This study demonstrated that a six-month MTX therapy reduced alpha diversity and modified the structure of the oral and gut microbiota in RA patients. Nonsurgical periodontal treatment partially reversed the impact of MTX treatment on the oral microbiota. Periodontal therapy improved periodontal parameters individually, whereas MTX treatment had no impact on the periodontium. Finally, MTX treatment modified the relationship between periodontal and RA clinical factors. Collectively, MTX appeared to act by simplifying correlations between the oral and gut microbiota.
Without early and efficient treatment, RA can result in progressive disability and extra-articular symptoms, leading to increased morbidity and mortality2. MTX targets the inflammation pathways responsible for joint swelling and damage40. However, a significant non-response rate of 30 to 40%, as well as the impossibility of predicting the patient's response to treatment, are evident impasses62. Recent literature has indicated multiple factors involved with MTX responsiveness, including the participation of the gut microbiota in drug metabolism and efficiency63,64,65. Understanding how the microbiota affects the response to MTX and, in turn, what impact this drug has on the microbiota is important not only because the modulating of the gut microbiota may offer a novel therapeutic or preventive approach for patients with RA, but also because it may be helpful in predicting the response to treatment. Herein, we observed that a decrease in the alpha diversity of both the oral and gut microbiota occurred after MTX treatment. Previous studies also observed partial changes in the diversity of the oral and gut microbiota after antirheumatic treatment. Zhang et al. (2015)16 reported that the use of MTX was related to the restoration of the gut microbiota due to the increase of Lactobacillus salivarus, Lachnospiraceae bacterium, and Clostridium asparagiforme, and the decrease of some Firmicutes of the Veillonellaceae family of the oral microbiota. Other studies also observed the impact of antirheumatic treatment of RA on the gut microbiota, causing an increase in the Firmicutes phyla66,67 and Bacteroidetes67,68, which can be considered a positive effect. However, the fine line that separates this effect from dysbiosis may depend on the intrinsic ability to metabolize the drug or on factors such as the pharmacological combination, the dose administered, or the length of time the drug is prescribed63.
At baseline, numerous positive and negative correlations were observed between oral and gut bacteria, in addition to correlations with clinical parameters of both periodontitis and RA. It is interesting to note the positive correlation between the clinical parameters plaque index and DAS 28 and the pathobiont oral bacterium Campylobacter that was previously identified to be abundant in individuals with RA and periodontitis69. Also, a negative correlation was observed between the oral pathogenic bacteria Leptotrichia and treatment time and a positive correlation with CRP. In the context of systemic lupus erythematosus, we demonstrated positive correlations between Leptotrichia and levels of proinflammatory interleukins70. Furthermore, Sher et al. (2012)71 demonstrated that Leptotrichia species were predominant in individuals with recent-onset RA; thus, the important role of this oral bacterium in the onset of RA30. Several other correlations were also demonstrated between gut bacteria and periodontal parameters. For example, gut bacteria of the genus Bacteroides, which were only observed abundantly in individuals with RA and which may be associated with disease progression26, positively correlated with the clinical parameters, probing depth, and CAL. On the other hand, a negative correlation was noted between Bacteroides and the oral bacterium Capnocytophaga, which a priori was associated with lower probing depth in individuals with RA21.
After treatment with MTX, there were changes in the correlation network, mainly in gut bacteria that exhibited numerous new positive correlations with DAS 28 and negative correlations with the number of teeth. Furthermore, it is important to note that correlations between pathogenic bacteria such as Tannerella and Treponema have not been previously observed after MTX treatment. On the contrary, new correlations with the health-related bacteria Streptococcus, Capnocytophaga, and Tyzzerella emerged after MTX treatment. This scenario of new correlations reinforces the evidence that changes in microbiota diversity and disease parameters occur after MTX treatment. In fact, the mechanism of response to MTX in patients with RA is likely associated with the catabolic capacity of the drug in the gut microbiota65. However, it is important to note that no microbiological signature regarding response to MTX was observed in our study. Probably, individual biological variations, disease parameters, and genetic susceptibility may also explain responsiveness to MTX treatment.
