Pros and Cons of Causative Association Between Periodontitis and Rheumatoid Arthritis

Abstract

Research in recent decades has brought significant advancements in understanding of the molecular basis of the etiology of autoimmune diseases, including rheumatoid arthritis (RA), a common systemic disease, in which an inappropriate or inadequate immune response to environmental challenges leads to joint destruction. Recent studies have indicated that the classical viewpoint of the immunological processes underpinning the pathobiology of RA is restricted and needs to be expanded to include a more holistic and interdisciplinary approach incorporating bacteria-induced inflammatory reactions as an important pathway in RA etiology. Here, we discuss in detail data showing the clinical and molecular association of RA development with periodontal diseases (PD). We also describe the unique role of periopathogens, which have been proposed to be crucial in the initiation and progression of this autoimmune pathological disorder.

1. Introduction

In the context of a possible causative link between PD and RA, it is worth noting that while PD with associated dysbiotic microflora was prevalent in ancient populations as diverse as Egyptians, Scottish prehistoric, early precolumbian Americans, the Roman British and medieval European populations ^1–6, bioarchaeological/paleontological studies failed to reveal convincing evidence of RA. The first recognized description of RA dates to the beginning of the 19^th century.⁷ Due to a lack of conclusive evidence, there are three hypotheses: (i) RA has an ancient origin; (ii) RA has a modern origin and (iii) RA originated in the New Word and was transmitted to the Old World after Columbus discovered the Americas. Therefore, if PD (with pathogens dating back to the Neolithic epoch) underlines development of RA, the first hypothesis would gain support in and rent credibility to an anecdotal report contributed to Hippocrates of the successful treatment of joint pain by extraction of bad teeth. Here we present and critically discuss current research data, both supporting and negating the causative link between PD and RA in the context of the underlying inflammatory pathobiology of both diseases and possible involvement of periodontal pathogens.

2. Rheumatoid arthritis in the reflection of periodontitis?

Rheumatoid arthritis (RA) is a common, systemic autoimmune disease. The reported prevalence of RA ranges from 0.5% to 1% in the adult population worldwide with a female-to-male ratio of 3:1. This chronic disease is characterized by inflammation of the synovial lining of joints leading to the destruction of cartilage, erosion of bone, pain, and chronic disability. RA usually affects joints in the hands, feet, and wrists, but progression of the disease may also affect the kidneys, skin, lungs, and liver. The disease is strongly associated with cardiovascular damage and other systemic complications increasing mortality. The etiology of RA is complex and the pathogenic mechanisms underlying RA are not fully elucidated. However, this disease shares striking similarities in its immunological, biological, and genetic background with periodontal diseases (PD), suggesting a strong association between these two disorders.⁸

2.1. Genetic and environmental risk factors

Numerous studies have identified genes implicated in the genetic predisposition to RA, the strongest of which is the shared epitope motif (SE), a specific five amino acid sequence in the type II human leukocyte antigen (HLA) of the major histocompatibility complex class II protein (MHC II). Several alleles are considered to contain the SE, many of which are located in the HLA-DRβ1 region. Expression of specific HLA-DRβ1 polymorphisms is associated with increased risk of developing RA, and individuals possessing one or more susceptibility alleles are associated with increased disease severity. By themselves, SE alleles are thought to be main genetic susceptibility factors.^9,10

The possible role of DRB1 alleles in alveolar bone resorption has been recently studied. Gehlot et al. documented that transgenic mice carrying the human SE-coding allele DRB1*04:01 have increased susceptibility to spontaneous periodontal inflammation and bone-destruction.¹¹ Moreover, Sandal and coworkers showed that exposure of the gingiva of HLA-DR1 humanized C57BL/6 mice to Porphyromonas gingivalis results in systemic inflammation with elevated level of cytokines, differentiation of Th17 cells (both in the peripheral blood and cervical lymph nodes), decrease of femoral bone density, and the generation of anti-citrullinated protein antibodies (ACPA).¹²

Polymorphisms in other genes, including solute carrier family 22 member 3 (SLC22A3), runt-related transcription factor 2 (RUNX2), peptidylarginine deiminase (PAD)I4, TNFAIP3, and PTPN22, may also confer susceptibility to RA. Interestingly, some of these may also contribute to the pathogenesis of PD. For example, Schulz and coworkers examined the impact of genetic variants in the PTPN22 (rs2476601), PADI4 (rs2240340), and CTLA4 genes (rs3087243) on RA and PD. The authors showed that the T allele of rs2476601 (PTPN22) is associated with a higher susceptibility to PD within the RA group, and is a significant biomarker of RA and PD comorbidity.¹³ Recently, a case-control study revealed that the KCNQ1 rs2237892 allele is significantly associated with comorbidity of RA and PD. The authors showed that patients with RA and chronic periodontitis carrying the T allele had increased disease severity and more inflammation, compared with non-carries.¹⁴

Single nucleotide polymorphisms of the human tumor necrosis factor-alpha (TNF-α) gene, such as rs1800629 (−308 promoter polymorphism; A/G), leads to increased expression of TNF-α, an inflammatory cytokine, and has also been connected to autoimmune diseases, including RA. Recently, the TNF-α rs1800629 polymorphism was also found to be significantly associated with PD.¹⁵

Taken together, it seems that PD and RA may share similar genetic risk factors. This complicates the search for a causal relationship and can be considered as an argument against causative association between diseases. Nevertheless, it needs to be considered that, while SE alleles, PTPN22, and to a lesser degree PADI4 variants, are strongly associated with RA, their impact on PD is very weak.

