CRNDE mediated hnRNPA2B1 stability facilitates nuclear export and translation of KRAS in colorectal cancer

Abstract

Development of colorectal cancer (CRC) involves activation of Kirsten rat sarcoma viral oncogene homolog (KRAS) signaling. However, the post-transcriptional regulation of KRAS has yet to be fully characterized. Here, we found that the colorectal neoplasia differentially expressed (CRNDE)/heterogeneous nuclear ribonucleoprotein A2/B1 (hnRNPA2B1) axis was notably elevated in CRC and was strongly associated with poor prognosis of patients, while also significantly promoting CRC cell proliferation and metastasis both in vitro and in vivo. Furthermore, CRNDE maintained the stability of hnRNPA2B1 protein by inhibiting E3 ubiquitin ligase TRIM21 mediated K63 ubiquitination-dependent protein degradation. CRNDE/hnRNPA2B1 axis facilitated the nuclear export and translation of KRAS mRNA, which specifically activated the MAPK signaling pathway, eventually accelerating the malignant progression of CRC. Our findings provided insight into the regulatory network for stable hnRNPA2B1 protein expression, and the molecular mechanisms by which the CRNDE/hnRNPA2B1 axis mediated KRAS nucleocytoplasmic transport and translation, deeply underscoring the bright future of hnRNPA2B1 as a promising biomarker and therapeutic target for CRC. By hindering hnRNPA2B1 from binding to the E3 ubiquitin ligase TRIM21, whose mediated ubiquitin-dependent degradation was thereby inhibited, CRNDE protected the stability of hnRNPA2B1's high protein expression in CRC. Supported by the high level of the oncogenic molecule CRNDE, hnRNPA2B1 bound to KRAS mRNA and promoted KRAS mRNA nucleus export to enter the ribosomal translation program, subsequently activating the MAPK signaling pathway and ultimately accelerating the malignant progression of CRC.

By hindering hnRNPA2B1 from binding to the E3 ubiquitin ligase TRIM21, whose mediated ubiquitin-dependent degradation was thereby inhibited, CRNDE protected the stability of hnRNPA2B1’s high protein expression in CRC. Supported by the high level of the oncogenic molecule CRNDE, hnRNPA2B1 bound to KRAS mRNA and promoted KRAS mRNA nucleus export to enter the ribosomal translation program, subsequently activating the MAPK signaling pathway and ultimately accelerating the malignant progression of CRC.

Fig. 1. Upregulation of hnRNPA2B1 is associated with CRC poor outcomes.

Correlation analysis of hnRNPA2B1 expression with pathological grade (A) and clinical TNM stage (B) in TMA cohort (n = 87). C Kaplan–Meier survival curves of CRC patients on TMA grouped by hnRNPA2B1 expression (P = 0.0403). D EdU assays examined the difference in the role of hnRNPA2 and hnRNPB1 variants on the proliferative capacity of CRC cells. n = 3 independent biological replicates. E Representative images of mice-bearing tumors derived from xenograft nude mouse model injected subcutaneously with sh-Ctrl or sh-hnRNPA2B1 DLD1 cells. Growth curve (F), weight (G) and size (H) of subcutaneous xenograft tumors were indicated. n = 6 biologically independent samples. I Representative images of metastatic loci in the lung derived from mouse model that tail vein injected with WT or KO HCT116 cells. Black arrows indicate the tumor nodules on lung surfaces. J Representative HE staining of lung metastatic lesion. Scale bar, 500 μm (left×2panel); 20 μm (right×40 panel). K Quantification of lung metastatic foci from nude mice. n = 5 biologically independent samples. A two-tailed Student’s t-test and one-way ANOVA were used for statistical analysis, respectively. ^*P < 0.05, ^**P < 0.01, ^***P < 0.001. Data represent mean ± SD.

Fig. 2. CRNDE binds to hnRNPA2B1 and protects its protein stability in CRC.

