Mutations in ß-cardiac myosin cause cardiomyopathies, which are leading causes of sudden cardiac death.
There are currently no cures for cardiomyopathies. Drugs targeting ß-cardiac myosin would transform
treatment of cardiomyopathies and heart failure, the leading cause of hospitalizations in the United States.
Efforts to develop new therapeutics have been hampered by an incomplete understanding of how
mutations in ß-cardiac myosin cause cardiomyopathy and of how existing compounds targeting myosins
achieve specificity. Recent studies have found that several hypertrophic cardiomyopathy (HCM) variants
release myosin motors from an auto-inhibited state where two myosin motors fold back on each other.
Unfortunately, only a small subset of cardiomyopathy variants has been functionally characterized, even
though insight into a variant’s molecular effects is crucial for determining pathogenicity and predicting drug
response. Drugs targeting ß-cardiac myosin need to achieve a high degree of specificity to avoid disrupting
critical functions carried out by other myosin proteins. Though the field has focused on sequence differences at
compound binding sites, one myosin inhibitor discovered in high throughput screens is highly specific even
though sensitive and insensitive myosin isoforms share identical sequences at its binding site. This suggests
structural dynamics at the binding site are an important determinant of specificity.
We hypothesize that hidden structural states during equilibrium fluctuations of myosin motors determine
functional effects of mutations and the specificity of compounds targeting myosins. We propose using
molecular dynamics simulations to elucidate the effects of a large number of ß-cardiac mutations and to
compare pocket openings between myosin isoforms. We will leverage our unique combination of enormous
computational power and cutting-edge machine learning tools for analyzing protein allostery.
Aim 1 is to identify ß-cardiac myosin mutations that alter the balance of active and auto-inhibited states.
Specifically, we will investigate the effects of HCM variants without functional data, dilated cardiomyopathy
variants, and “variants of unknown significance.” Aim 1 directly addresses the NIH’s goals of investigating new
pathobiological mechanisms and leveraging data science to open new frontiers in biomedical research.
Aim 2 is to predict drug specificity based on differences in conformational ensembles of myosin isoforms.
In particular, we will compare pocket opening probabilities at the binding sites of three clinically relevant
compounds before designing and experimentally testing novel compounds that bind specific pockets in ß-
cardiac myosin. Aim 2 directly addresses the NIH’s goals of developing novel therapeutic strategies.
Together, these results will inform our understanding of how mutations in ß-cardiac myosin cause
cardiomyopathies and advance drug discovery for cardiac disease.