High-Throughput Screening of Kinase siRNAs and Small Molecule Compounds Identify Novel Candidates for the Development of Myotonic Dystrophy Type 1 Therapies: A Step Towards Therapeutic Advancements in DM1

Abstract

Myotonic dystrophy type 1 (DM1) is the most common form of adult muscular dystrophy (1:8000) and is caused by an abnormal expansion of CTG repeats in the 3’ untranslated region of the dystrophia myotonica protein kinase (DMPK) gene. The expanded repeats of the DMPK mRNA forms hairpin structures which sequester RNA-binding proteins (RBP) in intranuclear foci, such as the splicing regulator muscleblind-like 1 (MBNL1), which results in aberrant splicing of several mRNAs and underlie, at least in part, DM1 pathogenesis. It has been previously shown that disaggregating these RNA foci repletes free and thus functional MBNL1, rescuing DM1 spliceopathy and alleviating associated signs and symptoms such as myotonia. Importantly, the direct upregulation of MBNL1 has comparable beneficial outcomes. The focus of this thesis was to develop novel and practical therapeutic avenues for DM1 by employing high-throughput screening technology to identify key pathways and small molecule candidates which reduce CUG foci in patient cells, and ultimately correct DM1 spliceopathy and associated signs in vivo. First, a high-throughput kinome screen using an siRNA library targeting 692 kinase subunits identified PACT, HIPK4, and PKA2β as candidates for reducing CUG foci in patient fibroblasts. Knockdown of each gene resulted in a partial reduction in CUG foci, but ultimately did not correct aberrant splicing of insulin receptor (IR) or sarcoplasmic/endoplasmic reticulum Ca2+-ATPase (SERCA1), two genes which are typically misspliced in DM1. A second set of screens focused on testing small molecules, several of which are FDA-approved for clinical use, in an effort to expedite drug discovery. One approach was to data-mine from a previously completed chemical screen, which used system-wide RNA sequencing to establish drug-gene interactions in mouse neuronal cultures treated with blood brain barrier-penetrant drugs, and specifically look for compounds which downregulate DMPK mRNA or upregulate MBNL mRNA (MBNL1 and MBNL2). No compounds were found to downregulate DMPK mRNA. However, several compounds upregulated MBNL mRNAs; the activity of one of these, nilotinib, was validated in human DM1 fibroblasts and converted myoblasts, mediating a small correction in SERCA1 spliceopathy. Administration of nilotinib to unaffected mice did not result in in vivo MBNL gene upregulation in mouse skeletal muscle, as was seen in vitro. Further testing of nilotinib in DM1 in vivo models is required. A final set of chemical screens in patient myoblasts using an FDA-approved drug library and a chemogenomic drug library identified several HDAC inhibitors which reduced foci and rescued SERCA1 spliceopathy in vitro in DM1 differentiated myoblasts. Of these, vorinostat (SAHA) was further tested in a mouse model of DM1 (HSALR), proving safe and effective in correcting aberrant muscle pathology as well as splicing defects of RYR1, SERCA1, and CLCN1. Functional validation, such as myotonia, remains to be completed, but given the strong evidence for CUG foci reduction and splicing correction, vorinostat has emerged as a promising novel candidate for DM1 therapy.