Rheumatoid arthritis (RA) is a chronic, inflammatory, autoimmune disease that affects mainly the diarthrodial joints and is the cause of progressive and incapacitating joint destruction. Systemic manifestations are also present in individuals suffering from the disease, which can be classified as extraarticular manifestations. It is characterized by the presence of inflammatory immune cells in the joints. RA evolves toward a state of synovial hyperplasia, which is the cause of loss of bone structure and functional impairment. Several studies in North America and northern Europe show a prevalence of 0.5–1.1%, whereas in south of Europe, the reported prevalence is of 0.3–0.7%. Few studies have been performed in developing countries using the American College of Rheumatology 1987 criteria (1), showing a prevalence of 0.1–0.5%. However, it is difficult to establish whether this low prevalence is true or represents a referral bias due to the characteristics of healthcare in developing countries (2).

Rheumatoid arthritis affects multiple dimensions of the quality of life of not only patients but also caregivers, such as the physical, emotional, affective and the economy of the family and the health care institutions, all of this accompanied of productivity loss (3). It has been estimated in Mexico that the financial burden of RA on the patient’s families is of 610 ± 302.2 USD/year, whereas the direct costs of RA for the health institutions is of 1,724.00 USD/year/patient generating a great burden in the financial viability of public health care institutions (4). In addition, there is also an increased frequency in the use of biologics such as etanercept, infliximab, adalimumab, and rituximab, increasing significantly the cost of each patient’s treatment.

Several risk factors for RA have been identified, such as allelic variants of HLA-DR 01, 04, and others of the shared epitope (5). Also, several genes like the PAD4 or CTLA4 have been identified as independent risk factors for the disease (6). Modifiable risk factors such as smoking habits have been associated with severity of RA and with the presence of autoantibodies (7). Other diseases such as periodontal disease and gingivitis associated with several bacterial species such as PAD-like-producing Porphyromonas gingivalis have been speculated to increase the availability of citrullinated antigens and therefore the probability of autoantibody generations against such antigens (8). Differences in age for the onset of RA and its symptoms have been proposed to be not only due to the exposure to UV radiation (9) but also due to pesticides, in both the domestic and work environments altering some mechanisms of the oxidative stress response and contributing to tissue damage and RA initiation (10).

The etiology of RA is complex; however, several components of the immune response have been linked to RA physiopathology (11). It has been described the important role of macrophages, dendritic cells, and neutrophils in joint inflammation and synovial destruction in the initial phases of the disease, as the result of type I and II interferons (IFNs) activating myeloid cells and helping to sustain the inflammatory process through the production of cytokines such as IL-6 and TNF-α (12). Synovial fibroblasts also contribute to this process due to increased responsiveness to inflammatory stimuli and reduced apoptosis, perpetuating it (13). Also, this inflammatory milieu and the increased availability of citrullinated and carbamylated antigens lead to the production of autoantibodies (14). Neutrophils contribute to the aforementioned process through the induction of RANK/RANKL activation of osteoclasts, and this has been associated with an increased severity of RA (15). Khandpur et al. have documented that neutrophils of RA patients are more prone to neutrophil extracellular trap formation, and given that these neutrophil-derived structures are rich in citrullinated antigens, they propose that these mechanisms are the ones responsible for autoantibody generation (16).

T cells, both CD4+ and CD8+ are also important actors in the inflammatory response in the synovia, the latter recently linked to increased citrullination through pore-forming mechanisms (17). In a similar fashion, the function of CD4+ Th2 lymphocytes has been suggested to be linked to autoantibody generation and to response polarization through interaction of CD40 and CD40L (18), as well as that mediated by OX-40 and OX40L (19) appear to have a preponderant role in autoantibody production, and these antibodies have an important role in the pathogenesis of disease (20). In the ACR/EULAR2010 criteria, anticitrullinated protein antibodies (ACPA) are considered an important tool to define the disease (21). The increase in ACPA levels is known to precede the onset of symptoms in first-degree relatives of RA patients, and there is also an increased risk of developing RA for the ACPA+ individuals (22). Recently, the presence of anticarbamylated protein (anti-CarP) has also been associated with RA severity and RA in undifferentiated arthralgia (23). This suggests that both ACPA and anti-CarP antibodies might have an important participation in the initiation of symptoms, in the early stages of the disease, and also as markers of tolerance loss (23, 24). Moreover, it has been suggested that in the early stages of RA, type I IFNs might have an important role in the phenomenon of loss of tolerance, and in the chronic or established phase of RA, these molecules contribute to the perpetuation of the inflammatory response (25). Several of the type I IFN genes have been linked to antiviral and immunomodulatory activities (26). Type I IFNs induce dendritic cell maturation and the expression of both co-stimulatory and MHC I/II molecules as well as an increase in cytokine expression (27). Type I IFNs are produced mainly by plasmacytoid dendritic cells in response to a plethora of stimuli such as viral DNA and RNA or due to immune complexes of DNA/RNA antibody (28). IFN-α promotes dendritic cell differentiation and Th0 cell polarization to Th1 and also the cytotoxic response of CD8+ lymphocytes and NKT cells. It has also been described to induce B cell differentiation and antibody production and class switch to IgG (26). In RA, it has been described that immune complexes are produced due to an increased susceptibility to netosis of neutrophils (16), and also autoantibodies could be mediating this process acting as endogenous activators of IFN-α production (29). Although such mechanisms were first described for systemic lupus erythematosus (SLE) (30, 31), recent evidence suggest their participation in other autoimmune diseases such as RA (32). The genes activated through the described mechanism are known as IFN signature and comprise genes such as MXA and MXB associated with antiviral response (33, 34); IFIT1 and IFIT2 linked to posttranscriptional regulation in viral infections (35); ISG15 and 55 as part of the proteasomal degradation machinery (36); LY6E and RSAD2 are involved in metabolism and proliferation regulation of B cells (37); also, EPSTRI1, IFI44L, and IFI35 (38) are thought to participate in IFN signature response modulation. IFI6 has been described to participate in apoptosis inhibition mediated by TRAIL (39).

