Before the clinical onset of rheumatoid arthritis (RA), a preclinical period exists during which genetic and environmental factors interact to initiate the autoimmune process. The ensuing autoimmune response is characterized by the production of rheumatoid factor (RF) and/or anti-citrullinated protein antibodies (ACPA). According to EULAR Standing Committee, this preclinical phase could be divided in three ≪at risk≫ stages: genetic and environmental risk, including first-degree relatives (FDR-RA) of patients with RA, systemic autoimmunity associated with RA and symptomatic preclinical phases [1].

This review focuses on FDR-RA, as defined by the EULAR terminology, who are good candidates for clinical and biomarker profiling, providing insight in the etiology of RA [2]. Our aim is to discuss the available evidence on mechanisms linking periodontitis and RA onset in this population.

Periodontitis is included in the “two-hit” model for RA etiology, introduced by Golub et al. [32]. The first hit represents ACPA production due to chronic periodontitis followed by a second hit in the joint that induces RA. In other words, abnormal and bacterial citrullination by P. gingivalis within the periodontal tissue results first in a local autoimmune response to citrullinated proteins followed by the systemic production of ACPA in the joints that can induce RA [32].

ACPAs are the most specific antibodies associated with RA [33]. They are found in approximately 80% of the RA patients [34]. Their presence is highly associated with the HLA-shared epitope (SE), which is linked to the risk of developing RA and in particular for ACPA-positive RA. Interestingly, ACPA can appear years before the onset of RA, thus being a strong predictor of the disease [35].

Recent research has focused on the identification of external factors that could trigger such autoantibody production. The autoantibodies are produced following the excess formation of citrullinated proteins [36]. Citrullination refers to the post-translational process of the modification of the amino acid arginine into citrulline. The process is mediated by the peptidyle arginine deimase enzyme (PAD) of various immune cells such as the neutrophils, macrophages, monocytes and T and B lymphocytes. Five PAD enzymes have been identified in humans [37]; two isoforms of the PAD family, the PAD2 and PAD4 are expressed in inflamed periodontal tissues [38]. In addition to the human PADs, the periodontal pathogen P. gingivalis, has been shown to express a PAD enzyme (referred to as PPAD to distinguish from the human PAD) capable of citrullinating host and bacterial peptides. In particular, citrullinated fibrinogen and citrullinated alpha-enolase are targeted by anti-citrullinated protein antibodies. These autoantibodies are found, respectively, in up to 60% and 40-60% of patients with RA [39].Thus, citrullination associated with the host-derived PAD is further increased by the bacterial-derived PADs, leading to an enhanced production of ACPA [40].

In addition to its ability to express PPAD, P. gingivalis can induce the production of various pro-inflammatory cytokines by the immune cells, stimulate a Th17 response and accelerate the development of RA. It has been established that Th17 cell-related cytokines are strong inducers of arthritis and that IL-17 plays important role in the osteoclast differentiation and bone erosions [41, 42].

A. actinomycetemcomitans, another important periodontal pathogen, has been also proposed as a potential trigger for the pathogenesis of RA. The bacteria possess a virulence factor, a pore-forming leukotoxin A (LtxA) which can dysregulate the activation of citrullinated enzymes and induce hypercitrullination in host neutrophils [43].

Aside from formation of ACPA due to citrullination, other molecular pathways have been also studied to link PD with RA. First, uncontrolled generation of neutrophil extracellular traps (NETs) has been found in several autoimmune diseases in response to periodontal pathogens. Accumulated NETs provide a source of autoantigens in both PD and RA [44]. A second mechanism, described as molecular mimicry, involves the capacity of P. gingivalis and some other bacteria in dental plaque to express antigens, that are structurally similar to host antigens, and can therefore cross-react with ACPAs. Bacterial enolase and bacterial heat shock protein 60 are the strongest and mostly studied candidates that trigger an immune response and generate antibodies [45]. Other potential mechanisms linking PD to RA focused on the capacity of P. gingivalis, as shown in an experimentally-induced periodontitis animal model, to modulate the gut microbiota composition [46] and to be hematogenously disseminated to synovial joints [47].

Genetic and environmental risk factors are common between PD and RA resulting in the progressive destruction of bone and connective tissue [48]. The shared epitope (SE) coding HLA-DRB alleles are potential genetic elements connecting RA and PD [49]; they have been associated with bone erosions in RA and alveolar bone destruction and PD progression [50]. Moreover, family transmission of putative periodontal pathogens between family members has been documented [51].

