LncRNA SNHG4 promotes tumour growth by sponging miR-224-3p and predicts poor survival and recurrence in human osteosarcoma

Abstract

Objective: Accumulating data show that dysregulation of long noncoding RNAs (lncRNAs) acts a critical role in a variety of malignancies. Among these lncRNAs, small nucleolar RNA host genes (SNHGs) are associated with tumour growth and progression. But, the molecular mechanisms by which SNHG4 contributes to osteosarcoma remain undocumented.

Methods: The association between lncRNA SNHG4 expression and clinicopathologic characteristics and prognosis in patients with osteosarcoma was analysed by TCGA RNA-sequencing data. Cell viability and colony formation abilities were respectively assessed by MTT and colony formation assays. LncRNA SNHG4-specific binding with miR-224-3p was verified by bioinformatic analysis, luciferase gene report, and RNA immunoprecipitation assays. Regulation relationship between SNHG4 and miR-224-3p expression was further evaluated by the rescue experiments.

Results: The expression level of lncRNA SNHG4 was significantly elevated in osteosarcoma samples and cell lines as compared with the adjacent normal tissues, and SNHG4 high expression was associated with tumour size (TS) and poor prognosis in patients with osteosarcoma. Knockdown of SNHG4 suppressed cell viability and invasive potential, whereas ectopic SNHG4 expression displayed the opposite effects. Moreover, we found that lncRNA SNHG4 acted as a sponge of miR-224-3p, and miR-224-3p mimic reversed SNHG4 induced tumour-promoting effects in osteosarcoma cells. The expression of miR-224-3p depicted a negative correlation with SNHG4 in osteosarcoma samples and miR-224-3p low expression was associated with TS and poor survival in patients with osteosarcoma.

Conclusion: Our findings demonstrated that LncRNA SNHG4 promoted tumour growth by sponging miR-224-3p and represented a poor prognostic factor in patients with osteosarcoma.

Figure 1

LncRNA SNHG4 was associated with tumour size and poor prognosis in patients with osteosarcoma. (A) TCGA cohort analysis of the expression levels of SNHG4 in osteosarcoma samples as well as in the pair‐matched osteosarcoma tissues. (B) TCGA cohort analysis of the expressed levels of SNHG4 in patients with TS ≥5 cm or TS <5 cm. (C) The cut‐off value, sensitivity, and specificity of SNHG4 was evaluated in osteosarcoma samples (n = 136). Kaplan‐Meier analysis of the association between SNHG4 high or low expression and (D) the overall survival and (E) tumour recurrence in patients with osteosarcoma

Figure 2

Knockdown of SNHG4 inhibited cell proliferation and colony formation. (A) qRT‐PCR analysis of the expression level of SNHG4 in osteosarcoma cell lines and normal tissues. (B) qRT‐PCR analysis of the transfection efficiency of si‐SNHG4 in Ssos‐2 and U‐2 OS cell lines. (C) MTT analysis of the effects of SNHG4 knockdown on cell viability in Ssos‐2 and U‐2 OS cell lines. (D) Colony formation assay assessment of the effects of SNHG4 knockdown on cell colony formation abilities in Ssos‐2 and U‐2 OS cell lines. (E) Western blot analysis of the expression level of PCNA protein in si‐SNHG4 transfected Ssos‐2 and U‐2 OS cell lines. Data are the means ± SEM of three experiments. *P < 0.05, **P < 0.01

Figure 3

Overexpression of SNHG4 promoted cell proliferation and colony formation. (A) qRT‐PCR analysis of the transfection efficiency of pcDNA3.1‐SNHG4 in MG‐63 and HOS cell lines. (B) MTT analysis of the effects of SNHG4 overexpression on cell viability in MG‐63 and HOS cell lines. (C) Colony formation assay assessment of the effects of SNHG4 overexpression on cell colony formation abilities in MG‐63 and HOS cell lines. (D) Western blot analysis of the expression level of PCNA protein in SNHG4 transfected MG‐63 and HOS cell lines. Data are the means ± SEM of three experiments. **P < 0.01

Figure 4

LncRNA SNHG4 acted as a sponge of miR‐224‐3p in osteosarcoma cells. (A) Pearson correlation analysis of the correlation of SNHG4 expression with miR‐224‐3p in patients with osteosarcoma. (B) Schematic representation of the binding sites of miR‐224‐3p with WT or Mut SNHG4. (C) The luciferase activity of WT or Mut SNHG4 was assessed after transfection with miR‐224‐3p mimic in Ssos‐2 and U‐2 OS cell lines. (D) qRT‐PCR analysis of the transfection efficiency of miR‐224‐3p mimic in Ssos‐2 and U‐2 OS cell lines. (E) qRT‐PCR analysis of the effects of SNHG4 overexpression on the expression of miR‐224‐3p in Ssos‐2 and U‐2 OS cell lines. (F) Ago2 RIP assay analysis of the enrichment of SNHG4 and miR‐224‐3p pulled‐down from the Ago2 protein in Ssos‐2 and U‐2 OS cell lines, and the expression levels of SNHG4 and miR‐224‐3p were examined by qRT‐PCR analysis. Data are the means ± SEM of three experiments. *P < 0.05, **P < 0.01

Figure 5

miR‐224‐3p mimic reversed SNHG4‐induced tumour‐promoting effects. (A) MTT analysis of the cell viability after transfection with SNHG4 and (or) miR‐224‐3p mimic in Ssos‐2 and U‐2 OS cell lines. (B) Colony formation analysis of the cell colony abilities after transfection with SNHG4 and (or) miR‐224‐3p mimic in Ssos‐2 and U‐2 OS cell lines. (C) Schematic representation of the binding sites of miR‐224‐3p with WT or Mut 3′UTR of DOCK7. (D) The luciferase activity of WT or Mut 3′UTR of DOCK7 was evaluated after transfection with miR‐224‐3p mimic in Ssos‐2 and U‐2 OS cell lines. (E) MTT analysis of the cell viability after transfection with DOCK7 and (or) miR‐224‐3p mimic for 120 h in Ssos‐2 and U‐2 OS cell lines. (F) Western blot analysis of the expression of DOCK7 protein after transfection with DOCK7 and (or) miR‐224‐3p mimic in Ssos‐2 and U‐2 OS cell lines. Data are the means ± SEM of three experiments. *P < 0.05, **P < 0.01

Figure 6

Low expression of miR‐224‐3p was associated with tumour size and poor survival in patients with osteosarcoma. (A) TCGA cohort analysis of the expression levels of miR‐224‐3p in osteosarcoma samples as well as in the pair‐matched osteosarcoma tissues. (B) TCGA cohort analysis of the expressed levels of miR‐224‐3p in patients with TS ≥5 cm or TS <5 cm. (C) The cut‐off value of miR‐224‐3p was estimated in osteosarcoma samples (n = 136). Kaplan‐Meier analysis of the association between miR‐224‐3p high or low expression and (D) overall survival and (E) tumour recurrence in patients with osteosarcoma