR-154-5p overexpression, followed by qRT-PCR to detect the level of MAPKAPK5-AS1 or miR-154-5p associated with Ago2. f The expression level of miR-154-5p in HCC tissues and adjacent non-tumor tissues. g The relativity between MAPKAPK5-AS1 and miR-154-5p in HCC tissues. *p < 0.05

Fig. 6

MiR-154-5p mediates the function of MAPKAPK5-AS1. a-c MTT, colony formation and EdU assays were conducted to explore the effect of miR-154-5p inhibitors on cell proliferation suppressed by MAPKAPK5-AS1 knockdown. d Flow cytometry showed the effect of miR-154-5p inhibitors on apoptosis induced by MAPKAPK5-AS1 knockdown. e-f Transwell assay were performed to evaluate the effects of miR-154-5p inhibitors on the tumor cells abilities of migration and invasion impaired by MAPKAPK5-AS1 kmockdown. c-d Expression of EMT markers was detected by western blotting to assess the effect of miR-154-5p inhibitors on EMT process inhibited by MAPKAPK5-AS1 knockdown. *p < 0.05

Fig. 7

PLAGL2 is the direct functional target of miR-154-5p. a 10 potential targets of miR-154-5p was predicted by applying three online bioinformatics tools. b mRNA levels of three potential targets in Hep3B cell with miR-154-5p overexpression. c The effect of miR-154-5p on PLAGL2 protein level evaluated by western blotting. d The predicted binding sites between miR-154-5p and PLAGL2 and mutant sequences of the potential miR-154-5p binding site in PLAGL2. In addition, the result of luciferase reporter assay is exhibited. e Western blotting confirmed that MAPKAPK5-AS1 and miR-154-5p could reverse the regulatory effect on PLAGL2 expression reciprocally. f The expression level of PLAGL2 in HCC tissues and adjacent non-tumor tissues. g Expression association between PLAGL2 and miR-154-5p in HCC tissues. h Expression association between PLAGL2 and MAPKAPK5-AS1 in HCC tissues. *p < 0.05

Fig. 8

PLAGL2/EGFR/AKT pathway mediates the oncogenic effect of MAPKAPK5-AS1. a-c MTT, colony formation and EdU assays showed the effect of PLAGL2 on cell proliferation suppressed by MAPKAPK5-AS1 knockdown. d Flow cytometry showed the effect of PLAGL2 on cell apoptosis induced by MAPKAPK5-AS1 knockdown. e Transwell assay showed the effects of PLAGL2 on migration and invasion of the tumor cells suppressed by MAPKAPK5-AS1 knockdown. f Expression of EMT and EGFR/AKT pathway markers was detected by western blotting to assess the effect of PLAGL2 on EMT process and pathway activity inhibited by MAPKAPK5-AS1 knockdown. *p < 0.05

Fig. 9

Hypoxia triggers a MAPKAPK5-AS1/PLAGL2/HIF-1α signaling loop. a Heat map of RNA-seq data exhibited the upregulation of MAPKAPK5-AS1 under hypoxia. b The upregulation of MAPKAPK5-AS1 and PLAGL2 induced by hypoxia was suppressed by HIF-1α knockdown. c The recognition motif of HIF-1α from JSAPAR database. d The hypoxia responsive element (HRE) in MAPKAPK5-AS1 promoter. In addition, the result of luciferase reporter assay is exhibited. e ChIP assay was conducted to validate the binding between HIF-1α and the HRE of the lncRNA MAPKAPK5-AS1 promoter under normoxia and hypoxia. f The regulatory effect of MAPKAPK5-AS1 on HIF-1α and PLAGL2 level. d The effect of PLAGL2 on HIF-1α and MAPKAPK5-AS1 expression under hypoxia. *p < 0.05

Fig. 10

Schematic of the findings of the present study