Inhibiting lncRNA NEAT1 Increases Glioblastoma Response to TMZ by Reducing Connexin 43 Expression

Abstract

Objectives: Glioblastoma multiforme (GBM) is considered the most assailant subtype of gliomas, presenting a formidable obstacle because of its inherent resistance to temozolomide (TMZ). This study aimed to characterize the function of lncRNA NEAT1 in facilitating the advancement of gliomas.

Methods: The expression level of NEAT1 in glioma tissues and cells was detected by qRT-PCR. RNA interference experiment, cell proliferation assay, FITC/PI detection assay, immunoblotting, bioinformatics prediction, a double luciferase reporter gene assay, RNA immunoprecipitation (RIP) assay, SLDT assay and correlation analysis of clinical samples were performed to explore the regulatory effects of NEAT1, miR-454-3p and Cx43 and their role in malignant progression of GBM. The role of NEAT1 in vivo was investigated by an intracranial tumor formation experiment in mice.

Results: The results showed that recurring gliomas displayed elevated levels of NEAT1 compared to primary gliomas. The suppression of NEAT1 led to a restoration of sensitivity in GBM cells to TMZ. NEAT1 functioned as a competitive endogenous RNA against miR-454-3p. Connexin 43 was identified as a miR-454-3p target. NEAT1 was found to regulate gap junctional intercellular communication by modulating Connexin 43, thereby impacting the response of GBM cells to TMZ chemotherapy. Downregulation of NEAT1 resulted in enhanced chemosensitivity to TMZ and extended the survival of mice.

Conclusions: Overall, these results indicated that the NEAT1/miR-454-3p/Connexin 43 pathway influences GBM cell response to TMZ and could offer a potential new strategy for treating GBM.

Keywords: Connexin 43; NEAT1; chemotherapy sensitivity; miR‐454‐3p; temozolomide.

FIGURE 1

NEAT1 knockdown enhances the sensitivity of GBM to TMZ. (A) Relative expression of NEAT1 in normal brain tissues, primary and recurrent GBM tissues, normalized to GAPDH. *p < 0.05, ***p < 0.001. (B) Relative expression of NEAT1 in NHA, U87MG, and A172 cells. **p < 0.01, ***p < 0.001. (C, D) The expression of NEAT1 in U87MG and A172 cells, NEAT1 was knockdown by transfecting with si‐NEAT1‐1, si‐NEAT1‐2, or NEAT1‐3. **p < 0.01. (E, F) CCK‐8 assay analysis revealed the effect of shNC and shNEAT1 on the U87MG and A172 cells after TMZ treatment at the indicated concentrations for 48 h. (G, H) Cell proliferation assays of shNC and shNEAT1 in U87MG and A172 cells with TMZ (100 μM for U87MG cells, 180 μM for A172 cells, the same below) or equal volume DMSO treatment. *p < 0.05, **p < 0.01, ***p < 0.001, compared to si‐NC + DMSO group; ^# p < 0.05; ^### p < 0.001, compared to si‐NEAT1 + DMSO group; ^& p < 0.05; ^&&& p < 0.001, compared to si‐NC + TMZ group. (I–K) Flow cytometric analysis of si‐NC and si‐NEAT1 in U87MG and A172 cells with TMZ or equal volume DMSO treatment (48 h), *p < 0.05, **p < 0.01. (L–N) The protein levels of ABCG2, BCL2, and BAX in U87MG and A172 cells with shNEAT1 or shNC. GAPDH was used as the control. Relative density values were counted using image J software. *p < 0.05; **p < 0.01; ^# p < 0.05; ^## p < 0.01; ^& p < 0.05; ^&& p < 0.01.

