Rheumatoid Arthritis and Periodontal Disease: A Complex Interplay

Recent advances in understanding the dynamic pathways involved in the pathogenesis of rheumatoid arthritis have emphasized the pivotal role of pro-inflammatory cytokines, inflammatory cells, endothelial cell activation and matrix degradation, acting in a genetically predisposed environment. On the other hand, there are significant amounts of data highlighting the potential role of bacteria (leading periodonthopatic pathogen Porfiromonas gingivalis) in promoting different types of arthritis, as well as the influence of periodontis (an infectious-inflammatory condition) as etiological or modulating factor in different pathologies, including cardio-vascular disorders, diabetes, respiratory disease and inflammatory rheumatic disorders (such as rheumatoid arthritis, ankylosing spondylitis and lupus). The present chapter deals with the possible association between rheumatoid arthritis and periodontitis as entities with common pathological events.

We systematically reviewed data from literature focusing on inflammation and tissuedamaging aspects, oral microbiota, antibodies against bacteria and autoantibodies and treatment for both RA and periodontitis, aiming to explore the relationship between both diseases.
To better understand, P. gingivalis encourages the immune subversion by several strategies including: IL-8 inhibition, complement manipulation and toll-like receptor 4 (TLR4) antagonism, leading to impaired host defense. Symbiotic gingival microbiota with tissue homeostasis swaps to dysbiotic environment, with inflammation and bone loss as main attributes of periodontal disease [1,8,11,12].
Of particular interest, P. gingivalis-dependent protein citrullination and the immune response to the periodontal infective aggression involving both innate and adaptive immunity (e.g., anti-citrullinated immunity and IgG anti-P. gingivalis antibodies) additionally outline the complexity of pathogenic pathways in periodontitis [1-5, 7-10, 15].

Porphyromonas gingivalis as a key player in periodontitis
Chronic periodontitis is, therefore, typically initiated by the colonization of dental plaque by a core set of periodontal pathogens, specifically P. gingivalis, T. forsythia and T. denticola, documented as the "red complex", prototype of polybacterial pathogenic consortium in periodontitis [1-5, 7-11, 19].
P. gingivalis, a leading microorganism in this dysbiotic setting, is a gram-negative anaerobic bacteria, involved in the onset of inflammation and tissue destruction during periodontal disease [1][2][3][4][5][6][7][8][9][10][11]. It is ordinarily found in small number in a healthy oral cavity [11], but the pathology occurs when P. gingivalis binds to and accumulates on the tooth surface, leading to the development of a mixed biofilm, the expansion of the bacteria into the gingival sulcus and the formation of a periodontal pocket [11,19].
P. gingivalis is an outstanding example of a bacteria able to shape the periodontal microenvironment and to destabilize the homeostatic immune system endorsing chronic inflammation and tissue damage as well [11].
It is worth mentioning that the switching of a charged arginine residue to an uncharged citrulline results in evident, irreversible posttranslational protein modifications, meaning changes in the three-dimensional conformation and abnormal function [8,10,17]. Citrullinated antigens as neo-epitopes in the periodontal tissues may selectively activate adaptive immune response and break the self-tolerance leading to severe aggressive periodontitis, anti-cyclic citrullinated peptides, anti-P. gingivalis antibodies as well, and, potentially, to the development of RA in genetically predisposed hosts [1-3, 20, 25].
PAD-mediated citrullination is also responsible for evading host defense against P. gingivalis by two additional mechanisms: generation of ammonia that result in a favorable microenvironment for anaerobic bacteria and inactivation of certain chemokines contributing to neutrophil recruitment into the gingiva [8,10,11,17].
Besides, P. gingivalis stimulates a true "local chemokine paralysis" by inhibiting the expression of IL-18 on epithelial cells [8,11], delaying the traffic of neutrophils and, therefore, permitting its persistence and/or proliferation in the periodontal tissue with subsequent modification of the subgingival microbiota [11,25].
This apparently distinctive bacterial enzyme is effective in generating citrullinated proteins (bacterial and human) and, furthermore, has the ability to autocitrullinate, eliciting the loss of immune tolerance and synthesis of autoantibodies in susceptible individuals [1-5, 7-11, 22].
The main differences between human and bacterial PAD focus on the mechanism of activation and substrates undergoing citrullination [1,10,11]. Thus, PPAD is a calcium-independent enzyme that becomes active at higher pH that PAD; its action is facilitated by the co-localization with arginine-specific gingipains in the outer membrane of P. gingivalis [1,2,10,11,20,25]. PPAD act specifically on C-terminal arginine residues and free arginine [1,2], events enabled by the proteolytic activity of gingipains. In fact, we talk about a two-step process in the inflamed-infected periodontal sites, initiated by cysteine protease gingipains that cleave protein chains and expose the C-terminal arginine, facilitating the secondary intervention of PPAD [1,2,10,11,25].
Finally, PPAD allows, either directly or indirectly (via enhancement of inflammatory reactions and release of host PADs), the abnormal citrullination of bacterial and human proteins with the assembly of the so-called cryptic-epitopes and subsequent antibody production [1,2,10,11,20,25].

