• inflammation
• autoimmunity
• systemic lupus erythematosus

Millions of micro-organisms populate our orifices from the moment we are born until we die. The microbiome represents the totality of micro-organisms that coexist with us rather peacefully and which we acquire purposefully from our environment. Health is associated with a balance between commensal and pathogenic microbes, which control the intestinal barrier integrity through the production of mucus and lipid metabolites, such as short-chain fatty acids. In the case of a disbalance (dysbiosis), leakage of these microbial products negatively affects the immune system, thereby contributing to a chronic proinflammatory state or worse, cancer. Accumulating evidence implicates the microbiota to a plethora of diseases among which rheumatological musculoskeletal disease (RMD).

For instance, the idea that the microbiota is associated with the development of rheumatoid arthritis (RA) has been put forward by several investigators, many of those who have sought the link to intestinal microbes.1–3 On a different take, the oral microbiota, more specifically the presence of Porphyromonas gingivalis, in patients with RA with accompanying periodontitis has been linked to the presence of anticitrullinated protein antibodies.4 In addition, dendritic cells from patients with RA were reported to have a blunted response towards Porphyromonas gingivalis further supporting the notion of an altered systemic immune response to microbiota in RA.5 More recently, the proposed relation between the microbiota and RMD has been extended to systemic sclerosis (SSc) and primary Sjogrens syndrome (pSS).6 7

Hitherto, all these studies associate gut and/or oral dysbiosis to pathology—most likely due to the altered production of metabolites associated by the altered microbiota. However, this puts us with a conundrum, as such postulate does not explain why patients with RA, SSc or pSS present with symmetrical arthritis, organ fibrosis and exocrine gland dysfunction, respectively.

The paradox mentioned above is, however, challenged by a recent study by Manfredo Vieira et al published in Science.8 In this study, the authors hypothesised that the loss of the gut barrier function leads to translocation of pathobionts to organs, thereby leading to autoimmunity. The authors focused on systemic lupus erythematosus, given the relationship between disease pathology, excessive signalling of RNA-sensing Toll-like receptor (TLR7) and the presence of a type I IFN signature in the majority of patients. Lupus prone (NZW × BXSB)F1 mice, at 16 weeks of age, showed marked bacterial growth in the mesenteric veins, mesenteric lymph nodes and liver, followed by the presence in the spleen as well 2 weeks later. Subsequent 16S sequencing revealed translocation of Enterococcus gallinarum to the livers of these mice. Mortality, lupus-related autoantibodies and autoimmune manifestations were relieved in these mice after oral administration or vancomycin or ampicillin. The authors provide multiple molecular mechanisms on how E. gallinarum induces pathology in these mice, for instance via upregulation of Enpp3, a protein known to increase plasmacytoid dendritic cells considered key cells in lupus via the production of type I IFNs9 10 as well as activation of the aryl hydrocarbon receptor, a known innate antimicrobial defence mechanism potentially via the induction of IL-17 producing T cells.11 Importantly, longitudinal stool samples from patients with systemic lupus erythematosus (SLE) showed increased levels of albumin and calprotectin, reflective of impaired gut barriers function as seen in the lupus prone mice. Moreover, E. gallinarum was readily detected in the livers of patients with SLE and patients with autoimmune hepatitis sharing autoimmune features with the former, and most of these patients showed increased serum levels of antibody titres directed against E. gallinarum. In line with these observations, the same group published a concurrent study in Science Translational Medicine showing that commensal orthologs of the autoantigen Ro60 trigger autoimmunity as observed in SLE.12

Taken together, these studies provide strong rational for a change in the existing paradox where intestinal/oral microbiota elicit distal immunological effects by metabolites such as short chain fatty acids. In contrast, these studies indicate that certain microbes might migrate to organs and tissues thereby potentially leading to immune reactions and possibly tissue-specific breach of tolerance or, in other words tissue-specific autoimmunity. In fact, this is in corroboration with earlier studies highlighting the presence of bacterial DNA in RA synovial tissues suggestive for early migration of microbes to the synovial compartment perhaps far before actual disease onset.13 14 There are, however, potential caveats to consider. For instance, the field of human development has witnessed a fundamental paradigm shift in the ‘sterile womb paradigm’ when several studies applied molecular techniques suggesting bacterial communities in the placenta and amniotic fluid.15 16 Today, several studies report evidence for this latter as extremely weak due to issues with regard to contamination of materials with bacteria, use of inferior molecular techniques and lack of appropriate controls (reviewed in Ref.17). Moreover, although the experimental models in the papers discussed provide strong evidence for pathobiont translocation, the attempts to find support for this in men is merely suggestive as it does not deal with the big question to whether pathobiont translocation is cause or consequence. Also, the paradox proposed by the work of Manfredo Vieira et al focused primarily on the liver and does not discuss how E. gallinarum would get to the kidney and/or skin, more usual locations or lupus involvement nor does it deal with the fact that in case of a general disturbance of barrier function species other than E. gallinarum would be involved. Last, several studies now provide evidence for the efficacy of faecal microbiota transplant (FMT) in multiple conditions. The FOCUS trial has reported a strong effect in patients with ulcerative colitis (n=81), 27% of the treated patients who received steroid-free remission compared with 8% of the patients treated with placebo.18 It is thought that this is effect is mainly sorted through changes in the gut microbiome leading to changed local immunology and barrier function as well as a change in circulating immune-metabolites. It is harder to understand that the translocation of pathobionts that have led to autoimmunity is affected by FMT.

Taking these potential caveats into account, it is still tempting to speculate that migration of microbiota to organs and tissues is disease-specific and dependent on host factors, including genetic susceptibility, environment and smoking, thereby contributing to disease heterogeneity (figure 1). Such considerations are important for the rheumatologist, as many argue that shaping of the microbiome in disease is worth pursuing. Maybe one has to start in the diseased tissue rather than the gut—it is just a gut feeling.

• Download figure
• Open in new tab
• Download powerpoint

Figure 1

Microbes populate the intestine, oral cavity and skin. Based on the observations by Manfredo Vieira et al 8 it is tempting to speculate that certain events that disturb epithelial barrier defence (eg, infection, trauma) could lead to pathobiont translocation to specific sites and/or organs. Several host factors might favour the translocation of specific microbiota to certain organs thereby eliciting organ/tissue-specific immune reactions and/or autoimmunity.