Revealing the molecular history of the transition from paroxysmal to permanent atrial fibrillation

This editorial refers to ‘Transcriptome and proteome mapping in the sheep atria reveal molecular features of atrial fibrillation progression’ by A. Alvarez-Franco et al., pp. 1760–1775 After more than 50 years of intense research activity on the most frequent cardiac arrhythmia, atrial fibrillation (AF), black holes remain. First, the precise role of each of the numerous molecular players found to be associated throughout the history of the arrhythmia is not yet clearly established. Second, mechanisms underlying the strong tendency of AF to progress from paroxysmal to permanent remain unidentified. Clearly, classical research strategies focusing on one experimental condition or one molecular player associated with AF have reached their limits. More holistic approaches are necessary. Another major pitfall is the difficulty to reproduce experimentally the molecular complexity in the natural history of AF, which is species dependent and takes years. Given the above considerations, the article of Alvarez-Franco et al.1 is an important breakthrough in the field. In a well-characterized model of AF in sheep that reproduces the progression of AF observed in humans.2 These authors used an unbiased approach to identify the molecular signature associated with the distinct progression phases of the arrhythmia. Two important concepts emerge from this study: first, during AF progression most changes occur during the transition phase but not in later stages; and second, atrial ageing. Using RNA-seq and proteomic analyses at three different time points (control, transition, and persistent), Alvarez-Franco et al. identified a total of 1367 and 2479 differentially expressed genes in left and right atrial tissue and isolated cardiomyocytes, respectively, between control and AF sheep. They also identified 581 proteins having differential abundance in the atrial cardiomyocytes of control vs. AF sheep. They then made comparisons of both transcriptomic and proteomic profiles between control, transition and chronic groups from both atrial appendages and demonstrated that changes to the control condition increased with disease progression. Notably, they showed that most changes occurred in the transition, but no changes occurred in the chronic group. Moreover, the molecular dynamics closely paralleled the temporal changes in dominant frequency (DF) shown previously in the sheep,2 which closely predicted the time to persistent AF in the same sheep model, and in a recently conducted observational study3 that included 216 patients undergoing transition from paroxysmal to persistent AF. Like the experiments in the sheep model, AF progression showed patient-specific patterns of atrial electrical remodelling, in that the rate of DF increase predicted the time at which AF stabilized and became persistent. Based on the foregoing animal, molecular and human studies most of the commotion associated with AF progression occurs during the transition between paroxysmal and persistent AF. Nothing much happens after this molecular earthquake once the atria remodel and AF becomes permanent (Figure 1). This finding argues for aggressive interventions to restore sinus rhythm at all cost, rather than just do rate control. This is in line with recent clinical studies showing that early rhythm-control therapy was associated with a lower risk of adverse cardiovascular outcomes than traditional care in patients with early AF.4 For further analysis, Alvarez-Franco et al. selected 4410 genes based on their extreme expression value and pathophysiological relevance. From such genes, they were able to define 31 independent clusters that correlated with the main biological sources of variations during AF progression: sinus rhythm vs. AF, left/right atria and transition phases. Among those genes, some are highly relevant to the pathophysiology of AF such as the calcineurin regulator RCAN1, the galectin-3 binding protein or the transcription factor and PITX2, which governs left atrial specialization and occupies the top locus identified by genome-wide association study (GWAs). Other clusters contain genes related to atrial physiology such as those encoding for ion channels or sarcomeric proteins. Diagram depicting the progression of atrial fibrillation (AF) from paroxysmal to persistent AF and the associated molecular remodeling (circle). With each episode of paroxysmal AF, there is a progressive accumulation of molecular changes underlying the electrical and metabolic remodeling of the atrial myocardium reaching an irreversible state when AF becomes permanent. Then, after this transition from paroxysmal to persistent AF, no more molecular change occurs, except changes in chromatin and DNA indicating the senescence of the atria. TTN, titin; KCNH2, potassium voltage-gated channel subfamily H; PUITX2Paired like homodomain 2; KCNJ, potassium voltage-gated channel subfamily J; KCNH, potassium voltage-gated channel subfamily H; HCN, hyperpolarization activated cyclic-nucleotide-gated channels; MYH2, Myosine heavy chain. The other major finding of the study of Alvarez-Franco et al. is the disorganization and de-compaction of chromatin in cardiomyocytes from sheep in persistent AF and more pronounced in left than right atria. Such epigenomic-related changes of chromatin have been observed already during tissue decline as it occurs with ageing. Accordingly, there is strong epidemiological evidence for a close link between ageing and AF. Furthermore, several markers of cellular ageing, such as reduced telomere length, SA-β-gal and p16a are associated with AF.5 DNA damage, a mechanisms underlying cell senescence, has been shown in tachy-paced atrial myocytes and patients in persistent AF. Senescence is caused by the activation of poly(ADP)-ribose polymerase 1 (PARP1), the depletion of nicotinamide adenine dinucleotide (NAD+), and oxidative stress.6 In patients in permanent AF, but also those in heart failure and sinus rhythm, almost 45% of atrial myocytes are dystrophic, characterized by extensive sarcomere loss together with DNA cleavage.7 These dystrophic myocytes express a low level of the anti-death protein, BCL-2 as indicated by the reduced p26-BCL-2/BAX ratio indicating that they are more susceptible to death signals than control cells as observed during degenerative neuronal disease and ageing. Altogether, the above data constitute a strong body of evidence that AF is a major accelerator of ageing of the atrial myocardium. Furthermore, ageing itself favours the progression of the AF substrate8 and several clinical condition can accelerate ageing of the atrial tissue, such as hypertension or heart failure.9 The ageing atrial myocardial may also be part of a general ageing process of the cardiovascular system involving arterial rigidity and atherosclerosis plaque formation and sharing similar risk factors such as obesity, metabolic disorders, or hypertension. The adipose tissue around the atria or coronary vessel could be one interface of cardiovascular ageing by regulating inflammation, metabolic, and oxidative stresses of neighbouring tissues.10,11 Therefore, the article of Alvarez-Franco et al. together with the recent literature on cardiovascular ageing open wide research perspectives on new medical care and therapies of AF patients taking into account the different stage of disease progression, targeting multiple molecular players, as well as cardiovascular risk factors. Conflict of interest: none declared. The project has received funding from European Union’s Horizon 2020 research and innovation programme under the Grant agreement number 965286, MAESTRIA. The opinions expressed in this article are not necessarily those of the Editors of Cardiovascular Research or of the European Society of Cardiology.