In other autoimmune diseases such as SLE, the immune complexes have been described to be internalized by pDC through the Fc-γRIIA into endosomes and interact with TLR7 and TLR9. After this activation, MyD88 induces activation and phosphorylation and translocation to the nucleus of the IFN-regulating factor 7 nuclear factor, inducing the transcription of the type I IFN genes (40).

Van der Pouw Kraan and colleagues identified several changes in gene expression in the IFN signature in established RA compared to control individuals (25), but the early stages of RA [early RA (eRA)] or the preclinical autoimmune phase of the disease was not the focus of that publication, and there are no available data regarding the role of the induced genes in these preclinical or early stages of the disease. Because of the direct association of several genes of the IFN signature with the physiopathology of RA, the natural history of the disease, and the expansion, regulation, and function of B cells, it is important to understand the relationship of the IFN signature activation and gene expression with the levels of ACPA and anti-CarP. The aims of this article were to analyze whether type I IFN-regulated genes were underexpressed or overexpressed in several groups of individuals at high risk of developing RA and in people with early or established disease and also to evaluate their relationship with autoantibodies in serum.

The analysis of gene expression of the type I IFN pathway has been associated with increased inflammatory responses in other autoimmune disease such as SLE. There is limited information regarding the role of the transcriptional pathway and the relationship that this might have with the inflammatory response in RA in the early stages of the disease and in the high-risk ACPA+ individuals. Therefore, we evaluated the gene expression of several genes associated with the IFN signature in the Mexican population. Given that several genes regulated by type I IFN can be grouped into several functions, such differences are described below. No differences in gender proportions were identified, and all patients belonged to a Mexican Mestizo population, which was not characterized for the presence of HLA-risk Alleles. All patients were treated (in case of RA patients) with DMARD, and methotrexate was used most frequently.

Rheumatoid arthritis is an inflammatory disease that is characterized by the generation of autoreactive clones of T and B cells and, in consequence, by the generation of autoantibodies (45). It has been proposed that autoantibodies could be involved in the generation of IFN through the activation of the pDC, and it could be implicated in the perpetuation of the inflammatory process as has been previously described for LES (46, 47). The first report that linked the IFN signature with autoimmunity in RA came from Van der Paw; he found an elevation in the expression pattern of several of the genes of the pathway in the established disease (25). However, little has been done to try to replicate such data in other populations and settings. Here, we report the analysis of several genes of the type I IFN signature and its relationship with high-risk individuals (ACPA+), eRA, and established RA and the participation of such genes in the physiopathology of the disease in different stages and their possible use as biomarkers in early stages of disease. Also, we discuss the association between the presence of anti-CarP and ACPA and the potential role of the type I IFN signature genes in the generation of such antibodies.

It has been reported that some chronic viral infections might be the trigger or at least are associated with, of autoimmunity, in particular due to cross-reactivity, as has been observed in several tropical viral infections such as Chikungunya (48), Epstein-Barr (49), and Zika (50), which have been associated with acute arthritis of unknown origin and have been poorly characterized (51, 52). In this regard, several genes of the IFN signature such as MXA and MXB have been associated with protection against viral infection (53, 54). Our data clearly show an increased expression of the MXA and MXB genes providing evidence of the role that these molecules might have in the induction of inflammation in the early phases of RA or even previous to the onset of symptoms in individuals at high risk, with the possible involvement of such molecules even in the preclinical autoimmune phase of RA, explaining just in part the lack of differences in the established phase of the disease. The risk of developing RA associated with increased expression of these genes needs to be further evaluated in the light of the induction of tolerance processes and determine whether MXA and MXB gene expression increase is due to viral infection and/or the generation of interferogenic complexes.