The intriguing relationship between PD and RA is further supported by related environmental risk factors. They can shape the immune response to bacterial infections omicrobial dysbiosis in a manner that might break tolerance against host antigens. The process can occur on the mucosal surfaces, specifically the lungs, gut and periodontium, confirming observation that unhealthy diet, high alcohol consumption, poor oral hygiene or cigarette smoking are risk factors of RA development.^16–19

2.2. Common pathophysiological mechanisms of inflammation

Although the etiology of RA and PD is different (one is autoimmune, and the other of an infectious nature), both diseases share several common features. Local chronic inflammation in the synovial compartment or gingival pockets is a typical clinical hallmark of both RA and PD, respectively. In both cases, chronic inflammation is the result of inappropriate response of the innate and acquired immune systems at various stages of the disease. Both disorders are characterized by local accumulation of cytokines (e.g., TNF-α, IL-1β, IL-6, and MCP-1), reactive oxygen species, nitric oxide, and lipid mediators, such as prostaglandin.²⁰ Uncontrolled inflammatory reactions result in increased infiltration of neutrophils, macrophages, and T- and B-lymphocytes, which continuously migrate to affected sites in response to locally released chemokines. The persistence of activated phagocytes (macrophages, dendritic cells, and their secretion of inflammatory mediators, including chemokines) strongly promotes self-sustaining inflammation. Neutrophils play an important role in the pathogenesis of PD and RA, constituting 80–90% of the synovial fluid cells in RA patients, and 80–95% of the leukocytes present in the gingival crevicular fluid (GCF) of periodontitis cases.²¹ Their massive influx, combined with prolonged life-span or impaired elimination, leads to their necrosis, degranulation, or formation of neutrophil extracellular traps (NETs). NETs have been found in copious amounts in the GCF, purulent crevicular exudates, and biopsies of the pocket epithelium from periodontitis patients.²² NETs lead to the increase and uncontrolled activity of enzymes released from the secondary granules of neutrophils, such as elastase, cathepsins, and proteinase 3, along with excessive production of reactive oxygen species (ROS). Together, these enhance the destruction of the gingiva, periodontal ligament, and alveolar bone in patients with PD, and the subchondral bone and cartilage in patients with RA. The process is augmented by the activity of collagenolytic matrix metalloproteinases.²³ The upregulation of receptor activator of nuclear factor-kB ligand (RANKL) expression by fibroblasts and lymphocytes leads to osteoclast formation by macrophages in both the inflamed periodontium and joints.^24,25 T- and B-cells have both been shown to be a major source of RANKL in the diseased synovium and periodontium.

Both RA and PD are inflammatory and Th-driven diseases. Recent evidence suggests that the two types of T helper cells, Th1 and Th17, play a major role in RA and PD pathogenesis.²⁶ The secretion of IL-17A cytokines by Th17 cells activates a number of pathways, such as fibroblast-like synoviocytes (FLS), maturation and function of osteoclasts, activation of neutrophils, macrophages, and B-cells.²⁷ Taken together, chronic inflammatory events and immunoregulatory imbalance eventually leads to the destruction of both the soft and hard tissues of the joint and alveolar bone in RA and PD, respectively. For example, a local inflammatory reaction in the periodontal tissue or synovial microenvironment accompanies the systemic chronic inflammation. It has been reported that individuals with PD or RA suffer from underling systemic dysregulation of the inflammatory response, as manifested by elevated levels of TNF-α and IL-6 detected in the serum.²⁸

Collectively, both diseases are characterized by a complex network of impaired interactions between components of the immune system, leading to a loss of homeostasis in the tissues and resulting in irreversible damage. The similarity of pathogenic mechanisms between PD and RA is another complication when investigating the reciprocal comorbidity of these two diseases. Due to the inflammatory nature of PD and RA, each can exacerbate the other through the production of proinflammatory cytokines.

3. Association between periodontitis and rheumatoid arthritis

Recent data from numerous clinical and epidemiological studies provide evidence for a bidirectional association between RA and PD. Clinical trials show that patients with long-standing active RA have a substantially higher frequency of PD than healthy subjects.

On the other hand, some studies report that periodontitis is a risk factor for developing or even enhancing the severity of RA. Below, we describe the most convincing reports from the clinical field, which are supported with data obtained using animal models.

3.1. Clinical signs

Evaluation of periodontal status in RA patients revealed that PD prevalence was at least 2-fold higher in RA patients, and that they had a worse periodontal condition then healthy controls.^29,30 The conclusion comes from the detailed dental examination assessed by probing pocket depth, clinical attachment level (CAL), and bleeding index (BI) determined at six sites per tooth, and the presence of supragingival plaque (PI) at four sites per tooth. The data presented by Eriksson revealed that the majority of tested RA patients (75%) had moderate or severe periodontitis, and the remainder had no or mild periodontitis ³⁰, strongly corroborating results of earlier clinical studies.³¹ These data were further supported by an observation of arthralgia in patients who had never been treated with anti-rheumatic drugs or glucocorticoids.³² Moreover, extended analysis of systemic mediators of inflammation revealed increased levels of sCD30/TNFRSF8, IFN-α2, IL-19, IL-26, MMP-1, gp130/sIL-6Rβ, and sTNF-R1, in the serum or GCF of RA patients.

Based on the hypothesis that periodontitis is a risk factor for the development of RA, it is important to examine the association between periodontitis and the risk of developing RA (pre-RA) or early signs of RA (eRA). A report by Bello-Gualtero studying 119 individuals with pre-RA showed significantly higher levels of plaque index (PI), bleeding on probing (BOP), and severity of periodontal disease. In the study, periodontitis was associated with pre-RA but not with eRA, and the appearance of anti-citrullinated protein antibodies (ACPA) preceded the onset of RA symptoms.³³ These findings were validated by Terao and coworkers who examined PD status and the presence of ACPA and IgM-rheumatoid factor (RF) in a cohort of 9554 adult healthy subjects.³⁴ These authors reported significant associations between PD parameters, disease status, and level of ACPA, thereby supporting the involvement of ACPA production in PD. Further, a study investigating the first-degree relatives (FDRs) of RA patients showed higher prevalence of periodontitis (79% vs 56%) in this group than in healthy subjects. Moreover, 15% of FDRs had severe periodontitis.³⁵ This observation was confirmed by the increased severity of periodontitis in FDRs patients, which was associated with seropositivity to ACPAs. In ACPA positive (+) individuals, the mean PI, probing depth (PD), BOP, CAL, and number of sites per person with PD >4 mm were significantly higher than in the ACPA negative (−) group. All ACPA+ subjects had periodontitis: 44.1% presented with moderate and 47.1% with severe periodontitis.³⁶

The association of RA with PD has also been examined in case-control studies enrolling PD patients. The data revealed that compared with the general population, subjects with PD are at an increased risk of developing RA. Moreover, the clinical course of PD in RA patients is more severe and is independent of age, gender, ethnicity, or smoking history, when compared with non-RA individuals.^37,38 The association between PD and RA is proposed to be based on the presence of ACPA antibodies and RF in the serum and gingiva of patients with periodontitis.³⁹ Using samples collected from 39 PD patients, Lappin et al. showed that patients with untreated periodontitis had higher ACPA titers than healthy controls.⁴⁰ Of note, the presence and titer of ACPA in serum is routinely measured using commercial kits utilizing immobilized cyclic citrullinated peptides (CCP) as antigens. Due to the limited evidence and inconsistent findings on whether periodontitis increases the risk for RA, Qiao and coworkers performed a meta-analysis of 13 published studies including a total of 706611 periodontitis patients and 349983 control subjects. The data from this meta-analysis indicated that patients with a periodontitis duration >5 years are more likely to develop RA and have a 69% greater risk for RA than people in the control group.⁴¹ This study showed that periodontitis represents a risk factor for RA.