Finally, tobacco consumption is an established risk factor for periodontal destruction and RA. Case-control studies have shown that in smokers, the risk of developing seropositive RA was twice higher compared to non-smokers and the risk was dose-dependent on lifetime exposure to smoking [52]. Likewise, smoking affects in a dose dependent way all aspects of periodontal health such as prevalence of PD, severity of periodontal destruction and unfavorable results to periodontal treatment [53]. Exogenic risk factors such as nutrition, socioeconomic status, psychological factors (stress) and obesity are common between the two pathologies [44].

As mentioned above, periodontopathic microorganisms may trigger, deteriorate and perpetuate RA [54]. Citrullinated proteins have been detected in periodontal tissues and in inflammatory exudates [38, 55]. Thus, PD could act as an environmental stressor for ACPA-positivity. The hypothesis is that in genetically susceptible individuals, citrullination associated with periodontitis may cause a localized oral mucosal response, which can lead to a systemic ACPA production and the onset of RA.

The clinical evidence for the relation of PD and RA is large and variant. The Nagahama study included 9,575 subjects with no connective tissue disease and showed significant associations between periodontal parameters and ACPA seropositivity. In this population 27.9% were ACPA-positive, supporting the involvement of PD in ACPA production [56]. However, when serum levels were analyzed for ACPA quantification in subjects with or without RA, and with or without PD, no correlation was found between ACPA and the clinical parameters of PD [57].

Conflicting data and opinions have been reported regarding the relationship between periopathogenic bacteria and RA. In some studies, the subgingival presence of P. gingivalis and A. actinomycetemcomitans and the levels of serum anti- Porphyromonas gingivalis and anti- Aggregatibacter actinomycetemcomitans immunoglobulins were not associated with RA [58]. While other publications reported a weak but significant correlation between anti- Porphyromonas gingivalis outer membrane levels and ACPA titers [59]. The study of Schmickler et al. [60], revealed a higher number of Fusobacterium nucleatum and P. gingivalis in ACPA seropositive patients with RA. In cases of untreated new-onset RA, P. gingivalis was identified in 55% of the patients [60] whereas in the study of Bello-Gualtero et al. [61], P. gingivalis specific IgG was found to be associated with ACPA in early-RA, but not in pre-RA. This supports the hypothesis that P. gingivalis infection plays a role in the early loss of tolerance to potential self-antigens during the RA pathogenesis [61]. The study of Fisher et al. [62] agreed that P. gingivalis was not associated with pre-RA autoimmunity or risk of RA in an early phase before the disease onset.

All the above mechanisms linking PD and RA may potentially also apply to the susceptibility of FDR-RA of developing RA [2, 63]. However, only a limited number of studies evaluated the prevalence or indicators of periodontitis in FDR-RA [36, 64, 65]. These studies evaluated mainly the role of ACPA and oral microbiota, as main mechanisms.

RA is a considerable health problem that affects all areas in the life of the diseased patients. Early diagnosis and preventive measures may decrease the prevalence and severity of RA, and improve the quality of life of patients. Because of the association between RA and PD, especially in the early phases of the disease, PD should be assessed in the baseline assessment of patients at high risk. Health care practitioners should be aware of the association and active screening of at-risk individuals should be considered [73].

FDR-RA are at higher risk for developing RA, but only limited data exist for the role of PD. ACPA status seems significantly associated with deteriorated periodontal conditions and higher risk of RA. Factors, such as smoking, increased age and high BMI are associated with ACPA seropositivity and with deteriorated periodontal conditions, thus increasing the risk of developing RA.

P. gingivalis and has been identified as a possible trigger for RA via the breakdown of tolerance against citrullinated proteins and the formation of ACPAs. However, in the literature, contrasting evidence for its role in the etiopathogenesis of RA is found. A. actinomycetemcomitans is another potential candidate.

Larger studies evaluating all the potential mechanisms linking RA and periodontitis are needed. In particular, longitudinal studies evaluating the effect of PD on the risk of developing RA in FDR-RA are required to confirm the role of PD as a trigger for RA development. And last but not least, the effect of periodontal treatment on the development of RA needs to be established before preventive oral health interventions can be definitely established in at-risk populations for RA.

CG and AZ performed the literature research. All authors contributed to the article and approved the submitted version.

AF's work is supported by research grant from the Swiss National Science Foundation (no. 310030E_205559/1; no. 320030_192471/1; no. 3200B0_120639).

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.