Musashi 2 (MSI2) is an RNA binding protein (RBP) that regulates asymmetric cell division and cell fate of several cell types including hematopoietic stem and progenitor cells. MSI2 was recently shown to be transcriptionally up-regulated in aggressive myeloid leukemia including chronic myelogenous leukemia in blast crisis (CML-BC) and acute myeloid leukemia (AML). Loss-of-function studies have confirmed an important role for MSI2 in the aggressive biology of these leukemia, but the precise molecular mechanisms remain undefined. MSI2 is thought to bind to the 3' untranslated region (3'UTR) of target mRNA in a sequence-specific fashion and inhibit their translation, but functional targets of MSI2 are not known. We implemented two transcriptome-wide techniques to define these targets: iCLIP (individual nucleotide-resolution cross-link and immunoprecipitation) and polysome profiling. We first defined the RNA interactome of MSI2 in a CML-BC cell line (K562) by iCLIP of FLAG-tagged MSI2. iCLIP relies on the ability of ultraviolet (UV) radiation to cross-link RBPs to interacting RNA allowing for stringent immunoprecipitation of RBP-RNA complexes and subsequent analysis by next generation sequencing. iCLIP analysis revealed direct binding of MSI2 to transcripts of 5036 genes, specifically enriched in the 3'UTR. These iCLIP peaks were also enriched in UAG, the minimal sequence motif of MSI2 previously defined through biochemical assays. Given that MSI2 is ubiquitously expressed, but has very specific cellular functions, we hypothesized that only a small fraction of the direct targets demonstrable by iCLIP are functionally relevant. To determine those target mRNA, we performed translatome profiling of K562 cells (MSI2 knock-down and control cells) with polysome profiling. Polyribosomes or polysomes are aggregates of ribosomes assembled on efficiently translated transcripts. Comparing change in abundance of polysome-associated transcripts (polysome profiling) allows determination of quantitative changes to translation. Using polysome profiling, we found that transcripts from only 413 genes underwent translational up-regulation (without a corresponding change in total transcript abundance) in response to MSI2 knock-down, of which 319 were deemed to be direct MSI2 targets from iCLIP analysis. Translational targets thus defined were highly enriched in biological pathways such as differentiation, cell-cycling, DNA damage response and apoptosis. Specific transcripts thus identified included transcription factors critical to myeloid and leukemia biology (cJUN, MYC and SP1) and cell-cycle regulators (CDK1, CDKN1B, RAD21 and RB1). Using expression profiles of 91 CML patient samples (reported in Radich JP et al PNAS 2006 Feb 21;103(8)) we tested if changes in gene-expression during disease progression (from chronic phase to blast crisis) reflect the change in translation of transcription factors as predicted by the translatome analysis. Accordingly, we found that transcripts down-regulated in CML-BC were highly enriched in targets of JUN and SP1, confirming a primary role for MSI2 targets in progression of CML. In summary, our novel approach that integrates two transcriptome-wide techniques show that while thousands of transcripts are directly bound by MSI2, only a small proportion of these undergo translational repression. Functional targets thus identified include transcription factors and cell-cycle regulators critical to progression of myeloid leukemia. Our results will form the basis of further studies to explore therapeutic targeting of MSI2 in myeloid leukemia.


No relevant conflicts of interest to declare.

Author notes


Asterisk with author names denotes non-ASH members.