PPAD and PAD autocitrullination
Although PPAD preferentially citrullinates different bacterial and/or host peptides bearing Cterminal arginine, it becomes clear that this enzyme may undergo autocitrullination and, hence, stimulate a specific pattern of autoimmunity and request for autoantibody response in permissive genetic background (carriers of HLA-DRB1 shared epitope) [2,10,20]. There is a heightened immune response to P. gingivalis and IgG specifically targeting autocitrullinated PPAD in chronic periodontal disease and rheumatoid arthritis which could perpetuate the immune response through epitope spreading and cross-reactivity with citrullinated human proteins suggesting a potential mechanistic linkage between the two disorders [10,11,25,29,30]. Of interest, PAD4 also experiences autocitrullination, and antibodies recognizing this isoform of human PAD embrace predictive value in RA [10,11,26,27,31].
To summarize, the infection with P. gingivalis triggers a broad spectrum of local and systemic inflammatory, destructive and immune events in susceptible individuals. Strategic steps in the complex pathobiology of periodontitis comprise early release of a true arsenal of specialized virulence factors and bacteria-danger signals (LPS, fimbriae, peptidoglycan, various proteolytic enzymes) exerting both direct and indirect effects (via stimulation of inflammatory and immune pathways, release of host PAD), followed by aberrant PPAD-dependent citrullination of different substrates such as host-derived and bacterial proteins, as well as PPAD-autocitrullination, with subsequent generation of criptic epitopes and autoantibody synthesis. Finally, the immune subversion endorses loss of tolerance to structurally similar host proteins and outcomes the induction of anti-cyclic citrullinated antibodies suggesting the relationship between chronic periodontitis and rheumatoid arthritis [1,2,10,13,17,19,20,25].

Role of citrullination in rheumatoid arthritis development
Citrullination is a posttranslational modification of proteins at arginine residues reported in a wide range of tissues and settings, holding relevant physiologic as well as pathologic implications [10,17,27,32]. Although considered a critical event in different inflammatory conditions, currently available data emphasize the link between citrullination and autoimmune diseases, suggesting that immunity towards citrullinated self-proteins rather than citrullination itself is a specific event in RA [10,17,27].

Citrullination and inflammation
Abnormal citrullination of endogenous proteins is typically acknowledged as an initial, tactical process in the complex pathobiology of RA, driving the induction of altered self-epitopes and extensive autoimmune response [10,17,27].
Moreover, immunogenicity of these new antigens explains the presence of antibodies against citrullinated peptides (ACPA) considered not only diagnostic biomarkers for RA and prognostic factors for severe aggressive disease, but also consuming a pathogenic role [10,17,27]. Nevertheless, only individuals with definite HLA polymorphisms, such as the conserved region of HLA-DRB1 alleles are at risk to recognize as "nonself" such citrullinated proteins and to develop RA [1,2,10,17,27].
Along with glycosylation and carbamylation, citrullination is basically involved in many physiological processes in a variety of tissues, biochemical enzymatic reactions leading to structural and functional protein/peptide modification [2,10,17,27]. Conversely, dysfunctional posttranslational protein modifications are often reported during autoinflammatory pathology, particularly RA, evading the immunological tolerance and stimulating a specific profile of humoral response [2,10,17,26,27].
As mentioned before, citrullination or deamination strictly encompass for a change of arginine with citrulline, meaning that an imine nitrogen of arginine-residues is replaced by an atom of oxygen; accordingly, histone modification, genomic regulation and neutrophils extracellular trap (NET) formation are described and engaged in development of newly citrullinated antigens [10,17,27].
It is well-known that citrullination constantly requires the activation and intervention of PADfamily enzymes at intra and/or extracellular levels in a Ca 2+ -dependent manner, and only two out of the five human PAD subtypes (PAD 2 and PAD4) are relevant for RA based on their expression in joints, immune cells, neutrophils and mast cells [1,10,11,12,17,26,27]. Thus, PAD-mediated citrullination could be easily classified as a basic cellular damage but also an inflammatory process occurring in a range of inflammatory backgrounds such as synovial tissue, gingival and periodontal sites and lungs, challenging local immunity [10,17,27].
On the other hand, recent advances in understanding the pathophysiology of chronic periodontal infection with P. gingivalis, and the potential connection between periodontitis and systemic disorders including RA, have reemphasized the role of this periodonthopathic microbe as one of the risk factors for RA in a permissive genetic context [1-11, 17, 27]. P. gingivalis is the only known periodontal bacteria in the gingival microbiota that expresses a PPAD able to initiate and perpetuate a gradual autoimmune process through epitopespreading by peptide citrullination (bacterial, PPAD itself as a citrulllinated bacterial protein), cross-reactivity with citrullinated human proteins and ACPA synthesis [1,2,10,11,25].
The break of tolerance to one citrullinated epitope predisposes to break of tolerance to additional citrullinated epitopes; the initial mechanism might be more related to recognition of citrullinated antigen per se than to a particular citrullinated autoantigen [10,11,26,32,33].

