Intra-genus dysbiosis of Streptococcus in tonsillar microbiota is associated with host immune features in rheumatoid arthritis

doi:10.21203/rs.3.rs-3141503/v1

Download PDF

Research Article

Intra-genus dysbiosis of Streptococcus in tonsillar microbiota is associated with host immune features in rheumatoid arthritis

https://doi.org/10.21203/rs.3.rs-3141503/v1

This work is licensed under a CC BY 4.0 License

Version 1

posted

You are reading this latest preprint version

Background

Palatine tonsils are mucosa-associated lymphoid organs that constantly engage in crosstalk with commensal microorganisms and the immune system. Focal infections at tonsils have been implicated in the pathogenesis of autoimmune diseases including rheumatoid arthritis (RA), but the underlying mechanisms through which tonsils contribute to host autoimmunity remain poorly defined.

Results

We identified a significant dysbiosis of tonsillar microbiota in RA patients, which was largely associated with disease activity. RA tonsillar microbiota was featured by an expansion of opportunistic pathogenic Streptococcus species including S. pyogenes, S. dysgalactiae and S. agalactiae, along with a contraction of numerous commensal Streptococcus members like S. salivarius. By defining a Streptococcus dysbiosis index, we found that RA patients, especially those without medication, were overrepresented in the Streptococcus dysbiotic set. Moreover, the intra-genus dysbiosis of Streptococcus in tonsillar microbiota was closely correlated with abnormal expression of circulating anti-streptolysin O, LPS-binding protein, soluble CD14, T helper 17 and natural killer cells. Finally, we demonstrated that the RA-deficient S. salivarius inhibited arthritis development and autoimmune responses.

Conclusions

Collectively, our study uncovers the functional link between host immune responses and tonsillar microbiota, and demonstrates that intra-genus dysbiosis of Streptococcus species contribute significantly to host autoimmunity.

Palatine tonsils

rheumatoid arthritis

tonsillar microbiota

Streptococcus dysbiosis

Streptococcus salivarius

Human palatine tonsil represents one of the mucosa-associated lymphoid tissues at the entrance of the aerodigestive tract ^1,2. The tonsils serve as the first handing sites against microbial antigens, and engage in continuous immune responses that help maintaining the local and systemic immunological homeostasis ¹. Infections at the tonsils can exert impacts on distant organs ^1,2, including the heart ³, kidney ⁴, and joints ^5–7. The well-known autoimmune condition acute rheumatic fever is caused by infection with the group A Streptococcus (i.g., Streptococcus pyogenes) at the tonsils, which is characterized by varying degrees of inflammation of the joints and the heart ³. In addition, the tonsils have been implicated in the pathogenesis of certain chronic and systemic autoimmune diseases including IgA nephropathy ⁴ and rheumatoid arthritis (RA) ^5–7.

RA is an autoimmune inflammatory disorder with high morbidity and a high disability rate, and characterized by persistent synovitis with joint destruction, systemic inflammation and autoantibody production ^8,9. Previous studies have highlighted the functional significance from the tonsils in the immunopathogenesis of RA ^5–7. The tonsillar lymphocytes can migrate to subcutaneously-engrafted human rheumatoid synovial tissue in mice ⁵, suggesting the possibility that inflammatory immune cells primed at the tonsils could migrate to the joints. Then clinical data reported that clonally identical T cell expansion occurs simultaneously in the tonsils and synovium of the same RA patients ^6,7, lending further support to the idea of immune cell migration between tonsils and synovial tissue. However, the underlying mechanisms through which tonsils contribute to RA pathogenesis remain poorly defined.

Emerging evidence is emphasizing the functional impacts from both the host mucosal immune system and from the microbiome on the etiopathogenesis of autoimmune diseases ^8–10. Studies on experimental arthritis models have demonstrated that RA initiation is dependent on the participation of microorganisms, as evidenced by the fact that germ-free or antibiotics-treated mice could not develop arthritis unless particular bacteria are introduced (e.g., segmented filamentous bacteria or Lactobacillus Bifidus, or Prevotella copri) ^11–13. Moreover, RA patients showed alterations in microbial compositions in multiple body sites including the gut, the oral cavity and the lung ^14–18. The dysbiotic microbiota could trigger arthritis through activation of specific T effector cells ^12,13. Despite these advances in understanding the biological connections between the microbiome and joints, many issues remain to be addressed between the microbiome and RA pathogenesis.

The tonsils are colonized by diverse microbes that are in intimate contact with abundant immune cells (e.g., T, B, and dendritic cells) ^1,2. Given the frequent migration of lymphocytes between the tonsils and the circulation, it is believed that changes of human-microbiota mutualism at tonsils can trigger autoimmune disorders such as RA, but little is known about the underlying mechanisms through which the tonsillar microbiota contributes to host autoimmunity. Previously, we have uncovered that the tonsillar microbiome-derived lantibiotic peptides salivaricins could directly bind to and inhibit IL-6 and IL-21 receptors and beneficially modulate host immunity in RA ¹⁹. In the present study, we profiled the tonsillar microbiota, and identified a dysbiosis of tonsillar microbiota in RA patients, especially the imbalance in genus Streptococcus, which play important roles in modulation of host autoimmunity and provide mechanistic understanding for future intervention and treatments.

Dysbiosis of tonsillar microbiota in RA patients

To profile the tonsillar microbiota, we began with 16S rRNA gene sequencing of tonsillar swab samples from 121 RA patients and 99 healthy controls (Supplemental Table 1). Compared to healthy controls, the tonsillar microbiota of RA patients displayed similar community richness denoted by α-diversity, yet showed significant increases in between-sample β-diversity (Supplemental Fig. 1), indicating a more heterogeneous microbial community at tonsils among RA patients. Principle coordinate analysis revealed a significant separation of RA samples from controls (Fig. 1A, Adonis test, P = 0.001). We then examined seven demographic factors with the inter-individual variation, diversity and richness of tonsillar microbiota. In healthy subjects, the age, gender and height were significantly associated with the variation of microbial composition and diversity or richness (Fig. 1B and Supplemental Table 2). In addition, weight and smoking showed evident correlations with microbial diversity (Fig. 1B and Supplemental Table 2). Nevertheless, these associations were almost completely diminished in RA patients (Fig. 1C and Supplemental Table 2), further supporting the fact that the tonsillar microbiota was significantly alternated in disease.

