Impaired immune tolerance mediated by reduced Tfr cells in rheumatoid arthritis linked to gut microbiota dysbiosis and altered metabolites

Rheumatoid arthritis (RA) is a highly disabling autoimmune disease characterized by persistent synovitis and joint damage [1, 2]. Excessive production of abnormal autoantibodies, including anti-cyclic peptide containing citrulline (anti-CCP), play a pivotal role in the onset and progression of RA, which can even be detected in the stage of pre-clinical RA (Pre-RA) [3, 4]. A comprehensive exploration of the intricate process of antibody generation is vital to achieving targeted therapy, early remission, and the prevention of RA. The complex etiopathogenesis of RA involves the loss of immune tolerance resulting from genetic and environmental factors, which has been confirmed to be a crucial contributor to the over-production of abnormal antibodies, and it always occurs in the Pre-RA stage to promote the further progression of RA [5, 6]. Therefore, inducing and restoring immune tolerance are promising in the prevention and treatment of RA.

The CD4 + CD25 + Forkhead Box 3 (FoxP3) + regulatory T (Treg) cells play a critical role in maintaining immune tolerance not only by producing the anti-inflammatory cytokine (such as interleukin-10) but also by suppressing the activation and proliferation of effector T cells, specifically helper T(Th)17 cells, which secret the pro-inflammatory cytokine, interleukin-17 [7]. It has been reported that RA patients had an imbalance of Th17/Treg cells, especially the reduced Treg cells [8, 9]. Exploiting the suppressive capacities of Treg cells to enhance immune tolerance has been an emerging field to treat autoimmune diseases such as the application of low-dose interleukin-2 (IL-2) in RA [10, 11]. However, the breakdown of immune tolerance mediated by the aberrant Treg cells seems to be difficult to explain the over-production of antibodies in RA.

The autoantibody production in RA depends on the response of lymphoid follicular germinal centers (GCs), which are essential for B cells to complete the series of reactions including affinity maturation, class switch recombination, and somatic hypermutation to produce a large number of high-affinity antibodies and memory B cells finally [12]. The discovery of a novel CD4 + T subset cells localized in lymphoid follicular GCs is termed follicular regulatory T (Tfr) cells, which is characterized by the expression of C-X-C chemokine receptor type 5 (CXCR5, a chemokine receptor homing to the T-cell zone) and Foxp3 [13‐15]. While the exact differentiation mechanism of Tfr cells is still poorly defined, existing evidence supports that they originate de novo from thymic-derived FoxP3 + Treg precursors under multiple stimulations. As a specific subpopulation of Treg cells, Tfr cells contribute to maintaining immune tolerance by inhibiting follicular helper T (Tfh) cells, another GC-residing cell type to facilitate the production of antibodies by promoting the formation and response of GC [16, 17]. The function of Tfr cells provides a new understanding of maintaining immune tolerance and antibody production, which may contribute to the further exploration of RA pathogenesis. It has been found that RA patients had imbalanced Tfr/Tfh cells [18, 19], especially those with decreased Tfr cells exhibited high disease activity and antibodies, suggesting that the aberrant Tfr cells could lead to the over-production of antibodies to destroy the immune tolerance. And those with active RA showed higher Tfh cells which was associated with the enhanced IL-6/pSTAT3 signaling [20]. Of note, the imbalance was altered after treatment and patients with RA in stable remission with lower levels of autoantibodies exhibited increased Tfr cells [21, 22], indicating that targeting Tfr cells to restore immune tolerance had significant therapeutic potential for RA. It has become a consensus that impaired immune tolerance and antibody production are the core pathogenesis of RA and the Pre-RA, and Tfr cells play an important role in inhibiting the production of antibodies and maintaining immune tolerance. Therefore, the impaired immune tolerance and the over-production of antibodies caused by aberrant Tfr cells play a crucial role in the pathogenesis of RA. Nevertheless, the potential upstream factors regulating Tfr cell-mediated immune tolerance remain to be fully elucidated, which is the aim of our study.

