Identifying miRNAs in normal and malignant B cells. (A) Analytic pipeline for the discovery and validation of known and novel miRNAs in B Cells. Small RNAs from B-cell subsets were subjected to massively parallel sequencing. Raw sequence reads (328 934 149) filtered, mapped to the genome, analyzed by miRDeep, and annotated. Identified miRNA loci were cross-referenced with miRBase13 to distinguish known from candidate novel miRNA. (B) Frequency of reads for miR-20b and their alignment to the genome. The sequence surrounded by a rectangle corresponds to the most abundant and longest mature miRNA sequence that perfectly matched the genome. (C) Outputs of the RNAshapes program that computed the abstract structure of the miRNA precursor, the probability of the predicted structure, and the minimum free energy of folding for the precursor structure for miR-20b. (D) Energetically favorable folding of the miR-20b precursor, with the miRNA mature sequence and the microRNA* sequence highlighted in yellow.

Identifying novel miRNAs through high-throughput sequencing. (A) Distribution of known and candidate novel miRNAs in normal and malignant B cells. Samples are listed in the same order as Table 1. (B) Distribution of known miRNA expression among the normal B-cell subsets. The majority of miRNAs are shared between the B-cell types. (C) Distribution of candidate novel miRNA expression among the normal B-cell subsets. The majority of miRNAs are shared between the B-cell types. (D) Proportion of samples in which candidate novel miRNAs were identified. The majority of miRNAs are expressed in more than one sample. (E) Proportion of candidate novel miRNAs identified in 20 additional sequencing datasets. (F) Conservation of novel human miRNAs in chimpanzee, rhesus monkey, mouse, and dog. Ninety percent of the 286 novel miRNAs were found to be conserved across humans and one or more additional species, and more than 80% were found to be conserved across humans and 2 or more additional species. (G) Validation of 86 novel miRNAs candidates by an alternative method of measurement, real-time PCR. Maximum measured values for miRNAs expressed in at least 10% of the samples are shown in log2-scale. Detection threshold is the expression level corresponding to C_T ≤ 35. (H) miRNA profiling distinguishes the molecular subgroups of DLCBL that were defined based on gene expression profiles. ABC refers to activated B-cell–like DLBCL. GCB refers to germinal center B-cell–like DLCBL. Unclassified cases are those that did not meet criteria for either group. Black bars indicate candidate novel miRNAs that were differentially expressed between the 2 groups.

Distribution of small RNAs in normal and malignant B cells. (A) Expression of small RNAs, including miRNAs, tRNA, piRNA, snRNA, rRNA, snoRNA, and tiRNA in each sample are shown more than a 2-fold range. Samples are listed in the same order as in Table 1. (B) Expression of miRNAs measured by sequencing across 2 biologic replicates for naive, germinal center, memory, and plasma B cells. The P values were computed using a correlation test against the null hypothesis that the data were not correlated; P < 10⁻⁶ in all 4 cases. (C) Relative expression of miRNAs differentially expressed in the naive to germinal center B-cell transition as measured by real-time PCR and sequencing. (D) Relative expression of miRNAs differentially expressed in the germinal center B-cell to plasma transition as measured by real-time PCR and sequencing. (E) Relative expression of miRNAs differentially expressed in the germinal center B to memory B-cell transition as measured by real-time PCR and sequencing.

Discovery and functional validation of a novel miRNA cluster. (A) Genomic locus of the mir-2355 cluster (chromosome 9q34.3), conservation across 3 primates and one additional mammal (mouse), and predicted secondary structure of each miRNA hairpin, with the mature sequences highlighted in yellow. The sequencess are conserved only in primates. (B) Knockout of the primary miRNA transcript processing enzyme, Drosha, by RNA interference results in accumulation of a known miRNA cluster primary transcript (pri-hsa-mir-17) and the novel miRNA cluster primary transcript (pri-hsa-mir-2355). (C) Expression of hsa-miR-2355 is higher in GCBs compared with memory B cells. (D) Chromatin immunoprecipitation with an antibody to H3K4me3 of GCBs and memory B cells shows enrichment of DNA from the predicted core promoter region of the miR-2355 cluster in GCBs compared with memory B cells. Three sets are shown: input (positive control for PCR using chromatin before immoprecipitation), immunoprecipitation with an antibody for H3K4me3, and immunoprecipitation with an antibody for CD20 (negative control). (E) Gene set enrichment analysis score and distribution of TGF-β pathway genes along the rank of transcripts differentially expressed in high versus low expressors of the miR-2355 cluster. Expression of the TGF-β pathway genes were inversely correlated with expression of the miRNA.