Long non-coding RNA growth arrest specific transcript 5 acts as a tumour suppressor in colorectal cancer by inhibiting interleukin-10 and vascular endothelial growth factor expression

Abstract

Long non-coding RNAs (lncRNAs) are highly involved in diverse biological processes of human malignancies. The expression profile and underlying mechanism of lncRNA growth arrest specific transcript 5 (GAS5) in colorectal cancer (CRC) is poorly understood. In this study, we found that GAS5 was commonly downregulated in CRC tissues, serum of CRC patients and CRC cell lines. Knockdown of GAS5 promoted CRC cell proliferation and colony formation, whereas overexpression of GAS5 produced the opposite result. We further demonstrated that knockdown of GAS5 increased the expression and secretion of interleukin-10 (IL-10) and vascular endothelial growth factor (VEGF-A) via NF-κB and Erk1/2 pathways. Neutralization of IL-10 and VEGF-A reduced tumour proliferation caused by GAS5 knockdown. Moreover, GAS5 expression showed a statistically significant correlation with the mRNA levels of IL-10 and VEGF-A in CRC tissues. We further illustrated that GAS5 was markedly downregulated and negatively correlated with the cytokine expression in a mouse model of colitis-associated cancer (CAC). These results delineate a novel mechanism of lncRNA GAS5 in suppressing colorectal carcinogenesis. The cytokines IL-10 and VEGF-A inhibited by GAS5 may provide targets for lncRNA-based therapies for CRC.

Keywords: GAS5; VEGF; colorectal cancer; interleukin-10; long non-coding RNA.

Figure 1. lncRNA GAS5 is downregulated in tissues and serum of patients with CRC

(A) RT-qPCR analysis of the relative expression of GAS5 in CRC tissues (n = 24) and in paired adjacent normal tissues. GAS5 expression was normalized to GAPDH expression. The data are presented as a fold-change in the tumour tissue relative to the normal tissue. ***P < 0.001 by paired t-test (B) Circulating RNA was extracted, and GAS5 was measured by RT-qPCR and normalized to GAPDH. Serum GAS5 level showed an obvious clear decrease in patients with CRC (n = 109) compared to the normal controls (n = 99). The p value was determined by two-sided t test. ***P < 0.001. (C–E) Clinical significance of GAS5 in CRC serum. (C) GAS5 expression in serum was significantly lower in patients with higher pathological stages. p = 0.0268. (D) GAS5 expression in serum was significantly lower in patients with tumours ≥ 3.5 cm. p = 0.0314. (E) GAS5 expression level was not related to lymphatic metastasis. p = 0.0508. The p values were determined by two-sided t test. *P < 0.05, **P < 0.01; ***P < 0.001.

Figure 2. The expression of GAS5 in CRC cell lines and the effect of GAS5 on CRC cell proliferation and colony formation

(A) GAS5 showed a relatively low expression level in five CRC cell lines (DLD-1, HCT-116, HT-29, SW620, and SW480) compared to FHC, the normal colon tissue cells. The results are shown as the mean ± SEM relative to GAPDH levels from three experiments. *P < 0.05, one-way ANOVA followed by Bonferroni post hoc test. (B–C) Efficacy of stable GAS5 overexpression and knockdown in HCT-116 and HT-29 cells were determined by RT-qPCR. GAPDH was used as an internal control. The results are shown as the mean ± SEM relative to GAPDH levels from three experiments. *P < 0.05, **P < 0.01, ***P < 0.001, compared to the control group by one-way ANOVA followed by Bonferroni post hoc tests. (D–G) MTT assays were performed to determine the proliferation of HCT-116 and HT-29 cells after GAS5 manipulation. Stable GAS5 manipulated cells and controls were seeded in 96-well plates and analysed daily for 4 days using the MTT assay. Overexpression of GAS5 (D and E) substantially reduced colorectal cancer cell proliferation, whereas knockdown of GAS5 (F and G) significantly enhanced the proliferation of both HCT-116 and HT-29 cells. The results are the mean ± SEM from three experiments, and each experiment had six replicates. *P < 0.05; **P < 0.01; ***P < 0.001 compared with control by one-way ANOVA followed by Bonferroni post hoc test. (H) Effect of GAS5 overexpression or GAS5 knockdown on the colony formation of HCT-116 and HT-29 cells. The error bar results present the colony formation ability relative to control cells (set to 100%). The values are expressed as the mean ± SEM from three experiments. *P < 0.05 compared to control cells by one-way ANOVA followed by Bonferroni post hoc test.