3.2. Microbiological data: Periopathogens in RA patients

It is considered that the causal association between RA and PD may be related, in great part, to periodontal pathogens. However, serological studies of antibodies specific for periopathogens yielded opposing observations when RA patients were compared with healthy controls. Some reports have shown that immunity to P. gingivalis, but not other periopathogens (Prevotella intermedia, Aggregatibacter actinomycetemcomitans, and Eikenella corrodens), is significantly associated with the presence of RA-related autoantibodies in individuals at risk of RA, which suggests that infection with P. gingivalis may play a central role in the early loss of tolerance to self-antigens that occur in the pathogenesis of RA.^42–44 Arvikar et al. noted a similar observation in a subset of eRA patients who had positive antibody responses to P. gingivalis. The responses correlated with ACPA reactivity. Compared with P. gingivalis antibody-negative patients, eRA patients with positive P. gingivalis responses had more anti-cyclic citrullinated peptide (anti-CCP) antibody reactivity and significantly higher anti-CCP levels, and the levels of anti-P. gingivalis antibodies correlated directly with anti-CCP levels.⁴⁵ Conversely, another study found that anti-P. gingivalis antibody titers did not significantly differ between patients with RA and controls, nor did they significantly differ with ACPA, RF, or HLA SE status.⁴⁶ In yet another study, IgGs specific for P. intermedia, Prevotella melaninogenica, and Tannerella. forsythia were significantly higher in RA patients than in controls.⁴⁷ Moreover, a case-control study by Johansson and coworkers showed that anti-P. gingivalis antibody concentrations were significantly higher in RA patients than in controls, and were detectable years before onset of symptoms of RA.⁴⁸ A significant association was found between IgG against P. gingivalis and ACPA in individuals at risk for RA, and markers of RA activity in individuals with eRA.³³ This observation was confirmed by Mankia and coworkers, who found an increased prevalence of periodontitis and P. gingivalis in anti-cyclic citrullinated protein (anti-CCP) antibody-positive at-risk individuals without arthritis.⁴⁹ Taken together, it seems that P. gingivalis positive-periodontitis is more likely to occur in ACPA+ individuals without any arthritis, suggesting that PD may precede RA. In a recent large population study in Korea, the authors suggested that increasing periodontal indices could be used as a marker of RA development, and that anti- P. gingivalis antibody titer could inform about RA severity in patients suffering from PD.⁵⁰ The discrepancy in these results, pertinent to the correlation between anti-P. gingivalis antibody levels and ACPA, prompted Bae and Lee to examine this issue using meta-analysis. The authors examined anti-P. gingivalis or anti-RgpB antibody levels reported in 15 separate studies comprising a total of 3829 RA patients and 1239 controls. The meta-analysis demonstrated that the anti-P. gingivalis antibody is significantly higher in patients with RA, and that a positive relationship exists between anti-P. gingivalis antibody levels and ACPA.⁵¹

In addition to serological studies, several reports have described an association between the level of specific bacterial periopathogens in the subgingival biofilm and RA. For example, Schmickler et al. reported higher P. gingivalis and F. nucleatum levels in aCCP-positive RA patients.²⁹ In a study conducted by Eriksson and coworkers, a detailed analysis of plaque and saliva microbiota revealed an altered subgingival microbial profile between investigated groups of patients with PD and RA. RA patients with moderate or severe periodontitis had a significantly higher abundance of Desulfobulbus sp., Prevotella sp., Bulleidia sp., Capnocytophaga sp., T. forsythia, and a single NA sp. in plaque, than RA patients with no or mild periodontitis. By contrast, Prevotella oris and Porphyromonas sp. were more abundant in patients with no or mild periodontitis. The results showed that P. gingivalis was present more frequently (62%) in the moderate/severe periodontitis group than in the no/mild periodontitis group (50%). In saliva, there were no significant differences in the detected bacterial species in RA patients with or without PD based on periodontal diagnosis.³⁰ By contrast, the number of P. gingivalis in tongue biofilm was significantly associated with the severity of RA disease expressed as DAS28. This observation suggests that the oral cavity microbiological status could play a role in the pathogenic mechanisms of RA-associated inflammation, leading to a more active disease state.⁵²

4. Therapy

Because PD and RA are associated, it is assumed that successful treatment of one disease should prevent or improve the outcomes of the other. In the section below, we discuss the results of published studies.

4.1. Treatment of PD and the influence on RA severity

The role of PD in RA development is supported by evidence from studies that evaluated the effects of periodontal treatment on RA severity in terms of the serum erythrocyte sedimentation rate (ESR), DAS28, C-reactive protein (CRP), and TNF-α levels.^37,53 Ribeiro et al. revealed that in RA patients who underwent full-mouth scaling and root planing (SRP), ESR was reduced with no changes in the level of RF.⁵⁴ By contrast, a recent report reported that SRP had no significant effects on ESR in patients with moderately active RA diagnosed with PD.⁵⁵ Another study showed that SRP improved periodontal conditions in PD in patients with and without RA. Moreover, the authors of this study suggested that in patients with RA, eradication of P. gingivalis in conjunction with a high level oral hygiene can transiently decrease disease activity of RA.⁵⁶ Together, these data corroborate previous observations showing successful treatment of RA patients with antibiotics against bacterial anaerobic infections.⁵⁷