A LncRNAs differentially expressed in CRC were screened in the GSE9348 dataset and TCGA database at P < 0.01 and with a fold change>2, which were then intersected with hnRNPA2B1-bound ncRNAs that obtained from the online databases ENCORI and POSTAR3. CRNDE and LINC00294 were enriched. B Comparison of the differential expression of CRNDE and LINC00294 in CRC in TCGA and various GEO databases, as well as their binding scores to hnRNPA2B1 in ENCORI and POSTAR3. C Correlation analysis between expression levels of hnRNPA2B1 and CRNDE in CRC patients (R = 0.46, P = 0.024, n = 24). D RNA pulldown assays were performed using biotin-labeled CRNDE and its antisense, followed by western blot analysis. E RIP assays were performed using hnRNPA2B1 specific antibody and IgG, and the co-precipitated RNA was subjected to qRT-PCR to measure the enrichment of CRNDE. n = 3 independent biological replicates. F Western blot detected the expression changes of hnRNPA2B1 protein in SW480 and DLD1 cells with stable CRNDE knockdown or overexpression. The effects of CRNDE (G) and hnRNPA2B1 (H) alterations on each other’s RNA levels were examined by qPCR. I Western blot detection of hnRNPA2B1 in CRNDE silencing cells treated with CHX (20 μg/mL) for the indicated times. n = 3 independent biological replicates. J CRNDE knockdown cells that were treated with MG132 (25 μmol/L) for 6 h and then the hnRNPA2B1 levels were tested by western blot. K Immunoprecipitated hnRNPA2B1 in extracts of CRC cells with stable CRNDE deletion, DUB inhibitor WP1130 (5 μM for 6 h) treatment, or USP33 knockdown was immunoblotted with an anti-Ubiquitin antibody. WP1130 and si-USP33 were employed as the positive controls. L Ubiquitin plasmids with either only K48 or K63 retained, or with only K48 or K63 mutated, along with their wild-type ubiquitin controls, were transfected into CRC cells. Ubiquitin levels in hnRNPA2B1-IP samples were analyzed by immunoblotting. A two-tailed unpaired Student’s t-test and Pearson correlation analysis were used for statistical analysis, respectively. ^**P < 0.01, ^***P < 0.001, n.s. not significantly. Data represent mean ± SD.

Fig. 3. CRNDE stabilizes hnRNPA2B1 by preventing TRIM21-mediated ubiquitination.

Co-IP for hnRNPA2B1 (A) and TRIM21 (B) were performed in indicated CRC cells. IgG was used as negative control. The co-IPs were analyzed by western blot using anti-hnRNPA2B1 or anti-TRIM21 antibodies. C The influence of knocking down TRIM21 on the expression levels of hnRNPA2B1 was examined by western blot. The binding of hnRNPA2B1 to CRNDE in protein extracts from indicated cells transiently transfected with si-TRIM21 or si-NC was measured by CRNDE-pulldown (D) and hnRNPA2B1-RIP (E). n = 3 independent biological replicates. F Immunoprecipitated endogenous hnRNPA2B1 from cells with TRIM21 depletion was immunoblotted with anti-Ubiquitin antibody. G The levels of CRNDE were detected by qPCR after knockdown of TRIM21. n = 3 independent biological replicates. Protein expression of TRIM21 detected by western blot after down-regulation of hnRNPA2B1 (H) or CRNDE (I). J TRIM21 was attenuated in CRNDE-silenced cells and IP of hnRNPA2B1 was carried out. Protein changes of total ubiquitin, k63 and K48 ubiquitin, and TRIM21 in immunoprecipitates were identified by western blot, respectively. K IP assays were conducted by immobilizing antibodies against hnRNPA2B1 or IgG on Protein A magnetic beads, and the precipitated TRIM21 and hnRNPA2B1 proteins from cell lysates with or without CRNDE overexpression were detected by western blot. A two-tailed unpaired Student’s t-test was used for statistical analysis. ^**P < 0.01, ^***P < 0.001, n.s. not significantly. Data represent mean ± SD.

Fig. 4. hnRNPA2B1 is required for CRNDE regulation of CRC cell proliferation and metastasis.