FIGURE 2

NEAT1 acts as a molecular sponger for miR‐454‐3p. (A) Venn diagram was used to achieve the potential miRNAs that may target the binding region of NEAT1. (B) Relative expression of miR‐454‐3p in normal brain and GBM tissues, normalized to U6. (C) The Pearson's correlation analysis was performed to analyze the correlation between NEAT1 and miR‐454‐3p. (D) The mRNA expression level of NEAT1. (E) Schematic illustration of the predicted binding sites between NEAT1 and miR‐454‐3p, and mutation of potential miR‐454‐3p binding sequence in NEAT1. (F, G) Dual luciferase assay was applied to assess the seed‐matching sites or mutant sites between NEAT1 and miR‐454‐3p in HEK‐293T cells and A172 cells. (H–J) QRT‐PCR was employed to analyze the expression of NEAT1 in nuclear and cytoplasmic in U87MG and A172 cells. (K, L) Anti‐Ago2 RIP analyses were applied in U87MG and A172 cells transfected with miR‐NC mimics or miR‐454‐3p mimics, the enrichment of NEAT1 was detected by qRT‐PCR. (K, L) *p < 0.05; **p < 0.01; ***p < 0.01.

FIGURE 3

miR‐454‐3p targeted GJA1 (Cx43). (A) Venn diagram was used to achieve the potential genes that miR‐454‐3p may target the binding region. (B) Schematic illustration of the predicted binding sites between GJA1 and miR‐454‐3p, and mutation of potential miR‐454‐3p binding sequence in GJA1. (C) Dual luciferase assay was applied to assess the seed‐matching sites or mutant sites between GJA1 and miR‐454‐3p in HEK‐293T. (D, E) The protein levels of Cx43 in U87MG and A172 cells transfected with miR‐454‐3p mimics, miR‐NC mimics or miR‐454‐3p inhibitor, miR‐NC inhibitor. GAPDH was the control. Relative density values were counted using image J software. (F) The mRNA expression level of GJA1 transfected with miR‐NC mimics or miR‐454‐3p mimics. (G) Relative expression of GJA1 in normal brain and GBM tissues, normalized to GAPDH. (H) The Pearson's correlation analysis was performed to analyze the correlation between GJA1 and miR‐454‐3p. *p < 0.05; **p < 0.01; ***p < 0.01.

FIGURE 4

Cx43 is responsible for NEAT1‐mediated TMZ resistance. (A) The Pearson's correlation analysis was performed to analyze the correlation between NEAT1 and GJA1. (B) The mRNA expression level of GJA1 in U87MG cells with shNC and shNEAT1 group, **p < 0.01. (C, F) The protein levels of Cx43, BCL2, and BAX in U87MG cells with shNC + Vector, shNEAT1 + Vector or shNEAT1 + Cx43. ***p < 0.001, compared to shNC + Vector group; ^### p < 0.001, compared to shNEAT1 + Vector group. (D) The mRNA expression level of NEAT1 transfected with GJA1 siRNA or NC siRNA. ***p < 0.01. (E) Cell proliferation assays of shNC and shNEAT1 in U87MG and A172 cells transfected with Cx43 plasmid or Vector, with TMZ treatment. ***p < 0.001, compared to shNC + Vector group; ^## p < 0.01, ^### p < 0.001, compared to shNEAT1 + Vector group. (G, H) Flow cytometric analysis of si‐NC and si‐NEAT1 in U87MG and A172 cells transfected with Cx43 plasmid or vector, with TMZ treatment (100 μM, 48 h). **p < 0.01, ***p < 0.001. (I) The SLDT assay was conducted to detect the influence of shNC + Vector, shNEAT1 + Vector or shNEAT1 + Cx43 on gap junction intercellular communication, which was indicated by the dye spreading area. Scale bar = 100 μm.

FIGURE 5

NEAT1 depletion enhances TMZ sensitivity in vivo. (A) Representative bioluminescence images of intracranial xenografts bearing U87MG‐Luc/shNC or U87‐Luc/shNEAT1 with TMZ (20 mg/kg) or same volume of physiological saline at indicated time points, n = 6. (B) Intracranial tumors in brain were shown by H&E staining. Scale bar = 100 μm. (C) Kaplan–Meier survival curve of nude mice is shown. (D, E) The total proteins were extracted from xenografts and subjected to Western blot analysis for Cx43 expression. GAPDH was the control. *p < 0.05; ***p < 0.001.