4.2. Treatment of RA and the influence on PD severity

Among the clinical biomarkers currently used for the diagnosis of RA are RF, anti-perinuclear factor (APF), anti-keratin antibodies (AKA), anti-filaggrin antibodies (AFA), and anti-cyclic citrullinated peptide antibodies, which are thought to be identical to ACPA. An investigation directed toward identifying new biologic markers revealed an association between increases in the levels of systemic proinflammatory mediators and RA, including IL-1, IL-6, IL-8, IL-17, IL-21, TNF-α, and GM-CSF.⁵⁸ Therefore, patients suffering from RA are treated with anti-inflammatory drugs, which can also strongly influence their periodontal status. Among anti-inflammatory drugs routinely applied to treat RA are the inhibitors of TNF, such as infliximab and adalimumab (anti-TNF neutralization antibodies), and etanercept (an anti-TNF receptor antibody). Mayer et al. evaluated the influence of infliximab on the periodontal health of patients with RA, and the association between GCF TNFα levels and periodontal inflammatory parameters. They found that treatment of RA patients with infliximab significantly reduced the levels of TNFα in GCF and decreased periodontal inflammation.⁵⁹ Apparently, such treatment has no influence on the level of anti-P. gingivalis antibody, as shown by a recent study performed with 79 infliximab-treated RA patients.⁶⁰ Therefore, the effects of TNF inhibition on PD parameters in RA patients remains an open question, but is of interest given that TNF promotes osteoclastogenesis and inhibits osteoblastogenesis. The effect of IL-6 inhibition with rituximab or tocilizumab was also intensively studied.⁶¹ Recent data demonstrated that treatment with tocilizumab (an anti-IL-6 receptor antibody) might not only improve clinical and biochemical RA-related parameters but also ameliorate PD. The authors observed that tocilizumab therapy significantly improved the gingival index scores and decreased the number of sites with BOP after only 3 months, while the probing pocket depth decreased significantly after 6 months. The authors linked the positive clinical outcome to a decrease in the level of IL-6 in the periodontal microenvironment and reduced systemic inflammation.⁶²

Among the drugs used to prevent RA development are conventional synthetic disease-modifying anti-rheumatic drugs (DMARDs). However, there are limited data on the influence of DMARDs on PD. Some studies have reported no association between anti-rheumatic treatment and periodontal parameters ⁶³, whereas others have shown beneficial effects of DMARDs on periodontal clinical parameters following non-surgical periodontal treatment.⁶⁴

Collectively, the data from the majority of studies suggest that treatment of periodontitis and eradication of P. gingivalis could be a good approach to prevent RA. The same is apparent for treatment of RA, which has a beneficial effect on the clinical outcome of PD. Together, these observations confirm the clinical association between PD and RA, but do not provide enough evidence for a causative relationship between these diseases. This is because both treatments target inflammation that is directly responsible for progression of both PD and RA.

5. Murine models

The association between PD and RA revealed by clinical and epidemiological studies is strongly supported by animal studies. The involvement of periodontal pathogens in the development of RA is clearly documented by studies demonstrating that pre-existing periodontitis exacerbates experimental arthritis in rat and mouse models.⁶⁵ Additionally, Bartold and coworkers demonstrated that the presence of pre-existing chronic inflammation exacerbated the development of adjuvant arthritis induced by injecting a mycobacterium cell wall in complete Freund’s adjuvant into rats.⁶⁶ These data indicated that synovial inflammation, with the pathobiology similar but not identical to periodontitis, promoted the development of experimental PD. Further, several groups have confirmed the relationship between gingival inflammation and arthritis by infection with live P. gingivalis using the oral gavage model. In a collagen antibody-induced arthritis model, severe arthritis developed faster in infected mice than in controls.⁶⁵

More recently, reports have been published implicating P. gingivalis as an etiologic factor contributing to disease severity in experimentally induced RA models.^12,67–69 A similar effect to infection with P. gingivalis was observed with other periodontal pathogens, including P. nigrescens ⁷⁰, F. nucleatum and A. actinomycetemcomintans.⁷¹ Two studies have examined the effects of multispecies infection in an experimental model of RA, including the use of P. gingivalis, T. denticola, and T. forsythia ⁷², or P. gingivalis, F. nucleatum, and A. actinomycetemcomintans.⁷¹ In the latter study, mice inoculated with a mixture of all three pathobionts showed less alveolar bone loss than mice inoculated with P. gingivalis alone, suggesting that F. nucleatum and A. actinomycetemcomitans somehow attenuate P. gingivalis-induced alveolar bone loss. Their finding suggests that the interaction of bacteria-forming dysbiotic biofilm in periodontitis patients may have a strong influence on RA development and progression. Moreover, it was shown recently that therapeutic eradication of P. gingivalis with chlorhexidine and metronidazole reduced the incidence and alleviated the severity of collagen-induced arthritis comparable to methotrexate in a murine model of periodontitis. ⁷³

As RA is strongly associated with genetic risk factors, it is unsurprising that periopathogen infection has a significant impact on RA development in mice strains genetically susceptible for RA, such as SKG ⁷⁴, HLA-DR1 humanized C57BL/6 mice ¹², DBA/ 1JJmsSlc ⁶⁹, DBA/ 1J ⁶⁷, DBA/1 × B10.Q F1 ⁷¹, and B10.RIII.⁷² Together, these results indicate that periodontitis may be the underlying condition facilitating RA development in genetically susceptible humans. This hypothesis is strongly supported by Courboun et al., who demonstrated that RA is triggered by P. gingivalis oral infection alone without additional induction of experimental arthritis. The authors showed that P. gingivalis induced severe PD, and that this was accompanied by elevated levels of inflammatory mediators (IL-17 and CXCL1) and antibodies against citrullinated proteins (anti-CCP2), with subsequent synovial inflammation and bone destruction. Eight months after oral P. gingivalis infection, the development of spontaneous arthritis in the rat ankle bone was slower than experimentally-induced arthritis.⁷⁵ Despite differences in experimental design, these studies produced similar results and showed that mice with pre-existing periodontitis developed more severe arthritis. Enhanced severity of arthritis was manifested by swelling and erythema in the fore and hind paws; a massive influx of leukocytes; elevated RANKL expression in the joints and periodontal tissues; the accumulation of osteoclasts, cartilage damage and bone erosion; and increased serum CRP levels indicating systemic inflammation. Importantly, the abovementioned arthritic symptoms in animal models closely resemble the clinical signs of RA in humans.

