Autoimmunity-Associated Gut Commensals Modulate Gut Permeability and Immunity in Humanized Mice (original) (raw)

Abstract

Objective

Although the etiology of rheumatoid arthritis (RA) is unknown, recent studies have led to the concept that gut dysbiosis may be involved in onset. In this study, we aimed to determine if human gut commensals modulate the immune response and gut epithelial integrity in DQ8 mice.

Methods

DQ8 mice were orally gavaged with RA-associated (Eggerthella lenta or Collinsella aerofaciens) and non-associated (Prevotella histicola or Bifidobacterium sp.) on alternate days for 1 week in naïve mice. Some mice were immunized with type II collagen and oral gavage continued for 6 weeks and followed for arthritis. Epithelial integrity was done by FITC-Dextran assay. In addition, cytokines were measured in sera by ELISA and various immune cells were quantified using flow cytometry.

Results

Gut permeability was increased by the RA-associated bacteria and was sex and age-dependent. In vivo and in vitro observations showed that the RA-non-associated bacteria outgrow the RA-associated bacteria when gavaged or cultured together. Mice gavaged with the RA-non-associated bacteria produced lower levels of pro-inflammatory MCP-1 and MCP-3 and had lower numbers of Inflammatory monocytes CD11c+Ly6c+, when compared to controls. E. lenta treated naïve mice produce Th17 cytokines.

Conclusions

Our studies suggest that gut commensals influence immune response in and away from the gut by changing the gut permeability and immunity. Dysbiosis helps the growth of RA-associated bacteria and reduces the beneficial bacteria.

INTRODUCTION

Rheumatoid arthritis (RA) is a debilitating disease of unknown etiology that leads to joint destruction. According to a report from the American Autoimmune Related Diseases Association in 2011, an estimated 50 million people in the USA are affected by autoimmune diseases. The incidence of autoimmune diseases is increasing, especially in developed countries. Among these autoimmune diseases, RA and type 1 diabetes occur with the highest prevalence. RA is an organ-specific autoimmune disease affecting 1% of the general population and around 2% of the military population in the USA. A recent study has reported an increased prevalence of RA in U.S. veterans.1

RA is a complex disease that requires an interaction between genetic and environmental factors.2 Among genetic factors, the increased presence of certain HLA class II alleles in patients with RA suggests an association between class II alleles and RA. While various ethnic populations show variability in HLA-association, an increased presence of HLA-DRB1*0401 in RA patients is almost universal. Based on the variable class II associations with RA, a “shared epitope hypothesis” was proposed where an allele-sharing amino acids at the third hypervariable region (HVR) with HLA-DRB1*0401 were associated with RA.3 On the other hand, alleles sharing HVR3 amino acids with DRB1*0402 were considered to be non-associated with RA. Alleles of DRB1 loci occur in linkage with the alleles of DQ loci and form a haplotype. In humans, one haplotype is inherited from each parent. DR4 occurs in linkage with DQ8 thus DR4-DQ8 is a susceptible haplotype for RA.4

Besides genetic factors, many environmental factors have been suggested to be involved in the onset of RA. The most studied of the environmental factors is smoking.5,6 Smoking has been suggested to enhance autoantibody response to Vimentin in patients with RA. However, a study with humanized mice showed that smoking augmented the immune response to Vimentin in both arthritis-resistant and susceptible mice.7 Genetic and environmental factors influence gut microbiota suggesting intestinal microbiota as a common link for various factors involved in RA pathogenesis. Gut microbiota’s involvement in predisposition to develop RA has regained interest due, in part, to the recent observations suggesting an association of gut commensals with RA.8–10 While the etiology of RA remains unknown, recent studies describing the contribution of gut microbiota in adaptive immune response have led to the concept that interaction between the host microbiome and genetic factors influences autoimmunity.8,10

Collagen-induced arthritis (CIA) is an animal model where certain strains of mice develop arthritis when immunized with type II collagen (CII).11,12 Cellular and humoral responses to CII are also present in RA patients, providing evidence that the immune response to CII plays a significant role in the pathogenesis of this disease. We generated mice lacking endogenous class II molecules and expressing the RA-associated HLA-DQ8 gene.13,14 Immunization of DQ8 mice with CII leads to a robust antigen-specific cellular and humoral response and mice develop severe CIA with histopathological changes as observed in humans.14 Mice expressing DRB1*0401 and DRB1*0402 transgene but lacking endogenous class II molecules have been generated. Similar to humans, expression of *0401 made mice susceptible to CIA while *0402 did not.15

