Gastric cancer is the third largest cause of cancer related deaths in the world. GC is highly prevalent in Asia with a 5 year survival rate < 20% and the mainstay of current therapy includes surgery and chemotherapy. Anti Her2 Mab in Her2+ patients is the only available targeted therapy. While genomic mutations, alterations in gene expression and epigenetic features in gastric cancer have been extensively studied, little is known about the role of aberrant mRNA splicing and its upstream regulators in oncogenesis. We performed strand specific next-gen RNA sequencing in 20 gastric tumors and their matched adjacent normal tissues. Using transcriptomic and gene knock down studies we have developed a computational RNA splicing network model to test the functionality of dysregulated splice-factors and their associated aberrantly spliced isoforms in gastric cancer. Differential gene expression analysis for known splicing factors in our RNA-seq data sets followed by q-RT-PCR and western blot experiments across a panel of gastric cancer cell lines confirmed that 13 SFs are dysregulated in gastric cancer: SF3B3, HNRNPF, HNRNPL, HNRNPI, NRNPH2, HNRNPA1, HNRNPA2/B1, HNRNPK and p62 are up-regulated, while ESRP2, ESRP1, MBNL1 and SRP75 are down-regulated. Studies involving siRNA mediated knockdown of these 13 dysregulated splicing factors (SFs) in gastric cancer cell lines demonstrate an emerging relationship between splicing factor expression, alternative splicing events and cellular function.

Using the ‘MISO’ (mixture of Isoforms probabilistic model) platform on our tumor/normal matched pair RNA-seq dataset we identified 291 gastric cancer associated alternatively spliced events (ASEs). We validated the aberrant splicing and oncogenic activation of CD44, FGFR2, INSR, CD47, MST1R. We find that other genes such as MBNL1, CLSTN1, LAS1L, SPAG9, TPM1, RPS24 exhibit recurrent aberrant ASEs in gastric cancer. Through a computational co-correlation analysis we have linked 178/291 gastric cancer associated ASEs mainly with the dysregulation of ESRP2, HNRNPI, HNRNPF, SF3B3 and MBNL1, thereby developing a prototype network-model of aberrantly spliced genes linked to upstream differentially expressed splice factors. The altered splicing networks in gastric cancer are associated with changes in EMT status, cytoskeletal reorganization pathways, cell migration, adhesion as well as differentiation processes and protein phosphorylation/kinase cascades. In summary, our findings comprise of the first comprehensive map of deregulated splicing networks in gastric cancer. This network provides the experimental basis for determination of the role of altered splicing in tumorigenesis and in identifying novel oncogenes/tumor suppressors in gastric cancer.

Citation Format: Debleena Ray. Aberrant splicing network delineated from next-gen RNA sequencing analyses identifies activated oncogenes and pathways in gastric cancer. [abstract]. In: Proceedings of the 107th Annual Meeting of the American Association for Cancer Research; 2016 Apr 16-20; New Orleans, LA. Philadelphia (PA): AACR; Cancer Res 2016;76(14 Suppl):Abstract nr LB-332.