3' Untranslated Region (UTR) Alterations Are Frequently Targeted By MM-Related Mirnas and Affects the Clinical Outcome

More than half of human genes use alternative cleavage and polyadenylation to generate alternative 3' untranslated region (3'UTR) isoforms that play a role in gene expression regulation. The 3' untranslated region (3'UTR) of mRNA contains elements that play regulatory roles in polyadenylation, localization, translation efficiency, and mRNA stability. Although relative contributions of different regulatory mechanisms remain poorly understood, several recent studies showed that alterations in 3'UTRs might affect protein localization as well as their interactions with other proteins.

Here, we obtained global measurements of 3' UTR sequencing and RNA-seq in MM and normal plasma cells to study the effects of 3'UTR alterations in MM. Whole transcriptome sequencing (RNAseq) data from 420 uniformly treated newly diagnosed MM samples were compared with 18 normal plasma cell by using dynamic analyses of alternative PolyAdenylation and results were compared with alternative 3'UTR isoforms measured using 3'-seq, an established quantitative 3' end sequencing method from 15 samples pilot study.

We observed ~10000 different isoforms with a median frequency of 297 (range 1-2495) altered 3'UTRs per sample. We observed 563 isoforms that have distal alternative poly A (APA) site and 449 isoforms that have proximal APA site compared to normal plasma cells in at least 10% of our patients. Enrichment analysis showed that short UTR genes are significantly targeted by miR-506, miR-133, miR-130, miR-27, miR-181 and miR-200 (FDR q-value < 0.05). We also observed that expression of 90 genes significantly increased between those samples with shorter UTR vs. regular UTR. Longer UTR genes are enriched in mir-124, mir-186, mir-200, mir-302, mir-495 and expression of 123 genes were significantly downregulated between long UTR and others. There were no tumor suppressor genes and 8 oncogenes in short UTR genes and 4 tumor suppressor and 14 oncogenes in longer UTR genes.

Our results were confirmed using the 3'seq study which showed that 170 genes have longer UTR in MM compared to normal plasma cells and 187 genes have shorter UTR. ~60% of longer and ~65% of shorter UTR genes reported by 3'seq were also observed with RNAseq. Furthermore, we focused on common genes reported by both high-throughput sequencing methods and analyzed any connection between outcome and UTR shortening or lengthening. We observed 26 genes with altered UTR regions have impact on PFS or OS.

In conclusion, we report significant alternate UTR usage, including intronic UTRs in MM affecting the disease biology and clinical outcome. Our data suggests the need to further investigate the molecular impact of alternate UTR usage and its relationship with miR in myeloma.