Relevant clinical data refer to the impact of MTX therapy on periodontal parameters. Contrary to the results observed by our group and others using an animal model of arthritis42,43 in which MTX treatment protected against alveolar bone loss, in the present study there was apparently no effect of MTX on the periodontium. Other clinical studies have shown similar results, observing the stability of periodontal parameters in individuals with moderate to severe periodontitis after 16 months of antirheumatic treatment72. In contrast, an improvement in periodontal inflammation parameters following treatment with conventional DMARDs has been reported elsewhere73,74. Evidence indicates that DMARDs, including MTX, are able to improve the periodontal condition of individuals with RA and periodontitis; however, such an effect is unfeasible to be observed in the short term75,76. The short follow-up period and the initial 6 months period without combined professional dental prophylaxis of our study design likely explains the maintenance of individual periodontal parameters before and after MTX therapy.
Nonsurgical periodontal therapy resulted in the improvement of periodontal parameters and also influenced the diversity of the oral microbiota of our patients. Other studies have also observed the effect of periodontal therapy among individuals with RA51,77−79. Studies investigating the influence of periodontal treatment on the subgingival microbiota of systemically healthy individuals with chronic periodontitis initially observed a decrease in alpha diversity after two and six weeks of periodontal treatment, which was completely restored after 12 weeks80. In our study, a decrease, albeit non-significant, in microbial diversity was also demonstrated when compared to baseline. However, a reestablishment of this microbiota was noticed after periodontal treatment when compared to the time after the antirheumatic treatment. These findings corroborate the assumption that greater microbiota diversity is associated with increased ecosystem resilience and a healthier condition81,82.
Moreover, it was observed that periodontal treatment individually modified the structure of the oral microbiota; however, it was not possible to identify a pattern regarding these changes due to the non-formation of clusters. Changes in the composition of the microbiota after periodontal treatment have also been reported in the literature, mainly reflected by the significant decrease in the relative abundance of periodontopathogenic bacteria such as Porphyromonas and Treponema species, and an increase in health-associated bacteria such as Streptococcus and Rothia species shortly after periodontal treatment, but these changes were gradually reversed up to 12 weeks after treatment80. Other studies analyzing the salivary microbiota after periodontal therapy in systemically healthy individuals with aggressive periodontitis documented a trend towards reduced diversity as measured by the Shannon index after three and six months of periodontal therapy compared to baseline83. It is important to highlight the individuality of each patient's oral microbiota and that there is probably no single composition that represents a healthy periodontal state, and that recovery from periodontal disease appears to reflect a shift from a personalized disease state to a personalized healthy state. While there is consensus that specific communities must change with the response to disease, there may not be a “healthy amount” of these bacteria that is consistent across individuals84.
Herein, we observed that the severity of periodontitis was associated with a greater relative abundance of some pathogenic bacteria. In this line, changes in the subgingival microbial profile of individuals with RA directly associated with the severity of periodontitis were demonstrated22. Conversely, other authors have not demonstrated significant differences between the subgingival microbial profile of individuals with RA and the classification of periodontitis. Nonetheless, individual periodontal parameters such as deep periodontal pockets were directly associated with a greater abundance of Gram-negative anaerobic pathogens such as Selenomonas21, which was also observed more abundantly in our study. Likewise, bacteria of the genus Anaeroglobus were linked with severe periodontitis. In other studies, the bacteria of this genus were correlated with an increase in the number of swollen and painful joints and with the levels of circulating rheumatoid factor and ACPA in individuals with RA, indicating a possible role of this oral bacteria (i.e., their increase in severe periodontitis) in the progression of RA18,30,71. We also showed that individuals with severe periodontitis had a higher abundance of the Desulfobulbus genus when compared to the ones without periodontitis. Previous studies also reported similar results, with the Desulfobulbus genus being increased in individuals with RA and more severe periodontitis22. Other authors have also identified the severity of periodontitis as a determining factor in defining the diversity of the subgingival microbiota71 and have identified pathogenic bacteria of the “red complex” associated with more severe periodontitis in individuals with RA85 (Ziebolz et al., 2011). These findings suggest that differences in relative abundance of the subgingival microbiota characterize more severe forms of periodontitis and do not represent a specific signature for the oral microbiota of RA71 (Scher et al., 2012).
Despite its limitations, mainly related to the non-assessment of the influence of lifestyle and diet on the microbiota of the included individuals, our study addressed for the first time the influence of MTX treatment and periodontal treatment on the oral-gut microbiota of individuals with RA at three time points. Considering the effect of nonsurgical periodontal treatment on periodontal parameters and oral microbiota, it would be important to perform periodontal treatment at the beginning of RA treatment with MTX. However, studies that include robust samples with suitable experimental design are necessary to validate these data.