Smoking, citrullination and rheumatoid arthritis
Environment-triggered citrullination (smoking, bacterial infections) in articular and extraarticular locations essentially progress to specific immunity and disease in genetic predisposed individuals [1,2,10,12,16,32,34,35]. Smoking is a well-known risk factor for RA, particularly, in ACPA-positive subsets. Cigarette smoking leads to a sequence of events responsible for local increase in PAD enzyme expression and consequent generation of protein citrullination in the lung [10,12,17,27]. Chronic exposure to self-citrullinated epitopes could further contribute to the loss of immune tolerance and gradual occurrence of humoral immune response towards some of the citrullinated autoantigens (ACPA) [10,17,27]. However, smoking and other risk factors might lead to breaking of immune tolerance only in the presence of genetic risk factors for ACPA positive RA [10,12,17,26,27,36].

Anti-citrullinated proteins/peptide antibodies (ACPA) pathobiology
ACPAs, the autoimmune signature of RA, are sensitive and specific diagnostic, prognostic and therapeutic biomarkers in RA; they are detected from very early phases of the disease, up to a decade prior to the RA clinical onset, and feature a dynamic correlation with RA severity, clinical outcomes and efficacy of synthetic and biologic therapies [10,37,38]. In fact, ACPAs are a family of partly cross-reactive antibodies which target citrullinated self-proteins and peptides abundant in inflamed joints, such as fibrinogen, fibronectin, vimentin and vinculin, β-enolase, histone, biglycan, clusterin, collagen type II, keratin. So far, 53 such citrullinated members are identified, an assembly called "citrullinome" [10,12,17,27].
According to the ACPA status, RA could be classified in two distinct clinical phenotypes with particular genetics, risk factors, immunopathology, clinical and treatment outcomes; thus, ACPA-positive RA is defined by a specific genetic (HLA-DRB1*0401 and *0404)-environmental interface (e.g., smoking and P. gingivalis infection), higher disease severity (erosive RA, extra-articular involvement, comorbidities) and poor remission rates. Besides, high ACPA titers could predict clinical response with B-cell depletive agents (rituximab) or selective costimulation modulators (abatacept), independently of disease activity and, therefore, might help to define therapeutic RA profiles [10,11,16,17,27,28,36].
ACPAs are detected as antibodies against cyclic citrullinated proteins (anti-CCP antibodies) which are detectable by specific assays in RA patients, and their screening performance largely varies with different tests: sensitivity and specificity significantly improved with thirdgeneration assays, as well as the ability to identify those subjects with undifferentiated arthritis who are likely to develop RA [10,17]. Conversely, second-generation tests are classically more specific for longstanding, established disease [10,17,39].
The exact molecular pathways of anti-citrullinated immunity in RA are still obscure, but it is clear that ACPAs encourage the perpetuation of synovial inflammation via binding to citrullinated proteins positioned in the cellular membrane, attached to a variety of cellular receptors (such as TLR4 or FcγR) or incorporated in NETs [1-4, 10, 11, 17]. Furthermore, complement binding and activation may enhance local inflammation [7,12,17,27].
Overall, the effector functions of different well-defined ACPA may adjust according to antibody specificities. Thus, reactivity towards citrullinated vimentin as well as anti-cartilagespecific protein collagen II antibodies directly bind to osteoclast surface and modulate their activation, determining subchondral bone erosive lesions and osteoporosis [10,12,40]. On the other hand, antibodies to autoantigenic enolase and histones focus on local inflammation [10,12,17,27], as ACPAs are able to induce NETs.