Transgenic mice helped to answer the question about the endogenous factors that distinguish hosts carrying arthritis-susceptible and resistant genes. Naïve *0401 mice do not have any age-dependent difference in their gut microbiota suggesting HLA genes determine microbial colonization.16,17 On the other hand, *0402 mice show a dynamic change in microbial composition when they age. In addition, aged *0401 female mice show reduced gut epithelial integrity.17 These observations in mouse model suggested a role of gut microbiota in arthritis. This was further confirmed in another report where oral gavage with a human gut-derived commensal, Prevotella histicola, protected DQ8 mice from developing arthritis by enhancing expression of tight junction proteins.18

These reports were followed by investigations of gut microbiota in patients with RA. Patients with RA when compared with first degree relatives and random healthy controls showed alterations in fecal microbial profile.10 Predictive modeling showed an increase in the abundance of rare lineage taxa, Eggerthella lenta and Collinsella aerofaciens, belonging to phylum Actinobacteria with a decrease in Faecalibacterium. Gavaging C. aerofaciens to CIA-susceptible DQ8 mice augmented antigen-specific cellular and humoral response and led to severe arthritis compared to non-treated mice.10 This led to the hypothesis that autoimmunity-associated microbes enhance disease severity by modulating systemic immune response.

In this report, DQ8 mice were used to test how RA-associated gut microbes modulate gut permeability and immune response. The observations suggested that the gut microbiota may contribute to disease pathogenesis by modulating intestinal epithelial permeability and systemic immune response.

METHODS

Transgenic Mice

HLA-DQ8 transgenic mice were generated as described previously.4,15 All mice used in this study lacked endogenous class II molecules (AEo) and expressed DQB1*0302/DQA1*0301 (DQ8.AEo) on the B6/129 background.13 Both sexes (5–12 weeks of age) of DQ8 mice were used. All mice were fed standard mouse diet. Mice were bred and maintained in the pathogen-free Mouse Colony by the Department of Comparative Medicine at the Mayo Clinic, Rochester, MN, USA in accordance with the Institutional Animal Use and Care Committee. Transgene negative littermates were used as controls and all experiments were carried out with the approval of the Animal Use and Care Committee.

Induction and Evaluation of CIA

CIA was induced by injecting mice intradermally in the tail with type II collagen (CII) (100 μg of CII emulsified 1:1 with complete Freund’s adjuvant) as previously described.19 Mice were monitored for the onset and progression of arthritis. Arthritis was scored using an established grading system wherein each paw was scored with a range 0–3; 0 = no swelling, 1 = 1or 2 digit swollen, 2 = 2 or more digits swollen and 3 = swollen paw. The mean arthritic score was determined using arthritic animals only.

Treatment With Commensal Bacterium

Disease associated bacteria E. lenta (ATCC 25559) and _C. aerofaciens (_ATCC 25986) were purchased from ATCC, USA. RA-non-associated bacteria, P. histicola and Bifidobacterium sp., were isolated from gut biopsy of a human subject. Bacteria were cultured in tryptic soy broth (TSB) at pH 7.0 ± 0.2 and incubated in an anaerobic condition using a Bactron anaerobic chamber (SHEL LAB, USA) at 37°C for 48 hours, containing the anaerobic gas mixture (N2 90%, H2 5% and Co2 5%). After 48 hours of incubation, the prepared bacterial cultures were orally gavaged to 8–12 weeks old naïve transgenic mice (100 μL/mice at 109 CFU mL−1) on alternate days up to 6 weeks (N = 5 mice/group). Another set of bacterial treatment experiment was carried out with CIA-induced transgenic mice (n = 5 mice/group). Controls were maintained by gavaging phosphate buffer saline (PBS) (N = 5 mice/group). Intestinal permeability was examined in all mice once before starting the experiment (considered naïve status of the mice) and after stopping the bacterial treatment (treatment effect).

