Research ArticleImmunologyInfectious diseaseMicrobiology Open Access | 10.1172/JCI202262

Fang R. Zhao,1 Maksim Kleverov,2 Emma S. Winkler,1 Russell B. Williams,3 Hana Janova,1 Lindsay Droit,2,4 Leran Wang,4 Ting-ting Li,5 Leah Heath,4 Ana Jung,4 Matthias Mack,6 Megan T. Baldridge,1,4 Thaddeus S. Stappenbeck,7 Larissa B. Thackray,1 Chyi-Song Hsieh,1 Scott A. Handley,2,4 Chun-Jun Guo,5 Michael A. Fischbach,8 Maxim N. Artyomov,2 and Michael S. Diamond1,2,9

1Department of Medicine and

2Department of Pathology and Immunology, Washington University School of Medicine, St. Louis, Missouri, USA.

3Bioanalytical Chemistry Facility, Donald Danforth Plant Science Center, St. Louis, Missouri, USA.

4The Edison Family Center for Genome Sciences and Systems Biology, Washington University School of Medicine, St. Louis, Missouri, USA.

5Weill Cornell Medicine, Cornell University, New York, New York, USA.

6Department of Nephrology, University Hospital Regensburg, Regensburg, Germany.

7Department of Inflammation and Immunity, Cleveland Clinic, Cleveland, Ohio, USA.

8Department of Bioengineering and ChEM-H, Stanford University, Stanford, California, USA.

9Department of Molecular Microbiology, Washington University School of Medicine, St. Louis, Missouri, USA.

Address correspondence to: Michael S. Diamond, Department of Medicine, Box 8051, Washington University School of Medicine, 660 S. Euclid Ave, St. Louis, Missouri 63110, USA. Email: mdiamond@wustl.edu.

Find articles by Zhao, F. in: PubMed | Google Scholar |

1Department of Medicine and

2Department of Pathology and Immunology, Washington University School of Medicine, St. Louis, Missouri, USA.

3Bioanalytical Chemistry Facility, Donald Danforth Plant Science Center, St. Louis, Missouri, USA.

4The Edison Family Center for Genome Sciences and Systems Biology, Washington University School of Medicine, St. Louis, Missouri, USA.

5Weill Cornell Medicine, Cornell University, New York, New York, USA.

6Department of Nephrology, University Hospital Regensburg, Regensburg, Germany.

7Department of Inflammation and Immunity, Cleveland Clinic, Cleveland, Ohio, USA.

8Department of Bioengineering and ChEM-H, Stanford University, Stanford, California, USA.

9Department of Molecular Microbiology, Washington University School of Medicine, St. Louis, Missouri, USA.

Address correspondence to: Michael S. Diamond, Department of Medicine, Box 8051, Washington University School of Medicine, 660 S. Euclid Ave, St. Louis, Missouri 63110, USA. Email: mdiamond@wustl.edu.

Find articles by Kleverov, M. in: PubMed | Google Scholar

1Department of Medicine and

2Department of Pathology and Immunology, Washington University School of Medicine, St. Louis, Missouri, USA.

3Bioanalytical Chemistry Facility, Donald Danforth Plant Science Center, St. Louis, Missouri, USA.

4The Edison Family Center for Genome Sciences and Systems Biology, Washington University School of Medicine, St. Louis, Missouri, USA.

5Weill Cornell Medicine, Cornell University, New York, New York, USA.

6Department of Nephrology, University Hospital Regensburg, Regensburg, Germany.

7Department of Inflammation and Immunity, Cleveland Clinic, Cleveland, Ohio, USA.

8Department of Bioengineering and ChEM-H, Stanford University, Stanford, California, USA.

9Department of Molecular Microbiology, Washington University School of Medicine, St. Louis, Missouri, USA.

Address correspondence to: Michael S. Diamond, Department of Medicine, Box 8051, Washington University School of Medicine, 660 S. Euclid Ave, St. Louis, Missouri 63110, USA. Email: mdiamond@wustl.edu.