Figure 3. GAS5 inhibits the expression and secretion of cytokines in vitro

(A–F) RT-qPCR results showed that GAS5 participated in regulating inflammatory gene expression: GAS5 overexpression significantly inhibited IL-10, TNF-α and VEGF-A expression (*P < 0.05; **P < 0.01, two-sided t tests), while GAS5 knockdown significantly promoted IL-10, TNF-α and VEGF-A expression in both HCT-116 and HT-29 cells (*P < 0.05; **P < 0.01, one-way ANOVA followed by Bonferroni post hoc test). The results are shown as the mean ± SEM relative to GAPDH levels from three experiments. (G–J) Secretion of inflammatory cytokines was measured by ELISA assay. IL-10 and VEGF-A secretion were clearly decreased in GAS5-overexpressing cell lines (*P < 0.05; **P < 0.01, two-sided t tests). IL-10 and VEGF-A secretion was highly increased in GAS5 knockdown cell lines (*P < 0.05; **P < 0.01, one-way ANOVA followed by Bonferroni post hoc test). The results are shown as the mean ± SEM relative to the control group from three experiments.

Figure 4. GAS5 inhibits the secretion of inflammatory cytokines via the NF-κB and Erk1/2 MAPK pathways

(A) GAS5 overexpression inhibited p-NF-κB p65, and GAS5 knock-down and significantly enhanced p-NF-κB p65 in HCT-116 cells. (B) GAS5 overexpression inhibited phosphorylated Erk1/2, and GAS5 knockdown significantly enhanced phosphorylated Erk1/2 in HCT-116 cells. p38 MAPK remained stable. (C) HCT-116 cells were treated with 10 μM U0126 or 10 μM BAY 11-7082 or DMSO, and the levels of p-Erk, total Erk1/2 (t-Erk), p-NF-κB p65 and total NF-κB p65 were measured. GAPDH was used as a loading control. (D–E) ELISA assays were performed to evaluate the secretion of IL-10 and VEGF-A in the supernatant of HCT-116 cells after different treatments. The results are shown as the mean ± SEM relative to the NC group from three experiments. *P < 0.05; **P < 0.01, compared to the control group, and all P values are from one-way ANOVA followed by Bonferroni post hoc tests.

Figure 5. Neutralization of IL-10 and VEGF-A reduces tumour proliferation caused by GAS5 knockdown

Neutralization antibody against IL-10 (1.5 μg/ml) or VEGF-A (0.1 μg/ml) was incubated with tumour cells, and MTT assays were performed at different time points (48 h, 72 h, and 96 h) to test tumour cell proliferation. The results are shown as the mean ± SEM from three experiments. *P < 0.05; **P < 0.01, compared to the control group, and all P values are from one-way ANOVA followed by Bonferroni post hoc tests.

Figure 6. Correlations of GAS5 levels with IL-10, VEGF-A and TNF-α in colorectal tumour tissues and GAS5 downregulation in the CAC mouse model

(A–C) The RNA levels were determined by RT-qPCR relative to GAPDH. A significant negative correlation was observed between GAS5 expression and mRNA levels of IL-10, VEGF-A and TNF-α in CRC samples (n = 24). The r values and P values are from Pearson's correlation analysis. (D) Diagram of AOM/DSS administration schedule for establishing the colitis-associated colon cancer model. (E) GAS5 downregulation in the colon of wild type (WT) mice on day 71 of AOM/DSS administration compared with the control group (n = 10). The results are shown as the mean ± SEM relative to GAPDH levels from three experiments. *P < 0.05, compared to the control group by two-sided t tests. (F–G) Negative correlations between GAS5 and IL-10 or TNF-α were detected in the colon of the CAC mouse model (n = 10). The inset shows the Pearson's r correlation and corresponding P value.