Alternative Polyadenylation Regulates Patient-specific Tumor Growth by Individualizing the MicroRNA Target Site Landscape.
Kim, S.; Bai, Y.; Diergaarde, B.; Tseng, G. C.; Park, H. J.
Show abstract
BackgroundAlternative polyadenylation (APA) shortens or lengthens the 3-untranslated region (3-UTR) of hundreds of genes in cancer. While APA genes modify microRNA target sites in the 3-UTRs to promote tumorigenesis, previous studies have focused on a subset of the modification landscape.\n\nMethodFor comprehensive understanding of the function of global APA events, we consider the total target site landscape of microRNAs that are significantly and collectively modified by global APA genes. To identify such microRNAs in spite of complex interactions between microRNAs and the APA genes, we developed Probabilistic Inference of MicroRNA Target Site Modification through APA (PRIMATA-APA).\n\nResultsRunning PRIMATA-APA on TCGA breast cancer data, we identified that global APA events concentrate to modify target sites of particular microRNAs (target-site-modified-miRNA or tamoMiRNA). TamoMiRNAs are enriched for microRNAs known to regulate cancer etiology and treatments. Also, their target genes are enriched in cancer-associated pathways, suggesting that APA modifies target sites of tamoMiRNAs to progress tumors. Knockdown of NUDT21, a master 3-UTR regulator in HeLa cells, confirmed the causal role of tamoMiRNAs for tumor growth.\n\nConclusionsFurther, the expressions of tamoMiRNA target genes, enriched in cancer-associated pathways, vary across tumor samples as a function of patient-specific APA events, suggesting that APA is a novel regulatory axis for interpatient tumor heterogeneity.
Matching journals
The top 6 journals account for 50% of the predicted probability mass.
Similar papers in this journal
- LMSM: a modular approach for identifying lncRNA related miRNA sponge modules in breast cancer 95%
- Exploring tumor-normal cross-talk with TranNet: role of the environment in tumor progression 94%
- Inferring latent temporal progression and regulatory networks from cross-sectional transcriptomic data of cancer samples 94%