Abstract 2960

Poster Board II-936

Processing of the pre-miRNA through Dicer1 generates a miRNA duplex, consisting of a miRNA and miRNA* strand. While the functional roles of miRNAs are now well established, the potential roles of miRNA* species remain unclear. However, recent evidence suggests that the star strand of some miRNAs can be abundant and enter the RISC complex. Since the abundance of miRNA*s has not been comprehensively assessed in mammals and we took advantage of 10 deep sequencing libraries from a variety of human and murine cells to determine the most abundant complementary strand for non-annotated miRNA*s. We then calculated the ratio of miRNA/miRNA* for each miRNA duplex. In contrast to previous assumptions that one strand is highly dominant, we found that approximately 50% of the investigated miRNA duplexes exhibit high ratios with a dominating strand (ratio >100), 20% have intermediate ratios (ratio between 100-10) and a remarkable 10% show low ratios (ratio <10), indicating comparable expression of both strands. In addition, we found that ∼10% of all miRNA/miRNA* duplexes display inverse ratios (ratio<1), indicating incorrect annotation in miRBase. Comparing miRNA/miRNA* ratios across the miRNA sequence libraries revealed that most ratios remain constant across tissues and species. This could possibly allow for a novel classification of miRNAs into a-duplexes, miRNAs duplexes with a dominant strand and b-duplexes with both strands being abundant. However, certain ratios were highly variable across the libraries examined as exemplified for the ratio of miR-223/miR-223* which ranged from 0.11 (317:2684 read counts) to 19.6 (13006:660 read counts) in murine and human leukemia cell lines. Bioinformatics analysis on predicted miR-223* targets showed an enrichment for cancer associated genes (p<0.05), suggesting a tumor suppressor-like role for miR-223. Consistent with this, an analysis of samples from 94 AML patients with normal karyotype revealed an inverse correlation of miR-223* with CD34 expression (p=0.018), a negative prognostic marker in AML. In addition, in vitro experiments with mutated miR-223 and miR-223* constructs revealed regulatory potential for miR-223* in myeloid progenitor cells.

Taken together, we propose a new classification for miRNA duplexes and provide evidence for a possible role a miRNA* in the development of acute myeloid leukemia.

Disclosures:

No relevant conflicts of interest to declare.

Author notes

*

Asterisk with author names denotes non-ASH members.

Sign in via your Institution