Antioxidant Properties of Oral Antithrombotic Therapies in Atherosclerotic Disease and Atrial Fibrillation

The thrombosis-related diseases are one of the leading causes of illness and death in the general population, and despite significant improvements in long-term survival due to remarkable advances in pharmacologic therapy, they continue to pose a tremendous burden on healthcare systems. The oxidative stress plays a role of pivotal importance in thrombosis pathophysiology. The anticoagulant and antiplatelet drugs commonly used in the management of thrombosis-related diseases show several pleiotropic effects, beyond the antithrombotic effects. The present review aims to describe the current evidence about the antioxidant effects of the oral antithrombotic therapies in patients with atherosclerotic disease and atrial fibrillation.


Introduction
Oxidative stress is due to the discrepancy between the rate of reactive oxygen species (ROS) generation and elimination. ROS are products of oxygen metabolism; and some environmental factors can trigger its production, tipping the scales toward an unbalanced state that results in cellular and tissue damage. Oxidative stress plays a key role in the onset of vascular thrombosis, both arterial and venous [1].
Oral antiplatelet therapies are the mainstay of the pharmacological treatment in patients with atherosclerotic disease since they have significantly reduced cardiovascular morbidity and mortality and achieved a class I recommendation both for coronary artery disease (CAD) [2][3][4] and symptomatic peripheral artery disease (PAD) [5].
Antithrombotic drugs have deeply modified the natural history of thrombosis-related diseases. Nevertheless, thrombotic disorders continue to pose a tremendous burden on healthcare systems, suggesting the need for a deeper understanding of thrombotic mechanisms.
The translational research offers interesting pathophysiology tips, gradually shedding light on the pivotal role of oxidative stress in the genesis of cardiovascular diseases. In this

Role of Oxidative Stress in Venous Thrombosis
Oxidative stress seems to be implicated in venous thrombosis since some evidence suggests an increased tissue factor (TF) expression in ECs and VSMCs after ROS exposition [75][76][77].
ROS produced by the mitochondrial respiratory chain and NOX induced a proinflammatory and procoagulant state in experimental models [78,79]. Moreover, as a feedback mechanism, thrombin was found able to further elicit ROS production, worsening ED [80].
The oxidative changes of proteins involved in the modulation of the hemostatic process have been proposed as an additional mechanism. The tissue factor pathway inhibitor (TFPI) is the main physiological regulator of TF activity and may be blocked by oxidative stress, impacting the thrombotic process [81]. ROS may directly inactivate both the protein C and its upstream agonist thrombomodulin [82,83]. Conversely, other studies suggested an indirect inhibition preventing the binding between these endogenous anticoagulants and thrombin [84]. Finally, fibrinogen, once oxidized, is more easily cleaved into fibrin [85] (Figure 2).

Role of Oxidative Stress in Venous Thrombosis
Oxidative stress seems to be implicated in venous thrombosis since some evidence suggests an increased tissue factor (TF) expression in ECs and VSMCs after ROS exposition [75][76][77].
ROS produced by the mitochondrial respiratory chain and NOX induced a proinflammatory and procoagulant state in experimental models [78,79]. Moreover, as a feedback mechanism, thrombin was found able to further elicit ROS production, worsening ED [80].
The oxidative changes of proteins involved in the modulation of the hemostatic process have been proposed as an additional mechanism. The tissue factor pathway inhibitor (TFPI) is the main physiological regulator of TF activity and may be blocked by oxidative stress, impacting the thrombotic process [81]. ROS may directly inactivate both the protein C and its upstream agonist thrombomodulin [82,83]. Conversely, other studies suggested an indirect inhibition preventing the binding between these endogenous anticoagulants and thrombin [84]. Finally, fibrinogen, once oxidized, is more easily cleaved into fibrin [85] ( Figure 2).