There is substantial evidence suggesting that the initiation of RA might originate in mucosal sites far away from joints, such as the gut, which emphasizes the effect of the gut-joint axis exerted in RA [23, 24]. The gut microbiota as the most critical component in the gut is important for balancing health and disease, and it is associated with autoimmune diseases including RA [25, 26]. Some significant studies have reported that gut microbiota dysbiosis in RA was correlated with disease activity [27‐29]. For instance, patients with RA exhibited elevated Prevotella copri, particularly those with high disease activity [27, 28], while the reduced Haemophilus spp. in untreated RA patients was associated with high-level abnormal autoantibodies negatively [29]. Gut microbiota plays an immunomodulatory role in maintaining immune homeostasis under normal conditions [30], while gut microbiota dysbiosis could activate both innate and adaptive immune cells, which may serve as the mechanistic connection between mucosal changes and arthritis development [31, 32]. Moreover, gut microbiota-derived metabolites including short-chain fatty acids (SCFAs), bile acids and tryptophan and its derivatives have been recognized as the crosstalk of the gut-joint axis to exert effects [33, 34]. Altered metabolite profiles have been observed in RA patients indicating their significant role in the pathological mechanism of RA [35, 36]. The main mechanism driving the onset of RA through the gut-joint axis revolves around the impact of gut microbiotas and their metabolites to activate pro-inflammatory immune cells and promote their trafficking from the gut to joints [37, 38]. Notably, the altered gut microbiotas and their metabolites may also play a role in the gut-joint axis by destroying immune tolerance. The present studies focus on whether and how gut microbiotas and their metabolites affect immune tolerance mediated by Treg and Tfr cells to exert critical components within the gut-joint axis. To date, some studies on collagen-induced arthritis (CIA) and SKG arthritis models have found that microbiota-derived butyrate might suppress autoantibody production and ameliorate arthritis by enhancing the Treg and Tfr cells [39, 40]. However, there are numerous gut microbiotas and their metabolites in RA, whether other gut microbiotas and metabolites are involved in the pathogenesis of RA, and how their relationship with Treg cells, especially Tfr cells, still needs further systematic research. Thus, it is necessary to find the disease biomarkers of RA from the numerous gut microbiotas and their metabolites and to analyze their relationship with Treg and Tfr cells, which is the focus of our study.

Therefore, we performed the study to explore the association of gut microbiotas and their metabolites with the immune tolerance mediated by Tfr cells in RA. First, we assessed the immune tolerance status in RA by detecting the expression of Th17, Treg, Tfr, and Tfh cells in the peripheral blood via modified flow cytometry. And then, considering that the detection of gut microbiota and metabolites in stool samples is an ideal method to study the direct correlation between gut microbiotas and their metabolites, we identified the characteristics of gut microbiotas and their metabolites in RA by the combination of 16S rDNA sequencing and ultra-performance liquid chromatography-tandem mass spectrometry (UPLC-MS)-based untargeted metabolomic profiling. Subsequently, we explored the association of gut microbiotas and their metabolites with the immune tolerance mediated by circulating Tfr cells. The results of our study aimed to provide a novel insight into the pathogenesis of RA from the perspective of gut microbiota-metabolite-immune tolerance.

Circulating Tfr cells play a role in maintaining immune tolerance, and the reduced Tfr cell may contribute to the breakdown of immune tolerance to participate in the progression of RA. However, the upstream mechanism of regulating Tfr cell-mediated immune tolerance remains unclear. Numerous studies have shown that gut microbiota dysbiosis and altered metabolites are closely related to the development of RA [35, 36], which may be caused by the interactions of gut microbiotas and their metabolites with the immune system. It still lacks systematic studies on the relationship between gut microbiota dysbiosis and altered metabolites with the Tfr cell-mediated immune tolerance in RA. Our study was the first to investigate the association of gut microbiotas and their metabolites with the immune tolerance mediated by Tfr cells in new-onset RA. The results of our study revealed the following characteristics of the new-onset patients with RA: (i) The reduced Treg and Tfr cells in RA were associated with the disease activity and the over-production of autoantibodies. (ii) Gut microbiota dysbiosis (especially at the genus levels) and altered gut microbiota-derived metabolites exhibited in new-onset RA patients were related to the disease. (iii) Ruminococcus 2 as well as arachidonic acid might be the potential biomarkers of RA. (iv) Gut microbiota interacted with their metabolites, and gut microbiota dysbiosis as well as the altered metabolites in RA were associated with the breakdown of immune tolerance mediated by reduced Tfr cells. The association between gut microbiotas and their metabolites with immune tolerance mediated by Tfr cells we pointed out may participate in RA, which provided a theoretical basis for further exploring the effect of specific gut microbiota and its metabolites on Tfr cells.

