Rheumatoid arthritis (RA) is a chronic autoimmune disease characterized by persistent inflammation of the synovium, which leads to the degeneration of bone and cartilage [1]. The complex and multifaceted pathophysiology of RA involves a combination of environmental triggers, dysregulated immune responses, and genetic predisposition [1–3]. The family of enzymes known as protein arginine deiminases (PADs) is a major contributor to this complex process [4,5]. This review examines the current understanding of PADs’ roles in RA, with particular attention to their function in disease pathophysiology, its consequences, and the ongoing challenges in developing effective therapeutics that target PADs.

PADs constitute a family of calcium-dependent enzymes responsible for an irreversible post-translational modification known as citrullination, or deamination [6,7]. This enzymatic reaction involves the hydrolysis of the guanidinium group of peptidylarginine residues, converting them into citrulline [8]. The resulting citrullinated proteins act as neo-antigens, prompting the production of ACPAs. ACPAs are highly specific to RA and frequently appear before the onset of clinical symptoms, making them valuable biomarkers for diagnosis and prognosis [9]. Their presence often precedes overt disease, suggesting their involvement in disease initiation or exacerbation [10].

ACPAs are not merely biomarkers; they can actively contribute to disease pathogenesis. By interacting with citrullinated proteins [11], they can activate immune cells and promote inflammation in the synovial joint. Moreover, ACPAs can directly interact with synovial lining cells, including fibroblast-like synoviocytes (FLSs), altering their activity and causing joint degradation [11]. Specifically, ACPAs can bind to citrullinated antigens expressed on the surface of FLSs. This interaction triggers intracellular signaling cascades, particularly the phosphorylation of Focal Adhesion Kinase (FAK) and the subsequent activation of the PI3K/Akt pathway, which ultimately enhances FLS migration and promotes synovial invasion [11]. The substantial correlation between the existence and levels of ACPAs and the prognosis and severity of the disease demonstrates their key role in RA pathophysiology [12,13]. Fig. 1 illustrates the central role of PAD4-driven citrullination in initiating this pathogenic cycle, highlighting how it simultaneously drives ACPA production and NETosis to amplify RA inflammation.

PAD4 functions as the core enzymatic engine in RA pathogenesis, catalyzing the conversion of arginine to citrulline, which drives two major intertwined pathways. Firstly, PAD4 creates neo-antigens (e.g., citrullinated fibrinogen), leading to the production of ACPAs that precipitate synovial inflammation and joint damage. Secondly, PAD4 translocates into the neutrophil nucleus to citrullinate histones, driving NETosis. The released NETs—rich in citrullinated proteins—further amplify the inflammatory response and ACPA production, establishing a pathogenic vicious cycle.

Protein citrullination by PADs significantly influences RA pathophysiology beyond being merely a consequence of inflammation. Citrullination generates neo-antigens from self-proteins [14,15], which the immune system recognizes as foreign targets. PADs themselves, particularly PAD4, can also serve as immune targets [16], leading to ACPA production. Certain HLA-DR alleles, particularly those carrying the “shared epitope” such as HLA-DRB1*04:01, have been linked to the occurrence of anti-PAD4 antibodies in RA patients, potentially via a hapten carrier mechanism where the PAD4 enzyme itself is presented as an autoantigen [7,17]. In this specific context, it is hypothesized that the PAD4 enzyme itself functions as the carrier protein. According to the hapten-carrier model, the PAD4 enzyme functions as a carrier protein that forms complexes with citrullinated substrates (acting as haptens). In this scenario, autoreactive T cells recognize PAD4 peptides presented by HLA-DR molecules and provide help to B cells that have internalized the PAD4-citrullinated protein complex, thereby driving the production of both ACPAs and anti-PAD4 antibodies [7,17]. The specific interaction between autoreactive T cells and B cells via this hapten-carrier mechanism, leading to the production of dual autoantibodies, is schematically represented in Fig. 2. Furthermore, PAD gene polymorphisms have been associated with RA in existing literature [18–20].

Beyond ACPA production, PADs contribute to RA pathogenesis through other mechanisms. PAD enzymes are crucial for the development of NETs [5,21,22]. NETs are web-like structures composed of DNA, histones, and other granule proteins released by activated neutrophils [23,24]. While NETs participate in host defense against infections, their overproduction in inflammatory diseases like RA causes tissue damage and inflammation [25]. Notably, PAD4’s citrullination of histones is essential for NET formation, highlighting the direct link between PAD activity and NET-mediated inflammation in RA [23]. Mechanistically, upon neutrophil activation by stimuli such as calcium influx, PAD4 translocates to the nucleus [26]. There, it catalyzes the hypercitrullination of histones (particularly H3), which reduces their positive charge and weakens DNA-histone binding; this enzymatic step is crucial for driving chromatin decondensation and the subsequent release of NETs [26,27].