RA tonsillar microbiota varied with disease activity

Then, we explored factors that associated with the composition variation, diversity and richness of tonsillar microbiota in RA patients. Among the 26 clinical and immunological indicators, disease activity was determined to be the top contributor to the microbial composition variation, and the autoantibodies glycosyl phosphatidyl inositol (GPI) and rheumatoid factor (RF) were significantly related with tonsillar microbiota diversity (Fig. 2A and Supplemental Table 2). Although the medication status explained small proportions of variation in RA tonsillar microbiota, RA patients with medication have relatively more similar microbiota compositions to the healthy controls than untreated patients (Fig. 2, A and B). Nevertheless, the detected difference between treated patients and healthy controls was still significant (Fig. 2B). Among the nine examined medications, methotrexate (MTX) and biologicals were the top two drugs that explained the most proportions of the microbial variation (Fig. 2C and Supplemental Table 2). Moreover, among the eight concomitant diseases we examined, interstitial lung disease (ILD), one of the common extra-articular manifestations and the second leading cause of death of RA ²⁰, showed the strongest associations with the variation of RA tonsillar microbiota (Supplemental Fig. 2 and Supplemental Table 2). Taken together, these results suggested the involvement of tonsillar microbiota in the immune responses and disease progression of RA.

RA tonsillar microbiota showed an imbalance in Streptococcus species

To investigate more detailed alternations in RA tonsillar microbiome, we additionally performed whole-metagenome sequencing of tonsillar microbiota from 32 RA patients and 30 healthy controls (Supplemental Table 3). Consistent with the results from 16S rRNA gene sequencing analysis, multivariate analyses for both taxonomical composition and functional capacities revealed a clear deviation of the tonsillar microbiome of RA patients from healthy controls (Fig. 3, A and B). Moreover, the tonsillar microbiota in patients with medication also showed partial recovery to the healthy state (Supplemental Fig. 3, A and B).

A detailed comparison of the tonsillar microbiome of RA patients with healthy controls identified 67 species (Wilcoxon rank-sum test, P < 0.05) and 85 KEGG (Wilcoxon rank-sum test, FDR < 0.1) functional modules with differential abundance between the two groups (Supplemental Tables 4 and 5). Streptococcus taxa comprised 25.4% (17/67) of the RA-associated species, followed by Neisseria (13, 19.4%) and Bacteroides (8, 11.9%) (Fig. 3C, Supplemental Fig. 3C and Supplemental Table 4). Among the Streptococcus species, S. pyogenes, S. dysgalactiae, and S. agalactiae were enriched in RA, whereas numerous commensal Streptococcus species (e.g., S. sanguinis, S. salivarius, and S. cristatus) were deficient (Fig. 3C). These Streptococcus species showed close correlations with RA-associated functional modules (Fig. 3D and Supplemental Table 5), most of which were in accordance with the observations in the gut and oral microbiota of RA patients ¹⁴. By contrast, the biosynthesis and transport of lantibiotics, a series of polycyclic peptide antibiotics known to function in maintaining microecological homeostasis ^21,22, were deficient in RA tonsils but not reported in the gut or oral cavity. In particular, the reduced capacity of lantibiotics biosynthesis and transport significantly associated with the deficiency of lantibiotics-producing commensal Streptococcus members and the enrichment of pathogenic Streptococcus (Fig. 3D). In addition, all the altered Neisseria members were depleted, while the Bacteroides species were enriched in RA (Supplemental Fig. 3D), without intra-genus shifts compared to that of Streptococcus. Together, these results suggest a potential role of multiple bacterial species within the same genus Streptococcus in RA development.

The Streptococcus dysbiosis in tonsillar microbiota was closely correlated with host immuno-inflammatory features

To dig further the intra-genus shifts of Streptococcus species in tonsillar microbiota, we defined a Streptococcus dysbiosis index as the value of [total abundance of Streptococcus members decreased in RA] minus [total abundance of Streptococcus species enriched in RA] for all samples (Supplemental Table 6 and Fig. 4A). The threshold for Streptococcus dysbiosis was calculated as 2.971% by ROC analysis (Figure S4A). Samples with Streptococcus dysbiosis index less than 2.971% were classified into the dysbiotic group (Fig. 4B, and Supplemental Table 6). Notably, the dysbiotic group was significantly overrepresented in RA samples (56%) than controls (10%) (Fisher's exact test, P < 0.001, Fig. 4B, and Supplemental Table 7).`In addition, patients with DMMARDs treatment (Fisher's exact test, P = 0.082), especially leflunomide (Fisher's exact test, P = 0.071) or hydroxychloroquine (Fisher's exact test, P = 0.069), were more prone appeared in the non-dysbiotic group compared with naïve-treatment patients (Supplemental Table 7), suggesting that the Streptococcus dysbiosis in tonsillar microbiota might be ameliorated after medication.

Subsequently, we detected that an increase of Streptococcus dysbiosis index was associated with a decrease in the level of anti-streptolysin O (ASO) (Figure S4B), a specific antibody against group A beta-haemolytic streptococcal infection. Moreover, the Streptococcus dysbiotic group was more abundant with serum LPS-binding protein (LBP) and soluble CD14 (sCD14) (Fig. 4C and D). Additionally, the dysbiotic group was deficient with circulating T helper 17 (Th17) and natural killer (NK) cells, which displayed positive associations with the Streptococcus dysbiosis index (Fig. 4E and F). These results indicate that the intra-genus dysbiosis of Streptococcus in tonsillar microbiota was closely correlated with host immuno-inflammatory responses, and might play key roles in RA immunopathogenesis.

The RA-deficient probiotic S. salivarius exert anti-arthritic and immunosuppressive effects in experimental arthritis

S. salivarius is a safe and well-tolerated probiotic that predominantly colonized in the human oral cavity and upper airways ^23,24. Previous studies have reported that S. salivarius-produced antimicrobial peptides salivaricins can selectively inhibit pathogenic bacteria S. pyogenes and S. dysgalactiae^21,25, two pathogenic bacteria that enriched in RA tonsillar microbiota and showed arthritogenicity in RA. More importantly, we previously have uncovered that S. salivarius-produced salivaricins could directly bind to and inhibit IL-6 and IL-21 receptors and exerted anti-arthritic effects in RA ¹⁹. To further determine the role of S. salivarius in RA, we introduced S. salivarius strain K12 cultures to CIA mice. In the initial preventative studies, S. salivarius K12 was administered intra-nasally, intra-orally, or in combination, and all three of these deliveries significantly decreased the incidence and severity of arthritis in the CIA mice without affecting body weight (Fig. 5, A-D). Consistent with clinical observations, visual evaluation by micro-computed tomography (micro-CT) and histopathological scoring of the paws confirmed significantly reduced signs of synovial inflammation, bone destruction and cartilage depletion in mice receiving S. salivarius K12 (Fig. 5, E and F). S. salivarius K12 administered intra-nasally and/or intra-orally successfully colonized in the oropharyngeal mucosa of mice (Fig. 5G). In contrast, administration of Lactobacillus salivarius exhibited no apparent protection from CIA (Supplemental Fig. 5, A and B), which might be due to lack of colonization. Additionally, S. salivarius K12 supplemented by intragastric gavage also failed to protect against CIA (Supplemental Fig. 5, C and D), additionally supporting that S. salivarius is only effective against RA in the oral cavity and upper airways.

