Genome-wide discovery and characterization of long noncoding RNAs in patients with multiple myeloma

Abstract

Background: Long noncoding RNAs (lncRNAs) are involved in a wide range of biological processes in tumorigenesis. However, the role of lncRNA expression in the biology, prognosis, and molecular classification of human multiple myeloma (MM) remains unclear, especially the biological functions of the vast majority of lncRNAs. Recently, lncRNAs have been identified in neoplastic hematologic disorders. Evidence has accumulated on the molecular mechanisms of action of lncRNAs, providing insight into their functional roles in tumorigenesis. This study aimed to characterize potential lncRNAs in patients with MM.

Methods: In this study, the whole-transcriptome strand-specific RNA sequencing of samples from three newly diagnosed patients with MM was performed. The whole transcriptome, including lncRNAs, microRNAs, and mRNAs, was analyzed. Using these data, MM lncRNAs were systematically analyzed, and the lncRNAs involved in the occurrence of MM were identified.

Results: The results revealed that MM lncRNAs had distinctive characteristics different from those of other malignant tumors. Further, the functions of a set of lncRNAs preferentially expressed in MM were verified, and several lncRNAs were identified as competing endogenous RNAs. More importantly, the aberrant expression of certain lncRNAs, including maternally expressed gene3, colon cancer-associated transcript1, and coiled-coil domain-containing 26, as well as some novel lncRNAs involved in the occurrence of MM was established. Further, lncRNAs were related to some microRNAs, regulated each other, and participated in MM development.

Conclusions: Genome-wide screening and functional analysis enabled the identification of a set of lncRNAs involved in the occurrence of MM. The interaction exists among microRNAs and lncRNAs.

Keywords: CeRNA; Expression profile; lncRNA; microRNA.

Fig. 1

Expression profiling changes of the known lncRNAs in patients with MM compared with the control. a Expression quantity distribution box. b Volcano plot of the differentially expressed known lncRNAs. c Heat map of lncRNAs showing the hierarchical clustering of the changed known lncRNAs in the MM group versus the control group. In the clustering analysis, red and blue colors are used for the upregulated genes and downregulated genes, respectively

Fig. 2

GO statistics of the differential expression of known lncRNAs. Molecular function, cellular component and biological brocess were described respectively. Including the proportion of up-regulated differentially expressed genes in this subgroup, and the proportion of down-regulated differentially expressed genes in this subgroup

Fig. 3

Enrichment of the pathways of the q-value distribution in known lncRNAs

Fig. 4

Expression profiling changes of the novel lncRNAs in the MM group compared with those in control. a Expression quantity distribution box. b Volcano plot of the differentially expressed novel lncRNAs. c Heat map of the lncRNAs showing hierarchical clustering of changes in the known lncRNAs in the MM group compared with the control group. In the clustering analysis, the upregulated genes and downregulated genes are colored in red and blue, respectively

Fig. 5

The expression profiling changes of the miRNA in MM compared with those in control. a Express the quantity distribution box of known miRNA; b Express the quantity distribution box of novel miRNA; c Volcano plot of differentially expressed miRNA; d Heat map of miRNAs showing the hierarchical clustering of changed miRNAs in the MM group vs the control. In the clustering analysis, upregulated genes and downregulated genes are colored in red and blue, respectively

Fig. 6

P value distribution thermal map of the KEGG enrichment pathway of the known miRNAs. The abscissa represents each comparison group, and the ordinate represents the KEGG path. Q < 0.05 indicates that the results were significant; the closer to the red color, the more significant the enrichment effect. The gray indicates no enrichment

Fig. 7

Verification of dysregulated known lncRNAs expression in multiple myeloma (MM). Fourteen known lncRNAs were selected according to the significance of difference. Except for CTB-61 M7.2, the Q-PCR results of 13 lncRNAs consistent with the sequencing results of the lncRNAs

Fig. 8

Verification of dysregulated novel lncRNAs expression in multiple myeloma (MM). Ten novel lncRNAs were selected according to the significance of difference. Q-PCR results consistent with the sequencing results of the 10 novel lncRNAs

Fig. 9

Verification of dysregulated miRNAs expression in multiple myeloma (MM) .Seven known and three novel small miRNAs were selected according to the significance of difference. The relative expression level of each miRNA was showed by 2-△Ct value. P < 0.05 was considered statistically significant. Q-PCR results consistent with the sequencing results of the ten miRNAs

Fig. 10

Aiming at MEG3 and CCDC26 as targets, the ceRNA analysis was carried out to construct interaction maps of the two known lncRNAs and nine microRNAs; the diagram depicts the interactions (blue for known lncRNAs, purple for microRNAs, and green for mRNAs)

Fig. 11

Seven novel lncRNAs were selected from differentially expressed novel lncRNAs, and eight differentially expressed microRNAs were selected as follows. Aiming at novel lncRNAs, ceRNA analysis was carried out to construct interaction maps. The diagram depicts the interactions (blue for known lncRNAs, purple for microRNAs, and green for mRNAs)