Several publications emphasize the involvement of PAD4, a specific PAD isozyme, in RA pathophysiology [2,16,28]. Neutrophils, as key inflammatory cells in the RA synovium, exhibit significant PAD4 expression [16,28–30]. PAD4 activity facilitates the production of NETs; specifically, PAD4-mediated citrullination of histones is necessary for chromatin decondensation, a crucial step in NET formation [22,23]. Since NETs can amplify inflammation and aggravate joint injury, PAD4 inhibition can reduce NET production, thereby attenuating inflammation. Strong evidence from animal models, such as the collagen-induced arthritis (CIA) model, also supports the role of PADs in RA pathogenesis. Studies showing a considerable reduction in disease severity using pan-PAD inhibitors, such as Cl-amidine, underscore the therapeutic potential of targeting PADs [31,32].

The PADs family comprises calcium-dependent enzymes that catalyze the post-translational conversion of peptidyl-arginine residues to peptidyl-citrulline [4,5]. This process, known as citrullination, alters protein structure and function, potentially leading to the generation of neo-epitopes that trigger autoimmunity [14]. The citrullination of various proteins, including fibrinogen and vimentin [1,14], is linked to the formation of ACPAs in RA.

The human genome encodes five PAD isozymes (PAD1–4 and PAD6) [21,33], each displaying distinct tissue distribution, subcellular localization, and substrate specificity [34]. While all PAD isozymes catalyze the citrullination of arginine residues, their specific roles in RA pathogenesis vary. The prominence of PAD4 in RA pathogenesis, compared to the more ubiquitously expressed PAD2, is thought to stem from several key differences. Critically, PAD4 possesses a unique nuclear localization signal [35,36], allowing it to enter the nucleus and citrullinate histones—the requisite step for the formation of pro-inflammatory NETs. In contrast, other isozymes like PAD2 are primarily cytoplasmic [37]. The exact contributions of other PAD isozymes, such as PAD2 and PAD3, to RA pathophysiology remain unclear compared to PAD4, despite their documented involvement in citrullination processes in the RA synovium [21,38]. Furthermore, intriguing clinical data suggest that while anti-PAD4 antibodies correlate with severe disease, anti-PAD2 antibodies are paradoxically associated with milder outcomes, implying distinct and non-redundant roles for these enzymes [39]. Further research is necessary to fully elucidate the interactions between various PAD isozymes and their overall effects on RA onset and progression.

The catalytic mechanism of PAD enzymes involves a unique reverse protonation mechanism, distinct from other hydrolases [34,40]. This mechanism critically involves an active site cysteine residue and a histidine residue, functioning as a nucleophile and general acid-base catalyst, respectively. PAD activity is tightly regulated, primarily by calcium ions [34,41]. An increase in intracellular calcium levels, often observed in inflammatory conditions like RA, is a critical determinant for PAD activation [23]. Calcium binding induces a series of conformational changes in PADs, leading to the formation of a catalytically competent active site [42]. However, PAD activity regulation extends beyond calcium. Recent studies highlight the role of bicarbonate as a pH-independent regulator of PAD activity. Bicarbonate, at physiological concentrations, enhances PAD activity independently of calcium levels and pH, suggesting a more complex regulatory network than previously appreciated. Further research is crucial to fully elucidate the interplay among calcium, bicarbonate, and other potential regulatory factors in controlling PAD activity within the context of RA. Understanding these regulatory mechanisms is vital for developing effective PAD inhibitors with enhanced specificity and therapeutic potential. The precise mechanisms controlling PAD isozyme expression and activity in the inflamed synovium of RA patients remain an active area of investigation, necessitating further studies to define the specific contributions of each PAD isozyme to RA pathogenesis and identify potential targets for selective inhibition.

Autoantibodies against PADs were discovered more than two decades ago [35]. Among them, autoantibodies against PAD4 (anti-PAD4 Abs) are the most well-documented. Anti-PAD4 Abs can exist before the clinical diagnosis of RA, and these antibodies correlate with the development of ACPAs.

The role of autoantibodies against PADs, particularly anti-PAD4 antibodies, represents one of the most significant ongoing debates in RA research. The evidence is highly contradictory, leading to the ‘foes or friends’ question. On one hand, a large body of evidence positions these antibodies as ‘foes’. Multiple studies have convincingly linked anti-PAD4 antibodies to more severe disease [43–45]. Furthermore, their presence has been associated with persistent disease activity and greater radiographic progression [46–49]. Furthermore, recent studies have provided direct mechanistic evidence for this pathogenic role, demonstrating that anti-PAD4 autoantibodies from RA patients can activate monocytes and exacerbate inflammatory arthritis [50].

In stark contrast, other studies present a paradoxical picture. A recent study focusing on RA-associated ILD found that anti-PAD4 antibodies were associated with less lung fibrosis and better overall survival, suggesting a context-dependent or even protective role in certain extra-articular manifestations [51].

This apparent contradiction is a critical area of investigation. The most likely explanation is that the anti-PAD4 antibody response is not uniform. These antibodies are heterogeneous; it is hypothesized that antibodies targeting different epitopes on the PAD4 enzyme may have opposing functions—some may activate the enzyme, while others may inhibit it [49,52].

