Predictive genotype-phenotype correlation of four tropomyosin-1 variants of unknown significance.

Abstract

Cardiomyopathies affect nearly 20% of humanity worldwide and are often caused by mutations in heart muscle genes that encode cardiac thin filament (TF) proteins. The on-off switching of cardiac muscle contraction is controlled by the TF proteins troponin and tropomyosin responding to alterations in sarcoplasmic calcium levels. In the absence of calcium, troponin-tropomyosin inhibits actomyosin interactions and induces muscle relaxation, whereas the addition of calcium relieves this constraint leading to contraction. Many mutations in TF proteins that are linked to cardiomyopathy appear to disrupt this regulatory switching. Here, we applied an interdisciplinary approach to correlate atomic-level alterations in tropomyosin structure to myofilament and tissue-level function to better understand mechanisms of disease. Twenty variants of unknown significance (VUS) in TPM1 from ClinVar were screened computationally using molecular dynamics to predict effects on tropomyosin structure and dynamics. Four candidate variants (A102D, D258E, K233N, and A239T) were identified and selected for further functional studies using in vitro motility assays and human engineered heart tissue (EHT) mechanics. The variants associated with hypertrophic cardiomyopathy, D258E and A102D, increased calcium sensitivity, while K233N, a VUS associated with dilated cardiomyopathy, lead to a decrease in calcium sensitivity. EHTs expressing A102D displayed a nearly 50% increase in peak isometric twitch force and delayed relaxation and D258E-EHTs also slowed relaxation. These phenotypes are consistent with the increased calcium sensitivity observed in the in vitro motility assay and with clinical manifestations of hypertrophic cardiomyopathy. Meanwhile, DCM-associated variants K233N and A239T caused dramatic and significant reductions in isometric twitch force when expressed in EHTs. Our approach reveals an important step toward predictive genotype-phenotype correlation and has the potential to benefit patients by accelerating the characterization and classification of tropomyosin variants.