Role of Oxidative Stress in Atrial Fibrillation
Atrial fibrillation (AF) is the most common sustained arrhythmia [6] and the major cause of cardioembolic stroke. Rising prevalence of AF has gradually paved the way for a shift in stroke etiology [8]; therefore, the cardioembolic stroke is now the most frequent type of ischemic stroke [8].
AF shares the same risk factors as atherosclerosis [86], and its prevalence increases with aging [87]. Therefore, it has been hypothesized that oxidative stress contributes to the pathophysiology of AF [88,89]. Indeed, previous models have shown enhanced oxidation of myocardial proteins and oxidative damage of atrial tissue [90,91].
The impaired intracellular Ca 2+ regulation plays a key role in the modification of electrical homeostasis. AF-derived atrial myocytes exhibit enhanced diastolic sarcoplasmic reticulum (SR) Ca 2+ leak through the ryanodine receptor (RyR2) [92,93]. In addition, a mutation-induced intracellular Ca 2+ leak may trigger AF in knock-in mice [94,95]. Some Antioxidants 2023, 12, 1185 5 of 24 data suggest that RyR2 may be a target of oxidized calmodulin-dependent protein kinase II (CaMKII), that phosphorylates the receptor, leading to a Ca 2+ leak [96]. Xie et al. [97] showed that RyR2 is oxidized by mitochondrial-produced ROS in atrial myocytes, increasing the intracellular Ca 2+ leak. Interestingly, the decrease of ROS levels limited the Ca 2+ leak and stopped AF.
The oxidative changes of proteins involved in the modulation of the hemostat cess have been proposed as an additional mechanism. The tissue factor pathway inh (TFPI) is the main physiological regulator of TF activity and may be blocked by oxi stress, impacting the thrombotic process [81]. ROS may directly inactivate both the p C and its upstream agonist thrombomodulin [82,83]. Conversely, other studies sug an indirect inhibition preventing the binding between these endogenous anticoag and thrombin [84]. Finally, fibrinogen, once oxidized, is more easily cleaved into [85] (Figure 2).

Aspirin
The antioxidant properties of aspirin are huge and extremely heterogeneous, since it seems to act at different levels.
Several preclinical studies [98][99][100] have showed the antioxidant effects of aspirin in bovine pulmonary artery endothelial cells and human umbilical vein endothelial cells (HUVECs). Aspirin exerted direct endothelial protection, mitigating H 2 O 2 -induced toxicity [98] and stimulating the ferritin synthesis [100][101][102]. These effects occurred at therapeutically relevant concentrations and after several hours of pretreatment, suggesting a triggered gene transcription as a possible mechanism. Moreover, aspirin enhances both the expression and activity of heme-oxygenase-1 (HO-1), which provides a strong defense against oxidative tissue injury through several mechanisms. One of these is enhancing bilirubin and carbon monoxide, which exert an antioxidant activity. HO-1 involvement was shown in either in vitro or in vivo models [103][104][105][106][107]. Several preclinical studies added details on aspirin-enhanced HO-1 activity. The Nrf2-ARE pathway plays a critical role in upregulating the antioxidant and detoxifying enzymes in the body, with HO-1 being the main downstream product [108].
Aspirin leads to an antioxidant property also through inhibition of NF-kB. NF-kB is a key modulator of inflammatory reactions and affects many different facets of both the innate and adaptive immune systems. NF-kB is involved in the control of the inflammasome and the induction of the expression of several proinflammatory genes, such as those encoding cytokines and chemokines [109].
Reduction of NF-kB transcription was observed in many in vitro [110,111] and in vivo models [112]. The final effect was a reduction of ROS levels and proinflammatory cytokines.
Yang et al. explored a novel mechanism by which aspirin limits NF-kB effects. They demonstrated that aspirin increases activator protein 2α (AP-2α) phosphorylation, upregulating the inhibitor of nuclear factor kappa B (IkBα). This effect was found both in apoE-/-mice and in humans and resulted in a reduction of oxidative stress and of the plaque instability [112].
The antioxidant implication of aspirin seems to also be related to the reduction of lipid peroxidation, and this was a common finding in many investigations, in vitro [113][114][115][116] and in vivo [117][118][119].
The balance between pro-oxidant and antioxidant systems is complex and aspirin plays a crucial role, interfering with activities of different enzymatic systems. Wròbel et al. [120] examined redox homeostasis in the liver and brain of BALB/c mice after aspirin delivery. There was an enhanced activity of 3-mercaptopyruvate sulfur transferase and g-cystathionase, i.e., enzymes involved in the production of H 2 S, and the GSH/GSSG ratio was higher, thus reflecting a higher antioxidant capacity [120].
In other in vivo models, aspirin improved glutathione peroxidase (GP), glutathione transferase (GT) [124] and CAT function [125][126][127]. CAT is an antioxidant enzyme which converts H 2 O 2 into water and oxygen, reducing the toxicity related to ROS production. The antioxidant property was also related to the ability of aspirin to enhance the transcription of genes encoding for these antioxidant enzymes (CAT, SOD) [126]. In two models of ischemia-reperfusion injury administration, aspirin led to a significant reduction in nitrooxidative stress and an improvement of endothelial function [128,129]. Moreover, in one study, aspirin antioxidant properties protected from oxidative stress-induced DNA strand breaks [130].
Four studies evaluated the antioxidant effects of aspirin in humans [131][132][133][134], and two of them were conducted in healthy middle-aged subjects.
In a pilot study, Ristimae et al. [131] evaluated the serum products of lipid peroxidation in 25 healthy men at baseline and after two weeks of treatment with 100 mg of aspirin. No significant modifications were observable in SOD and CAT activity nor in GSH levels; however, in the aspirin group, a higher serum antioxidant capacity (AOC), expressed by the ability of serum to inhibit peroxidation of linoleic acid, has been shown.
Kurban et al. [132] confirmed this preliminary observation among 30 healthy volunteers receiving 100 or 150 mg of aspirin. After two months of treatment, patients on 150 mg of aspirin showed a significantly lower total oxidative status (TOS) and serum oxLDLs levels. These results suggest that ASA treatment may contribute to the prevention of atherosclerosis, a beneficial effect which is dose-and time-dependent.
Cheng et al. [134] evaluated the effects of aspirin in two clinical scenarios: patients with stable angina (SA) and those referred for a first-time coronary artery bypass graft (CABG).

