1. Role of NADPH Oxidases and Oxidative Stress in the Onset of Paroxysmal AF
After cardiac surgery, paroxysmal AF (PAF), also known as postoperative AF or POAF, commonly occurs and typically peaks between days 2 and 4 following the operation
[1]. Clinical studies have shown that oxidative stress plays an essential role in the development of PAF
[2][3]. Kim et al.
[4] examined the sources of O
2•− production in right atrial appendage (RAA) homogenates or isolated myocytes from 15 patients with AF (six with persistent AF, four with persistent AF developing from PAF). Dipenyleneiodonium (DPI), an inhibitor of flavin-containing oxidases or apocynin, a NOX inhibitor that blocks p47phox translocation, inhibited O
2•− production in AF, suggesting a pathogenic role for NOX-derived O
2•− in AF
[4]. Similarly, a borderline, yet significant, association was observed between O
2•− generated by NOX2 in left atrial appendages (LAA) and left atrial enlargement in AF patients
[5]. The serum level of soluble NOX2-derived peptide increased significantly in patients with persistent/PAF and significantly contributed to the increased production of serum isoprostane
[6]. Additionally, urine F2-isoprostanes and isofurans were 20% and 50% higher, respectively, in patients with POAF at the end of the surgery
[3]. POAF patients were found to have significantly higher basal myocardial O
2•−, NADPH-stimulated O
2•−, and ONOO
− production, as reported by Casadei and colleagues
[7].
Studies on animals have revealed that Rac1-dependent NOX-derived ROS plays a role in AF pathogenesis
[8][9][10]. Clinical studies have also established a correlation between AF, NOX2 upregulation, and oxidative stress
[6][11][12]. Inhibition of Rac1 and NOX2 activity was observed in right atrial samples of POAF patients treated with atorvastatin. However, atorvastatin did not affect nitric oxide synthase uncoupling, tetrahydrobiopterin levels, or ROS in patients with permanent AF
[13]. This suggests that the activation of atrial NOXs is an early and transient event in the natural history of AF. It is possible that changes in ROS sources during atrial remodeling could explain why statins are effective in preventing AF, but not in treating it. In diet-induced obese mice, AF was induced by pacing in all cases, while only 25% of controls exhibited AF
[14]. In these mice, cardiac
INa expression, and atrial APD were decreased while Kv1.5 potassium channel expression and corresponding current (
IKur), F2-isoprostanes, NOX2, and PKC-a/d expression, and atrial fibrosis significantly increased. A mitochondrial antioxidant restored
INa,
ICa,L,
IKur, APD, reversed atrial fibrosis, and attenuated AF burden in diet-induced obese mice.
Oxidative stress in mitochondria contributes significantly to cardiac dysfunction in the pathogenesis of POAF
[15][16]. In a prospective study, Montaigne et al. examined mitochondrial respiration and myocardial oxidative stress levels in the right atrial tissues of patients with metabolic disease before cardiac surgery
[15]. Those with reduced mitochondrial respiration, high sensitivity to calcium-induced mPTP opening, increased ROS production, and upregulation of MnSOD and catalase had a 44% incidence of POAF. Case-control studies have supported this finding, showing lower oxidative phosphorylation and altered gene expression in cardiac mitochondrial energy production in human atrial biopsy specimens from patients with chronic AF
[17][18]. NOX4 is a major source of mitochondrial ROS in cardiomyocytes and vascular cells
[19][20][21]. Although mitochondrial oxidases are not a major source of ROS in POAF, NOX-derived ROS can damage mitochondrial DNA and protein complexes, leading to mitochondrial dysfunction and, ultimately, cardiac arrhythmia
[15][22][23][24].
2. Role of NADPH Oxidases and Oxidative Stress in Permanent AF
Patients with persistent AF experience high atrial activity, leading to excessive energy consumption by cardiac myocytes to maintain excitation-contraction coupling, which can affect their structure and function. Mitochondria, the primary energy source, are also subject to metabolic stress. In individuals with AF, fragmented mitochondria are present in the atrial tissue. When HL-1 atrial myocytes are subjected to tachypacing, mitochondrial Ca
2+ handling is impaired, mitochondrial membrane potential decreases, and ATP production is reduced
[25]. Increased atrial activity alters the mitochondrial redox balance as well. For instance, a study by Yoo and colleagues found that ROS production increases in isolated atrial swine myocytes in a frequency-dependent manner
[26]. Additionally, they examined the sources of ROS in atrial samples from dogs subjected to tachypacing and found that mitochondria and NOX2 were significant sources of ROS. Mitochondrial DNA is highly vulnerable to ROS damage, as evidenced by 4977-base deletions of mtDNA and elevated mitochondrial 8-OHdG levels in atrial samples from patients with permanent AF, indicating oxidative stress-induced DNA damage
[27]. In a sheep model of persistent AF, Rac1 expression and NOX-stimulated O2
•− production increased, while eNOS expression and circulating NO levels decreased
[28].