Bacterial Co-existence

In order to determine the growth and co-existence of all four selected bacteria together in vitro, equal quantity (~103 CFU ml−1) of each bacterium was inoculated into 20 mL of TSB. Cultures were incubated as mentioned above. Each bacterium was quantified by using quantitative polymerase chain reaction (qPCR). Also, the colony forming unit (CFU) enumeration was done by using serial dilution method. To analyze in vivo co-existence of the four bacteria, culture with equal concentrations (~108 CFU ml−1) of each bacterium was mixed together (Final mixed culture concentration was ~109 CFU ml−1) and gavaged to transgenic mice every other day up to two weeks (n = 4/group). Fecal samples were collected at the end of the experiment to quantify each selected bacterium. Triplicates were maintained at all stages of the experiments.

Quantitative PCR

Bacterial quantification for in vitro and in vivo co-existence was evaluated using qPCR technique. Bacteria cultured in broth (in vitro) and fecal samples from mice (in vivo) were subjected to genomic DNA isolation using Power-DNA isolation kit (QIAGEN, USA). The qPCR reactions were performed as reported previously using RT2 SYBR Green qPCR Master mix (QIAGEN, USA)20. List of specific primers used in the study is given in Table I.

TABLE I.

qPCR Primers Used for Specific Bacteria

TABLE I.

qPCR Primers Used for Specific Bacteria

Intestinal Permeability

Intestinal permeability was determined using 4 kDa FITC-labeled dextran (Sigma, USA). Sera were collected from mice before bacterial gavage. Mice were deprived of food for 3 hours, then gavaged with FITC–labeled dextran (0.6 mg/g body weight). Three hours later, mice were bled and serum collected with a tail nick. FITC-dextran was determined at 490 nm in sera collected before gavage and after bacterial gavage. In a separate group of experiments, gut permeability was analyzed for age and sex-specific differences in mice bred and reared in specific pathogen-free colony (SPF) and those moved from SPF to conventional colony (CC) at 4 weeks of age. Gut permeability was assessed at the age of week 5 and week 9 (N = 5/group).

Flow Cytometry

Mice were characterized for the presence of the transgene DQ8 before experiments. IVD12 (anti-DQ) was used to analyze the expression of DQ in transgenic mice by flow cytometry. Dendritic cells (DCs) and B cells were visualized using conjugated antibodies for CD11b, CD11c, Ly6c and B220 (BD Biosciences, CA). Collected immune cells were pooled from 2 mice/strain for each experiment and repeated 2–3 times.

CD4 and DC Culture

Mice were immunized with 200ug of CII emulsified 1:1 in CFA (Difco) intradermally at the base of the tail. For sorted CD4 and DCs, mice were sacrificed as described.19 CD4 cells (5 × 106) were sorted from lymph nodes of CII-primed mice that were treated with P. histicola, Bifidobacterium sp. or PBS only. Sorted CD4 cells were cultured in vitro in the presence or absence of the antigen and splenic CD11c+ DCs (5 × 105). The assay was done with combinations of DCs and CD4 cells from all treated mice. Cells were challenged in vitro with 100 μg/ml of the CII and incubated for 48 h at 37°C. Supernatants were stored at −80°C for measuring cytokines.

Cytokines

Cytokines were measured using the MILLIPLEX MAP MOUSE CYTOKINE/CHEMOKINE MAGNETIC BEAD PANEL (Millipore Sigma, USA) as per manufacturer’s instructions. The results were analyzed using Bio-Plex manager 2.0 software (Bio-Rad Laboratories, USA) as per recommendations.

Statistical Analysis

The comparison between different groups in gut permeability, cytokines, and various cells was compared using non-parametric Student’s _T_-test. p < 0.05 was considered significant.

RESULTS

RA-Associated Gut Bacteria Enhance Gut Permeability

Patients with RA have an expansion of species belonging to Phylum Actinobacteria and taxa E. lenta and C. aerofaciens.10 However, probiotic-like bacteria, Faecalibacterium Prausnitzii, was reduced. This suggested that E. lenta and C. aerofaciens are either pro-inflammatory or are abundant due to disease activity. This was tested in CIA model in humanized mice where arthritis was observed to be augmented when mice are gavaged with C. aerofaciens, supporting its role in disease severity.

To determine if RA-associated gut microbes change gut permeability, naïve DQ8 mice were gavaged with PBS or gut commensals, E. lenta and C. aerofaciens, and gut permeability was tested in sera by measuring FITC-Dextran (Fig. 1A). Mice gavaged with E. lenta showed an increase in gut permeability as compared to PBS gavaged mice, similar to that shown for C. aerofaciens before10. Bifidobacterium sp. and P. histicola were used as control commensals since both have been shown to have probiotic properties and are able to provide protection from autoimmune conditions.23 The observations showed that gavaging mice with Bifidobacterium sp. did not alter the gut permeability significantly (Fig. 1A). Similar results were obtained with P. histicola, another gut commensal that was shown to be protective against arthritis.18 Together these data suggest that gut permeability may play a role in disease onset.