P2Y12 Inhibitors
The interaction between the P2Y12 receptor and adenosine diphosphate (ADP) is essential for thrombus formation, leading to the release of platelet-dense granules, the inhibition of the intracellular prostacyclin pathway, and finally, the promotion of platelet aggregation. Thus, P2Y12 has become an attractive target for the latest generation of antiplatelet therapies [135]. The potential antioxidative property of P2Y12 inhibitors has so far been evaluated mainly with clopidogrel. Clopidogrel is a second-generation thienopyridine that irreversibly binds P2Y12, inhibiting platelet aggregation. As an inactive prodrug, clopidogrel needs to undergo hepatic bioactivation. However, this step involves only 15% of the administered prodrug, as the remaining 85% is extensively converted into a non-functional metabolite [136]. Despite the pharmacokinetic profile and the rise of more potent third-generation P2Y12 inhibitors (prasugrel and ticagrelor), clopidogrel is still widely prescribed [137,138]. Some preclinical studies show that clopidogrel diminishes lipid peroxidation, and therefore the MDA level, as a marker of oxidative stress damage. This effect was observed in several in vivo models [139][140][141]. Clopidogrel was able to not only reduce ROS toxicity but also to prevent a GSH decrease, enhancing natural and endogenous antioxidant systems. In two studies, clopidogrel was evaluated in the context of ischemia reperfusion injury models [139,140]. Treatment with clopidogrel reduced oxidative stress toxicity. In BALB-c mice, clopidogrel administration resulted in a higher total antioxidant capacity, despite the unchanged levels of CAT, SOD, and GSH, and it reduced renal cell apoptosis [140].
In human aortic endothelial cells (HAECs), clopidogrel has proven to have a vasoprotective action: it counteracts oxidative stress through the activation of CaMKKβ/AMPK/ Nrf2 pathways, and by increasing GSH and HO-1 expression [142].
Clopidogrel also exerted anti-inflammatory and antioxidant functions in wild-type mice infused with angiotensin II. Angiotensin II causes vascular dysfunction and platelet activation, enhancing the inflammation response. Clopidogrel was shown to reduce platelet deposition and the platelets-monocytes interaction induced by angiotensin [143].
In another study, sixty male mice fed a high-fat diet were evaluated after isoproterenolinduced myocardial ischemia. Pretreatment for eight weeks was performed either with clopidogrel or aspirin. Mice in the clopidogrel group had better protection against myocardial infarction. Indeed, as compared with the aspirin group, metalloproteinases, hsCPR, and troponin C were further reduced. Conversely, CAT activity and the redox state improved [144].
Three studies assessed ticagrelor's antioxidant properties [145][146][147]. Kang et al. [145] induced apoptosis in HUVECs through oxLDLs administration. Ticagrelor diminished apoptosis in a dose-dependent fashion, proving protection against oxidative damage. In the study of El-Mokadem et al. [146], ticagrelor was evaluated in renal ischemia-reperfusion injury. Ischemia was induced in Wistar rats by clamping renal arteries. Animals in the treatment group received ticagrelor for three days after reperfusion, showing a significant reduction of lipoperoxidation, TNF-alpha, and NF-kB expression [146]. Finally, Bitirim et al. [147] showed that the extracellular vesicles, derived from ticagrelor-exposed high-glucoseincubated H9c2-cells (H9c2), dramatically reduced ROS generation; in addition, the oxidative stress-associated miRNA expression profile was mitigated. Grzesk et al. [148] demonstrated that among the P2Y12 inhibitors, only ticagrelor avoided the ADP-induced VSMC contraction. This effect, derived from the interaction with extra-platelet-located P2Y12 receptors, enriches the evidence of pleiotropic effects of this new potent antiplatelet drug.
Several in-human studies support the hypothesis of the pleiotropic effects of the P2Y12 inhibitors.
Heitzer et al. [149] evaluated daily clopidogrel administration after a 300 mg loading dose in patients with stable coronary artery disease (CAD) already receiving aspirin. Participants were randomized to take a P2Y12 inhibitor or a placebo. After five weeks, patients receiving clopidogrel had lower levels of urinary 8-iso-PG F2α and sCD40L, reducing proinflammatory stimuli. Clopidogrel ameliorates ED, thus improving NO bioavailability.
