Abstract
Transcription factors (TFs) are important oncogenic regulator and are altered during tumor initiation and progression. Our oncogenomic analysis of gene expression data from large clinically-annotated patient samples identified TF Dp1 as one of the most important gene affecting both overall and event-free survival in multiple myeloma (MM). Elevated Dp1 expression was predictive of adverse clinical outcome, independent of Dp1 protein partners, E2Fs and RB, suggesting direct impact of Dp1 and providing the rationale to further evaluate its specific role in MM.
We have observed high level of Dp1 expression and activity in MM cells which was further induced after interaction with bone marrow stromal cells (BMSC). Moreover, Dp1 knock-down using specific sh-RNA decreased MM cell growth in 5 MM cell lines with different genetic background, with a concomitant G1 arrest and late induction of apoptosis. These data suggest a role for Dp1 in MM cell proliferation and survival and established a rationale to identify its molecular impact.
We have further characterized Dp1 activity using chromatin immunoprecipitation with Dp1 or E2F1 specific antibody followed by genome wide sequencing (ChIP-Seq) to identify Dp1-binding regions in MM. We have identified 2783 exclusive Dp1 binding regions in two MM cell lines. Examination of Dp1 and E2F1 binding revealed that Dp1 co-occupies 65% of the binding sites with E2F1. The DAVID gene set enrichment analysis showed that identified genes were related to cell cycle, as well as to transcriptional and translational processes. To assess the functional consequences of Dp1 DNA binding, the ChIP-Seq data were supplemented with gene expression profile of MM1S cells following shRNA-mediated Dp1 and E2F knock-downs. Integrated analysis incorporating ChIP-seq and expression data identified Dp1 response program in MM. 805 (46%) of 1752 differentially expressed genes also have binding sites for Dp1 and likely are direct transcriptional targets of Dp1 in MM. Enrichment analysis of direct targets revealed that the most strongly enriched pathways for both Dp1 and E2F1 genes combined were related to the cell cycle, especially DNA replication, repair, and metabolism. Interestingly, pathway analysis identified ‘‘regulation of RNA metabolic processes’’ (40 target genes), ‘‘RNA processing’’ (93 target genes) ‘‘RNA splicing’’ (95 genes), and ‘‘RNA binding’’ (53 genes) as statistically significant RNA-related categories enriched among Dp1 target genes, suggests role of Dp1 in RNA splicing.
Based on our previous data showing that dysregulated alternate splicing (AS) has significant impact on overall clinical outcome MM, we evaluated the expression of Dp1-modulated splicing factors in our clinically annotated cohort of MM patients and 5 normal PCs. We identified 23 SFs upregulated in MM compared to normal plasma cells. Importantly, the increased expression of 12 of these SFs was linked with poor prognosis in this cohort of myeloma patients. Our data show for the first time that SFs are upregulated in myeloma and link to clinical outcome. To evaluate the impact of Dp1 on alternate splicing (AS), we performed genome-wide analysis of alternate splicing in total RNA from Dp1 silenced MM1S cells using Human Exon1 ST arrays. Splicing profiles showed that Dp1 knock down causes widespread changes in AS. We have identified 3683 genes whose one exon has splicing index more than 1.5 in in shDP1 compared to control pLKO.1-transduced MM1S cells, suggesting impact of Dp1 silencing on alternate splicing. We are now evaluating impact of a peptide able to disrupt Dp1-E2F1 binding with consequent effect on MM cell growth and alternate splicing.
In conclusion, our investigation showed that the Dp1/E2F1 signaling pathway plays significant role in myeloma and can directly activate transcription of specific SFs with effect on alternate splicing and potential functional, clinical and therapeutic implications in myeloma.
Disclosures:
No relevant conflicts of interest to declare.
BICORN: An R package for integrative inference of de novo cis-regulatory modules
