Characterization of Complete Lncrnas Transcriptome Reveals Expression of Lncrnas As a Prognostic Factor and Linc-Smilo As a Potential Therapeutic Target in Multiple Myeloma

Deregulation of long non-coding RNAs (lncRNAs) is emerging as a common feature of different human tumors and their investigation may uncover novel biomarkers and oncogenic mechanisms. Previous studies have suggested that the alteration of some lncRNAs may play an important role in the pathogenesis of multiple myeloma (MM); however, the complete expression landscape of lncRNAs has not been elucidated. In the present work we characterized the lncRNAs transcriptome of MM and determined the potential involvement of lncRNAs in the pathogenesis and clinical behavior of MM.

To characterize the MM transcriptome, we performed paired-end strand-specific RNA sequencing (ssRNA-seq) in 38 purified plasma cell (PC) samples from MM patients and in 3 bone marrow PCs (BMPCs) of healthy donors, as well as in distinct normal B-cell populations (Naïve, Centroblasts, Centrocytes, Memory and Tonsilar PCs). We identified 40,511 novel lncRNAs that were expressed, accounting for more than half of MM transcriptome (56%). This group of novel lncRNAs together with previously annotated lncRNAs comprised most (82%) of the MM transcriptome. We studied the transcriptional heterogeneity in MM samples and observed that lncRNAs showed a much more heterogeneous expression than coding genes, suggesting that these elements could contribute to the biological heterogeneity of the disease. Moreover, to determine differentially expressed genes, each MM patient was compared to normal BMPCs, detecting 19,886 lncRNAs deregulated (10,351 overexpressed and 9,535 downregulated) in more than 50% of patients. We then analyzed the transcriptional dynamics of MM considering the different stages of B-cell differentiation and focused on a group of 989 lncRNAs that were upregulated specifically in plasma cells from MM in comparison with the rest of B-cell stages (MM-specific lncRNAs).

Next, we aimed to determine whether upregulation of MM-specific lncRNAs in MM was under epigenetic control so we analyzed the distribution of six histone modifications with non-overlapping functions (H3K4me3, H3K4me1, H3K27ac, H3K36me3, H3K27me3, and H3K9me3) of within the lncRNAs of interest by ChIP-seq in MM cases as compared to normal B cell subtypes. We detected 89 lncRNAs with de novo epigenomic activation. These data suggest an epigenetic rewiring in MM where the loci of most MM-specific lncRNAs are in an inactive state in normal cells and become active in MM. We focused on a specific lncRNA, LINC-SMILO, de novo epigenetically active and expressed in MM cells to determine whether upregulation of this lncRNA could play a role in the pathogenesis of the disease. Knockdown of LINC-SMILO in 3 different MM cell lines (MM.1S, MM.1R and KMS-11) using two different shRNAs, resulted in reduced proliferation and induction of apoptosis of myeloma cells. Using low input RNA-seq (MARS-seq), we found that inhibition of LINC-SMILO was associated with activation of ERVs (Endogenous retroviruses) and increase in interferon (IFN) induced genes and activation of IFN pathways, essential for MM cells survival.

Finally, we aimed to determine whether the use of specific lncRNAs could improve the current prognostic stratification of MM patients using the IA11 release of CoMMpass data. We analyzed the prognostic value of lncRNAs using COX regression analysis and Backward elimination of Stepwise regression analysis, obtaining that the overexpression of the lncRNA PDLIM1P4 together with 1q amplification and 17p deletion stratified MM patients in three different risk groups (Figure 1).

In summary, our study shows that the lncRNA transcriptome is widely altered in MM and suggests that some of the identified lncRNAs have marked prognostic influence and can be used as potential therapeutic targets for MM.

View largeDownload slide

View largeDownload slide

Close modal