By contrast, there are many reports showing that arthritis can influence the development of periodontitis, which confirms the hypothesis of a bidirectional relationship between these two diseases. In one study, Ramamurthy and coworkers demonstrated that induction of arthritis is associated with spontaneous loss of alveolar bone and increased levels of IL-1β, TNFα, and metalloproteinases in the gingival tissues of an adjuvant arthritis (AA) animal model.⁷⁶ In another study, Trombone at al. documented that a hyperinflammatory genotype aggravates bacteria-induced PD in a model of pristane-induced RA (PIA). The authors showed that PIA induction resulted in alveolar bone loss, which was dependent on the presence of A. actinomycetemcomitans and P. gingivalis.⁷⁷ Further interesting data were obtained in a chronic Ag-induced arthritis (AIA) model, whereby the authors observed the spontaneous development of PD. They found that AIA resulted in severe alveolar bone loss, migration of osteoclasts, and release of proinflammatory cytokines in the maxillae.⁷⁸

Taken together, animal models of PD and RA provide the most conclusive evidence that not only can periodontitis induce RA, but also that the reverse is true. This supports the hypothesis that inflammation is the most important causative factor for both RA and PD (Table 1).

Table. 1.

A compilation of data showing the pros and cons of causative relationship between periodontitis and rheumatoid arthritis;

PROS

CONS

• PD patients are at greater risk for RA than healthy individuals 41 • Inflammatory nature of both diseases; bidirectional exacerbation through inflammatory mediators without causative relationship 20,23–25 • PD prevalence and severity is higher in individuals with pre-RA, first-degree relatives and RA patients 29,30,33–35 • Significantly higher level of antibodies specific for periodontal pathogens in RA patients, including individuals at early stages of RA 33,42–45,47,48 • Association of PD clinical parameters and anti-P. gingivalis antibodies with ACPA in RA and individuals at high-risk of RA development 49,50 • Increased prevalence of PD and P. gingivalis in ACPA positive RA patients and individuals at-risk of RA development 33,49 • PD treatment and eradication of P. gingivalis improved RA status 54,56, while RA treatment ameliorated severity of chronic PD 59,62 • Pre-existing periodontitis exacerbates experimental arthritis 65,71,72 • The inflamed periodontium is a source of PTMs proteins; A.a is a potent stimulator of PADs; PPAD citrullinates bacterial and host proteins 118,130,135 • Citrullinated P. gingivalis proteins and peptides reacts with ACPA 135,140,141 • P. gingivalis and P. nigrescens promote arthritis progression via IL-17 signaling 67,70 • Molecular mimicry of bacterial enolase, Hsp60 and RgpA; bacterial proteins are recognized by antibodies from RA patients 145,148,150

• RA risk alleles within the HLA (particularly HLA-DRB1) predispose to PD12 • Polymorphism of PTPN22, KCNQ1 and TNF-α is proposed as a marker for RA and PD comorbidity13–15 • Anti-P. gingivalis antibody titer did not differ between RA and healthy control 46 • The RA treatment has no influence on anti- P. gingivalis antibodies level 60 • A.a is considered pathogen of rare aggressive forms of PD • P. gingivalis derived citrullinated peptides are not recognized by ACPA in early RA 138 • Pre-existing chronic inflammation induced by other than periopathogens factors exacerbates the development of adjuvant arthritis 66 • The spontaneous alveolar bone loss in adjuvant arthritis model 76

PD - periodontitis, RA - rheumatoid arthritis, PTMs - posttranslational modifications, PADs – peptidylarginine deiminases, PPAD - P. gingivalis PAD, A.a - Aggregatibacter actinomycetemcomitans

6. Molecular mechanism of PD-induced RA

Because most of the research in the field has focused on the infectious etiology of RA, including periodontitis, scientists are actively searching for the causative role of bacteria in the initiation and/or progression of RA. Autoantibodies are drivers of damaging immune responses and are strongly and specifically associated with autoimmune diseases. In the case of RA, the most important autoantibodies are directed against post-translationally modified proteins, which are often produced by the inflamed tissues. The most common modifications include citrullination, carbamylation, or proteolytic fragmentation of proteins. In genetically susceptible individuals, these can be recognized as epitopes by the aberrant humoral immune response. Therefore, elucidation of the immunotolerance breakdown mechanism is crucial in understanding the pathology of autoimmune diseases, including RA. In the case of RA, despite very extensive research, the pathological conditions and anatomical locations where immunotolerance breakdown against citrullinated and carbamylated epitopes occurs are still disputable.

6.1. Post-translational modified proteins

An autoimmune reaction in RA patients is characterized by generation of ACPA, which is detected long before the clinical onset of the disease.⁷⁹ This antibody is both highly specific and sensitive for the diagnosis of RA.⁸⁰ It is produced in response to the increased level of citrullinated proteins, which are considered to be the main factors causing the breakdown of tolerance observed during RA development. Of note, an inflamed periodontium is reported to be a rich source of citrullinated proteins.^81,82 These proteins have been found in the gingiva, GCF, and saliva of PD patients and include citrullinated histones, which are targets of autoantibodies in RA patients.^83–85

There are several reports indicating that anti-carbamylated protein (CarP) antibodies are also present in the sera of patients with RA.^86,87 Because the level of this autoantibody is predictive of joint damage, it has also been proposed as a diagnostic marker for RA. ^{87, 88} Studies examining experimental arthritis have also shown that the appearance of anti-CarP antibodies in the sera precedes the onset of RA.^89,90 Of note, inflamed human periodontal tissue constitutes a significant source of CarPs.⁸² It is tempting to propose that carbamylation in the inflamed periodontium constitutes a plausible mechanism for the generation of antigens involved in the development and progression of RA.

Apart from carbamylation and citrullination, malondialdehyde-acetaldehyde (MAA) adduct formation is significantly increased in RA patients.⁹¹ IgG and IgA anti-MAA autoantibodies were detected in RA patients prior to clinical diagnosis. As they appear later in the preclinical course than ACPA or RF, the authors proposed that MAA adduct formation and anti-MAA immune responses could play a role in the transition from subclinical autoimmunity to clinically apparent arthritis.⁹² Notably, MAA-modified proteins have been found in inflamed periodontal tissue as they are generated as a result of inflammation-associated oxidative stress. However, because current data suggest that anti-MAA autoantibodies may play a role in the development of RA, further investigations are needed.⁹³

All PTMs described above can initiate the generation of autoantibodies, thus promoting the breakdown of immune tolerance, which apparently occurs outside of the synovium on mucosal surfaces. Therefore, below we focus on the inflammatory reaction in the periodontium with emphasis on oral pathogens, which may trigger the production of disease-specific autoantibodies and arthritis development in susceptible individuals.

6.2. Inflammation in the periodontium as a source of autoantigens

The chronic inflammation in PD constitutes a perfect niche for the generation of neo-epitopes due to the local release and/or accumulation of protein substrates and enzymes that catalyze PTMs.

