Activation of the LINC00242/miR-141/FOXC1 axis underpins the development of gastric cancer

Abstract

Background: Long non-coding RNAs (LncRNAs) are a class of newly identified transcripts recognized as critical governors of gene expression during human carcinogenesis, whereas their tumor-suppressive or tumor-promoting effects on gastric cancer (GC) are required for further investigation. In the study, we identify the expression pattern of a novel lncRNA LINC00242 in GC and its possible permissive role in the development of GC.

Methods: The study included 68 pairs of GC and adjacent normal gastric tissue samples. The viability, migration, and invasion of cultured human GC cells HGC27 were evaluated by CCK-8 and Transwell chamber assays. In vitro tube formation of human brain microvascular endothelial cells (HBMVECs) in HGC27 cell coculture was detected. The regulatory network of LINC00242/miR-141/FOXC1 was verified using dual luciferase reporter gene assay and RNA immunoprecipitation (RIP) assay. Subcutaneous xenografts of HGC27 cells were performed in nude mice.

Results: LINC00242 was highly expressed in GC tissues and cells and contributed to poor prognosis. LINC00242 knockdown inhibited HGC27 cell viability, migration and invasion, and tube formation of HBMVECs. LINC00242 interacted with miR-141 and positively regulated FOXC1, a target gene of miR-141. LINC00242 knockdown was partially lost in HGC27 cells upon miR-141 inhibition or FOXC1 overexpression. The tumor-promoting effect of LINC00242 on GC was demonstrated in nude mice.

Conclusion: Taken together, the present study demonstrates the oncogenic role of the LINC00242/miR-141/FOXC1 axis in GC, highlighting a theoretical basis for GC treatment.

Keywords: FOXC1; Gastric cancer; LINC00242; MicroRNA-141.

Fig. 1

Expression pattern of LINC00242 in GC tissues and cells. a–c, Heat maps depicting the DEGs obtained from the GC-related datasets GSE79973, GSE19826 and GSE65801. The abscissa represents the sample number, and the ordinate represents the differentially expressed gene; the left dendrogram represents gene expression cluster; each rectangle corresponds to a sample expression value; the histogram at the upper right refers to color gradation. d Venn analysis of DEGs in GC microarray, three circles in the figure respectively represent the DEGs from the three datasets, and the middle part represents the overlaps. e The expression of LINC00242 in 68 pairs of GC and adjacent normal gastric tissues was determined by RT-qPCR. f The area under the ROC curve according to LINC00242 expression. The threshold of LINC00242 expression was 0.803. g The Kaplan–Meier method was used to plot GC patient survival according to LINC00242 expression. f The expression of LINC00242 in GES, SGC7901, BGC823, MKN45 and HGC27 cell lines was determined by RT-qPCR. Results are presented as mean ± S.D. of three technical replicates. *(compared with adjacent normal gastric tissues) indicates p < 0.05 by paired t-test. ** (compared with GES cells) indicates p < 0.01 by one-way ANOVA followed by Tukey’s test

Fig. 2

LINC00242 knockdown represses GC cell migration, invasion, and in vitro angiogenesis. a The expression of LINC00242 in HGC27 cells upon LINC00242 knockdown was determined by RT-qPCR. b HGC27 cell viability was measured by CCK-8 assay. c Representative view of HGC27 cells with LINC00242 knockdown invading from Matrigel-coated chambers into lower ones and statistics of invading cells. d Representative view of HGC27 cells with LINC00242 knockdown migrating from upper transwell chambers into lower ones and statistics of migrating cells. e Representative Western blots of N-cadherin, Vimentin, MMP-2, MMP-9 and their quantitation analyses in HGC27 cells upon LINC00242 knockdown. fIn vitro tube formation of HBMVECs in HGC27 cells upon LINC00242 knockdown. Results are presented as mean ± S.D. of three technical replicates. *indicates p < 0.05 compared with si-NC by unpaired t-test (panel a, c–f) or repeated measurements ANOVA with Bonferroni corrections (panel b)

