Targeting the Mechanism of c-Mpl Exon-Inclusion through Antisense Oligonucleotides Yields a Dominant Negative Isoform

Activating mutations of the Thrombopoietin (THPO) receptor, c-MPL, have been revealed as drivers in myeloproliferative diseases (MPDs).In addition, wildtype c-MPL signaling has been shown to be required for the pathogenesis of Calreticulin-mutated MPDs and implicated in leukemic stem cell survival in CML and AML. Therefore, the ability to impair c-MPL signaling has the potential to provide a novel, targeted approach toward these difficult to treat diseases. Alternative splicing of c-MPL gives rise to a truncated dominant negative isoform, MPL-TR, which skips exons 9 and 10 and is conserved between species. Our prior work using a knockout mouse model identified the spliceosomal protein, Ott1(Rbm15), as a key regulator of alternative splicing between Mpl-TR and the full length isoform, Mpl-FL. Loss of Ott1 increased the fraction of Mpl-TR from 33% to 73% in hematopoietic stem cells (HSCs) and abolished Thpo response. Furthermore, overexpression of Mpl-TR impaired HSC function in vivo. To further delineate the global effects of Ott1 on alternative splicing within the HSC compartment, we performed RNA-seq on sorted Ott1-deleted Lin-Sca1+Kit+ (LSK) cells. DEXseq analysis of significant differential exon expression identified 1821 exon exclusion and 158 exon inclusion events in the Ott1-deleted cells compared to wildtype controls, establishing Ott1 as primarily an exon-inclusion factor. The splicing factor Srsf2 has been observed in the same complex as Ott1 and also functions primarily as a cassette exon inclusion factor, therefore we compared previously identified Srsf2 exon targets and found 15-19% overlapped with Ott1-affected exons, suggesting a functional link between the proteins in the spliceosome.

Srsf2 utilizes the exon splicing enhancer (ESE) sequence to promote exon inclusion. Scanning the c-Mpl sequence revealed an ESE within exon 10 that is conserved between mice and humans. We hypothesized blocking the exon 10 ESE with an anti-sense oligonucleotide (AON) would interfere with the spliceosomal machinery responsible for c-Mpl exon-inclusion and lead to exon skipping, thus resulting in production of the dominant negative Mpl-TR isoform. An advantage of AON-targeted therapy is the ability to produce highly specific agents based on sequence complementarity. We synthesized chemically modified AONs with phosphorothioate backbones, either fully modified with 2'- O -methoxyethyl RNA (MOE) and thus designed to induce splice-switching, or designed as gapmers to elicit cleavage of their targets by RNase H. MOE-based AONs targeted towards the exon 10 ESE were the most effective at increasing the Mpl-TR isoform fraction, increasing them from 30-40% to 60-70% in murine cell lines and primary bone marrow/fetal liver cells. AONs directed toward the splice acceptors and donors of exons 9 and 10 were less effective, suggesting a key role for the exon 10 ESE in regulating c-Mpl splicing decision. To test the ability of AONs to successfully target in vivo, fluorophore-labelled AONs were used in a dose-escalation study in mice. 18 hours after tail vein injection with 1-6 mg/kg un-encapsulated AON dose, bone marrow and spleen were harvested and lineage labelled. High levels of dose-dependent uptake were observed, particularly within the myeloid compartment (granulocyte >99%; megakaryocyte 55%; macrophage 20.9%; erythroid 44.7%; B cell 5% and CD4 T 45.6% and CD8 T 77%). To assess uptake and intracellular localization of AONs within the HSC compartment, AMNIS flow cytometry was performed and showed robust nuclear localization within 64% of LSK cells and 42% of myeloid progenitors. The high levels of uptake in a wide range of hematopoietic lineages, including HSCs, confirms IV injection of naked AONs can efficiently target the hematopoietic compartment. In summary, through uncovering the pivotal cis- and trans-factors controlling splicing at the c-Mpl locus, we were able to develop a specific, effective agent for impairing the Thpo/Mpl axis.