Taher et al. [150] documented a rise in GSH levels in diabetic patients randomized to clopidogrel therapy without a loading dose. Uncontrolled diabetic patients were randomized into two groups: clopidogrel and placebo groups, both already receiving hypoglycemic therapy. After two months, a reduction of oxidative stress as well as a reduction of glycemic parameters was observed in the clopidogrel group.
Circulating endothelial cells (CECs) are an established marker of vascular injury in people with diabetes [154]. Treatment with clopidogrel seems to reduce the number of CECs. Participants of the study received clopidogrel at 75 mg for 30 days. Blood specimens were collected before and after the treatment. CECs returned to almost normal levels; moreover, clopidogrel increases the level of phosphorylated Akt and phosphorylated adenosine monophosphate kinase, which increases the Enos activity and provides a beneficial endothelial function [155].
Another observational study evaluates the antioxidant effects of clopidogrel in patients undergoing percutaneous coronary intervention (PCI). Bundhoo et al. examined 58 patients with SA. All of them underwent PCI with stent implantation and received aspirin before the procedure. One group received clopidogrel at a 600 mg loading dose, while the second group was already on a clopidogrel maintenance dose of 75 mg for at least three days. Both loaded patients and chronic therapy patients showed a beneficial trend of the plasma total antioxidant capacity (TAC) [156].
More recently, the antioxidant effects of two different P2Y12 strategies were evaluated in stable CAD patients requiring PCI [152,153].
The use of ticagrelor on top of aspirin, for at least one month, in CAD patients with chronic pulmonary obstructive disease (COPD) showed a significant reduction in the apoptosis rate and ROS production when patients' serum was added to HUVECs [152]; moreover, an increased expression of sirtuin1 (SIRT1) and hairy enhancer of split-1 transcription factor (HES-1), two antioxidant transcription factors, was shown [153].
The antioxidant effect of prasugrel has been evaluated in a single, randomized, activecontrolled study including patients with unstable angina in need of PCI. After three months of 10 mg of prasugrel, a significant reduction in MPO levels, sCD40L, and nitrite levels was shown [151] (Table 2).
Three preclinical studies employed HUVECs as experimental models [158,161,163]. The exposition of human plasma and advanced glycation end products to HUVECs led to ROS production and adhesion molecules' expression. These effects were reverted by rivaroxaban in a dose-dependent fashion [158]. Reduced levels of adhesion molecules, oxidative stress biomarkers, proinflammatory cytokines, and chemokines were also demonstrated in the other two HUVECs models [161,163]. Rivaroxaban seems to mediate these functions through inhibition of the PAR pathway, activated by thrombin signaling.
Most preclinical studies, both in vitro and in vivo, were conducted on Wistar rats [160,162,164,165,[167][168][169]. Table 3. Main studies evaluating the antioxidant effects of rivaroxaban. MDA: malondialdehyde; ROS: reactive oxygen species; MCP-1: monocyte chemoattractant protein-1; ICAM-1: intercellular adhesion molecule 1; VCAM-1: vascular cell adhesion protein 1; MPO: myeloperoxidase; NOX: NADPH oxidase; NOS: nitric oxide synthase; SOD: superoxide dismutase; GP: glutathione peroxidase; NF-kB: nuclear transcription factor-κB; TBARS: thiobarbituric acid reactive substances; GR: glutathione reductase; HUVECs: human umbilical vein endothelial cells. Gul Utku et al. demonstrated the healing effects of rivaroxaban in a model of induced colitis due to its favorable activity on MPO and SOD [160]. In two models of liver fibrosis, Vilaseca et al. showed that rivaroxaban improved the ED and reduced oxidative stress [162]. Rivaroxaban also exerted a protective effect in a model of testicular injury, where rats pretreated with rivaroxaban maintained higher levels of endogenous antioxidants and showed lower expression of NF-kB [164]. Rivaroxaban also mitigates the lipid peroxidation, as demonstrated by the lower levels of MDA in an ischemia-reperfusion model of Sprague-Dawley rats [157].