6.2.1. Soluble mediators of inflammation and enzymes

A complex network of inflammatory mediators, comprising cytokines, complement molecules, enzymes (proteases and peptidylarginine deiminases, PADs) and alarmins, is involved in the immunological processes that promote autoimmunity and ultimately tissue destruction in RA, among. Some of these play roles in the recruitment of leukocytes or the release of enzymes, while some directly modify the antigens released from dying cells. Those factors coordinate the extracellular conversion of proteins into immunodominant epitopes, known as remnant epitopes. Below, we describe the most convincing data from this field, linking the etiology of RA with PD.

Anaphylatoxin C5a is the most important factor in the complement system and plays a crucial role in the inflammation underlying the pathogenesis of RA and PD.⁹⁴ The C5a molecule was identified in alveolar and synovial tissue.⁹⁵ Inhibition of the C5a–C5aR axis limits the inflammation and bone destruction observed in murine models of RA and PD, and C5aR-deficient SKG mice with reduced arthritis.^96–98 Furthermore, Arg-specific gingipains (Rgps) very efficiently release the active C5a molecule directly from C5.⁹⁹ The involvement of C5a in the progression of RA infection was investigated by Munenaga et al., which revealed that neutralization antibody against C5a suppresses osteoclast differentiation when mice with experimental arthritis are orally infected with P. gingivalis.¹⁰⁰ A potential role for P. gingivalis in the activation of the C5a signaling pathway and RA pathogenicity is additionally supported by the fact that C5a promotes the development of Th17 cells. Homeostasis in the periodontium is considered to be dependent on the T-cell population, as a proper balance of Th1, Th2, and Th17 regulates the immune events that prevent bone destruction. Among T-cells, Th17 seems to play a unique role in cytokine secretion, including IL-17, which promotes osteoclast differentiation and the development of bone erosions. Using a murine model, P. gingivalis and Prevotella nigrescens oral infection was shown to shift the balance of T-cells into a Th17 population. The Th17-driven response was manifested by elevated serum levels of IL-17 and IFNγ, increased osteoclast numbers in joints, and enhanced arthritis progression and development.^67,70 As the levels of IL-17 induced by periodontal pathogens directly correlated with the intensity of arthritic bone erosion, the authors suggested that IL-7 has a potential role in RA pathology.⁷⁰ The role of IL-17 signaling in arthritis progression in the presence of P. gingivalis was confirmed using IL-17 receptor A-knockout mice.¹⁰¹

The expression of matrix metalloproteinases (MMPs) is highly up-regulated in RA and PD. MMPs are essential for connective tissue homeostatic turnover, but are also responsible for tissue damage if not properly regulated. This happens at chronic inflammatory sites where the unrestrained activity of MMPs leads to the accumulation of a range of protein-degradation products, derived from collagen type I, II, and III; vimentin; and CRP.¹⁰² In the case of GCF and synovial fluid, it was shown that MMP-8 and MMP-9 released from accumulating neutrophils degrade the collagen fibers that generate neo-epitopes.¹⁰³ Moreover, in the GCF of periodontal sites infected with P. gingivalis, lysine-specific gingipain Kgp cleaves human IgG in vivo, releasing Fab fragments that form remnant epitopes.¹⁰⁴

Impaired elimination of leukocytes from inflammatory milieu leads to their necrosis followed by the release of cellular contents. Among the released factors are a group of molecules constituting endogenous damage-associated molecular patterns (DAMPs), including alarmins. Besides being spontaneously released, alarmins can be actively secreted by leukocytes, and oral and salivary tissue. To this end, it was shown that P. gingivalis, T. forsythia, T. denticola, and F. nucleatum induce the release of alarmins, including IL-1, ATP, HSP60, fibronectin, and high mobility group box protein 1 (HMGB1), from human cells.^105–107

The role of HMGB1 has been further studied. The concentration of this alarmin is elevated in GCF, and large numbers of cells in the gingiva of PD patients are HMGB1-positive. HMGB1 promotes the differentiation of osteoclasts as well as the activation of dendritic cells, T-cells, and endothelial cells. Systemic administration of anti-HMGB1 inhibited periodontal inflammation and alveolar bone loss in an oral gavage model of periodontitis ¹⁰⁸, highlighting a potential role for this alarmin in the pathology of PD and RA.

6.2.2. Neutrophils and PAD

The conversion of positively-charged arginine to the neutral citrulline residue is catalyzed by PADs. In humans, there are five PADs, each of which is distributed in different tissues. They play important physiological roles, including control of the immune system, skin keratinization, and the regulation of global gene expression.¹⁰⁹ Under inflammatory conditions, PAD activity is considerably increased with pathological effects. It was shown that accelerated necrosis, or formation of NETs, constitute the main source of active PAD4 enzyme.^110,111 Therefore, neutrophils are suggested to play an essential role in immune tolerance breakdown and ACPA generation, which eventually triggers the development of clinical RA.^112,113

Periodontitis is characterized by the accumulation of high numbers of neutrophils, as well as their overactivity and incomplete removal by efferocytosis from periodontal tissues. The uncontrolled activity of neutrophil proteases and excessive release of ROS contributes to the destruction of periodontal soft tissue and damage of organic components of the alveolar bone.¹¹⁴ NETs have been found in GCF, purulent crevicular exudates, and biopsies of the pocket epithelium of periodontitis patients.^22,115 The abundant neutrophil extracellular nets (NETs) formation (NETosis) is likely due to neutrophils interacting with the bacteria in periodontal pockets.¹¹⁶ Recently, it was shown that P. gingivalis plays a crucial role in NETosis as gingipains can directly induce NET generation in vitro by hijacking the protease-activated receptor-2 (PAR-2) signaling pathway.¹¹⁷ On the other hand, A. actinomycetemcomitans was identified as the potent stimulator of PAD activity in neutrophils.¹¹⁸ Among different periodontal pathogens and oral commensals, only A. actinomycetemcomitans has been reported to hijack activity of host PADs and induce hypercitrullination of proteins in the inflamed periodontium and rheumatoid joint. The process depends on secreted leukotoxin A (LtxA), a pore-forming toxin that induces calcium influx, and subsequent hyperactivation of PAD enzymes in the neutrophil. Exposure to leukotoxic A. actinomycetemcomitans strains is more prevalent both in periodontitis and in RA patients than in healthy individuals and when associated with ACPA.¹¹⁸ Contradictory results were published by Engstrom et al., who found no correlation between the increased level of citrullinated proteins and PAD2 or PAD4 enzymes, in the gingival tissue of PD patients, or the presence of periodontal pathogens, including P. gingivalis and A. actinomycetemcomitans.¹¹⁹ These results argue against a role for neutrophils, as well as dysbiotic bacteria, in the generation of citrullinated proteins. Another mechanism deserving attention is protein carbamylation facilitated by myeloperoxidase (MPO) of neutrophils infiltrating in large numbers into the inflamed periodontium.⁸⁶ MPO catalyzes the oxidation of thiocyanate in the presence of hydrogen peroxide to cyanate, which spontaneously reacts with N-terminal α-amino groups and ε-NH₂ of lysine to generate homocitruline (ε-carbamyl-lysine) in proteins and peptides. Thus, the breakdown of immunotolerance in RA patients is still ill-defined and requires more investigation.