In addition to testing its prophylactic effects, we also conducted experiments with CIA model mice that had already-developed arthritis to investigate the potential therapeutic application of S. salivarius. We found that S. salivarius K12-treated mice showed significantly relieved disease severity and reduced inflammatory cell infiltration in the joints compared to controls (Fig. 5, H-J), further supporting potential clinical applicability of S. salivarius K12 in treating RA.

We next investigated which immune effector molecules may contribute to the observed anti-arthritis efficacy of S. salivarius K12 by examining cytokines and lymphocyte subsets. S. salivarius K12 markedly decreased the frequencies of Tfh cells in both draining lymph nodes (DLN) and the spleen (Fig. 6A and Supplemental Fig. 6A). Accordingly, the expression of Bcl-6, a transcription factor essential for the differentiation and function of Tfh cells ²⁶, was also reduced (Fig. 6B and Supplemental Fig. 6A). Moreover, the frequency of germinal center B (GCB) cells in the lymph nodes and serum autoantibody (anti-CII) level were both significantly decreased (Fig. 6, C and D), consistent with the downregulated Tfh cells. However, the proportions of GCB cells in the spleen were not obviously changed (Supplemental Fig. 6B). Other immune cell subset including Th1, Th17 and Treg cells were unchanged in S. salivarius K12 treated mice (Supplemental Fig. 6, C-H), suggesting selective immunomodulatory functions for S. salivarius K12 in distinct CD4⁺ T cell subsets. In addition, decreased concentration of pro-inflammatory cytokine IL-6 and increased serum level of anti-inflammatory mediator IL-10 were detected in S. salivarius K12-treated mice (Fig. 6, E and F). Collectively, these results show the active immunosuppressive effects of S. salivarius K12 in oropharyngeal mucosa in protecting against arthritis progression.

Focal infections in the tonsils have long been implicated in autoimmune diseases but lack mechanistic insights. Here, our study uncovers the dysbiotic composition and function of tonsillar microbiota contributes to disease etiology in RA patients. In particular, we demonstrate that species-level imbalance within one genus of Streptococcus contributes significantly to this autoimmune disorder. Our work substantially identifies a functional connection between tonsillar microbiota and autoimmune disease, and provides mechanistic insights into the contribution of tonsillar bacteria in maintaining immune homeostasis.

The three RA-enriched Streptococcus species (S. pyogenes, S. dysgalactiae, and S. agalactiae) were known to be opportunistic pathogens, and two of them (S. pyogenes and S. dysgalactiae) have previously been reported to contribute to the autoimmune conditions of rheumatic fever and heart disease ^3,27. Besides the enrichment of pathogenic Streptococcus members in RA tonsillar microbiota, numerous commensal Streptococcus species including S. salivarius were in turn deficient in RA. S. salivarius is one the core colonizers of the human oral cavity and upper airways where it remains a predominant commensal inhabitant ²⁴. In particular, S. salivarius can produce antimicrobial peptides salivaricins, belonging to lantibiotics and known to selectively inhibit oral and upper-respiratory-tract pathogenic bacteria including the RA-enriched S. pyogenes and S. dysgalactiae^21,25. This intra-genus, species-level competition using peptides has been demonstrated previously in skin commensals, for example Staphylococcus hominis-derived Sh-lantibiotics can exclude pathogenic strains of Staphylococcus aureus in the skin microbiota ²⁸, and Staphylococcus epidermidis-produced serine protease Esp inhibits Staphylococcus aureus biofilm formation and nasal colonization ²⁹. In gut although no intra-genus competition by peptides have been demonstrated, recent study found Blautia producta BP_SC5K produces the lanthipeptide nisin A and inhibits vancomycin-resistant Enterococcus faecium²² to maintain microbial stability and prevent infections.

Furthermore, antibacterial peptides produced by commensal bacteria not only have the capacity to eliminate specific colonizing pathogens, but also have a particularly important role in shaping the microbial community ³⁰. We have uncovered that the tonsillar microbiome-derived lantibiotics salivaricins are able to directly act on IL-6 and IL-21 receptors and beneficially modulate host autoimmunity by suppressing IL-6R/IL-21R-STAT3 signaling pathway and then inhibiting Tfh cell differentiation and IL-21 production ¹⁹. The deficiency for salivaricins in the tonsillar microbiome of RA may reduce the capacity for proper modulation of immune responses, thereby inducing autoimmune disorders ¹⁹. Therefore, the lack of lantibiotics-producing members like S. salivarius may be an important factor in the dysbiosis of tonsillar microbiota and RA pathogenicity.

Previous studies have shed light on the positive multifactorial implications of S. salivarius K12 on host health ^23,24. Small clinical trials showed that administration of S. salivarius K12 could reduce the occurrence of tonsillitis and otitis media in children and treat halitosis in adults ^31,32. Moreover, S. salivarius as well as its supernatants of culture medium exhibited an anti-inflammatory effect on human epithelial cells through inhibition of NF-κB activation ³³. Nevertheless, the host immune responses to S. salivarius K12, which might contribute to its commensal or probiotic properties, have not been studied yet. Here, we demonstrated that S. salivarius K12 exerted immunosuppressive and anti-arthritis effect in mice. Notably, S. salivarius K12 is only effective against RA in the oral cavity and upper airways, consistent with its niche- preference. An advantage of targeting the oral cavity or tonsils is that relatively high concentrations of viable microorganisms can be directly delivered to the preferred niches ²⁴. In addition, samples of the tonsillar microbiota can readily be obtained to determine the colonization and efficacy ²⁴. Our study paves the way for further clinical trials to be carried out to evaluate the therapeutic effect of S. salivarius in RA patients, with the hope of expanding the current scope of treatment for this important morbidity.

In summary, our work identifies specific disease-related tonsillar species and the functional molecules, and uncovers their central roles in shaping the etiology of RA development. We provide an interesting scheme of microbiota-host cross-talks in which autoimmunity occurs as an unfortunate consequence, and pave the way for novel regime of treatment for RA and perhaps other autoimmune disorders.

