It was only recently that the crucial importance of the cytoskeleton function to maintain balanced protein (i.e., proteostasis [3,6,17,18]) and cardiomyocyte function has been recognized. In cardiomyocytes, the cytoskeleton not only provides a communication highway by transporting proteins throughout the cell, but it also capacitates its contractile function [19,20,21]. In cardiomyocytes, the cytoskeleton is highly specialized, consisting of actin filaments, desmin (intermediate) filaments, and microtubules, interacting with membrane-associated proteins, sarcomeric and nuclear proteins, and proteins of the intercalated disk. This complex network, which also interacts with various organelles, including sarcoplasmic reticulum (SR), mitochondria, and the nucleus, plays an important role in the transmission of signals and the transport of (ubiquitinated) proteins within the proteostasis network [6,19,22] (Figure 2). Within the proteostasis network, especially the microtubules are of vital importance. As the microtubules, SR/endoplasmic reticulum (ER), sarcomeres, and mitochondria are in contact with each other, loss in contact results in Ca²⁺ overload in the organelles, unfolded protein response in the ER, and, consequently, excessive autophagic protein degradation and mitochondrial dysfunction [5,6,23,24,25]. Contact between organelles and trafficking through the cells is mediated by acetylated microtubules [3,6,26,27]. Consequently, AF-induced histone deacetylase 6 (HDAC6) activation and the subsequent deacetylation and degradation of the microtubule network have detrimental effects on the trafficking of proteins as well as on the contractile function of the atrial cardiomyocytes [3,28,29]. Therefore, a functional cytoskeletal network underlies a balanced communication within the proteostasis network and ensures proper contractile function. The conservation of this network is of utmost importance to ensure proper cardiac function. As such, attenuation of the disruption of the cytoskeleton protects against cardiac diseases such as AF and heart failure [3,26,30].

In approximately 15% of AF patients, AF occurs in the absence of common risk factors and at a younger age [1]. In these patients, AF may be familial, suggesting a heritable genetic predisposition. To explore the role of gene variations in AF, the emergence of exome and genome sequencing data has provided extensive new data and revealed a previously unsuspected link between AF and several cytoskeletal (-associated) proteins (Table 1).

Recently, several AF families have been identified to carry a mutation in genes encoding the intermediate filament proteins lamin A/C (LMNA), desmin (DES), and titin (TTN) [31,32,33,34] (Table 1). Intermediate filament proteins integrate the outer cell membrane via desmoplakin (DSP), with sarcomeric proteins such as titin, Z-disk, and the nuclear membrane, thereby regulating the sarcomere architecture and function as well as the nuclear morphology, DNA stability, and gene expression (Figure 2) [6,35,36,37]. Variants in these cytoskeletal proteins are known to be associated with the development of dilated cardiomyopathy (DCM), hypertrophic cardiomyopathy (HCM), and peripartum cardiomyopathy (PPCM) [38,39,40,41]. Of note, various studies have revealed that early-onset AF often represents the initial manifestation of the cardiac phenotype, sometimes even preceding cardiomyopathy by several years, and in the absence of common risk factors as well as gross structural changes in the heart. This observation indicates that AF is a direct consequence of the mutation rather than being caused by structural abnormalities in the ventricle [42,43]. Over 50% of LMNA [31,42,44], 60% of DES [32] and 30–60% of TTN [34] mutant carriers develop cardiac conduction disorders, arrhythmias, and atrial tachycardia, including AF. Furthermore, these patients reveal a more progressive form of cardiac disease with poor outcomes. These findings are further supported by other studies. In a Chinese family, four subjects revealed an AV block, and three of them suffered from AF, which was related to a new frameshift insertion in the LMNA gene (c.825_826insCAGG) [45]. In addition, in a study in Italian families with LMNA variants [46], a total of 30 subjects were included, of which 19 were positive and 11 were negative for LMNA variants. Of the 19 positive study subjects, 11 showed early AF versus none of the 11 negative subjects. These studies reveal a clear association between LMNA variants and the development of AF.

Interestingly, a recent study including early-onset familial AF patients uncovered the phosphodiesterase-4D-interacting-protein (PDE4DIP) p.A123T mutation as a genetic modifier of DES p.S13F [47]. PDE4DIP p.A123T increases the penetrance of cardiac arrest and early-onset AF in the DES mutation carriers. These findings suggest an epistatic interaction of DPE4DIP with the DES gene, leading to increased penetrance of both traits and AF promotion [47].

 

Interestingly, a recent study identified multiple index patients with DCM carried an identical mutation (c.59926 + 1G>A) in the TTN gene, encoding the giant titin protein, which is the major gene underlying inherited DCM. The identified variant likely leads to truncated titan (TTNtv), which is associated with AF onset [34,53]. In an important subset (53%) of index patients and their family members carrying this founder mutation, atrial tachyarrhythmias, in particular AF, were demonstrated. In three patients, AF preceded the development of TTNtv-associated DCM by 11–14 years [34]. In patients with TTNtv-associated DCM and AF, left atrial enlargement was not noticed in half of them. In addition, the traditional risk factors for AF were absent, suggesting a potential intrinsic effect of TTNtv [34], which is in line with the study of Ahlberg et al. [53]. Together, these data strongly suggest that (paroxysmal) AF is an important part of the clinical disease spectrum caused by TTNtv, even if major structural heart abnormalities are still absent, and the individual does not present traditional risk factors for AF. However, the underlying mechanism of TTNtv causing cardiomyocyte structural and contractile remodeling and, consequently, AF remains to be elucidated.

Thus, emerging evidence has identified novel variants in cytoskeletal proteins to underlie early AF (Table 1). Currently, no detailed mechanistic explanation, treatment modalities, or diagnostic screening tools for genetic AF are available. To overcome this, we first need to understand the mutation-specific molecular pathways that drive genetic AF. This knowledge may ultimately lead to the identification of novel drug targets and therapeutic and diagnostic strategies.