Find articles by Winkler, E. in: PubMed | Google Scholar

1Department of Medicine and

2Department of Pathology and Immunology, Washington University School of Medicine, St. Louis, Missouri, USA.

3Bioanalytical Chemistry Facility, Donald Danforth Plant Science Center, St. Louis, Missouri, USA.

4The Edison Family Center for Genome Sciences and Systems Biology, Washington University School of Medicine, St. Louis, Missouri, USA.

5Weill Cornell Medicine, Cornell University, New York, New York, USA.

6Department of Nephrology, University Hospital Regensburg, Regensburg, Germany.

7Department of Inflammation and Immunity, Cleveland Clinic, Cleveland, Ohio, USA.

8Department of Bioengineering and ChEM-H, Stanford University, Stanford, California, USA.

9Department of Molecular Microbiology, Washington University School of Medicine, St. Louis, Missouri, USA.

Address correspondence to: Michael S. Diamond, Department of Medicine, Box 8051, Washington University School of Medicine, 660 S. Euclid Ave, St. Louis, Missouri 63110, USA. Email: mdiamond@wustl.edu.

Find articles by Williams, R. in: PubMed | Google Scholar

1Department of Medicine and

2Department of Pathology and Immunology, Washington University School of Medicine, St. Louis, Missouri, USA.

3Bioanalytical Chemistry Facility, Donald Danforth Plant Science Center, St. Louis, Missouri, USA.

4The Edison Family Center for Genome Sciences and Systems Biology, Washington University School of Medicine, St. Louis, Missouri, USA.

5Weill Cornell Medicine, Cornell University, New York, New York, USA.

6Department of Nephrology, University Hospital Regensburg, Regensburg, Germany.

7Department of Inflammation and Immunity, Cleveland Clinic, Cleveland, Ohio, USA.

8Department of Bioengineering and ChEM-H, Stanford University, Stanford, California, USA.

9Department of Molecular Microbiology, Washington University School of Medicine, St. Louis, Missouri, USA.

Address correspondence to: Michael S. Diamond, Department of Medicine, Box 8051, Washington University School of Medicine, 660 S. Euclid Ave, St. Louis, Missouri 63110, USA. Email: mdiamond@wustl.edu.

Find articles by Janova, H. in: PubMed | Google Scholar

1Department of Medicine and

2Department of Pathology and Immunology, Washington University School of Medicine, St. Louis, Missouri, USA.

3Bioanalytical Chemistry Facility, Donald Danforth Plant Science Center, St. Louis, Missouri, USA.

4The Edison Family Center for Genome Sciences and Systems Biology, Washington University School of Medicine, St. Louis, Missouri, USA.

5Weill Cornell Medicine, Cornell University, New York, New York, USA.

6Department of Nephrology, University Hospital Regensburg, Regensburg, Germany.

7Department of Inflammation and Immunity, Cleveland Clinic, Cleveland, Ohio, USA.

8Department of Bioengineering and ChEM-H, Stanford University, Stanford, California, USA.

9Department of Molecular Microbiology, Washington University School of Medicine, St. Louis, Missouri, USA.

Address correspondence to: Michael S. Diamond, Department of Medicine, Box 8051, Washington University School of Medicine, 660 S. Euclid Ave, St. Louis, Missouri 63110, USA. Email: mdiamond@wustl.edu.

Find articles by Droit, L. in: PubMed | Google Scholar

1Department of Medicine and

2Department of Pathology and Immunology, Washington University School of Medicine, St. Louis, Missouri, USA.

3Bioanalytical Chemistry Facility, Donald Danforth Plant Science Center, St. Louis, Missouri, USA.

4The Edison Family Center for Genome Sciences and Systems Biology, Washington University School of Medicine, St. Louis, Missouri, USA.

5Weill Cornell Medicine, Cornell University, New York, New York, USA.

6Department of Nephrology, University Hospital Regensburg, Regensburg, Germany.

7Department of Inflammation and Immunity, Cleveland Clinic, Cleveland, Ohio, USA.

