A positive feedback loop of lncRNA-RMRP/ZNRF3 axis and Wnt/β-catenin signaling regulates the progression and temozolomide resistance in glioma

Abstract

Drug resistance strikingly limits the therapeutic effect of temozolomide (TMZ) (a common drug for glioma). Long non-coding RNA (lncRNA) RMRP has been found to be implicated in glioma progression. However, the effect of RMRP on TMZ resistance along with related molecular mechanisms is poorly defined in glioma. In the present study, RMRP, ZNRF3, and IGF2BP3 were screened out by bioinformatics analysis. The expression levels of lncRNAs and mRNAs were measured by RT-qPCR assay. Protein levels of genes were detected by western blot and immunofluorescence assays. ZNRF3 mRNA stability was analyzed using Actinomycin D assay. Cell proliferative ability and survival rate were determined by CCK-8 assay. Cell apoptotic pattern was estimated by flow cytometry. The effect of RMRP knockdown on the growth of TMZ-treated glioma xenograft tumors was explored in vivo. The relationships of IGF2BP3, RMRP, and ZNRF3 were explored by bioinformatics prediction analysis, RNA immunoprecipitation, luciferase, and RNA pull-down, and chromatin immunoprecipitation assays. The results showed that RMRP was highly expressed in glioma. RMRP knockdown curbed cell proliferation, facilitated cell apoptosis and reduced TMZ resistance in glioma cells, and hindered the growth of TMZ-treated glioma xenograft tumors. RMRP exerted its functions by down-regulating ZNRF3 in glioma cells. IGF2BP3 interacted with RMRP and ZNRF3 mRNA. IGF2BP3 knockdown weakened the interaction of Argonaute 2 (Ago2) and ZNRF3. RMRP reduced ZNRF3 expression and mRNA stability by IGF2BP3. RMRP knockdown inhibited β-catenin expression by up-regulating ZNRF3. The inhibition of Wnt/β-catenin signaling pathway by XAV-939 weakened RMRP-mediated TMZ resistance in glioma cells. β-catenin promoted RMRP expression by TCF4 in glioma cells. In conclusion, RMRP/ZNRF3 axis and Wnt/β-catenin signaling formed a positive feedback loop to regulate TMZ resistance in glioma. The sustained activation of Wnt/β-catenin signaling by RMRP might contribute to the better management of cancers.

Fig. 1. Expression profile analysis of lncRNAs in glioma and screening of interested lncRNAs.

A The top 100 up-regulated lncRNAs in glioma versus normal brain tissues. B Expression analyses of 3 lncRNAs (RMRP, HOXA-AS3, and CASC9) in glioma versus normal tissues in the GEPIA database. C Expression levels of RMRP, HOXA-AS3, CASC9 in tumor tissues isolated from patients with recurrent glioma (n = 12) and glioma patients without any treatment (n = 20) were measured through RT-qPCR assay. *P < 0.05. **P < 0.01.

Fig. 2. RMRP knockdown curbed cell proliferation, facilitated cell apoptosis, and reduced TMZ resistance in glioma cells.

A U251 and LN229 cells were transfected with si-NC, si-RMRP#1, si-RMRP#2, or si-RMRP#3. Forty-eight hours later, RMRP level was examined through RT-qPCR assay (n = 3). B Cell proliferative ability was estimated through CCK-8 assay in U251 and LN229 cells transfected with si-NC or si-RMRP#3 at 0, 24, 48, 72 h post-plating (n = 3). C Cell apoptotic rate was determined by flow cytometry in U251 and LN229 cells transfected with si-NC or si-RMRP#3 at 48 h after transfection (n = 3). D U251, LN229, U251/TMZ, and LN229/TMZ cells were treated with different doses of TMZ for 24 h, followed by the measurement of TMZ IC₅₀ values via CCK-8 assay (n = 3). E and F U251, LN229, U251/TMZ, and LN229/TMZ cells were transfected with si-NC or si-RMRP#3. Transfected cells were seeded into 96-well plates for 24 h and then treated with different concentrations of TMZ for an additional 24 h. Next, the cell survival rate was determined through the CCK-8 assay (n = 3). *P < 0.05. **P < 0.01. ***P < 0.001.

Fig. 3. Effects of RMRP knockdown on the growth of TMZ-treated glioma xenograft tumors.

A The volumes of xenograft tumors were measured every 2 days after TMZ treatment. B Tumor weight was determined on day 14 after TMZ treatment. C Representative photographs of xenograft tumors in control, sh-NC, and sh-RMRP groups. D RMRP level was measured by RT-qPCR assay in xenograft tumors of sh-NC and sh-RMRP groups. Each group contains 6 mice. **P < 0.01. ***P < 0.001.

Fig. 4. ZNRF3 was a target of RMRP3.

A Survival analysis of glioma patients with high or low expression of TOMM6, ZNRF3, EEF1G, ADORA2A, KIAA0408, KREMEN1, and XBP1. B RMRP level in the nucleus and cytoplasm of U251 and LN229 cells was measured by RT-qPCR assay after the isolation of nuclear and cytoplasmic RNA (n = 3). C, D The effects of RMRP knockdown on ZNRF3 mRNA and protein expression were measured by RT-qPCR and western blot assays at 48 h after transfection in U251 and LN229 cells (n = 3). E The effect of RMRP knockdown on ZNRF3 mRNA stability was examined through ActD and RT-qPCR assays in U251 and LN229 cells (n = 3). **P < 0.01.