FIGURE 1.

Gut permeability in DQ8 mice gavaged with various bacteria shows that RA-associated bacteria augment gut permeability. (A) Bifidobacterium sp. gavage led to a decrease in gut permeability as compared to E. lenta and control. (B) Sex-dependent gut permeability, male (M) CIA mice increased when treated with E. lenta. But in female (F) CIA mice did not show any response even after E. lenta treatment. (C) In both SPE and CC, mice show significant decrease in gut permeability as they age. Sex-specific increase in gut permeability was observed in mice in CC when they were 9 weeks old. Females had higher gut permeability than males. (D) RA-associated bacteria, C. aerofaciens, significantly increase gut permeability in naïve females compared to males, p < 0.01. Mice gavaged with Bifidobacterium sp. did not show any sex-specific effect of gut permeability.

To further determine if E. lenta enhances gut permeability in mice induced for arthritis, mice were gavaged with E. lenta or PBS and then immunized with CII. Gut permeability was tested 6 weeks later. Overall there was a trend towards increased gut permeability in mice gavaged with E. lenta although the difference between mice treated with E. lenta or PBS control was not significant. However, when the data were analyzed according to the sex of animals, females clearly showed a much more significant increase in gut permeability when compared to males treated with E. lenta (Fig. 1B).

Gut Permeability in Susceptible Mice Is Age and Sex-Dependent

Gut microbial composition is dependent on the gut location, environmental and genetic factors. To determine if specific pathogen-free (SPF) and conventional colonies (CC) change the gut permeability, naïve mice at the ages of 5 weeks and 9 weeks in SPF and CC were tested (Fig. 1C). Interestingly, male mice at 5 weeks had higher gut permeability as compared to 9 weeks, both in SPF and CC, p < 0.04 and p < 0.004, respectively. None of the differences was significant for females at specific ages though females showed a trend of higher gut permeability compared to males. Females in a conventional colony at 9 weeks showed a significant increase in gut permeability as compared to males of the same age, p < 0.02.

To determine if the RA-associated gut commensals contribute to sex-bias in augmenting gut permeability, naïve DQ8 mice were orally gavaged with C. aerofaciens or PBS and gut permeability was tested by FITC-Dextran in sera. Bifidobacterium sp. was used as a control commensal. Both naïve and C. aerofaciens gavaged mice showed significant sex-dependent differences in gut permeability, p < 0.008, while mice gavaged with Bifidobacterium sp. did not show any significant difference between sexes (Fig. 1D).

Gut Commensals Modulate Gut Immunity

Previous work had shown an increase in regulatory T cells in mice treated with P. histicola suggesting CD4 cells regulate inflammatory response.18 To determine if commensals impact arthritis development by modulating local and extra-intestinal immune response, mice were immunized with CII and gavaged with microbes associated with probiotic-like features, Bifidobacterium sp. and P. histicola or PBS (Control). Splenic CD11c+ DCs were cultured in vitro with sorted splenic CD4 cells harvested from mice in various combinations. CD4 cells from mice gavaged with bacteria were co-cultured with self or control DCs and MCP-1 (CCl2) and MCP-3 (CCL7) were measured (Fig. 2A). MCP-1 and MCP-3 are both involved during inflammation and are increased in patients with RA.24 CD4 cells isolated from P. histicola and Bifidobacterium sp. treated mice cultured with DCs of self or control mice produced low levels of both chemokines. On the other hand, no significant differences in production of MCP-1 and MCP-3 were observed when CD4 cells of control mice (gavaged with PBS only) were cultured with DCs isolated from self or mice gavaged with P. histicola and Bifidobacterium sp. However when CD4 cells from P. histicola or Bifidobacterium sp. gavaged mice were cultured with DCs from same mice, significantly lower MCP-1 and MCP-3 were produced, (MCP-1, P. histicola vs control, p < 0.01 and Bifidobacterium sp. vs control, p < 0.03; MCP-3 P. histicola vs control, p < 0.02 and Bifidobacterium sp. vs control, p < 0.06). Also, lower levels of MCP-1 were observed when compared between cultures of DCs from P. histicola gavaged mice with CD4 cells from control mice and CD4 cells from P. histicola gavaged mice, p < 0.04. Similar, comparison was not significant for MCP-3. On the other hand, DCs from Bifidobacterium sp. gavaged mice produced lower MCP-3 when cultured with CD4 from Bifidobacterium sp. gavaged as compared to CD4 cells from control mice, p < 0.03. These data suggest that DCs modulate gut immune response by producing MCP-1 and MCP-3.