Participant enrollment. Adult patients (n = 121) diagnosed with rheumatoid arthritis (RA) according to the American College of Rheumatology/European League Against Rheumatism (ACR/EULAR) 2010 classification criteria ³⁴ were recruited from Peking University People’s Hospital, Beijing, China. Healthy volunteers (n = 99) with no history of inflammatory arthritis and rheumatic diseases were enrolled. All the participants were Chinese and the vast majority of them were Han nationality. Subjects were excluded if they 1) had underwent tonsillectomy; 2) had any current infection (especially in the oropharynx); 3) were suffering from serious diseases (e.g., cancer, heart failure, renal failure); 4) had taken antibiotic treatment or probiotic supplements in the three months prior to sample collection; 5) were strict vegetarians, alcoholics or subject with other unusual dietary modes; 6) were pregnant or lactating women. Detailed information of the cohort was given in Supplemental Tables 1 and 3.

Sample collection and DNA extraction. The tonsillar microbiota samples were collected by following the procedure of Human Microbiome Project (http://hmpdacc.org/doc/HMP_MOP_Version12_0_072910). Briefly, sterile cotton swabs, pre-moistened in sterile saline, were inserted into the oral cavity and rubbed around the mucosal surface of palatine tonsil for 5 seconds without touching the uvula, tongue, or other oral structures by using of a tongue depressor. The sampling was performed twice in the left and right tonsils, respectively. The right and left sides were pooled together as a combined specimen. To detect possible contamination, several negative controls were prepared and subjected to the same procedures. Immediately after swabbing, each swab must be swirled in a garnet bead tube containing 750 µl buffers (MoBio Laboratories, Inc., Carlsberg, CA, USA). The swab sponge should be pressed against the tube wall multiple times for 20 seconds to ensure transfer of bacteria from swab to solution. Put the tubes in a Ziploc bag and place over ice. The samples were placed at -80°C until DNA extraction.

Genomic bacterial DNA was extracted using the MoBio PowerSoil® DNA Isolation Kit 12888-100 protocol (MoBio Laboratories) according to the manufacturer’s recommendations. DNA concentrations were determined using the Qubit® 3.0 fluorescent quantitation kit (Thermo Fisher Scientific, Waltham, MA, USA). Extracted DNA was stored at -80°C prior to sequencing.

16S rRNA gene sequencing and analyses. The V3-V4 hypervariable regions of the 16S rRNA gene were amplified (RA = 121, healthy controls = 99). PCR reactions were performed by using unique fusion primers designed based on the universal primer set, 338F (5’-GTACTCCTACGGGAGGCAGCA-3’) and 806R (5’-GTGGACTACHVGGGTWTCTAAT-3’), along with barcode sequences³⁵. The primers were designed incorporating the Illumina adapters and a sample barcode sequence. Triplicate PCR reactions were performed for each sample and visualized on 2% agarose gels. PCR amplicons were purified using Agencourt AMPure XP kit (Beckman Coulter, Inc, Brea, CA, USA) and quantified by Qubit® 3.0 fluorescent quantitation kit (Thermo Fisher Scientific), then pooled at equimolar amounts using the Illumina TruSeq Sample Preparation procedure. The amplicon library was constructed following the manufacturer’s instructions (Illumina, San Diego, CA, USA) and quantified by KAPA Library Quantification Kit KK4824 (KAPA Biosystems, Woburn, MA, USA). The completed library was sequenced on an Illumina MiSeq (Illumina, San Diego) platform using dual-index sequencing strategy according to the Illumina recommended protocol ³⁶.

16S rRNA amplicon sequences were processed based on the quantitative insights into microbial ecology (QIIME2, https://qiime2.org/) platform ³⁷. Briefly, the forward and reverse reads were truncated at the 260th and 250th base, respectively, to avoid sequencing errors at the end of the reads. Noisy sequences, chimeric sequences and singletons in sequencing data were eliminated using DATA2 ³⁸. Denoised paired-end reads were joined, setting maximum mismatch parameter of 6 bases, which means a minimum similarity threshold of 90% on the overlap zone of the forward and reverse reads. The representative sequences (named “feature” in QIIME2 nomenclature) were defined at 100% similar merged sequences, after then low occurrence (n < 5 in pool samples) sequences were removed. We used the term “operational taxonomic unit (OTU)” instead of “feature” in the whole article for convenience. Then, taxonomy of OTUs was identified using classify-sklearn classification methods based on the Greengenes v13.8 database ³⁹ via the “q2-feature-classifier” plugin. We aligned the OTU sequences into the SILVA rRNA gene databases (release 138)⁴⁰, and the mapped OTUs with > 98.5% similarity and > 90% coverage was considered as species level annotation. The phylogenetic analysis was performed in QIIME2 with “qiime alignment mafft”, “qiime alignment mask”, and “qiime phylogeny fasttree” commands. Lastly, all samples were rarefied to an even sampling depth of 20,000 sequences, when calculating the OTU and taxa relative abundances. To capture major changes across the whole cohort, we focused on common taxa out of all mapped ones, excluding those that were present in less than 0.01% in relative abundance.

Alpha and beta diversities were calculated in QIIME2 platform with “qiime diversity core-metrics-phylogenetic” command. The Shannon’s diversity index, observed OTUs, Faith’s phylogenetic diversity (a qualitative measure of community richness that incorporates phylogenetic relationships between the OTUs) and Pielou’s evenness were used to reflect community richness and evenness. The Jaccard distance, Bray-Curtis distance, unweighted and weighted UniFrac distances, were implemented to assess the similarity or dissimilarity between individuals.

Whole-metagenome shotgun sequencing and bioinformatic analyses. The fresh genomics DNA samples (RA = 32, HC = 30, including 13 RA and 11 HC from the former cohort) were mechanically fragmented to ~ 400 bp with Bioruptor Pico (Diagenode, Belgium). A magnetic beads-based method was used for DNA fragments selection following a standard protocol (Agencourt AMPure XP). Libraries were prepared by using the NEBnext® Ultra™ II DNA Library Prep Kit for Illumina® (New England BioLabs). The Illumina HiSeq X platform was then used for 2 × 150 bp paired-end whole-metagenome sequencing (WMS). Initial base calling of WMS data was performed based on the system default parameters under the sequencing platform. The quality control used the following criteria: (1) reads were removed if they contained more than 3 ‘N’ bases or more than 50 bases with low quality (< Q20); (2) no more than 10 bases with low quality (< Q20) or assigned as N in the tails of reads were trimmed. The remaining reads were then mapped to the reference human genome (GRCh38) using the Centrifuge algorithm⁴¹, to remove host DNA contamination.