Authors
Wistar rats were the preferred model for in vivo studies as well [167][168][169]. Two studies used sunitinib, an oral tyrosin kinase inhibitor, to induce oxidative stress, resulting in renal injury and cardiotoxicity. Intraperitoneal administration of sunitinib resulted in renal injury [168] and cardiotoxicity [167]. Rivaroxaban-treated rats exhibited lower levels of MDA and TNF-alpha; conversely, GSH and glutathione reductase decreases were aborted. Moreover, cardiac fibrosis and remodeling were mitigated due to diminished NF-kB expression and ROS production [166]. Rivaroxaban was also able to reduce oxidative stress in brain tissue, in an experimental model of depression [169].
A unique human ex vivo investigation evaluated the influence of rivaroxaban on abdominal aortic aneurysmal sites with intraluminal mural thrombus. In the rivaroxaban pretreated group, a decreased expression of ICAM-1, VCAM-1, and IL-6 was reported. Moreover, levels of NOX subunits were reduced [170].
In the COMPASS trial [171], the combination of 100 mg of aspirin and 2.5 mg of rivaroxaban twice daily significantly reduced the cardiovascular death, stroke, and myocar-dial infarction in patients with stable CAD and/or peripheral artery disease (PAD). These positive results led to several studies investigating the synergistic effects of antiplatelet and anticoagulant medications. Abedalqader et al. [165] examined rivaroxaban and aspirin effects in a model of isoproterenol-induced cardiac injury. Combination therapy was associated with a decrease in TBARS and IL-6 levels. However, no significant difference was observed between combination therapy and rivaroxaban, or aspirin administered as a monotherapy. Recently, Russo et al. [172] showed that dual-pathway inhibition with low-dose rivaroxaban and aspirin in patients with an established diagnosis of CAD and/or PAD was associated with a reduction in serum levels of some inflammation markers, such as IL-6 and fibrinogen. Moreover, this combined therapy showed little to no impact on hemoglobin values and renal function markers. These findings support the hypothesis of a pleiotropic anti-inflammatory effect of rivaroxaban, in addition to its anticoagulant effect, and partially explain the positive results of the COMPASS trial for the reduction of cardiovascular events in patients with stable atherosclerotic vascular disease.