Early work suggested that the activated CD4 + CD25 + CD69 − Treg cells were able to gain the expression of CXCR5 and migrate to the B-cell follicle to suppress B-cell responses [41, 42], which revealed the relationship between Treg cells and another new type of cells in lymphoid follicular. And then, three independent groups defined them as Tfr cells (CXCR5 + PD-1 + BCL6 + FoxP3 + cells), which originated de novo from thymic-derived FoxP3 + Treg precursors requiring multiple stimulations [13‐15]. Increasing data highlighted the significance of Treg and Tfr cells in maintaining immune tolerance, especially Tfr cells, which exert an essential effect on regulating antibody production as the subsets of Treg cells via suppressing Tfh cells and B cells in GCs [16]. Considering the difficulties of getting organ tissues from humans for clinical investigations, the circulating Tfr and Tfh cells are always discussed in regulating immune tolerance, which may be derived from GCs and have similar phenotypes and function to GC-Tfh and GC-Tfh cells [43]. By analyzing the expression of circulating Th17, Treg, Tfr, and Tfh cells, our study found that in the new-onset patients with RA, the ratio of Tfh/Tfr was aberrant, and Treg and c-Tfr cells were reduced and associated with the disease activity and the abnormal autoantibodies of RA negatively, but the expressions of Th17 and c-Tfh cells were not increased significantly. The results were consistent with the previous studies [9, 18‐20] and supported the previous spot that the immune tolerance breakdown mediated by reduced Treg and Tfr cells participated in the progression of RA rather than the over-immune response. Interestingly, our study found that both Treg and Tfr cells were decreased in RA and were related to disease activity and antibody production of RA, but Treg cells were mainly related to disease activity while Tfr cells were mainly related to antibody production specifically. The above might suggest that the function of Treg cells and Tfr cells in regulating antibody production and maintaining immune tolerance was different despite that Tfr cells were derived from Treg cells. The role of Tfr cells in the early stage of RA, even Pre-RA, needs further exploration.

The interactions between gut microbiotas and their metabolites with the host are important in health and disease, especially the role of gut microbiotas and their metabolites in RA is the focus of our research. Our study revealed that gut microbiota was dysbiosis in new-onset RA patients. Especially the increased Ruminococcus 2, one of the predominant gut microbiotas in RA, exhibited a positive correlation with disease activity and autoantibody production. And Ruminococcus 2 might be the potential biomarker of RA. Notably, our study found that Lachnospira was recognized as one of the 21 differential gut microbiotas from phylum to genus levels between RA and HCs, and it was downregulated in new-onset RA patients and negatively associated with RF-IgG of RA. Lachnospira is mainly present in the gut of most healthy individuals and may be a potential probiotic involved in the metabolism of a variety of carbohydrates. A two-sample Mendelian randomization study of the causal effects between gut microbiome and systemic lupus erythematosus (SLE) showed that Lachnospira was negatively correlated with the risk of SLE [44]. Although our study found the reduced Lachnospira in RA was only related to RF-IgG negatively, it still suggested that the reduced Lachnospira might promote RA, and the supplementation of it may be a potential treatment for RA. Of course, it still needs further exploration. Additionally, our results also showed that the gut microbiota-derived metabolites involved in the biosynthesis of unsaturated fatty acids, arginine biosynthesis, and tryptophan metabolism as well as alanine, aspartate, and glutamate metabolism were altered in RA, which was consistent with the prior studies [35, 36, 45]. We found that the altered metabolites involved in the biosynthesis of unsaturated fatty acids and tryptophan metabolism pathways were associated with the progression of RA. Especially, the increased arachidonic acid in RA showed positive correlations with disease activity and autoantibody production. And it exhibited good discrimination in distinguishing RA and HCs as the potential biomarker value of RA. In brief, the above supported that gut microbiota dysbiosis and altered metabolites in RA were associated with the development of RA.