Fig. 3

LINC00242 interacts with miR-141 and positively regulats FOXC1, a target gene of miR-141. a The predicted localization of LINC00242. The abscissa represents different cell lines and the ordinate represents the localization of LINC00242; the histogram above the 0 line represents the expression mainly in the cytoplasm, and the histogram below the 0 line represents the expression mainly in the nucleus. b A heat map illustrating differentially expressed miRNAs from GC-related GSE26595 dataset. c The intersecting miRNAs from the GSE26595 dataset and the RAID database (http://www.rna-society.org/raid/search.html). d A heat map illustrating some significantly upregulated genes from the GSE13861 dataset. e The intersecting genes from the mirDIP (http://ophid.utoronto.ca/mirDIP/index.jsp#r), TargetScan (http://www.targetscan.org/vert_71/) and DIANA (http://diana.imis.athena-innovation.gr/DianaTools/index.php?r=microT_CDS/index) databases and the GSE13861 dataset. f Subcellular localization of LINC00242 detected by FISH assay (× 400, scale bar = 25 μm). Green represents LINC00242 expression and blue represents the stained nucleus. g The binding of LINC00242 and FOXC1 to AGO2 assessed by RIP assay. * indicates p < 0.05 compared with IgG. h Putative miR-141 binding sites in the FOXC1 3′UTR by a web-available prediction (www.targetscan.org). i The luciferase activity at the promoter of the reporter gene containing FOXC1-WT and FOXC1-MUT upon miR-141 mimic transfection. * indicates p < 0.05 compared with mimic NC. j Expression of FOXC1 binding to using anti-AGO2 antibody in upon LINC00242 knockdown detected by RIP assay. * indicates p < 0.05 compared with si-NC. k Expression of miR-141 and FOXC1 in cells upon si-LINC00242 transfection detected by RT-qPCR, normalized to U6 and GAPDH respectively. * indicates p < 0.05 compared with si-NC. l Expression of FOXC1 in cells transfected with miR-141 mimic detected by RT-qPCR, normalized to GAPDH. * indicates p < 0.05 compared with mimic NC. m Expression of FOXC1 in cells transfected with si-LINC00242 or miR-141 inhibitor detected by RT-qPCR, normalized to GAPDH. Results are presented as mean ± S.D. of three technical replicates. * (compared with si-NC + inhibitor NC) and # (compared with si-LINC00242 + inhibitor NC) indicate p < 0.05 by unpaired t-test or one-way ANOVA followed by Tukey’s test

Fig. 4

LINC00242 knockdown retards HGC27 cell viability, migration and invasion, and tube formation of HBMVECs by regulating miR-141 and FOXC1. HGC27 cells were treated with si-LINC00242 alone or with miR-141 inhibitor or oe-FOXC1. a HGC27 cell viability was measured by CCK-8 assay. b Representative view of HGC27 cells invading from Matrigel-coated chambers into lower ones and statistics of invading cells. c Representative view of HGC27 cells migrating from upper transwell chambers into lower ones and statistics of migrating cells. d Representative Western blots of MMP-2, MMP-9, N-cadherin, Vimentin and their quantitation analyses in HGC27 cells. eIn vitro tube formation of HBMVECs in HGC27 cells. Results are presented as mean ± S.D. of three technical replicates. * (compared with si-LINC00242 + inhibitor NC) and # (compared with si-LINC00242 + oe NC) indicate p < 0.05 by unpaired t-test (panel b–e) or repeated measurements ANOVA with Bonferroni corrections (panel a)

Fig. 5

LINC00242 knockdown hampers the tumorigenesis of GC cells and angiogenesis in vivo. a–c, The volume and mass of GC xenografts by HGC27 cells upon LINC00242 knockdown. d, e, Representative Western blots of FOXC1, VEGF, CD31, and their quantitation analyses in tissues of GC xenografts by HGC27 cells upon LINC00242 knockdown. f Immunohistochemistry staining for VEGF and CD31 in tissues of GC xenografts by HGC27 cells upon LINC00242 knockdown (× 200, scale bar = 50 μm). Results are presented as mean ± S.D. of three technical replicates. * (compared with sh-NC) indicates p < 0.05 and ** (compared with sh-NC) indicates p < 0.01 by paired t-test

Fig. 6

Mechanism diagram illustrating the regulatory network and function of LINC00242 in GC. LINC00242 is upregulated in GC tissues. Overexpressed LINC00242 competitively binds to miR-141 and consequently promotes FOXC1 expression, thus increasing expression of N-cadherin, Vimentin, MMP-2 and MMP-9, ultimately accelerating the tumorigenesis and angiogenesis in GC