Recently, a strong association between RA and the unfolded protein response (UPR) was described (55–57), and therefore, the regulation of the transcription/translation process has been put on the spotlight of RA pathogenesis. It has been described that several regulators of the translation of proteins particularly GADD34 are associated with the UPR in RA (55); however, little attention has been put to the transcriptional process and their regulators such as IFIT1 and IFIT2 that are known to regulate the apoptotic process (58). In this regard, we found an increased expression of these genes in the established disease phase. Suggesting that the role of the transcriptional regulation in the chronic phase of RA might be associated with the activation of the immune response by type I IFN; also, given that due to medication several viral infections become recurrent in these patients (59), upregulation of IFIT1 and IFIT2 could be associated with these infections in RA. Also the role of other regulators in the expression of these genes such as miRNA needs to be further evaluated, and also the viral subclinical infection status of such individuals (Epstein-Barr virus and cytomegalovirus infection status) needs to be determined.

The mechanisms of ubiquitination are very important for the clearance of several proteins within the cell and therefore are related to the UPR (60). ISG15 is a small ubiquitin molecule associated with antiviral response in a process called “ISGlation” (61). It has been described that this molecule can regulate the activity of negative regulator IRF3 and in this way promotes the expression of genes of the IFN signature (62). Further research on the role of HERC5 and ISG15 in RA pathogenesis is needed to establish their participation in experimental models of RA and to evaluate how is regulated in the early phases of disease or if the phenomenon is exclusively associated with chronic inflammatory processes as observed in the established disease patients. Also, the potential use of these molecules as markers of treatment success has been described previously by Raterman and colleagues (63).

The regulation of apoptosis may also play an important role in the pathogenesis of RA (13). Little is known about the function of IFI6, and it has been described that IFI6 inhibits apoptosis trough the downregulation of caspase 3 and casase 9 (64). This effect might be associated with increased survival of inflammatory cells that perpetuate the activation of antibody-producing B cells, and this needs to be further evaluated. Also, Ly6E has been associated with an increased proliferation in several cell lineages (65) and could mediate the process of B cell expansion in inflammatory disorders through a TGF-beta-mediated mechanism, but this needs to be further explored. RSAD2 (also known as Viperin) is an antiviral protein associated with regulation of proliferation of immune cells, and it has been associated with Th2 polarization of immune cells through a mechanism mediated by GATA3, NF-κB, and IL-4 production, therefore inducing and modulating the production of autoantibodies (44). The increased expression of RSAD2 in ACPA+ individuals and cRA patients might contribute to autoantibody generation. The increased expression of this pathway needs to be further elucidated given that expression of these molecules is mainly regulated by viral infection (66–68).

There is no clear evidence if the perpetuation of the inflammatory process through the activation of inflammatory cells might be associated with increased levels of antibodies in several stages of the disease (69, 70). For this purpose, we analyzed the expression of several genes of the type IFN gene signature in groups of individuals who had no previous family history of the disease, HC, and also in individuals with ACPA−, ACPA+, eRA, and cRA, which accounts for the natural history of the disease. We found a significant correlation between the expression of these genes associated with proliferation control of B cell maturation and antibody production such as IFI6, IFI35, ISG15, EPSTRI1, MXA, and Ly6E, several of which are associated with promotion of B cell differentiation and antibody production (71). Our results differ from those of other groups in which no association has been found for autoantibody production of ACPA and other autoantibodies (RF, antinucleosome, etc.) in adalimumab-treated patients (72), which has been recently confirmed to modify the gene expression pattern of the type I IFN signature in neutrophils of patients with RA (73). Also the authors did not analyze the correlation effect of the IFN signature gene expression on the antibodies that were analyzed, neither did they analyze the anti-CarP antibodies. Another substantial difference that may account for the differences in our findings and that of Cantaert et al is that we included individuals (eRA) who were virgin to treatment (72). Our groups of individuals were recruited in an attempt to emulate the natural history of disease in a patient with RA, and therefore, the tendency is clear to increase the levels of autoantibodies in the late phases of disease. Ethnic and population differences may also account for such differences (74, 75).

Although our correlation analysis cannot be confirmed as causality, it clearly highlights the association of the increased expression of Type I IFN response genes with the master regulators of B cell development, and differentiation into plasma cells might be of pivotal importance in the generation of autoantibodies in RA, which has been already described in SLE (76) and some myopathies (77). Further mechanistic studies are needed to confirm such association and describe the detailed mechanism.

This article provides evidence of the limited use of the gene expression of these markers to differentiate the early stages of RA, although MXA, MXB, and IFI6 showed an increased expression when compared to HC. The participation of several genes of the IFN signature is also described in the context of the available data on the pathogenesis of the disease. We describe a plausible link between type I IFN-induced gene expression and the generation of anti-CarP antibodies and ACPA. Limitations of the study include sample size and previous DMARD use in the cRA group.

JC-D, YB-H, LE-M, DO-C, DS-A, and AA-N: manuscript, experiment, analysis, and idea conception; CR-R, JC, and PM-T: supervision and sample collection; JE-M: overall supervision, analysis, manuscript approval, and conception of idea.

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.