Impact of Oncogenic Targets by Tumor-Suppressive miR-139-5p and miR-139-3p Regulation in Head and Neck Squamous Cell Carcinoma

Abstract

We newly generated an RNA-sequencing-based microRNA (miRNA) expression signature of head and neck squamous cell carcinoma (HNSCC). Analysis of the signature revealed that both strands of some miRNAs, including miR-139-5p (the guide strand) and miR-139-3p (the passenger strand) of miR-139, were downregulated in HNSCC tissues. Analysis of The Cancer Genome Atlas confirmed the low expression levels of miR-139 in HNSCC. Ectopic expression of these miRNAs attenuated the characteristics of cancer cell aggressiveness (e.g., cell proliferation, migration, and invasion). Our in silico analyses revealed a total of 28 putative targets regulated by pre-miR-139 (miR-139-5p and miR-139-3p) in HNSCC cells. Of these, the GNA12 (guanine nucleotide-binding protein subunit alpha-12) and OLR1 (oxidized low-density lipoprotein receptor 1) expression levels were identified as independent factors that predicted patient survival according to multivariate Cox regression analyses (p = 0.0018 and p = 0.0104, respectively). Direct regulation of GNA12 and OLR1 by miR-139-3p in HNSCC cells was confirmed through luciferase reporter assays. Moreover, overexpression of GNA12 and OLR1 was detected in clinical specimens of HNSCC through immunostaining. The involvement of miR-139-3p (the passenger strand) in the oncogenesis of HNSCC is a new concept in cancer biology. Our miRNA-based strategy will increase knowledge on the molecular pathogenesis of HNSCC.

Keywords: GNA12; HNSCC; OLR1; expression signature; miR-139-3p; miR-139-5p; microRNA; passenger strand; tumor suppressor.

Figure 1

Expression of miR-139-5p and miR-139-3p in HNSCC tissues. (A) Volcano plot of the miRNA expression signature determined through RNA sequencing. The log₂ fold change (FC) is plotted on the x-axis, and the −log₁₀ (p-value) is plotted on the y-axis. The blue points represent the downregulated miRNAs with an absolute −log₂ FC < −2.0 and p < 0.05. The red points represent the upregulated miRNAs with an absolute −log₂ FC > 2.0 and p < 0.05. (B) The expression levels of miR-139-5p and miR-139-3p evaluated in an HNSCC dataset from TCGA.

Figure 2

Expression levels of miR-139-5p and miR-139-3p in the HNSCC cell lines. (A,B) Expression levels of miR-139-5p (A) and miR-139-3p (B) in normal oral tissues and the HNSCC cell lines evaluated with real-time PCR. RNU48 was used as the internal control. Gene expression levels are expressed relative to those in “#1 normal”.

Figure 3

(A–C) Functional assays of miR-139-5p and miR-139-3p in HNSCC cells (HSC-3 and Sa3). (A) Cell proliferation assessed with an XTT assay at 72 h after transfection of mature miRNAs (N.S.; not significant). (B) Cell migration assessed using a membrane culture system at 48 h after seeding miRNA-transfected cells into the chambers. (C) Cell invasion determined with a Matrigel invasion assay at 48 h after seeding miRNA-transfected cells into the chambers.

Figure 4

Flowchart of the strategy used to identify the putative oncogenes regulated by miR-139-5p and miR-139-3p in HNSCC cells.

Figure 5

Clinical significance of the expression of putative target genes of miR-139-5p and miR-139-3p in HNSCC tissues. (A) Expression levels of five putative target genes (PLEC, YWHAQ, GNA12, OLR1, and ACOT9) in the clinical specimens of HNSCC from a TCGA dataset. All genes were downregulated in HNSCC tissues (n = 518) compared with normal tissues (n = 44). (B) Kaplan–Meier curves of the five-year overall survival rate according to the expression of each gene. Low expression of all five genes was significantly predictive of a worse prognosis in patients with HNSCC. The patients were divided into two groups—the high- and low-expression groups—according to the median miRNA expression level. The red and blue lines represent the high- and low-expression groups, respectively. (C) Forest plot presenting the results of a multivariate Cox regression analysis of the prognostic value of two putative target genes (GNA12 and OLR1) identified in an HNSCC dataset from TCGA. The expression levels of GNA12 and OLR1 were determined to be independent prognostic factors in terms of the 5-year overall survival rate after adjustments for tumor stage, age, and pathological grade (p < 0.05).

Figure 5

Clinical significance of the expression of putative target genes of miR-139-5p and miR-139-3p in HNSCC tissues. (A) Expression levels of five putative target genes (PLEC, YWHAQ, GNA12, OLR1, and ACOT9) in the clinical specimens of HNSCC from a TCGA dataset. All genes were downregulated in HNSCC tissues (n = 518) compared with normal tissues (n = 44). (B) Kaplan–Meier curves of the five-year overall survival rate according to the expression of each gene. Low expression of all five genes was significantly predictive of a worse prognosis in patients with HNSCC. The patients were divided into two groups—the high- and low-expression groups—according to the median miRNA expression level. The red and blue lines represent the high- and low-expression groups, respectively. (C) Forest plot presenting the results of a multivariate Cox regression analysis of the prognostic value of two putative target genes (GNA12 and OLR1) identified in an HNSCC dataset from TCGA. The expression levels of GNA12 and OLR1 were determined to be independent prognostic factors in terms of the 5-year overall survival rate after adjustments for tumor stage, age, and pathological grade (p < 0.05).

Figure 6

Direct regulation of GNA12 and OLR1 expression by miR-139-3p in HNSCC cells. (A) Real-time PCR showing the significantly reduced expression of GNA12 mRNA at 72 h after miR-139-3p transfection in HSC-3 cells. Expression of GAPDH was used as an internal control. (B) Western blot showing reduced expression of the GNA12 protein at 72 h after miR-139-3p transfection in HSC-3 cells. Expression of GAPDH was used as an internal control. (C) One putative miR-139-3p binding site predicted within the 3′-UTR of GNA12 through TargetScanHuman analyses (upper panel). Dual luciferase reporter assays showed reduced luminescence activity after cotransfection of the wild-type GNA12 3′-UTR sequence (containing the miR-139-3p binding site) with miR-139-3p in HSC-3 cells (lower panel). Normalized data were calculated as the Renilla/firefly luciferase activity ratio (N.S., not significant). (D,E) Real-time PCR and Western blots showing reduced OLR1 mRNA and protein levels, respectively, after miR-139-3p transfection in HSC-3 cells. (F) One putative miR-139-3p binding site predicted within the 3′-UTR of OLR1 through TargetScanHuman analyses (upper panel). Luminescence activity was reduced after cotransfection of the wild-type OLR1 3′-UTR sequence (containing the miR-139-3p binding site) with miR-139-3p in HSC-3 cells.

Figure 7

Overexpression of GNA12 and OLR1 in clinical specimens of HNSCC. (A–D) Immunohistochemical staining of GNA12 in clinical specimens of HNSCC. High expression of GNA12 was detected in the nuclei and/or cytoplasm of cancer cells (A–C), and weak expression was detected in normal oral mucosa (D). (E–H) Immunohistochemical staining of OLR1 in clinical specimens of HNSCC. High expression of OLR1 was detected in the nuclei and/or cytoplasm of cancer cells (E–G), and weak expression was detected in normal oral mucosa (H).