FIGURE 2.

Gut bacteria regulate cytokine response via CD4 cells and DCs. DQ8 mice immunized with type II collagen were gavaged with P. histicola or Bifidobacterium sp. Sorted CD4 cells were cultured with DCs in various combinations and supernatants were used to measure MCP-1 and MCP3 chemokines. Control DCs cultured with P. histicola CD4 vs Control CD4;*p = 0.03 and #p = 0.01, Bifidobacterium sp. DCs cultured with Bifidobacterium sp. CD4 vs control CD4, **p = 0.03.

As MCP-1 and MCP-3 regulate chemotaxis of monocytes/macrophages, the presence of monocytes was compared between the control and P. histicola and Bifidobacterium sp. gavaged mice. Monocytes that are considered to be pro-inflammatory express CD11c+Ly6c+. To determine if probiotic-like commensals can control differentiation of these inflammatory cells, DQ8 mice immunized with CII were gavaged with PBS and commensals, Bifidobacterium or P. histicola, and CD11c+Ly6c cells were enumerated (Fig. 3A). Mice gavaged with P. histicola had much lower numbers of inflammatory CD11c+Ly6c+ cells, supporting the above data, p < 0.05. Prevotella histicola has been shown to be protective for arthritis in humanized mice, although some mice still develop mild arthritis.18 A comparison of the CD11b and CD11c cells in arthritic mice gavaged with PBS or P. histicola showed that arthritic mice in both groups had similar numbers of CD11b and. On the other hand, P. histicola treated arthritis-free mice had a higher percent of CD11b and CD11c cells as compared to control arthritic mice, p < 0.06 and p < 0.001, respectively (Fig. 3B). These data support previous observations with P. histicola that have shown an increase in myeloid suppressors and regulatory DCs in treated mice compared to non-treated mice.

FIGURE 3.

Probiotic-like commensal gavaged DQ8 mice have less pro-inflammatory monocytes than non-treated mice. (A) Pro-inflammatory, CD11c+Ly6c+ cells were reduced in numbers in mice gavaged with RA non-associated bacteria. (B) Antigen-presenting cells, CD11b, CD11c and B220+ cells were enumerated in control arthritic mice (CIA), mice gavaged with P. histicola and developed CIA (P. histicola CIA+) and not developed CIA (P. histicola CIA-).

Since arthritis is dependent on a humoral response, mice gavaged with PBS or P. histicola were used to compare B cell numbers. Interestingly, arthritic mice in both groups, PBS and P. histicola gavaged, had similar B cells. Non-arthritic P. histicola gavaged mice had lower numbers of B cells, though differences were non-significant.

Co-existence of Commensal Bacteria

Studies in RA patients have shown an expansion of E. lenta and C. aerofaciens. The next question addressed in this study was to understand co-existence and dominance of the microbes by culturing the four selected bacteria together in vitro. The observations showed that P. histicola dominated in growth when all four bacteria were cultured together (Fig. 4). Dominance of P. histicola was followed by Bifidobacterium sp. Growth of C. aerofaciens and E. lenta was suppressed by P. histicola and Bifidobacterium sp. In order to determine if this was true in vivo, naïve DQ8 mice were gavaged with the mixture of four bacteria as per methods. Our observations showed that in vivo growth of bacteria followed the trend observed in vitro with P. histicola dominating the growth followed by Bifidobacterium sp. (Fig. 4). However, RA-associated bacteria were not detectable in fecal samples. The detectable limit of the qPCR experiment was 103 CFU only.20 These data confirmed that RA non-associated bacteria can suppress the growth of RA-associated bacteria both in vitro and in vivo.

FIGURE 4.