A de novo gene catalogue was constructed based on the metagenomic data from the tonsillar samples of all individuals. High quality reads were used for de novo assembly via MEGAHIT ⁴², which generated the initial assembly results based on different k-mer sizes and then merged and extended them using the parameter “-precorrection”. Ab initio gene identification was performed for all assembled scaffolds taking advantage of MetaGeneMark ⁴³. The predicted genes were then clustered at the nucleotide level by CD-HIT (version 4.5.4) ⁴⁴, and genes sharing greater than 90% overlap and greater than 95% identity were treated as redundancies. A non-redundant gene catalogue was generated after removing the redundancy genes.

The abundance of genes in the non-redundant gene catalogue was quantified as the relative abundance of reads. First, the high-quality reads from each sample were aligned against the gene catalogue based on SOAP v2.21 ⁴⁵ using a threshold that allowed at most two mismatches in the initial 32-bp seed sequence and 90% similarity over the whole read. Then, only two types of alignments were accepted: (1) those in which the entirety of a paired-end read could be mapped onto a gene with the correct insert size; (2) those in which one end of the paired-end read could be mapped onto the end of a gene only if the other end of the read mapped outside the genic region. The relative abundance of a given gene in a sample was finally estimated by dividing the number of reads that uniquely mapped to that gene by the length of the gene region and by the total number of reads from the sample that uniquely mapped to any gene in the catalogue. The resulting set of gene relative abundances for all samples was termed a gene profile. To capture major changes across the whole cohort, we focused on common genes out of all mapped ones, excluding those that were present in less than 0.01% in relative abundance.

The Kyoto Encyclopedia of Genes and Genomes (KEGG) database was used for functional annotation of genes using Blast KOALA (v2) ⁴⁶. Each protein was assigned a KEGG orthologue (KO) on the basis of the best-hit gene (BLASTP e-value < 1e-10 and protein level similarity > 30%) in the database. For KO profiling, we used KO assignment of each gene from the original gene catalog and summed the relative abundance of genes from the same KO to yield the abundance of that KO. The relative abundance of a module was calculated from summation of the relative abundance of its corresponding KOs.

S. salivarius K12 culture. S. salivarius K12 (ATCC®BAA1024™) was grown in Todd-Hewitt broth (THB, Hopebio, Qingdao, China), then serial dilutions were inoculated onto M17 agar medium (Hopebio, Qingdao, China) and incubated at 37°C in a 5% CO₂ atmosphere⁴⁷. The bacterial plaque that plump, smooth and distributed independently on the plate was chosen to move into a centrifugal tube with sterile probe, and was shaken to dissolve sufficiently. Then repeat the above steps, and the first generation of the strain was obtained and identified by 16S rRNA gene sequencing.

Experimental arthritis induction. Male DBA/1 mice (6–8 weeks old) were purchased from Huafukang Co. Ltd. (Beijing, China) and fed under specific pathogen-free conditions. Experimental arthritis induction was established in the same way as collagen-induced arthritis (CIA) by following a previously published protocol⁴⁸. Briefly, DBA/1 mice were immunized intradermally at the base of the tail with 200 µg of bovine type II collagen (CII, Chondrex, Redmond, WA, USA) or different microbial peptides (A-R-EVLDS-R-GNPTVE, IYA-R-EVLDS-R-GNPTIE, and LGSMGESRRALQDSQR) emulsified in complete Freund’s adjuvant (Sigma-Aldrich, St Louis, MO, USA) in equal volumes. Three weeks later, a booster was delivered using 100 µg CII r different microbial peptides emulsified in Freund’s incomplete adjuvant (Sigma-Aldrich, St Louis, MO, USA). Mice were monitored for signs of arthritis after the booster immunization. Clinical score was assessed by using the following system as detailed previously⁴⁸: 0, normal; 1, erythema and swelling of one or several digits; 2, erythema and moderate swelling extending from the ankle to the mid-foot (tarsals); 3, erythema and severe swelling extending from the ankle to the metatarsal joints; and 4, complete erythema and swelling encompassing the ankle, foot and digits, resulting in deformity and/or ankyloses. The scores of all four limbs were summed, yielding total scores of 0–16 per mouse.

Experimental arthritis intervention. We randomized mice into treatment groups. S. salivarius K12 was inoculated intra-nasally (10⁸ cfu/mice), intra-orally (10⁸ cfu/mice), combined intra-nasally with intra-orally (10⁸ cfu/mice) or intragastrically (10⁹ cfu/mice) three times a week. Lactobacillus salivarius (ATCC 11741) was administrated both intra-nasally and intra-orally (10⁸ cfu/mice). Mice treated with sterile phosphate buffer solution (PBS) served as controls. The preventive group started on day 1, and the therapeutic groups commenced after the onset of CIA (approximately day 26).

Radiography evaluations and histological analyses. The paws from each mouse were collected and fixed in 4% paraformaldehyde (PFA) for 48 hours, then scanned using a Micro-CT scanner (Quantum FX, Caliper, USA). After that, the paws were decalcified in 5% EDTA, paraffin-embedded, sectioned and stained with haematoxylin and eosin (H&E). A microscopic assessment of sagittal sections was performed and histopathological changes were scored based on the following previously reported parameters ⁴⁹: 0, normal synovium; 1, synovial membrane hypertrophy and cell infiltrates; 2, pannus and cartilage erosion; 3, major erosion of cartilage and subchondral bone; and 4, loss of joint integrity and ankylosis. The scores of all four limbs were summed and divided by 4, yielding average scores of 0–4 per mouse.

Cytokine and auto-antibody detection. At the study endpoints, mice were euthanized and serum samples were collected for cytokine and auto-antibody detection. The concentrations of cytokines (IL-6 and IL-10) were measured using ELISA kits (Multisciences), and the titer of collagen-specific antibody was analyzed using a mouse anti-bovine type II collagen IgG antibody assay kit (Chondrex). Serum levels of Streptococcus enolase peptides were tested by indirect ELISA. In detail, chemosynthetic peptides (10 µg/ml in 0.05 M carbonate buffer) were coated in the 96 well plates (Corning, New York, USA) at 4˚C overnight. Then wash the 96 wells with PBS plus 0.05% Tween-20 (PBST) three times and block them with 5% BSA-PBST at 37˚C for 3 hours. All serum samples were diluted at 1:100 with 1% BSA-PBST, and 100 ul diluted samples were added to the 96-well plate and incubated 1 hour at 37˚C. 1% BSA-PBST was used as control nonspecific background. After incubation, the wells were washed four times with 0.05% PBST. Then, 100 ul of goat anti-mouse IgG conjugated to peroxidase (1:10000 in 1% BSA-PBST) was added to the wells and incubated at 37˚C for 40 minutes. The bound antibodies were tested by tetramethylbenzidine (TMB) as substrate after washing four times. This reaction was terminated by adding 50 ul of 2 M sulfuric acid to the wells. The absorbance density (OD) was read by a Bio-Rad plate reader at 450 nm and 630 nm. The OD values were converted into arbitrary units (AU) values, which was calculated as follows: AU = [(OD peptide – OD nonspecific background) test serum/(OD peptide - OD nonspecific background) positive control serum]×100. The AU value was considered positive if it was greater than the mean + 3 × standard deviation of the control group.