Gut microbiotas and their metabolites are the vital bridge of the gut-joint axis to contribute to the pathogenesis of RA [26, 34, 46], but the mechanism remains unclear. Exploring the mechanism of RA triggered by gut microbiotas and their metabolites through the gut-joint axis has great significance. The trafficking of activated pro-inflammatory immune cells from the gut to joints has been thought to be one of the main mechanisms of driving the RA onset through the gut-joint axis [37, 38]. T cells in the synovium of patients with RA have been found to express the gut-homing receptor αEβ7 integrin supporting the viewpoint that the trafficking of mucosa-derived immune cells (such as mucosal-associated invariant T cells, Th17 cells, γδT cells, Tfh cells, and so on) from gut to joints [37, 38]. Gut-residing segmented filamentous bacteria (SFB) was found to promote autoimmune arthritis in K/BxN mouse models via the migration of Tfh cells and Th17 cells suggesting the interactions of gut microbiota and effector T cells contribute to the development of RA [47, 48]. Meanwhile, it is worth noting that the gut microbiotas and their metabolites may also contribute to the pathogenesis of RA by regulating Treg and Tfr cell-mediated immune tolerance through the gut-joint axis.

The interactions between gut microbiotas and Treg cells have attracted extensive attention, which is important in establishing intestinal immune tolerance. On the one hand, gut microbiotas could regulate the function and expression of Treg cells by influencing the Treg cell-modulatory activity (such as transforming growth factor-β) directly or controlling signals coming from epithelial cells, dendritic cells, or other Treg cell-regulating cells indirectly [49, 50]. On the other hand, Treg cells exert the effect in establishing intestinal immune homeostasis by inducing the tolerance to symbiotic flora and the host defense against intestinal pathogens, and gut microbial-specific Treg cells have been confirmed to be the essential cells to induce intestinal immune tolerance [51‐53]. The effects of gut microbiotas on Treg cells are also researched in RA. The association of gut microbiota dysbiosis and reduced Treg cells in RA patients has been observed [54, 55]. And the animal experiments also confirmed that gut microbiotas and their metabolites affected Treg cells [56‐58]. The reduced Bacteroides fragilis in collagen-induced arthritis (CIA) mice inhibited the differentiation of CD4 + T cells into Treg cells, while the colonization of Bacteroides fragilis in germ-free mice promoted the proliferation of Treg cells and the production of anti-inflammatory cytokines [56, 57]. Lactobacillus casei CCFM1074 strain also upregulated the number of Treg cells in the CIA mouse model [58]. In addition, as one of the most abundant gut microbiota-derived metabolites, SCFAs could also regulate the Treg cell-mediated immune tolerance by upregulating the expression of Foxp3 in Treg cells and enhancing the ability of dendritic cells to induce differentiation of Treg cells [59, 60]. Butyrate as the component of SCFA, it has been found that patients with RA lacked the butyrate-producing species and the supplementation of dietary butyrate could exert anti-inflammatory effects to ameliorate RA by promoting Treg cells while suppressing effector T cells and osteoclasts [39]. The above evidence suggests that gut microbiota dysbiosis and altered metabolites may contribute to RA by influencing immune tolerance mediated via Treg cells, which seems to be one of the mechanisms of the gut-joint axis. Tfr cells are largely derived from Treg cells; therefore, Tfr cells are also the potential targets for gut microbiotas and their metabolites to regulate immune tolerance in patients with RA. However, few studies have revealed the relationship of gut microbiotas and their metabolites with Tfr cells directly. Although there had been some studies finding that arthritis induced by SFB was associated with the reduced expression of cytotoxic T-lymphocyte-associated protein 4 (CTLA-4) on the surface of Tfr cells [61] and microbiota-derived butyrate could suppress the development of autoimmune arthritis by enhancing the histone acetylation of Tfr cell to promote their differentiation [40], the effects of gut microbiotas and their metabolites on Tfr cells still value the in-depth study in the future, especially to explore the role of other microbiotas and metabolites on regulating Tfr cells to participate in RA. Our study was the first to investigate the association of gut microbiotas and their metabolites with immune tolerance mediated by Tfr cells and found that gut microbiota dysbiosis and altered metabolites were related to the reduced Treg and Tfr cells. Specifically, the increased Ruminococcus 2, the increased arachidonic acid involved in the biosynthesis of unsaturated fatty acid pathway and the increased 3-methyldioxyindole involved in the tryptophan metabolism pathway exhibited negative correlations with the reduced Treg and Tfr cells. It meant that RA patients with altered gut microbiotas and their metabolites were more likely to exhibit impaired immune tolerance mediated by reduced Tfr cells. Therefore, we suspected that the association of altered gut microbiotas and their metabolites in RA with the breakdown of immune tolerance mediated by the reduced Tfr cells was involved in the development of RA.