Co-existence of commensal bacteria showed the dominance of RA non-associated bacteria. (A) Coexistence of commensal bacteria when cultured in vitro in TSB media was measured by qPCR. (B) Co-existence of commensal bacteria in vivo measured by qPCR analysis using fecal samples of DQ8 mice gavaged with the mixture of all bacteria. In both in vitro and in vivo conditions, P. histicola dominated while RA-associated bacteria C. aerofaciens and E. lenta were not detected (ND) in fecal samples (p ≤ 0.05).

RA-Associated E. lenta Causes Production of Antigen-Specific Proinflammatory Cytokines

Naïve DQ8 mice were gavaged with PBS or E. lenta for 2 weeks. Spleen cells were harvested after 2 weeks and cultured in vitro with CII. Supernatants were used for measuring proinflammatory cytokines, IL-6, IL-21, and IL-23. Splenocytes from naïve mice gavaged with PBS only did not produce measurable cytokines in response to CII while mice gavaged with E. lenta produced all the three proinflammatory cytokines (Fig. 5).

FIGURE 5.

RA-associated bacteria E. lenta generates inflammatory response in naïve DQ8 mice. Eggerthella lenta gavage in DQ8 mice led to secretion of high levels of IL-6, IL-21, and IL-23 as tested in sera. Control mice produced very low levels of tested interleukins.

DISCUSSION

The observations presented in this manuscript support the role of gut microbiota in RA. Further, data show that the pathogenic gut commensals when expanded as shown in patients might contribute to disease severity by augmenting gut permeability. Epithelial integrity is important for maintaining the barrier between the luminal contents and extra-intestinal organs. Pathobionts can lead to the generation of macrophages that produce inflammatory chemokines causing influx of pro-inflammatory immune cells. This is supported by an increase in inflammatory macrophages in mice gavaged with E. lenta while P. histicola, that has been shown to protect from arthritis, led to decrease in these cells. Thus based on the microbial composition, gut immune system can be pro-inflammatory under certain conditions. Patients with RA have low microbial diversity and dysbiosis with expansion of certain taxa.10 The observations with E. lenta and C. aerofaciens support the contention that these gut microbes contribute to inflammation. This is further supported by recently published data showing a human gut-derived commensal protects humanized mice from arthritis by regulating intestinal and extra-intestinal immunity.18

The gut commensals, P. histicola and Bifidobacterium sp. change gut immunity by CD4 cells as DCs from P. histicola – gavaged mice cultured with CD4 cells from P. histicola – gavaged or control mice showed that the former produce lower proinflammatory cytokines. A similar phenomenon was observed with Bifidobacterium sp. suggesting CD4 cells regulate inflammatory response to gut commensals. This is further supported by the presence of CD11b and CD11c cells in arthritic control and P. histicola gavaged mice. Some mice gavaged with P. histicola develop arthritis have a similar percent of CD11b and CD11c cells as controls while P. histicola gavaged mice that do not develop arthritis, have much higher numbers. This study supports the previous results that gut-derived commensal regulates immune response and further shows that inflammatory cytokines and monocytes are also decreased in mice gavaged with P. histicola.18

The intestinal immune response is dependent on the microbial composition. A single microbe, segmented filamentous bacteria have been shown to lead to the development of arthritis.25 However, microbes do not exist in isolation rather they colonize based on the local conditions. All microbes grow at different rates, but generally there are checkpoints in place which help to maintain homeostasis. However, any disruption or alteration can cause certain bacteria to grow due to favorable conditions. Thus alterations in the growth of one microbe may affect the growth of another. The RA non-associated Bifidobacterium sp. and P. histicola outgrew both C. aerofaciens and E. lenta in vitro and in vivo. These data suggest that the beneficial bacteria such as Bifidobacterium sp. and P. histicola can reduce gut permeability, by reducing the growth of pathogenic bacteria. Whereas when RA-associated bacteria E. lenta and C. aerofaciens become abundant during certain conditions, it can cause dysbiosis and reduce the growth of beneficial bacteria, thus inducing increase in gut permeability and disease onset. The data presented here supports our previous study showing an association with E. lenta and C. aerofaciens with disease severity in RA patients.10 Furthermore, E. lenta gavaged mice produced Th17 cytokines in response to CII in vitro. However, the mechanism to explain why naïve E. lenta gavaged mice generate an inflammatory response to CII needs further evaluation. It could be molecular mimicry with CII or treatment with E. lenta can lead to an expansion of other microbes that have an epitope with structural homology to CII. One can speculate that similar to in vitro observations, deficiency or lower numbers of probiotic-like commensal may lead to an expansion of pathogenic/opportunistic bacteria in vivo. This may also explain an increase in gut permeability caused by an expansion of pathogenic bacteria and production of inflammatory cytokines as observed in naïve mice gavaged with E. lenta.