Apixaban
Few data are available about the antioxidant effects of apixaban. In a preclinical in vitro study, Torramade-Moix et al. [173] showed that the preincubation of HUVECs and human dermal microvascular endothelial cells (HMECs-1) with apixaban was associated with normalization of ROS levels, increasing eNOS expression, and the reduction of adhesion molecules' expression (Table 4).

Edoxaban
Two preclinical studies described the role of edoxaban on oxidative stress [174,175]. Edoxaban antioxidant activity was assessed in a model of human proximal tubular cells (human kidney 2 cells (HK-2 cells)) exposed to several oxidative stress stimulants. Edoxaban blunted ROS production induced by angiotensin II, indoxyl-sulfate, and factor Xa; moreover, it was able to effectively scavenge peroxynitrite and O2-. Notably, edoxaban scavenging activity was independent from its action on factor Xa, but it was derived from its peculiar molecular structure [174].
In a recent study by Fang et al. [175], wild-type mice underwent a subtotal nephrectomy at 8 weeks of age and were randomized to receive edoxaban or a normal diet. Edoxaban-treated mice showed lower levels of inflammatory and oxidative stress biomarkers; moreover, the HK-2 cells' preincubation with edoxaban resulted in a lower expression of NF-kB (Table 5).

Dabigatran
Two early preclinical studies [176,177], including ApoE−/−mice fed a high-fat diet, showed a considerable reduction in the mean plaque area and an increased thickness of the fibrous cap, in those randomized to receive dabigatran. These modifications were linked to lower NF-kB levels.
In the mouse models of neurodegenerative diseases, dabigatran significantly reduced the ROS levels and the expression of iNOS and NOX [178][179][180].
Treatment with dabigatran significantly inhibited the P65 of nuclear factor κB, tumor necrosis factor α, interleukin (IL)-1β, and IL-6 activities, and significantly enhanced the catalase and superoxide dismutase activities in the AMI rabbits [184].

Clinical Implications
The beneficial effects of aspirin, clopidogrel, ticagrelor, and rivaroxaban on the oxidative stress system have been shown in preliminary observational clinical studies including patients with CAD; however, the relationship between the antioxidant properties and the occurrence of cardiovascular events has not been demonstrated. The choice of the oral antithrombotic therapy based on the need of its antioxidant properties should follow a patient-centered approach. In the future, the use of oxidative stress biomarkers could help to identify these patients. The long-term management of antiplatelet and anticoagulant regimens may be complex in real-world settings, especially in elderly patients with comorbidities [189][190][191][192]. Bleeding risk remains a major issue, as clinicians have to carefully monitor the changing risk over time and take into account the residual thrombotic risk. In the setting of acute coronary syndrome and a high bleeding risk, the use of the P2Y12 inhibitor or aspirin monotherapy after a short period of dual-antiplatelet therapy (DAPT), or the early shift from newer P2Y12 inhibitors to clopidogrel, are considered potential strategies to reduce the bleeding risk [192,193]. DOACs emerged as the preferred choice for the prevention of stroke and systemic embolism in AF patients at an increased thromboembolic risk; moreover, DOACs are indicated in the treatment of VTE. Even if much lower than VKAs, some drug-food and drug-drug interactions have been described during DOACs treatment due to the interaction with P450 cytochromes, ABC transporters, and P-glycoprotein (P-gp) [194][195][196]. When starting a DOAC, knowledge of current kidney and liver functions is required. Importantly, kidney function should be assessed using the Cockcroft-Gault formula, as it was used in the pivotal phase III RCTs. Moreover, a baseline hematological profile should be obtained for reference during long-term follow-up. [195] The use of DOACs in daily clinical practice does not require monitoring of coagulation since all seminal RCTs comparing DOACs to VKAs have been conducted without dose adjustments based on plasma level measurements. However, conflicting results have emerged from the analysis of RCTs and long-term longitudinal studies [197][198][199][200]. Therefore, assessment of the anticoagulant effect of DOACs may be desirable in certain rare situations, such as extreme body weight, concomitant oncologic therapies, patients after transplantation, patients on HIV medication, etc. [201,202].

Conclusions
The oxidative stress plays a role of pivotal importance in the physiopathology of CAD, peripheral artery disease, venous thrombosis, and atrial fibrillation. Several experimental in vitro and animal studies have shown the pleiotropic antioxidant effects of both oral antiplatelet and anticoagulant therapies. In the clinical setting, the beneficial effects of aspirin, clopidogrel, ticagrelor, and rivaroxaban on the oxidative stress system have been demonstrated by preliminary observational clinical studies including patients with coronary artery disease. Whether this mechanistic evidence may have an impact on lowering cardiovascular events is an intriguing question to investigate.