Flow cytometry. The spleens and joint draining lymph nodes (popliteal and axillary lymph nodes, DLN) were obtained from mice, sieved through a 70 µm cell strainer (Corning) in RPMI 1640 medium with 10% FBS and single-cell suspensions (10⁶ cells/100µl) were prepared for flow cytometry. For intracellular cytokine analysis, cells were stimulated with 25 ng/ml PMA plus 1 µg/ml ionomycin for 4–5 hours in the presence of 10 µg /ml brefeldin A (PeproTech). Cell surface markers were first stained, and the cells were then fixed and permeabilized with an intracellular staining buffer set (Thermo Fisher Scientific) following the manufacturer's protocol and stained with intracellular or intranuclear markers. Antibodies were purchased from BD Biosciences (San Jose, CA, USA), eBiosciences (Thermo Fisher Scientific) or Biolegend (San Jose, CA, USA). Th1 cells were defined as CD4⁺IFN-γ⁺, Th17 cells as CD4⁺IL-17A⁺, Treg cells as CD4⁺CD25⁺FOXP3⁺, Tfh cells as CD4⁺Foxp3-CD44⁺CXCR5⁺PD-1⁺Bcl6⁺, GCB cells as B220⁺CD4⁻Fas95⁺GL-7⁺. Flow cytometry was performed using FACSAria^TMII (BD Biosciences) and the data was analyzed using FlowJo v10.0.7 software (Tree Star Inc., Ashland, OR, USA).

Determination of S. salivarius K12 abundance. Genomic DNA was extracted from the oropharyngeal mucosa using a Protease K Fasting Burst Technique. The DNA concentrations were evaluated spectrophotometrically. The quantity of S. salivarius K12 in the samples was estimated by Quantitative real-time PCR (qPCR) with SRBR Premix Ex TaqTM (2×) (Takara, Japan) and FTC-3000TM Real-Time Quantitative Thermal Cycler (Funglyn, Shanghai, China). All qPCR reactions were run with 3 replicates per DNA. Standard curves were set up by serially diluting plasmid of a pMD18-T vector (Takara, Japan) with the appropriate insert from 10⁷ to 10² target K12 gene copies/µl. The standard curve was obtained using linear regression of threshold cycle numbers (C_T) versus log copy numbers of targets. S. salivarius K12-specific primers were as follows: forward, 5’-AAGGGAGAATGATTGCCATGAA-3’; reverse, 5’-GAGTTTGGACAGTCATCAGTAATAGTTG-3’. As TaqMan® probe, K12Taq was designed as follows: 6-FAM-5’-AGAGGTACAGGTTGGTTTG-3’-MGB ⁵⁰. Q-PCR reactions were performed in a 25 µL reaction volume containing 12.5 µL SRBR Premix Ex Taq™ (2×) (Takara), 1 µl (10µM) each of the forward and reverse primer, 5 µL DNA, 2 µL probe and 3.5 µl of RNAse/DNAse free PCR water. PCR reaction conditions included an initial denaturation at 94 ℃ for 10 min, followed by 45 cycles of 94 ℃ for 1min and 60 ℃ for 105 s. Melting Curve analyses were performed from 60 ℃ to 96 ℃ with increments of 0.1℃ per cycle. The gene copy number in each sample was determined by comparing to Ct values/gene copy number of the standard curve.

Statistical analyses. The statistical analyses for microbiome were performed at the R v3.3 platform (https://www.r-project.org/). Permutational multivariate analysis of variance (Adonis analysis) was performed with the R vegan package, and the Adonis P-value was generated based on 999 permutations. Distance-based redundancy analysis (dbRDA) was performed on these taxonomic composition profiles with the vegan package, based on the Bray-Curtis dissimilarly, and visualized via the R ade4 package. The RA-associated species and modules were identified based on the Wilcoxon rank-sum test. The q value was used to evaluate the false discovery rate for correction of multiple comparisons, and was calculated based on the R fdrtool package. P value < 0.05 or q value < 0.1 was considered statistically significant. Correlation analysis was carried out by using the Spearman's rank correlation statistical measurement system ⁵¹. One RA patient was excluded from the clinical association analyses due to the missing information.

In the experiment part, the differences of arthritis scores and incidence between groups were determined by two-way ANOVA followed by Tukey's multiple comparisons test and Kaplan-Meier analysis with log-rank test, respectively. The difference of cytokines, antibodies, lymphocytes or DNA abundance between two groups was analyzed by Mann-Whitney U nonparametric test. The correlation between variables was evaluated using the Spearman’s rank correlation test. The statistical analyses were performed by the GraphPad prism v8.00 software (GraphPad, San Diego, CA, USA). A two-sided P-value of < 0.05 was considered statistically significant.

Ethics approval and Consent to participate

Informed consent was obtained from all subjects. This study was approved by Peking University People’s Hospital Ethics Committee. All experiments were carried out in accordance with guidelines prescribed by the Animal Care and Use Committee of Peking University People’s Hospital.

Consent for publication

Not applicable

Availability of data and materials

The raw sequencing dataset acquired in this study has been deposited to the China National GeneBank (CNGB) database under the accession codes CNP0002433 and CNP0002434. The authors declare that all other data supporting the findings of this study are available within the article or from the corresponding authors upon reasonable request.

Authors' contributions

ZG.L., J.W., and J.L, conceived and designed the project. J.L., JY.J., YB.J., Y.Z., H.Y., and Y.A. were in charge of participant enrollment and sample collection. RT.Y. and YJ.Z. carried out DNA extraction, PCR, and microbiota sequencing. SH.L., RC.G., YJ.Z., X.Z., and BB.Y. completed all the bioinformatic and statistical analyses of the microbiota. J.L., JY.J., YB.J., XH.X., and WJ.X. performed all the in vitro and in vivo experiments. J.L., SH.L., JY.J., RC.G., XL.S., and JP.G. conducted data acquiring and processing as well as figure preparations. J.L., SH.L., and JY.J. wrote the manuscript with input and edits from ZG.L., J.W., J.H., JJ.Q., and LJ.W. All the authors have revised and approved the manuscript submission.