Gut permeability in humanized mice was sex-dependent with female mice showing a trend to have an increase in permeability as they age. Interestingly, at 9 weeks of age, gut permeability showed a decrease from 5 weeks old mice in contrast to the previous data showing increased permeability as DR4 female mice aged.17 This could be due to the fact that mice in that study were older than 16 weeks. Increase in gut permeability in females as observed with gavage with RA-associated C. aerofaciens to naïve mice suggests that expansion of certain pathogenic microbes can have a sex-specific impact. A similar sex-specific influence of E. lenta was observed in mice immunized with CII. During aging, changes in immune response and gut microbiota have been suggested. In women, changes in hormones during menopause can impact gut microbiota. The median age for developing RA is around 50 years. According to these and previous observations, as women age, gut permeability might increase and this might also explain why women are more likely to develop RA. These observations provide an insight into the role of gut microbiota in gut permeability and disease susceptibility.

CONCLUSIONS

The observations suggest a role of gut microbiota in regulating gut immunity. Changes in pro-inflammatory cytokines and gut permeability by a single microbe suggest that alterations can result in inflammation and chemotaxis of inflammatory cells leading to changes in gut permeability and autoimmunity. Gut microbiota may provide an option to modulate intestinal and extra-intestinal immune response and provide an individualized treatment option.

Previous Presentation

Presented as an oral talk at the 2017 Military Health System Research Symposium.

Abstract

# MHSRS-17-1039

Platform

Utilization of the Microbiome for Diagnostics, Treatment, & Sustainment of Health & Performance

Location: 27–30 August 2017, Kissimmee, FL, USA.

Funding

This study was supported by grants from the Department of Defense, W81XWH-10-1-0257 and W81XWH-15-1-0213 and the Center of Individualized Medicine, Mayo Clinic. This supplement was sponsored by the Office of the Secretary of Defense for Health Affairs.

References

1

Jones

KA

,

Granado

NS

,

Smith

B

, et al. :

A prospective study of lupus and rheumatoid arthritis in relation to deployment in support of Iraq and Afghanistan: the millennium cohort study

.

Autoimmune Dis

2011

;

2011

:

741267

.

2

Taneja

V

:

Arthritis susceptibility and the gut microbiome

.

FEBS Lett

2014

;

588

(

22

):

4244

–

9

.

3

Gregersen

PK

,

Silver

J

,

Winchester

RJ

:

The shared epitope hypothesis. An approach to understanding the molecular genetics of susceptibility to rheumatoid arthritis

.

Arthritis Rheum

1987

;

30

(

11

):

1205

–

13

.

4

Taneja

V

,

David

CS

:

Role of HLA class II genes in susceptibility/resistance to inflammatory arthritis: studies with humanized mice

.

Immunol Rev

2010

;

233

(

1

):

62

–

78

.

5

Ummarino

D

:

Rheumatoid arthritis: smoking influences autoimmunity to vimentin

.

Nat Rev Rheumatol

2016

;

12

(

11

):

624

.

6

Wegner

N

,

Lundberg

K

,

Kinloch

A

, et al. :

Autoimmunity to specific citrullinated proteins gives the first clues to the etiology of rheumatoid arthritis

.

Immunol Rev

2010

;

233

(

1

):

34

–

54

.

7

Bidkar

M

,

Vassallo

R

,

Luckey

D

,

Smart

M

,

Mouapi

K

,

Taneja

V

:

Cigarette smoke induces immune responses to vimentin in both, arthritis-susceptible and -resistant humanized mice

.

PLoS One

2016

;

11

(

9

):

e0162341

.

8

Zhang

X

,

Zhang

D

,

Jia

H

, et al. :

The oral and gut microbiomes are perturbed in rheumatoid arthritis and partly normalized after treatment

.

Nat Med

2015

;

21

(

8

):

895

–

905

.

9

Vaahtovuo

J

,

Munukka

E

,

Korkeamaki

M

,

Luukkainen

R

,

Toivanen

P

:

Fecal microbiota in early rheumatoid arthritis

.

