AEP-cleaved DDX3X induces alternative RNA splicing events to mediate cancer cell adaptation in harsh microenvironments
Wenrui Zhang, … , Jiayi Chen, Yingying Lin
Wenrui Zhang, … , Jiayi Chen, Yingying Lin
Published November 21, 2023
Citation Information: J Clin Invest. 2024;134(3):e173299. https://doi.org/10.1172/JCI173299.
View: Text | PDF
Research Article Oncology

AEP-cleaved DDX3X induces alternative RNA splicing events to mediate cancer cell adaptation in harsh microenvironments

Abstract

Oxygen and nutrient deprivation are common features of solid tumors. Although abnormal alternative splicing (AS) has been found to be an important driving force in tumor pathogenesis and progression, the regulatory mechanisms of AS that underly the adaptation of cancer cells to harsh microenvironments remain unclear. Here, we found that hypoxia- and nutrient deprivation–induced asparagine endopeptidase (AEP) specifically cleaved DDX3X in a HIF1A-dependent manner. This cleavage yields truncated carboxyl-terminal DDX3X (tDDX3X-C), which translocates and aggregates in the nucleus. Unlike intact DDX3X, nuclear tDDX3X-C complexes with an array of splicing factors and induces AS events of many pre-mRNAs; for example, enhanced exon skipping (ES) in exon 2 of the classic tumor suppressor PRDM2 leads to a frameshift mutation of PRDM2. Intriguingly, the isoform ARRB1-Δexon 13 binds to glycolytic enzymes and regulates glycolysis. By utilizing in vitro assays, glioblastoma organoids, and animal models, we revealed that AEP/tDDX3X-C promoted tumor malignancy via these isoforms. More importantly, high AEP/tDDX3X-C/ARRB1-Δexon 13 in cancerous tissues was tightly associated with poor patient prognosis. Overall, our discovery of the effect of AEP-cleaved DDX3X switching on alternative RNA splicing events identifies a mechanism in which cancer cells adapt to oxygen and nutrient shortages and provides potential diagnostic and/or therapeutic targets.

Authors

Wenrui Zhang, Lu Cao, Jian Yang, Shuai Zhang, Jianyi Zhao, Zhonggang Shi, Keman Liao, Haiwei Wang, Binghong Chen, Zhongrun Qian, Haoping Xu, Linshi Wu, Hua Liu, Hongxiang Wang, Chunhui Ma, Yongming Qiu, Jianwei Ge, Jiayi Chen, Yingying Lin

×

Figure 6

tDDX3X-C interacts with spliceosome factors and regulates AS in a manner partially dependent on hnRNAPA1.

Options: View larger image (or click on image) Download as PowerPoint
tDDX3X-C interacts with spliceosome factors and regulates AS in a manner...
(A) Immunoblots of mCherry-tagged tDDX3X-C, lamin B and β-tubulin in the nuclear and cytoplasmic fractions of U87-MG-mCherry-tDDX3X-C cells. mCherry antibody (1:50) was used to precipitate mCherry-tDDX3X-C in the nuclear lysate of 2 cell samples; #1, sample 1; #2, sample 2; C, cytoplasmic; N, nuclear. (B) Venn diagram showing the intersection of 2 sets of proteins detected by IP-MS. There were 611 kinds of proteins in the intersection and 405 kinds of nuclear proteins are represented in green circle. (C) GO and KEGG term enrichment analyses of the DDX3X-interacting nuclear proteins identified from the IP-MS data. (D) String 10.5 program (http://string-db.org) analyses showing the interaction networks among 21 candidate proteins associated with mRNA splicing. (E) Expression correlation analysis of DDX3X and splicing factors based on the TCGA database. (F) HEK293T cells were cotransfected with eGFP-U2AF2/mCherry-tDDX3X-C. Scale bar: 40 μm. (G) Diagram of BIFC. (H) HEK293T cells were transfected with YN155-empty vector/YC156-tDDX3X-C, YN155-SRSF1/YC156-tDDX3X-C, or YN155-hnRNPA1/YC156-tDDX3X-C. Scale bar: 100 μm. (I) Co-IP assays of HEK293T cells transfected with the indicated plasmids. (J) U87-MG and MDA-MB-231 cells expressing tDDX3X-C (1 × 107) were harvested for Co-IP assays. (K) RT-qPCR assays of the indicated mRNAs in U87-MG and MDA-MB-231 cells. (L) PCR and AGE analyses for PRDM2 exon 2 in U87-MG and MDA-MB-231 in the following groups: NC, shSRSF1-3 and shhnRNPA1-1; quantification of percent spliced in (PSI). (M) PCR and AGE analyses for PRDM2 exon 2 and ARRB1 exon 13 in U87-MG and MDA-MB-231 grouped by NC, hnRNPA1 OE, and hnRNPA1 OE/shDDX3X-C-1. Data were plotted as the mean ± SEM. Statistical analysis was performed using 1-way ANOVA followed by Dunnett’s multiple comparisons test (K and L) and Tukey’s multiple comparisons test (M). *P < 0.05, ***P < 0.001, ****P < 0.0001. Data shown are representative of 3 independent experiments.