The paradigm of citrullination and ACPA in the development of ACPA positive rheumatoid arthritis
The environmental exposure (such as smoking, silica and other nanomaterials of air pollution, bacterial infection, e.g., P. gingivalis) and the genetic determinants (MHC class II alleles) enable abnormal citrullination and support immunity against citrullinated proteins in certain extraarticular sites (lung, periodontal tissue). The relationship between the emergence of local ACPAs as the first sign of autoimmunity and RA is further supported by the documentation of shared citrullinated peptides in the lungs, periodontal sites and joints as potential ACPA epitope bearing proteins [10,17,27].

The biological link between periodontitis and rheumatoid arthritis
Continuing periodontal disease as a trigger for chronic arthritis in susceptible individuals via dysregulation in oral microbiota and host immune barriers is a reliable, but still debatable concept [1,2,8,[10][11][12]. The old paradigm indicates that RA could be a consistent risk factor for chronic periodontitis: impaired periodontal health labels the multifactorial and synergistic result of compromised joint functionality (hand as well as temporomandibular joint), oral dryness (secondary Sjogren's) and specific RA medications [8].
In contrast, newer theories emphasize that periodontal disease is a risk factor for RA: both are multifaceted disorders with shared pathogenic mechanisms such as local and systemic excess of inflammation, irreversible bone injury, aberrant activation of the immune system with an extra dose of autoimmunity and a common genetic background [1-5, 8, 41-45].
Overwhelming evidence coming from cross-sectional studies [1-13, 10, 40, 46-48] showed a statistically significant correlation between RA and periodontitis (prevalence and/or severity); nevertheless, other cohort-based or case-control studies noticed no association between oral health and rheumatic inflammatory pathology [1,2,10,12]. The conflicting reports may be explained by differences between study populations, sample size, adjustments for confounders (age, gender, race/ethnicity, smoking status, RA activity and medication) and lack of uniformity in disease classification criteria for periodontal disease [49].
More specifically, a meaningful positive correlation between measures of RA activity and extent and severity of periodontitis was reported: high disease activity (DAS28 scores) is generally demonstrated in RA with aggressive periodontitis as compared to RA without or with moderate periodontitis, even after adjusting for multiple confounders [4,8,47].
A systematic review of cross-sectional trials further strengthens the evidence addressing periodontitis in relation to autoimmunity damage in RA. Overall, significantly higher ACPA, rheumatoid factor and antibodies to P. gingivalis were described among RA with periodontitis than those without periodontal disease [4,8,11,47], whereas increased alveolar bone loss was significantly associated with ACPA positivity in the same patient population [4]. Similarly, detailed analysis of ACPA specificities (citrullinated vimentin and histone) showed increased levels in RA associated with both moderate and high alveolar loss, without differences based on smoking status [4,15,16,26,[48][49][50]. It seems that periodontitis is a strong predictor of ACPA positivity in RA [8], whereas IgG and IgM ACPA subtypes are highly associated with antibodies against P. gingivalis [2,3,8,29]. Additionally, in refractory RA with periodontitis the most powerful anaerobes identified in synovial fluid are P. intermedia, P. gingivalis and T. denticola [1,40,41]. However, no difference in ACPA or rheumatoid factor levels for RA participants with mild periodontal lesions as compared to moderate or severe periodontitis were found among other studies, suggesting that ACPA status might not influence the prevalence and severity of periodontal disease [1,8,11,49].
In particular, only a small number of studies focused on periodontitis as a primary event, occurring before clinical onset of RA versus periodontitis as comorbidity, described after the diagnosis of rheumatic condition [1,8,11].