J Rheumatol

2008

;

35

(

8

):

1500

–

5

.

10

Chen

J

,

Wright

K

,

Davis

JM

, et al. :

An expansion of rare lineage intestinal microbes characterizes rheumatoid arthritis

.

Genome Med

2016

;

8

(

1

):

43

.

11

Taneja

V

,

David

CS

:

HLA class II transgenic mice as models of human diseases

.

Immunol Rev

1999

;

169

:

67

–

79

.

12

Taneja

V

,

David

CS

:

Autoimmunity versus tolerance: analysis using transgenic mice

.

Hum Immunol

2000

;

61

(

12

):

1383

–

9

.

13

Taneja

V

,

Taneja

N

,

Behrens

M

,

Griffiths

MM

,

Luthra

HS

,

David

CS

:

Requirement for CD28 may not be absolute for collagen-induced arthritis: study with HLA-DQ8 transgenic mice

.

J Immunol

2005

;

174

(

2

):

1118

–

25

.

14

Nabozny

GH

,

Baisch

JM

,

Cheng

S

, et al. :

HLA-DQ8 transgenic mice are highly susceptible to collagen-induced arthritis: a novel model for human polyarthritis

.

J Exp Med

1996

;

183

(

1

):

27

–

37

.

15

Taneja

V

,

Behrens

M

,

Mangalam

A

,

Griffiths

MM

,

Luthra

HS

,

David

CS

:

New humanized HLA-DR4-transgenic mice that mimic the sex bias of rheumatoid arthritis

.

Arthritis Rheum

2007

;

56

(

1

):

69

–

78

.

16

Gomez

A

,

Luckey

D

,

Taneja

V

:

The gut microbiome in autoimmunity: sex matters

.

Clin Immunol

2015

;

159

(

2

):

154

–

62

.

17

Gomez

A

,

Luckey

D

,

Yeoman

CJ

, et al. :

Loss of sex and age driven differences in the gut microbiome characterize arthritis-susceptible 0401 mice but not arthritis-resistant 0402 mice

.

PLoS One

2012

;

7

(

4

):

e36095

.

18

Marietta

EV

,

Murray

JA

,

Luckey

DH

, et al. :

Suppression of inflammatory arthritis by human gut-derived Prevotella histicola in humanized mice

.

Arthritis Rheumatol

2016

;

68

(

12

):

2878

–

88

.

19

Taneja

V

,

Behrens

M

,

Mangalam

A

,

Griffiths

MM

,

Luthra

HS

,

David

CS

:

New humanized HLA-DR4-transgenic mice that mimic the sex bias of rheumatoid arthritis

.

Arthritis Rheum

2007

;

56

(

1

):

69

–

78

.

20

Balakrishnan

B

,

Luckey

D

,

Marietta

E

, et al. :

Development of a real-time PCR method for quantification of Prevotella histicola from gut

.

Anaerobe

2017

;

48

:

37

–

41

.

21

Png

CW

,

Linden

SK

,

Gilshenan

KS

, et al. :

Mucolytic bacteria with increased prevalence in IBD mucosa augment in vitro utilization of mucin by other bacteria

.

Am J Gastroentrol

2010

;

105

:

2420

–

8

.

22

Lyra

A

,

Forssten

S

,

Rolny

P

, et al. :

Comparison of bacterial quantities in left and right colon biopsis and faeces

.

World J Gastroentrol

2012

;

18

(

32

):

4404

–

11

.

23

Balakrishnan

B

,

Taneja

V

:

Microbial madulation of the gut microbiome for treating autoimmune diseases

.

Epert Rev Gastroentrol Hepatol

2018

;

12

(

10

):

985

–

96

.

24

Szekanecz

Z

,

Vegvari

A

,

Szabo

Z

,

Koch

AE

:

Chemokines and chemokine receptors in arthritis

.

Front Biosci (Schol Ed)

2010

;

2

:

153

–

67

.

25

Wu

HJ

,

Ivanov

II

,

Darce

J

, et al. :

Gut-residing segmented filamentous bacteria drive autoimmune arthritis via T helper 17 cells

.

Immunity

2010

;

32

(

6

):

815

–

27

.

Author notes

Mayo Clinic and VT have a financial interest related to one of the product, Prevotella histicola, referenced in this manuscript.

© Association of Military Surgeons of the United States 2019. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com.