Ruminococcus is a kind of common commensal gut microbiota present in healthy individuals with low abundance, and the increase of it would directly lead to the disruption of intestinal barrier function [62], which may be involved in the pathogenesis of autoimmune diseases. Some studies have shown that Ruminococcus was elevated in SLE [63] and spondyloarthropathies [64]. Ruminococcus was also found to be positively correlated with RF-IgA and anti-CCP antibodies and the disease activity of RA [65, 66]. And the deletion of T-cell death-associated gene 8 (TDAG8) was found to significantly reduce local mucosal inflammation and relieve the disease severity of RA by decreasing the abundance of proinflammation-related Ruminococcus [67], which both suggested the important role of Ruminococcus in RA. Here, we focused on Ruminococcus 2, which was one of the abundant gut microbiotas detected in RA and was associated with the progression of RA. The role of Ruminococcus 2 in RA has not been well-studied. Only one study has found that Ruminococcus 2 was more abundant in RA patients with lower Treg cells indicating that Ruminococcus 2 was associated with Treg cells in RA [54]. And even fewer studies about Ruminococcus 2 and Tfr cells. Based on our results, we hypothesized that individuals with specific compositions of gut microbiota such as increased Ruminococcus 2 might be more susceptible to RA by reducing Tfr cells to destroy immune tolerance. Arachidonic acid, a polyunsaturated fatty acid, is an important inflammatory mediator in exerting regulatory effects as the direct precursor of various bioactive lipid mediators [68] and active substances such as prostaglandin E2, prostaglandin I2, and thromboxane A2 [69]. A recent study of the serum metabolites in Pre-RA showed that arachidonic acid was enriched in the Pre-RA group [45], and there also had been some studies about the role of arachidonic acid in RA [70, 71]. It showed that arachidonic acid could regulate calcium signaling in the T cells of patients with RA to promote synovial inflammation [70]. And further study showed that Ershiwuwei Lvxue Pill (ELP), a prescription of Tibetan medicine, could alleviate cartilage and bone injury by regulating host metabolites such as arachidonic acid [71]. The results of our study supported that arachidonic acid was the core metabolite of gut microbiota and the increase of it was associated with reduced Treg and Tfr cells. We suspected that the increased arachidonic acid in RA patients may be mainly caused by the gut microbiota dysbiosis, and it promoted the conversion of the immune balance towards autoimmunity to contribute to RA, which is related to the breakdown of immune tolerance mediated by reduced Tfr cells. In addition, our study showed that Ruminococcus 2 and arachidonic acid were positively related with each other and they were both associated with the symptoms of arthritis as the potential biomarker of RA. The increased Ruminococcus 2 might aggravate arthritis and pain by promoting the production of arachidonic acid to further generate the active substances including prostaglandin.