Acknowledgements

We are grateful to Yun Wang and Yan Ding (Department of Rheumatology and Immunology, Peking University Shougang Hospital, Beijing, China) for their assistance during participant enrollment.

Funding

This study was supported by Guangdong Basic and Applied Basic Research Foundation (2020B1515130005, ZG.L.); the National Natural Science Foundation of China (NSFC, U1903210, ZG.L.; 81901648, J.L.), Beijing Municipal Science & Technology Commission (Z191100006619109, H.Y.), and the General Financial Grant from the China Postdoctoral Science Foundation (2016M600018, J.L.).

Competing interests

The authors declare no competing financial interests.

Nave, H., Gebert, A. & Pabst, R. Morphology and immunology of the human palatine tonsil. Anat Embryol (Berl) 204, 367-373, doi:10.1007/s004290100210 (2001).
Johnston, J. J. & Douglas, R. Adenotonsillar microbiome: an update. Postgrad Med J 94, 398-403, doi:10.1136/postgradmedj-2018-135602 (2018).
Karthikeyan, G. & Guilherme, L. Acute rheumatic fever. Lancet 392, 161-174, doi:10.1016/S0140-6736(18)30999-1 (2018).
Watanabe, H. et al. Comprehensive microbiome analysis of tonsillar crypts in IgA nephropathy. Nephrol Dial Transplant 32, 2072-2079, doi:10.1093/ndt/gfw343 (2017).
Jorgensen, C. et al. In vivo migration of tonsil lymphocytes in rheumatoid synovial tissue engrafted in SCID mice: involvement of LFA-1. Autoimmunity 24, 179-185, doi:10.3109/08916939608995363 (1996).
Kawano, M. et al. Simultaneous, clonally identical T cell expansion in tonsil and synovium in a patient with rheumatoid arthritis and chronic tonsillitis. Arthritis Rheum 48, 2483-2488, doi:10.1002/art.11212 (2003).
Koo, J. et al. Increased Lymphocyte Infiltration in Rheumatoid Arthritis Is Correlated with an Increase in LTi-like Cells in Synovial Fluid. Immune Netw 13, 240-248, doi:10.4110/in.2013.13.6.240 (2013).
Zaiss, M. M., Joyce Wu, H. J., Mauro, D., Schett, G. & Ciccia, F. The gut-joint axis in rheumatoid arthritis. Nat Rev Rheumatol 17, 224-237, doi:10.1038/s41584-021-00585-3 (2021).
Holers, V. M. et al. Rheumatoid arthritis and the mucosal origins hypothesis: protection turns to destruction. Nat Rev Rheumatol 14, 542-557, doi:10.1038/s41584-018-0070-0 (2018).
Integrative, H. M. P. R. N. C. The Integrative Human Microbiome Project. Nature 569, 641-648, doi:10.1038/s41586-019-1238-8 (2019).
Abdollahi-Roodsaz, S. et al. Stimulation of TLR2 and TLR4 differentially skews the balance of T cells in a mouse model of arthritis. J Clin Invest 118, 205-216, doi:10.1172/JCI32639 (2008).
Teng, F. et al. Gut Microbiota Drive Autoimmune Arthritis by Promoting Differentiation and Migration of Peyer's Patch T Follicular Helper Cells. Immunity 44, 875-888, doi:10.1016/j.immuni.2016.03.013 (2016).
Maeda, Y. et al. Dysbiosis Contributes to Arthritis Development via Activation of Autoreactive T Cells in the Intestine. Arthritis Rheumatol 68, 2646-2661, doi:10.1002/art.39783 (2016).
Zhang, X. et al. The oral and gut microbiomes are perturbed in rheumatoid arthritis and partly normalized after treatment. Nat Med 21, 895-905, doi:10.1038/nm.3914 (2015).
Scher, J. U. et al. The lung microbiota in early rheumatoid arthritis and autoimmunity. Microbiome 4, 60, doi:10.1186/s40168-016-0206-x (2016).
Scher, J. U. et al. Expansion of intestinal Prevotella copri correlates with enhanced susceptibility to arthritis. Elife 2, e01202, doi:10.7554/eLife.01202 (2013).
Moentadj, R. et al. Streptococcus species enriched in the oral cavity of patients with RA are a source of peptidoglycan-polysaccharide polymers that can induce arthritis in mice. Ann Rheum Dis, doi:10.1136/annrheumdis-2020-219009 (2021).
Scher, J. U. et al. Periodontal disease and the oral microbiota in new-onset rheumatoid arthritis. Arthritis Rheum 64, 3083-3094, doi:10.1002/art.34539 (2012).
Li, J. et al. Tonsillar Microbiome-Derived Lantibiotics Induce Structural Changes of IL-6 and IL-21 Receptors and Modulate Host Immunity. Adv Sci (Weinh) 9, e2202706, doi:10.1002/advs.202202706 (2022).
Figus, F. A., Piga, M., Azzolin, I., McConnell, R. & Iagnocco, A. Rheumatoid arthritis: Extra-articular manifestations and comorbidities. Autoimmun Rev 20, 102776, doi:10.1016/j.autrev.2021.102776 (2021).
Barbour, A., Wescombe, P. & Smith, L. Evolution of Lantibiotic Salivaricins: New Weapons to Fight Infectious Diseases. Trends Microbiol 28, 578-593, doi:10.1016/j.tim.2020.03.001 (2020).
Kim, S. G. et al. Microbiota-derived lantibiotic restores resistance against vancomycin-resistant Enterococcus. Nature 572, 665-669, doi:10.1038/s41586-019-1501-z (2019).
Wescombe, P. A., Heng, N. C., Burton, J. P. & Tagg, J. R. Something Old and Something New: An Update on the Amazing Repertoire of Bacteriocins Produced by Streptococcus salivarius. Probiotics Antimicrob Proteins 2, 37-45, doi:10.1007/s12602-009-9026-7 (2010).
Wescombe, P. A., Hale, J. D., Heng, N. C. & Tagg, J. R. Developing oral probiotics from Streptococcus salivarius. Future Microbiol 7, 1355-1371, doi:10.2217/fmb.12.113 (2012).
Hyink, O. et al. Salivaricin A2 and the novel lantibiotic salivaricin B are encoded at adjacent loci on a 190-kilobase transmissible megaplasmid in the oral probiotic strain Streptococcus salivarius K12. Appl Environ Microbiol 73, 1107-1113, doi:10.1128/AEM.02265-06 (2007).
Ueno, H., Banchereau, J. & Vinuesa, C. G. Pathophysiology of T follicular helper cells in humans and mice. Nat Immunol 16, 142-152, doi:10.1038/ni.3054 (2015).