The hnRNPA2B1 overexpression plasmid or its empty vector was transfected into stable knockdown CRNDE CRC cells SW480 and DLD1 and their proliferative (A), invasive and migratory (B–D) abilities were assessed by functional rescue assays. n = 3 independent biological replicates. Scale bar, 100 μm (B); 200 μm (D). Functional rescue assays were performed in CRNDE overexpression cells with si-hnRNPA2B1 or si-NC transfection to examine the alterations in cell proliferation (E) and metastatic (F–H) capacity. n = 3 independent biological replicates. Scale bar, 100 μm (F); 200 μm (H). A two-tailed unpaired Student’s t-test and one-way ANOVA were used for statistical analysis, respectively. ^*P < 0.05, ^**P < 0.01, ^***P < 0.001. Data represent mean ± SD.

Fig. 5. CRNDE/hnRNPA2B1 axis targets KRAS and activates MAPK pathway in CRC.

Western blot analysis for the expression changes of MAPK signaling pathway-related marker P38, ERK1/2, JNK and their corresponding phosphorylated proteins in SW480 and DLD1 cells with stably silencing or overexpressing hnRNPA2B1 (A) or CRNDE (B). C, D qPCR measure of changes in KRAS mRNA levels following alterations in hnRNPA2B1 or CRNDE. n = 3 independent biological replicates. E Detection of hnRNPA2B1 protein expression by western blot in CRC cells after knockdown of KRAS. F-G Quantification of hnRNPA2B1 (F) and CRNDE (G) RNA levels in KRAS-silenced CRC cells by qPCR, respectively. n = 3 independent biological replicates. H–L Transfection of KRAS overexpression plasmids and their controls into stably silenced hnRNPA2B1 cells for functional rescue experiments. n = 3 independent biological replicates. Scale bar, 100 μm (I); 200μm (K). A two-tailed unpaired Student’s t-test and one-way ANOVA were used for statistical analysis, respectively. ^*P < 0.05, ^**P < 0.01, ^***P < 0.001. n.s. not significantly. Data represent mean ± SD.

Fig. 6. CRNDE cooperates with hnRNPA2B1 to promote KRAS mRNA nuclear export and translation.

A Flag-IP assays were performed in CRC cells with Flag-tagged hnRNPA2B1 overexpression plasmids transfection. B KRAS-pulldown assays were conducted and then were analyzed by western blot. C hnRNPA2B1-RIP-qPCR assays were undertaken to examine the enrichment of KRAS mRNA. KRAS-pulldown assays were performed in cells stably silenced (D) or overexpressing (E) CRNDE to evaluate the altered binding efficiency of hnRNPA2B1. F The hnRNPA2B1-RIP assays were executed after knockdown of CRNDE. G qPCR detection of KRAS variant KRAS4A and KRAS4B. H, I The pulldown products of CRNDE and KRAS were identified by qPCR. J SUnSET assays were employed to detect the puromycin-labeled proteins in CRC cells with decreasing or increasing of hnRNPA2B1 followed by treatment with 10 μg/mL puromycin for 15 min. K hnRNPA2B1-deficient cells with Biotin-dC-puromycin treatment for 24 h were collected for precipitation with streptavidin beads, and the products were measured by western blot. L The ribosomal fractions were separated on a sucrose gradient and their RNA was extracted. Amount of KRAS mRNA was analyzed by qRT-PCR. M Western blot detection of hnRNPA2B1 levels in the nuclear and cytoplasmic fractions of CRC cells stably overexpressing CRNDE. N–O Cytoplasm and nucleus of CRC cells with stable hnRNPA2B1 knockdown and overexpression were isolated, of which KRAS mRNA levels were subsequently detected via qPCR. P, Q Cytoplasm and nuclei of CRC cells with CRNDE upregulation were isolated and extracted, and then hnRNPA2B1-RIP-qPCR analysis for KRAS enrichment were performed to determine the effect of CRNDE on the distribution and efficiency of hnRNPA2B1-KRAS mRNA interaction. n = 3 independent biological replicates. A two-tailed unpaired Student’s t-test and one-way ANOVA were used for statistical analysis, respectively. ^*P < 0.05, ^**P < 0.01, ^***P < 0.001. n.s. not significantly. Data represent mean ± SD.