Additionally, tryptophan metabolism could exert effects in regulating the immune system, and it has also been confirmed that the abnormal tryptophan metabolism was related to autoimmune disease [72‐74]. There are three main metabolic pathways for tryptophan, and nearly 90% of tryptophan is metabolized by indoleamine-2,3-dioxygenase (IDO) to produce intermediate metabolite kynurenine [75]. And CD4 + CD25 − T cells can be transformed into Foxp3 + CD4 + CD25 + Treg cells through IDO-mediated tryptophan metabolism suggesting the role of tryptophan metabolism in regulating immune tolerance [76]. Gut microbiota participates in tryptophan metabolism directly or indirectly, and gut microbiota-derived tryptophan metabolites were found to be associated with autoimmune arthritis [72, 73]. IDO is highly expressed in intestinal epithelial cells, which may be affected by gut microbiota to further influence tryptophan metabolism and the transformation of Treg cells [77]. The interactions between gut microbiota and tryptophan metabolism may regulate Treg cell-mediated immune tolerance, but the effect on Tfr cells remains unclear. Considering that Tfr cells was differentiated from Treg cells, the tryptophan metabolism affected by gut microbiota might also regulate Tfr cells, which needed further exploration. Our study suggested that there was an altered distribution of fecal tryptophan metabolites of new-onset RA and they were correlated with the arthritis symptoms and the reduced Treg and Tfr cells. It provided support for exploring the effect of tryptophan metabolism on immune cells, especially Tfr cells, in RA. However, the exact effect of tryptophan metabolism on Tfr cells needs further exploration.

Our result elucidated that gut microbiotas and their metabolites were potential disease markers for RA, and there was a correlation between them with Treg and Tfr cells. The influence of gut microbiota, especially their metabolites, on immune cells is still a frontier of immunomicroecology. The effects of gut microbiota-derived metabolites including SCFAs, bile acids, and tryptophan and its derivatives on Treg cells have been demonstrated, and previous studies have also found that SCFA affected Tfr cells. Our study highlighted the association of the biosynthetic unsaturated fatty acid pathway and the tryptophan metabolic pathway with Tfr cells, which provided valuable insights for exploring the effect of other metabolites on the immune tolerance of RA in the future, especially tryptophan metabolism. Understanding these relationships may help to develop the potential therapeutic strategies targeted at modulating gut microbiotas and their metabolites to restore immune tolerance and improve RA management. However, there were some limits in our study. Firstly, the sample sizes were small and limited to a single center, which may affect the generalizability of the results. Larger and multi-center studies are needed to further verify the results, and it was also necessary to enroll in the different stages of RA (including pre-clinical RA, RA transition, early RA, and established RA) to further explore the characteristics of gut microbiotas and their metabolites during the progression of RA. Secondly, as an observational study, our study only demonstrated the relationship of gut microbiotas and their metabolites with Tfr cells in RA instead of the causality, it is necessary to conduct the vitro experiments to elucidate the causal relationship between them and identify whether the gut microbiotas and their metabolites contribute to the pathogenesis of RA by influencing Tfr cells as one of the mechanisms involved in the gut-joint axis. Finally, further studies are necessary to confirm the specific biological significance underlying these differences in the identified metabolites in RA, for example targeting the metabolism of tryptophan and unsaturated fatty acids may have significance in exploring the pathogenesis of RA.

Our study revealed that gut microbiota dysbiosis and altered metabolites were associated with the breakdown of immune tolerance mediated by reduced Tfr cells in RA. Based on it, we hypothesized that the altered gut microbiotas and their metabolites might result in the breakdown of immune tolerance by affecting Tfr cells. This finding highlighted the importance of gut microbiotas and their metabolites in the gut-joint axis. It also provides a foundation for further investigations into the potential role of gut microbiotas and their metabolites on Tfr cell-mediated immune tolerance in RA from the perspective of microecology-metabolism-immune. This could help us to better understand the pathogenesis of RA and develop new treatments for the disease.