Fig. 5. ZNRF3 down-regulation weakened the effects of RMRP knockdown on glioma cell proliferation, apoptosis, and TMZ resistance.

A Knockdown efficiency analysis of si-ZNRF3#1, si-ZNRF3#2, or si-ZNRF3#3 in U251 and LN229 cells by RT-qPCR assay at 48 h post-transfection (n = 3). B–D U251 and LN229 cells were transfected with si-NC, si-RMRP, or si-RMRP + si-ZNRF3, followed by the measurement of cell proliferative ability B (n = 3), cell apoptotic rate C (n = 3), and TMZ IC₅₀ values D (n = 3). E TMZ IC₅₀ values were determined by CCK-8 assay in U251/TMZ and LN229/TMZ cells treated with different doses of TMZ (n = 3). **P < 0.01. ***P < 0.001.

Fig. 6. RMRP regulated ZNRF3 expression and ZNRF3 mRNA stability through RBP IGF2BP3.

A Venn analysis for RMRP-RBPs, ZNRF3-RBPs, and differentially expressed genes in glioma versus normal tissues. RMRP-RBPs (green): RBPs with the possibility to bind with RMRP (predicted by the POSTAR database). ZNRF3-RBPs (blue): RBPs with the possibility to bind with ZNRF3 mRNA (predicted by the POSTAR database). Diffgenes (red): Differentially expressed genes in glioma versus normal brain tissues (data were downloaded from the TCGA/GTEx databases). B RIP assay was performed using IgG or IGF2BP3 antibody in U251 cells. Next, RMRP and ZNRF3 mRNA levels enriched by IgG or IGF2BP3 antibodies were measured by RT-qPCR assay (n = 3). C RNA pull-down assay was performed using biotin-labeled ZNRF3 mRNA 3’UTR fragment 1, fragment 2, and fragment 3 in U251 cells, followed by the detection of IGF2BP3 protein level via western blot assay. D U251 cells were transfected with si-NC or si-IGF2BP3. At 48 h after transfection, a RIP assay was performed using IgG or Ago2 antibody. E U251 and LN229 cells were transfected with pcDNA3.1 or pcDNA3.1-RMRP. Forty-eight hours later, the RMRP level was measured by RT-qPCR assay (n = 3). F, G U251 and LN229 cells were transfected with pcDNA3.1, pcDNA3.1-RMRP, pcDNA3.1-RMRP + si-NC, or pcDNA3.1-RMRP + si-IGF2BP3. F, G ZNRF3 mRNA and protein levels were determined by RT-qPCR and western blot assays (n = 3). H ZNRF3 mRNA stability was tested by Actinomycin D and RT-qPCR assays (n = 3). **P < 0.01. ***P < 0.001.

Fig. 7. RMRP knockdown inhibited β-catenin expression by up-regulating ZNRF3 in glioma cells.

A, B Cells were transfected with si-NC or si-ZNRF3 for 48 h, followed by the measurement of β-catenin protein level via western blot assay A (n = 3) and IF assay (n = 1) B. C, D Cells were transfected with si-NC, si-RMRP, or si-RMRP + si-ZNRF3. At 48 h post-transfection, western blot assay C (n = 3) and IF assay D (n = 1) were performed to examine the protein expression pattern of β-catenin. E U251/TMZ and LN229/TMZ cells were transfected with pcDNA3.1 or pcDNA3.1-RMRP for 24 h and then stimulated with or without DMSO/XAV-939 (10 µM) and different doses of TMZ for 24 h, followed by the measurement of TMZ IC₅₀ values via CCK-8 assay (n = 3).

Fig. 8. β-catenin facilitated RMRP transcription by TCF4 in glioma cells.

A, B U251 and LN229 cells were transfected with pcDNA3.1 empty vector, pcDNA-β-catenin, pcDNA-TCF4, or pcDNA-β-catenin+pcDNA-TCF4. Forty-eight hours later, the RMRP level was examined by RT-qPCR assay (n = 3). C LN229 cells were transfected with TCF/LEF1-Luc reporter, TCF/LEF1-Luc reporter+pcDNA3.1 empty vector, or TCF/LEF1-Luc reporter+ pcDNA-β-catenin, followed by the detection of luciferase activity at 48 h after transfection. D LN229 cells were co-transfected with pRL-TK Renilla luciferase plasmid, pGL3-luc-TBE1/pGL3-luc-TBE2/pGL3-luc-TBE3 and pcDNA3.1-TCF4 plasmid/pcDNA3.1 empty vector. At 48 h post-transfection, luciferase activities were measured. E LN229 cells were co-transfected with pRL-TK Renilla luciferase plasmid, pcDNA3.1-TCF4 plasmid/pcDNA3.1 empty vector, and pGL3-luc-TBE3(WT)/pGL3-luc-TBE(MUT) reporter. Forty-eight hours later, luciferase activities were measured. F LN229 cells were transfected with pcDNA3.1 empty vector or pcDNA-β-catenin. Forty-eight hours later, a CHIP assay was carried out using TCF4 or IgG antibody. The level of RMRP gene fragment enriched by IgG or TCF4 antibody was measured by PCR. **P < 0.01. ***P < 0.001.