Contraction in muscle occurs by the relative sliding of thick and thin filaments. The thick filaments are composed by myosin molecules, of which two main isoforms have been found in ventricles, the alpha- and beta-myosin heavy chain (MHC) homodimers. The expression of the two MHCs is developmentally regulated and controlled by numerous stimuli. Polarization Second Harmonic Generation (PSHG) enables the computation of the average helical pitch angle of peptides (AHPAP) of a number of proteins arranged in a fiber structure, like collagen or myosin. AHPAP of these proteins has been used to discriminate them in tissue. We also reported that AHPAP of cardiac tissue correlates consistently with aging of cardiac LV in rabbits and beta-MHC distribution. This study reports a molecular model based on public DNA sequences that directly explains the experimental differences of AHPAP of unlabeled histological sections of cardiac tissue expressing different MHC isoforms.
We computed all second-order tensor elements from the 3D molecular distribution of peptides and its associated hyperpolarizability of the S2 subfragment of the rod domain of MHC for different isoforms, species and mutations: for human, MYH6 (alpha), MYH7 (beta) and E924K mutation; for mouse, MYH6 and MYH7; for rabbit, MYH7. Based on a reported biophysical model we computed the expected AHPAP for each isoform and species. We finally acquired PSHG images of transversal histological sections in three animal models: healthy mouse, which naturally expresses MYH6; a mouse fed a iodine-deficient diet containing 0.15% 6-propyl-2-thiouracil for 4 weeks which induces hyperthyroidism and subsequent shift to MYH7, and 1 year old rabbit which naturally expresses 95% of MYH7 of the heart myosin. We also computed the average AHPAP for the experimental data.
We found agreement between experimental PSHG calculations and molecular model of 3D protein structure: for mouse, MYH6: 67,8° and 67,4°, MYH7: 71,3° and 71,2°; for rabbit MYH7: 73,3° and 72,5°. We then extended our results to 3D protein models of human isoforms and mutations: MYH6: 73,3°, MYH7: 63,7°, and E924K mutation: 66,0°.
This study supports the hypothesis that PSHG can measure subtle differences in molecular structure of myosin that are available only through much more complex techniques (i.e. crystallography). We report a direct explanation for the experimental differences in PSHG signal. This suggests that PSHG can provide in-situ and in-vivo tissue information that might help to quantify the expression and spatial distribution of myosin isoforms and mutations in tissue.