Furthermore, the complexity is highlighted by other PAD isozymes. For instance, anti-PAD2 antibodies have also been identified, and they are associated with less disease severity, directly challenging the simple narrative that all anti-PAD autoimmunity is pathogenic [39]. In clinical practice, resolving this heterogeneity is crucial. Anti-PAD4 Abs may serve as a potential biomarker for diagnosis and prognosis in the future [35], but larger cohorts and standardized assays are necessary for validation.

The dysregulation of PAD activity in RA is not limited to arthritis or joint inflammation. Emerging evidence suggests a link between PADs and various RA complications. RA is a systemic autoimmune disease that can affect multiple organs and systems, leading to a range of extra-articular manifestations [53]. These extra-articular manifestations (EAMs) affect multiple organ systems, are significant predictors of mortality, and include major complications such as interstitial lung disease (ILD) and cardiovascular disease (CVD), where PAD involvement is an active area of research [54].

The critical role of PADs in RA pathogenesis has driven significant efforts to develop PAD inhibitors as therapeutic agents [33,65]. Various classes of PAD inhibitors have been identified, including irreversible inactivators and reversible inhibitors [33]. Irreversible inactivators, such as Cl-amidine and F-amidine, covalently modify the active site cysteine residue of PAD enzymes, leading to irreversible inhibition [33]. These compounds have shown promise in preclinical models of RA, demonstrating a significant reduction in disease severity [31,32]. However, their irreversible nature raises concerns about potential off-target effects and toxicity. Reversible inhibitors, such as streptomycin and minocycline, bind to PADs non-covalently and are less likely to cause off-target effects [29]. However, their potency and selectivity may be lower compared to irreversible inactivators.

A significant challenge in PAD inhibitor development is achieving isozyme selectivity [34,65]. PADs exhibit distinct tissue distribution and substrate specificity, and inhibiting all isozymes simultaneously may lead to undesirable side effects. Therefore, the development of potent and selective inhibitors targeting specific PAD isozymes, particularly PAD4, is a major focus of current research efforts [32]. Significant progress has been made in designing PAD inhibitors with improved selectivity and potency. Structure-activity relationship (SAR) studies have been instrumental in optimizing the chemical structures of PAD inhibitors to enhance their affinity and selectivity for specific PAD isozymes [33,65]. For instance, the development of PAD2-selective inhibitors is of significant interest, considering the role of PAD2 in multiple sclerosis and other autoimmune diseases [65].

The use of high-throughput screening (HTS) and activity-based protein profiling (ABPP) has enabled the identification of novel PAD inhibitors with unique mechanisms of action [66]. These advanced techniques have significantly accelerated the drug discovery process and facilitated the identification of lead compounds for further optimization. A recent database study using in silico molecular docking simulations on the PAD4 enzyme found several potential FDA-approved medications [67], but further in vivo validation is crucial for clinical applications. A summary of these potential therapeutic agents is provided in Table 1.

The involvement of PADs in RA pathogenesis and complications has fueled significant interest in developing PAD-targeted therapies [4,74]. Several approaches to PAD inhibition are currently under investigation. Fig. 4 summarizes the distinct mechanisms of action for these therapeutic strategies, categorizing them into pan-PAD inhibitors, isozyme-selective inhibitors, biologics, and novel allosteric modulators.

PADs are central to the pathogenesis of rheumatoid arthritis (RA) and its complications, contributing to both joint inflammation and extra-articular manifestations. Specifically, PADs, particularly PAD4, facilitate the generation of ACPAs and NET formation, alongside potentially other disease manifestations. Significant progress has been made in developing PAD inhibitors, with several promising candidates demonstrating efficacy in preclinical models [56,57]. However, the development of effective PAD-targeted therapies still faces significant challenges. These include achieving optimal potency, high isozyme selectivity, improved bioavailability, enhanced in vivo stability, and ensuring overall safety and tolerability [15]. Future research is crucial to fully elucidate the complex roles of PADs in RA, understand their intricate regulation, and identify novel drug targets. This includes focusing on PAD4-selective inhibitors and thoroughly investigating the role of PADs in extra-articular manifestations. Advancements in drug design, sophisticated screening methodologies, and innovative drug delivery strategies hold promise for future PAD-targeted therapies. The continued exploration of drug repositioning strategies, exemplified by the investigation of Saquinavir’s potential as a PAD4 inhibitor, also presents a viable path towards effective and cost-efficient RA treatments [67]. Furthermore, a deeper understanding of the complex interplay between PADs, the immune system, and the microbiome (particularly Porphyromonas gingivalis and its PPAD enzyme), along with the potential role of bacterial PADs, will be crucial for developing truly effective and personalized therapies [15,91,92]. Investigating additional regulatory factors like bicarbonate could also pave the way for novel therapeutic strategies [42]. Ultimately, success in translating the promise of PAD-targeted therapy into effective clinical treatments for RA and related autoimmune disorders will hinge on a comprehensive understanding of these complex interactions and the development of more sophisticated biomarkers [82,83,89]. Ultimately, overcoming these challenges is essential for therapeutic progress rather than academic interest alone. The successful translation of PAD-targeted strategies into clinical practice would introduce a first-in-class mechanism in RA treatment, with the potential to benefit patients who are refractory to current therapies and those burdened by extra-articular disease.