Altogether, the uncontrolled inflammatory reaction, combined with a high degree of destruction observed in periodontal tissues, makes the mechanism of the bystander activation in PD-induced RA development quite plausible.

7. Oral–gut microbiome axis and RA pathogenesis

The gut microbiome has been intensively studied because, among many other functions, it contributes to the development and maintenance of the immune system. Therefore, there is no surprise that dysbiosis in this microbial community is linked to autoimmune diseases, including RA.¹⁷ In the case of RA, the focus of investigation is on the Prevotella species, especially Prevotella copri. In a study published in 2013, P. copri was abundantly detected in fecal samples of new-onset untreated RA (NORA) patients. The abundance of P. copri strongly correlated with disease severity, and in NORA subjects was associated with a decrease in Bacteroides and a deficit of allegedly beneficial microbes.¹²⁰ Until now, several reports have confirmed this ground-breaking finding strongly suggesting P. copri as a causative pathobiont in RA. Colonization of germ-free mice with C. copri alone induced arthritis in a Th17 cell-dependent manner that closely resembled clinical RA in human.¹²¹ The derived peptides form a 27 kDa protein of P. copri, which was identified in complex with HLA-DR in the synovial environment and stimulated TH1 responses in the NORA patients.^122,123 Finally, it was shown that P. copri is enriched in the gut microbiota before the onset of the clinical RA in the NORA cohort. ¹²⁴ Recently, a metagenome-wide association study (MWAS) revealed high abundance of other species belonging to the Prevotella genus, including P. denticola, P. marshii, P. disiens, P. corporis, and P. amnii, in RA patients but not in healthy controls.¹²⁵ Interestingly, P. gingivalis was also found in the feces of RA patients but at relatively low abundance. Nevertheless, P. gingivalis abundance was positively correlated with Prevotella spp. ¹²⁶, suggesting cooperation between P. gingivalis and Prevotella spp in the gut microbiome of RA patients. This finding adds an interesting twist to the interpretation of the mechanism of P. gingivalis-induced RA in mice models. Instead of exerting a direct effect on immunotolerance breakdown, P. gingivalis may initiate a chain of events that lead to the generation of ACPA and finally clinical RA, via changes in gut microflora that alter the gut immune system. Oral gavage of P. gingivalis impairs the function of the gut barrier and provokes dysbiosis in the gut microbiome through induced metabolic changes and endotoxemia.^127,128 Immunologically, this is manifested by increased IL-17 levels in the sera and conditioned media of cultured lymphocyte fractions obtained from the spleen, Peyer’s patches, mesenteric lymph nodes, and inguinal lymph nodes of mice orally administrated with P. gingivalis. Furthermore, P. gingivalis gavage caused a significant increase in the proportion of Th17 cells among mesenteric lymphocytes and impacted the composition of the gut microbiome. A similar response might happen in humans if P. gingivalis is swallowed and colonizes the gut. Acting as the keystone pathobiont, it may affect the gut microbiota composition and immune system despite its very low relative abundance, thus indirectly contributing to development of RA.

7.1. The direct effect of periodontal pathogens on immunotolerance breakdown

7.1.1. Enzymatic mimicry: P. gingivalis-derived PAD

Citrullination of key RA human antigens is catalyzed by a unique PAD enzyme: P. gingivalis-derived peptidylarginine deiminase (PPAD). Since the first report describing PPAD among proteins expressed by P. gingivalis ¹²⁹, its enzyme structure and function have been extensively studied.^130–133 PPAD catalyzes the same reaction as mammalian PADs; however, its activity is calcium independent, strongly favors C-terminal arginine, and also citrullinates free arginine. In protein citrullination, PPAD synergizes with Arg-specific gingipains (Rgps), which cleave a broad array of bacterial and host proteins to liberate C-terminal Arg residues. This cooperation is facilitated by the colocalization of PPAD and Rgps in the outer membrane. Both enzymes working together have been shown to liberate 37 and 11 C-terminally citrullinated peptides from the two best characterized autoantigens in RA, fibrinogen and α-enolase, respectively.¹³⁰ These results contradict the earlier finding by Abdullah et al. who showed that gingipains have no influence on citrullination of yeast enolase, human vimentin, or fibrin by PPAD.¹³⁴ Another controversial report published recently showed that PPAD efficiently citrullinated internal arginines at the RG or RGG consensus motif in major RA autoantigens, such as fibrinogen, vimentin, hnRnP-a2/B1, and histone H1. In these experiments, recombinant PPAD with the C-terminal domain (CTD) was used.¹³⁵ By contrast, the mature P. gingivalis PPAD has no CTD, which functions as a secretory signal and is cleaved-off during translocation across the outer membrane.¹²⁹ Apparently, the presence of the CTD in the recombinant PPAD facilitates autocitrullination of the E. coli expressed enzyme and its activity on internal Arg residues in proteins and peptides. In contrast, regarding the activity of P. gingivalis, fully processed PPAD is practically limited to C-terminal Arg residues and no citrullination of the native enzyme or internal Arg residues are observed.¹³⁰ Regardless of the controversy, RA patient sera screened for PPAD-citrullinated epitopes uncovered 16 RA autoantigens and nine autoantigens associated with lung diseases.¹³⁵ As the anti-RA-PPAD level correlated with the ACPA level and interstitial lung disease autoantigens, the authors proposed that treatment of P. gingivalis could be used as a standard routine to prevent interstitial lung disease at onset of RA.¹³⁵

The risk of tolerance breakdown is dependent on the citrullination burden at infected periodontal sites. To this end, P. gingivalis effectively citrullinates its own proteins.¹³⁶ The proteomic analysis of P. gingivalis citrullinome clearly indicated that secreted proteins, either released into the medium or associated with bacterial cell surface, constitute the majority of PPAD modified proteins.¹³⁷ Most of the modifications were detected at C-terminal Arg residues in accord with PPAD specificity and cooperation with Arg-gingipains. Of note, these P. gingivalis-derived citrullinated peptides are not recognized by ACPA in eRA.¹³⁸ This corroborates an earlier report showing that modified peptides derived from auto-citrullinated PPAD are not a target for ACPA.¹³⁹ The conclusion from both of these studies contradicts the findings by other groups ^135,140,141, which show that citrullinated P. gingivalis proteins and peptides react with ACPA. Taken together, it is clear that the role of the P. gingivalis citrullinome in driving the pathology of inflammatory arthritis is still uncertain and requires more investigation.