A parallel between chronic periodontal disease and rheumatoid arthritis
Exploring the association between chronic periodontal disease and RA has demonstrated striking similar immunological, biological and genetic underlying processes [1,2,4,12,22], although apparently totally different etiologies (autoimmune RA, infective periodontitis respectively) [10]. Figure 2 offers a detailed synopsis of similarities between RA and periodontitis. Both are complex multifactorial disorders, characterized not only by a dysfunction of basic inflammatory and tissue destructive mechanisms, but also by an altered adaptive and innate immune response in individuals at risk (genetic traits) [1,2,8,10,12,22].
The excess of local (periodontal tissue and gingival crevicular fluid in periodontitis, synovial microenvironment in RA) along with systemic chronic inflammation is driven by an imbalance between pro-inflammatory and anti-inflammatory cytokines, with persistent up-regulated pro-inflammatory signaling pathways (mainly IL-1, IL-6, TNFα, NF-kB) and higher amounts of inflammatory mediators including reactive oxygen species, nitric oxide and lipid mediators [1-3, 8, 10-12, 22]. In addition, altered connective tissue and bone homeostasis with irreversible damage of collagen-rich structures (gingiva, periodontal ligament and alveolar bone in periodontitis, subchondral bone and cartilage in RA) is essentially related to augmented activity of collagenolytic matrix metalloproteinases and other enzymes (elastase, bacterial cysteine proteases, neutrophil associated enzymes), but also to overexpression of the RANK/ RANKL/OPG system [1-3, 8, 10, 12, 13, 45].
The validity of the relationship between periodontitis and RA is, furthermore, supported by a particular genetic predisposition and related environmental risk factors (e.g., smoking) [1,8,10,12]; both entities share a common genetic profile (shared epitope HLA-DRB1 alleles) [1,10,12], while polymorphism of genes encoding inflammatory cytokines as well as IL-1 combined risk alleles which may cause a synergistic effect on bone destruction in joints and the periodontium might also confer susceptibility for RA and periodontitis [1,8,10,12,15,16].
Finally, abnormal PAD and PPAD-mediated citrullination of the synovial and host periodontium peptides/proteins (vimentin, keratin, α-enolase), respectively, with sequential loss of tolerance against neo-epitopes account for altered local and systemic immune reactions in susceptible patients with either periodontitis or RA; the generation of citrullinated new autoantigens and patterns of antibody response by ACPA, anti-autocitrullinated PPAD, antiautocitrullinated PAD4 and anti-heat shock proteins of P. gingivalis (anti-hsp70) antibodies are frequently encountered in both RA and periodontitis [1,10,12,15,25,30,45].
Overall, gingival citrullination and local induction of ACPA, as well as cross-reactivity between antibodies to P. gingivalis and ACPA may explain the link periodontitis-RA [1, 10, 12, 27].

Effects of therapy for periodontitis on rheumatoid arthritis disease activity and systemic inflammation
It was advocated that specific therapies for periodontitis not only improve disease activity and severity in RA by removing periodontal infection and gingival inflammation, but also improve patient response to therapeutic drugs [1,8,50,53,54].
Thus, nonsurgical periodontal therapy was directed to improve both signs and symptoms of active RA as well as periodontal status [1,8,12]. Similarly, advanced periodontal treatment with scaling and root planning significantly ameliorated gingival infection and severity of periodontitis, while promoting consistent reduction in acute phase reactants, mainly erythrocyte sedimentation rate, in the RA population [1,8,12,53,54]. The impact of periodontal treatment on RA seems to be greater in patients with substantial amount of systemic inflammation, particularly if periodontitis intervention therapy combined with adjunctive periodontal host-modulating therapy [4,12,53,54].
Furthermore, a positive influence on RA activity scores, rheumatoid factor and ACPA levels, but also on anti-P. gingivalis antibodies was reported following supportive periodontal therapy in patients having concomitant RA and chronic periodontitis [1,12,52,[54][55][56].
The consequences of therapy for periodontal disease on RA status actually reflect on the decrease in systemic inflammation, removal of joint exposure to the infectious trigger and the decline in microbial immune subversion [1,8,12].

Effects of conventional therapy for rheumatoid arthritis on periodontal status
On the other hand, different authors concluded that early aggressive RA management with synthetic disease modifying anti-rheumatic drugs (DMARDs) and/or biologic agents aiming to achieve and maintain disease remission according to treat-to-target strategies might also restrict periodontal damage in active periodontitis [1,8,12,54].

Conclusions
There is a significant interplay between periodontitis and rheumatoid arthritis as supported by emerging evidences. Despite common T cell activation, the imbalance between proinflammatory and anti-inflammatory cytokine profile, citrullination of endogenous proteins and the resultant bone destruction in specific environments (synovium and gingival tissue), further research is required to assess the exact mechanisms and causative link, as well as to define top therapy.