Marks and colleagues showed that mice with RyR2 mutations that cause intracellular Ca
2+ leak have an increased susceptibility to AF due to mitochondrial dysfunction, ROS production, and atrial RyR2 oxidation
[29]. In patients with chronic AF, RyR2 was found to be oxidized, phosphorylated, and depleted of calstabin 2, which plays a role in stabilizing the closed state of RyR2 during diastole
[29][30]. Genetic variants in
SCN5A have also been linked to AF in patients with congenital long QT syndrome and lone AF. Atrial myocytes from chronic AF patients showed a significant reduction in peak
INa density and lower expression of Nav1.5, along with an increase in late
INa [31][32][33]. By inhibiting
INaL-
INa peak ratio, ranolazine was found to reverse proarrhythmic activity in atrial myocytes and improve diastolic function. Studies by Avula et al. reveal that expressing human mitochondrial catalase in mice overexpressing human F1759A NaV1.5 channels can reduce cardiac structural remodeling, spontaneous AF incidences, and pacing-induced after-depolarizations despite the heterogeneously prolonged atrial action potential
[34].
Transcriptional oxidative stress response affects permanent AF pathophysiology. In their study, Kim et al. used cDNA microarrays to compare transcription levels of putative genes that might be involved in the oxidative stress response between control patients and patients with chronic atrial fibrillation AF
[35]. Results showed that eight genes associated with oxidative stress were upregulated in atrial samples from AF patients, including NOX, while six antioxidant genes were downregulated, such as those encoding glutathione peroxidase, glutathione reductase, and superoxide dismutase. Patients with persistent AF also had higher levels of oxidized glutathione and cysteine in their serum and plasma than those with normal sinus rhythm. Interestingly, EhGSH levels above the median were a stronger predictor of chronic AF than paroxysmal AF
[36][37]. In contrast to oxidative stress, no significant correlation was found between proinflammatory markers and AF
[37].
3. ROS in the Transition from Paroxysmal to Permanent AF
Paroxysmal AF begins with short episodes that become persistent and eventually permanent over time. As AF advances from short episodes to a persistent state, the atria undergo progressive structural and functional changes that ensure its long-term persistence
[38][39]. In the initial stages, electrical remodeling results in the shortening of atrial refractoriness
[40][41]. It is unclear how these changes stabilize atrial fibrillation despite known ion channel alterations in animal models and humans
[42][43][44]. Structural remodeling and fibrosis result in intra-atrial conduction disturbances. However, their role in the progression of atrial fibrillation from paroxysmal to persistent remains unclear
[38][41].
It is undeniable that oxidative stress plays a significant role in the development of AF. However, as AF progresses from paroxysmal to permanent, the source of ROS also changes. According to Riley et al., the mechanism responsible for NO-redox imbalance differs between atria, as well as the duration and substrate of AF
[45]. After two weeks of AF, left atrial O
2•− production increases due to increased NOX expression and activity. Conversely, in cases where AF or atrioventricular block has been persistent, mitochondrial oxidases and uncoupled NOS activity in the right atrium are responsible for the biatrial increase in increase in O
2•− production (due to a reduction in BH
4 and/or an increase in arginase activity). It was found that Rac1 activity in RA samples from persistent AF patients was not increased. In addition, atorvastatin did not reduce O
2•− production in a mevalonate-reversible manner, which indicates that statins may not be beneficial in secondary AF prevention. The group discovered that miR-31 upregulation in goats and patients with persistent AF depletes nNOS and reduces NOS activity due to mRNA decay and the translational repression of dystrophin, leading to the loss of nNOS in the sarcolemmal region
[46]. nNOS is constitutively expressed in cardiomyocyte sarcoplasmic reticulum and sarcolemmal membranes as part of the dystrophin-associated glycoprotein complex
[47]. nNOS-derived NO controls sarcolemmal ion conductance
[48][49] and calcium fluxes
[50] and prevents arrhythmic death in mice after myocardial infarction
[51]. Upregulation of miR-31 and disruption of nNOS signaling contribute to the electrical remodeling of atrial myocardium in mice and significantly increase AF inducibility. This is done by shortening APD and abolishing rate-dependent adaptation.
Martins et al. studied persistent long-standing AF in sheep and discovered that the rate at which dominant frequency (DF) increases can predict when AF stabilizes and becomes persistent
[38]. Transcriptome and proteomic analyses revealed changes in extracellular matrix remodeling, inflammation, ion channels, myofibril structure, and mitochondrial function during the early stages of AF, but not in later stages in sheep
[39][52]. CMs in AF have decreased expression in several potassium channel genes (
KCNJ3,
KCNJ5), calcium channel genes (
CACNA1C), or sodium channel genes (
SCN5A), most notably when comparing control and transition samples. In contrast, others, such as
HCN2 or
KCNH7, have increased expression
[39]. Some of the genes that are upregulated in AF, including
RCAN1 and
LGALS3BP, have a significant effect on its pathophysiology. Patients with persistent AF show significant DNA damage correlated with poly(ADP)-ribose polymerase activation, NAD
+ depletion, and oxidative stress in CMs
[53]. In patients with permanent AF, about 60% of atrial myocytes are dystrophic, with extensive sarcomere loss and DNA cleavage
[52]. These dystrophic myocytes express low levels of antiapoptotic death protein BCL-2, indicating increased susceptibility to death signals compared to control cells.