The importance of PPAD in the process of altered host epitope formation and promotion of autoimmune reactions is supported by the fact that the enzyme is heat stable and exhibits optimal activity under alkaline conditions similar to those present in the inflammatory environment.¹²⁹ Although the expression of human PAD enzymes is not affected by PPAD ¹⁴², the presence of both enzymes in the inflammatory milieu potentiates the chances of citrullination. This was verified recently by detecting PPAD in synovial tissue ¹³⁵ and an elevated level of antibodies to PPAD in RA sera.¹⁴⁰

The results of translational studies have shown that PPAD expression has a profound impact on the development and progression of RA. Using a murine model of collagen induced arthritis, the authors clearly showed that disease severity was dependent on the expression of PPAD.⁶⁸ This observation was confirmed using a collagen antibody-induced arthritis model.¹⁴³ Importantly, increases in the levels of autoantibodies to collagen type II and citrullinated proteins were observed only when patients were infected with bacteria expressing PPAD.⁶⁸ Collectively, the data strongly support an infection-based concept of induction of ACPAs via enzymatic mimicry, suggesting that PPAD might break immune tolerance in RA.

7.1.2. Molecular mimicry: Similarity of epitopes

Antigen similarity/mimicry due to structural similarities between P. gingivalis antigens and self-antigens has been hypothesized to explain the role of periodontitis in RA development. Among the specific antigens identified in RA patients is citrullinated enolase.¹⁴⁴ Lundberg et al. proposed that this protein plays a central role in the initiation of the PD-induced pathogenic pathway that leads to RA development.¹⁴⁵ The above hypothesis comes from studies showing that citrullinated α-enolase peptide 1, which is detected in the majority of RA patients (up to 62%), shows 82% sequence homology with P. gingivalis-derived enolase. Moreover, it was demonstrated that the antibody levels to citrullinated α-enolase peptide 1 correlated with those to the bacterial peptide, and that they cross-react with citrullinated recombinant P. gingivalis enolase.¹⁴⁶ Translational studies using DR4-IE–transgenic mice supported the role of the bacterial enolase. Kinloch and coworkers showed that bacterial enolase induces arthritis as efficiently as the human enolase, and induced the humoral response by producing antibodies to both citrullinated and unmodified human-enolase.¹⁴⁷ Based on these results, the unmodified form of P. gingivalis α-enolase may be important in initiating the corresponding subset of human RA. Further, these data provide a strong basis for the causative association between RA and PD, based on the molecular mimicry hypothesis.

Heat shock protein 60 (Hsp60) in considered to be another bacterial-derived molecule that is highly cross-reactive with human antibodies. The analysis of Hsp60 sequences from different pathogens, including Chlamydia pneumoniae and Mycobacterium tuberculosis, revealed that only the P. gingivalis epitope identified within this chaperone sequence is predominantly and frequently recognized by antibodies in the serum of RA patients.¹⁴⁸ The Hsp60 sequence is highly conserved; therefore, it cannot be excluded that other types of bacteria could contribute to molecular mimicry and the breakdown of immune tolerance in inflamed gingival mucosae.

Antigenic determinants of type II collagen play a role in collagen-induced arthritis. ¹⁴⁹ Recently, Peng et al. identified an amino acid sequence similar to the RgpA catalytic domain and rat type II collagen, suggesting a potentially immunogenic role of gingipain. The authors demonstrated that pre-immunization of rats with purified recombinant RgpA triggered a potent protective immune response that manifested as an increase in the level of type II collagen-specific antibodies that in-turn alleviated arthritis in the joints of the animals.¹⁵⁰ This evidence suggests that translocation of P. gingivalis expressing RgpA in the synovium may exacerbate the inflammatory response that promotes RA. Therefore, it may be beneficial to pre-immunize or apply small-molecule inhibitors of gingipain (Figure 1).

Figure 1. Scheme summarizing molecular pathways of RA development in the reflection of PD.

(1) Periodontal pathogens induce inflammation of the periodontium by expressing a set of virulence factors, including: egzo- and endotoxins, fimbriae or proteolytic enzymes. The process leads to the destruction of ligament cells, activation of leukocytes, activation of enzymes and releasing of soluble mediators, such as: cytokines, components of complement system or alarmins. Therefore, it is proposed that PD can promote RA developed by the mechanism of bystander activation; (2) Enhanced local and systemic inflammation in PD leads to the acceleration of PTMs. Among them are citrullination, carbamylation and malondialdehyde-acetaldehyde (MAA) adduct formation. Antibodies directed against modified proteins are found in RA patients, suggesting that the causative mechanism linking PA with RA is based on the formation of neoepitopes; (3) Periodontal pathogens express molecules including: enolase, Hsp70, RgpA and PPAD which structure and/or function resemble that of host proteins. This suggests that the mechanism of the molecular mimicry is underlying a causative association of PD with RA; (4) P. gingivalis influences the composition of the gut and lung microbiome. The alteration of immune system favors the development of RA.

8. Summary

Consistent with the hypothesis for the mucosal origins of RA development, it is clear that periodontal pathogens should be considered as important environmental players contributing to immune abnormality in RA patients. The detailed analysis of RA etiology in this review suggests that dysbiosis, responsible for chronic inflammation of the periodontium or in the gut, could trigger autoimmunity via several mechanisms, including bystander activation, amplification of autoimmunity by cytokines, epitope spreading, autoantigen overproduction, microbial translocation, and molecular mimicry. Considering that development of microbial dysbiosis in the gut and the oral cavity is preventable suggests that appropriate diet, the use of probiotics and strict oral hygiene should at least slow down the development of RA and lessen the disease severity.