Sikder, S., Rush, C. M., Govan, B. L., Alim, M. A. & Ketheesan, N. Anti-streptococcal antibody and T-cell interactions with vascular endothelial cells initiate the development of rheumatic carditis. J Leukoc Biol 107, 263-271, doi:10.1002/JLB.4MA0919-096RR (2020).
Nakatsuji, T. et al. Antimicrobials from human skin commensal bacteria protect against Staphylococcus aureus and are deficient in atopic dermatitis. Sci Transl Med 9, doi:10.1126/scitranslmed.aah4680 (2017).
Iwase, T. et al. Staphylococcus epidermidis Esp inhibits Staphylococcus aureus biofilm formation and nasal colonization. Nature 465, 346-349, doi:10.1038/nature09074 (2010).
Heilbronner, S., Krismer, B., Brotz-Oesterhelt, H. & Peschel, A. The microbiome-shaping roles of bacteriocins. Nat Rev Microbiol 19, 726-739, doi:10.1038/s41579-021-00569-w (2021).
Di Pierro, F. et al. Preliminary pediatric clinical evaluation of the oral probiotic Streptococcus salivarius K12 in preventing recurrent pharyngitis and/or tonsillitis caused by Streptococcus pyogenes and recurrent acute otitis media. Int J Gen Med 5, 991-997, doi:10.2147/IJGM.S38859 (2012).
Masdea, L. et al. Antimicrobial activity of Streptococcus salivarius K12 on bacteria involved in oral malodour. Arch Oral Biol 57, 1041-1047, doi:10.1016/j.archoralbio.2012.02.011 (2012).
Kaci, G. et al. Inhibition of the NF-kappaB pathway in human intestinal epithelial cells by commensal Streptococcus salivarius. Appl Environ Microbiol 77, 4681-4684, doi:10.1128/AEM.03021-10 (2011).
Aletaha, D. et al. 2010 Rheumatoid arthritis classification criteria: an American College of Rheumatology/European League Against Rheumatism collaborative initiative. Arthritis Rheum 62, 2569-2581, doi:10.1002/art.27584 (2010).
Caporaso, J. G. et al. Global patterns of 16S rRNA diversity at a depth of millions of sequences per sample. Proc Natl Acad Sci U S A 108 Suppl 1, 4516-4522, doi:10.1073/pnas.1000080107 (2011).
Kozich, J. J., Westcott, S. L., Baxter, N. T., Highlander, S. K. & Schloss, P. D. Development of a dual-index sequencing strategy and curation pipeline for analyzing amplicon sequence data on the MiSeq Illumina sequencing platform. Appl Environ Microbiol 79, 5112-5120, doi:10.1128/AEM.01043-13 (2013).
Bolyen, E. et al. Reproducible, interactive, scalable and extensible microbiome data science using QIIME 2. Nat Biotechnol, doi:10.1038/s41587-019-0209-9 (2019).
Callahan, B. J. et al. DADA2: High-resolution sample inference from Illumina amplicon data. Nat Methods 13, 581-583, doi:10.1038/nmeth.3869 (2016).
DeSantis, T. Z. et al. Greengenes, a chimera-checked 16S rRNA gene database and workbench compatible with ARB. Applied and environmental microbiology 72, 5069-5072, doi:10.1128/AEM.03006-05 (2006).
Quast, C. et al. The SILVA ribosomal RNA gene database project: improved data processing and web-based tools. Nucleic acids research 41, D590-596, doi:10.1093/nar/gks1219 (2013).
Kim, D., Song, L., Breitwieser, F. P. & Salzberg, S. L. Centrifuge: rapid and sensitive classification of metagenomic sequences. Genome Res 26, 1721-1729, doi:10.1101/gr.210641.116 (2016).
Li, D., Liu, C. M., Luo, R., Sadakane, K. & Lam, T. W. MEGAHIT: an ultra-fast single-node solution for large and complex metagenomics assembly via succinct de Bruijn graph. Bioinformatics 31, 1674-1676, doi:10.1093/bioinformatics/btv033 (2015).
Zhu, W., Lomsadze, A. & Borodovsky, M. Ab initio gene identification in metagenomic sequences. Nucleic Acids Res 38, e132, doi:10.1093/nar/gkq275 (2010).
Li, W. & Godzik, A. Cd-hit: a fast program for clustering and comparing large sets of protein or nucleotide sequences. Bioinformatics 22, 1658-1659, doi:10.1093/bioinformatics/btl158 (2006).
Li, R. et al. SOAP2: an improved ultrafast tool for short read alignment. Bioinformatics 25, 1966-1967, doi:10.1093/bioinformatics/btp336 (2009).
Kanehisa, M., Furumichi, M., Tanabe, M., Sato, Y. & Morishima, K. KEGG: new perspectives on genomes, pathways, diseases and drugs. Nucleic Acids Res 45, D353-D361, doi:10.1093/nar/gkw1092 (2017).
Geng, M., Austin, F., Shin, R. & Smith, L. Covalent Structure and Bioactivity of the Type AII Lantibiotic Salivaricin A2. Appl Environ Microbiol 84, doi:10.1128/AEM.02528-17 (2018).
Brand, D. D., Latham, K. A. & Rosloniec, E. F. Collagen-induced arthritis. Nat Protoc 2, 1269-1275, doi:10.1038/nprot.2007.173 (2007).
Nishikawa, M. et al. Prevention of the onset and progression of collagen-induced arthritis in rats by the potent p38 mitogen-activated protein kinase inhibitor FR167653. Arthritis Rheum 48, 2670-2681, doi:10.1002/art.11227 (2003).
Horz, H. P., Meinelt, A., Houben, B. & Conrads, G. Distribution and persistence of probiotic Streptococcus salivarius K12 in the human oral cavity as determined by real-time quantitative polymerase chain reaction. Oral Microbiol Immunol 22, 126-130, doi:10.1111/j.1399-302X.2007.00334.x (2007).
Zapala, M. A. & Schork, N. J. Multivariate regression analysis of distance matrices for testing associations between gene expression patterns and related variables. Proc Natl Acad Sci U S A 103, 19430-19435, doi:10.1073/pnas.0609333103 (2006).

No competing interests reported.

Download PDF

Version 1

posted

You are reading this latest preprint version

Intra-genus dysbiosis of Streptococcus in tonsillar microbiota is associated with host immune features in rheumatoid arthritis

Status:

Version 1

Abstract

Background

Results

Conclusions

Figures

Introduction

Results

Dysbiosis of tonsillar microbiota in RA patients

RA tonsillar microbiota varied with disease activity

Discussion

Materials and methods

Declarations

References

Additional Declarations

Supplementary Files

Status:

Version 1