Our results demonstrate a cell-autonomous role of arginylation in the development and function of the heart muscle, and identify arginylation as a novel mechanism that maintains cardiac health and prevents the development of dilated cardiomyopathy in mice (). It has been previously shown that arginylation is essential for cardiac morphogenesis and the development of heart muscle in embryogenesis[9
], however only the use of the current αMHC-Ate1 mouse model made it possible to identify the specific role of arginylation in cardiomyocytes and the heart muscle though the development and adult life. Our study has provided the first model of arginylation-related heart failure that can be used to address the corresponding mechanisms in human heart disease and to develop a new generation of arginylation-based cardiac therapeutics.
Model of the regulation of cardiac formation and function by protein arginylation
Unlike in the complete Ate1
mouse knockout, heart-muscle-specific deletion of Ate1
does not lead to visible morphogenic defects by impairing myocardium size, heart septation, or the formation of the outflow tract. Consistent with this observation, deletion of Ate1
in migratory neural crest cells – the embryonic cell lineage that contributes to the formation of these structures – does not lead to cardiac morphogenic defects [12
]. These facts suggest that Ate1
-dependent developmental defects in cardiac morphogenesis, unlike the impairments in the heart muscle structure and functions, originate from longer-range tissue signaling rather than directly from the heart-forming cells. It is also possible that arginylation affects specification and/or differentiation of cardiomyocytes during the stages of heart formation that precede the activation of the αMHC promoter.
It has been previously shown that during development α-cardiac actin is arginylated to a relatively high extent, suggesting that arginylation of this protein may be required for myofibril formation [10
]. Strikingly, our current data reveals no arginylated α-cardiac actin in the adult cardiac myofibrils, however we find a prominent set of other proteins that are arginylated, most of them directly relevant to the establishment and maintenance of the sarcomeric structure. Unlike in our previous studies [7
], where we had to employ special enrichment methods to identify arginylated proteins, arginylation detection in the myofibrils required no enrichment procedures, suggesting that each of these proteins was arginylated to a relatively high extent. It should be noted that mutations in these same proteins have been previously shown to result in cardiomyopathies in human patients (reviewed in [26
]), providing a possible functional link between arginylation and their role in heart disease. Addressing this link constitutes an important direction of further studies.
Our results show that posttranslational regulation is essential for maintaining both the active contractile forces and the passive-elastic forces of the heart. Both types of forces are essential for optimal contractile function. The reduction in the active contractile force suggests that the myosin motor function may itself be regulated by arginylation via direct and/or indirect mechanisms. In support of this, myosin heavy and light chain have been found arginylated in our experiments (). The reduction of passive forces in the mutants suggests that the mutant myofibrils are more compliant and therefore are weaker under stress associated with normal cardiac contraction. Such altered properties likely contribute to the longer-term heart dilation and weakening of the heart muscle observed in αMHC-Ate1 mice. Remarkably, similar changes, with more compliant myofibrils, have been observed in other studies for conditions of dilated cardiomyopathy [30
]. Since in our experiments myofibril preparations are removed from extracellular matrix components that could add to the stiffness measurements, the passive forces measured in these preparations are mostly attributed to the giant protein titin, which scaffolds multiple myofibril components and is a key protein in maintaining the structure of the myofibrils. Indeed, changes in titin’s characteristics have been observed in varying heart diseases [31
]. Moreover, it has been shown that the ratio between the titin isoforms N2BA:N2B is increased in dilated cardiomyopathy – the N2BA isoform is more compliant and is associated with lower stiffness levels in myofibrils. Like myosin, titin has been found among our arginylation targets (), suggesting that its structural role within the heart muscle may be directly regulated by arginylation.
Despite the prominence of structural and functional defects in the hearts of αMHC-Ate1 mice, only tropomyosin showed significantly abnormal localization compared to control. Lack of the visible defects for other proteins in our assays could be explained by the fact that arginylation may play highly specialized roles in fine-tuning of these proteins’ levels and their interaction with their binding partners within the myofibrils. Indeed, if striking mislocalization of any of these key proteins was observed in the absence of arginylation, such an effect would likely lead to more severe phenotypes, causing perinatal or even embryonic lethality. The fact that αMHC-Ate1 mice develop the symptoms of cardiomyopathy only later in life suggests that subtle but persistent mechanisms in the absence of arginylation gradually perturb the myofibrils and cardiomyocyte function under constant duress during heart contraction. This phenotype makes αMHC-Ate1 mice highly reminiscent of those human patients that are apparently healthy but start developing severe heart problems at or after ~60 years of age (which corresponds to 8–12 months old in mice).
The higher incidence of localization of tropomyosin inside, rather than adjacent to the α-actinin-stained Z-bands, may indicate that in such myofibrils this protein loses its tight association with the actin filaments – an effect that could severely affect myofibril integrity even if observed only in a small percentage of cases. However, such altered localization could also be due to hypercontraction, during which the increased force of the sarcomere contraction causes the thin filament ends to perforate the Z-bands, eventually destroying the myofibril structure. In support of this mechanism, we see visible Z-band diffusion in αMHC-Ate1 myofibrils at as early as the newborn stage, suggesting that these Z-bands either have not formed normally or are forced to endure constant duress of super-contracting muscle. Addressing these possible mechanisms of myofibril regulation constitutes an exciting direction of future studies.