Sequence Variations in miRNA Genes Are Common and May Affect Their Expression in Patients with Chronic Lymphocytic Leukemia

Unique miRNA expression signature is associated with CLL prognostic factors and disease progression (Calin et al., 2005; Mraz et al., 2009), and thus contributes to CLL pathogenesis. The mechanism through which miRNA expression is deregulated in cancer, including CLL, is not well characterized. However, the presence of mutations and single nucleotide polymorphisms (SNPs) in miRNA genes has been described (Calin et al., 2005; Ryan et al., 2010). Nevertheless, the impact and frequency of such variations in CLL patients have yet to be elucidated.

We searched for sequence variations in 109 pre-miRNAs with possible role in CLL pathogenesis. Additionally, we studied the effect of variations detected in miR-29b/29c genes on their expression in more details.

A custom resequencing microarray (Affymetrix) contained probes for 109 pre-miRNAs incl. ∼20 nt from 5′ and 3′ ends of pri-miRNA. MiRNAs were amplified by long range PCR from genomic DNA isolated from peripheral blood cells of 98 high-risk CLL patients (81.6 % unmut IgHV; 37.8 % mut TP53). Amplicons were pooled, fragmented and co-hybridized on the array according to the manufacturer's protocol. The presence of variations detected by microarray was confirmed by Sanger sequencing. The germline/somatic origin of novel variations was performed by direct sequencing of DNA isolated from buccal swabs of the patient carrying the variation. Pri-miR-29c and pri-miR-29b-2 were analysed by Sanger sequencing in additional cohort of 213 CLL patients. The effect of variations detected in miR-29 on their expression was studied by Real-Time PCR (TaqMan miRNA Assays, ABI) in 107 CLL patients. RNAfold tool was used to study the effect of variations on miR secondary structures. To see the difference between allelic frequencies of SNPs detected in miRNAs in CLL patients and healthy population, data from the 1000 genomes project (n=1092 samples) were used.

Within 98 CLL patients, resequencing microarrays detected 18 annotated SNPs (NCBI dbSNP Build 134, 135) in 15 miRNAs, and importantly, 5 novel variations in 5 miRNAs (pri-miR-29a, pre-miR-16–1, pre-miR-372, pre-miR-106b, mature miR-142–3p). To see whether the novel variation in pre-miR-16–1 is a true mutation or a very rare SNP, 193 CLL patients (105 pts. with del13q) were further analysed. Surprisingly, neither novel variation detected by microarray nor known mutation in pri-miR-16–1 previously detected in two CLL patients (Calin et al., 2005), were found. MiR-29 family, which is directly implied in CLL pathogenesis (Pekarsky et al., 2006; Santanam et al., 2010), has been analysed in more details. In the studied cohort of 213 CLL patients, 7 SNPs and one novel variation were found in pri-miR-29b-2 and two SNPs were found in pri-miR-29c. As shown by RNAfold, miR secondary structure was affected by 14 variations in total, suggesting their possible subsequent relevance for miR stability and processing. Moreover, the insertion in pri-miR-29b lowered its expression (fold change 0.7), but only in patients with unmut IgHV (p=0.036). All detected variations were of germline origin except for miR-655, whose mutation was present only in one B-cell subclone in a biclonal CLL case. The allelic frequency of miR-SNPs differed significantly in 8/62 SNPs (p<0.05). Interestingly, 5 out of these 8 SNPs, incl. the insertion in miR-29b-2, were statistically significantly more frequent in analysed CLL patients.

We herein confirm that both SNPs and new sequence variations, most of which were present outside mature miRNAs, are frequent in CLL patients. We also show that miR-16 variations are extremely rare (<0.5%) in CLL, which is in contrast to previously published data (Calin et al., 2005). On the contrary, some miR-SNPs, including the insertion in miR-29b, are more frequent in CLL patients than in healthy population. This suggests that not only mutations but also SNPs may be related to CLL pathogenesis; however, larger population studies of different cohorts are needed to verify their association and possible impact on CLL biology.

No relevant conflicts of interest to declare.