The interaction of myosin and the thin filament: how mutations cause allosteric dysfunction and their connection to genetic cardiomyopathy

点击翻译按钮获取中文摘要

Project Summary: The long-term goal of this research program is to develop a rigorously experimentally validated all-atom computational model of the cardiac thin filament (CTF) bound to myosin S1 which provides a unique and accessible platform to identify novel, high resolution disease mechanisms linked to Hypertrophic Cardiomyopathy (HCM). In the prior funding period, we refined and extended our existing CTF computational model and successfully employed it to identify unique and clinically relevant allosteric disease mechanisms including HCM mutation-induced changes in myofilament Ca2+ kinetics, mutation-specific molecular causes of differential cardiac remodeling and disease progression. This included an in vivo validation via the development of a novel transgenic mouse model of cTnT-linked dilated cardiomyopathy and a predictive algorithm to determine the pathogenicity of cTnT mutations that out-performed existing computational approaches in a preliminary test. The key to these advances has been the ability of the current model to precisely identify and locate allosteric changes caused by mutations throughout all components of the CTF followed by closely coupled experimental validation and eventual in vivo model correlation. We now propose to significantly expand the biological complexity of the model to include myosin S1, the molecular motor that drives contraction and the second most common genetic cause of HCM. This important and challenging advance will facilitate a deeper understanding of disease pathogenesis by, for the first time, incorporating the role of molecular allosteric mechanisms between myosin S1 and thin filament. This new computational – experimental platform will be used for both mechanistic insight (for example used for the identification of novel myofilament disease targets,) and the development of a comprehensive deep-learning predictive algorithm to assign pathogenicity to both myosin and thin filament HCM mutations. The latter represents the first use of high-resolution structure, dynamics and function to predict HCM disease allele pathogenicity, a central challenge in the clinical management of these complex patients. Both the training and testing components of the deep learning development will utilize data from the highly annotated and curated SHaRe HCM registry thus greatly improving translational power. Two Specific Aims will be pursued: Aim 1 will utilize state of the art rare event simulation methods developed in one of our groups and refinement of existing unstructured domains of the CTF via FRET to establish the new model. Aim 2 will employ an extensive program of computational analysis and subsequent in vitro validation using pathogenic, variants of unknown significance and non- pathogenic HCM alleles derived from SHaRe to provide inputs to the machine learning environment for algorithm development. Novel disease mechanisms for myosin and thin filament HCM that include crosstalk between the two components will also be explored. Elucidation of these mechanisms can be the basis for robust molecular approaches to disease.