8Department of Bioengineering and ChEM-H, Stanford University, Stanford, California, USA.

9Department of Molecular Microbiology, Washington University School of Medicine, St. Louis, Missouri, USA.

Address correspondence to: Michael S. Diamond, Department of Medicine, Box 8051, Washington University School of Medicine, 660 S. Euclid Ave, St. Louis, Missouri 63110, USA. Email: mdiamond@wustl.edu.

Find articles by Wang, L. in: PubMed | Google Scholar

1Department of Medicine and

2Department of Pathology and Immunology, Washington University School of Medicine, St. Louis, Missouri, USA.

3Bioanalytical Chemistry Facility, Donald Danforth Plant Science Center, St. Louis, Missouri, USA.

4The Edison Family Center for Genome Sciences and Systems Biology, Washington University School of Medicine, St. Louis, Missouri, USA.

5Weill Cornell Medicine, Cornell University, New York, New York, USA.

6Department of Nephrology, University Hospital Regensburg, Regensburg, Germany.

7Department of Inflammation and Immunity, Cleveland Clinic, Cleveland, Ohio, USA.

8Department of Bioengineering and ChEM-H, Stanford University, Stanford, California, USA.

9Department of Molecular Microbiology, Washington University School of Medicine, St. Louis, Missouri, USA.

Address correspondence to: Michael S. Diamond, Department of Medicine, Box 8051, Washington University School of Medicine, 660 S. Euclid Ave, St. Louis, Missouri 63110, USA. Email: mdiamond@wustl.edu.

Find articles by Li, T. in: PubMed | Google Scholar

1Department of Medicine and

2Department of Pathology and Immunology, Washington University School of Medicine, St. Louis, Missouri, USA.

3Bioanalytical Chemistry Facility, Donald Danforth Plant Science Center, St. Louis, Missouri, USA.

4The Edison Family Center for Genome Sciences and Systems Biology, Washington University School of Medicine, St. Louis, Missouri, USA.

5Weill Cornell Medicine, Cornell University, New York, New York, USA.

6Department of Nephrology, University Hospital Regensburg, Regensburg, Germany.

7Department of Inflammation and Immunity, Cleveland Clinic, Cleveland, Ohio, USA.

8Department of Bioengineering and ChEM-H, Stanford University, Stanford, California, USA.

9Department of Molecular Microbiology, Washington University School of Medicine, St. Louis, Missouri, USA.

Address correspondence to: Michael S. Diamond, Department of Medicine, Box 8051, Washington University School of Medicine, 660 S. Euclid Ave, St. Louis, Missouri 63110, USA. Email: mdiamond@wustl.edu.

Find articles by Heath, L. in: PubMed | Google Scholar

1Department of Medicine and

2Department of Pathology and Immunology, Washington University School of Medicine, St. Louis, Missouri, USA.

3Bioanalytical Chemistry Facility, Donald Danforth Plant Science Center, St. Louis, Missouri, USA.

4The Edison Family Center for Genome Sciences and Systems Biology, Washington University School of Medicine, St. Louis, Missouri, USA.

5Weill Cornell Medicine, Cornell University, New York, New York, USA.

6Department of Nephrology, University Hospital Regensburg, Regensburg, Germany.

7Department of Inflammation and Immunity, Cleveland Clinic, Cleveland, Ohio, USA.

8Department of Bioengineering and ChEM-H, Stanford University, Stanford, California, USA.

9Department of Molecular Microbiology, Washington University School of Medicine, St. Louis, Missouri, USA.

Address correspondence to: Michael S. Diamond, Department of Medicine, Box 8051, Washington University School of Medicine, 660 S. Euclid Ave, St. Louis, Missouri 63110, USA. Email: mdiamond@wustl.edu.

Find articles by Jung, A. in: PubMed | Google Scholar

1Department of Medicine and

2Department of Pathology and Immunology, Washington University School of Medicine, St. Louis, Missouri, USA.

3Bioanalytical Chemistry Facility, Donald Danforth Plant Science Center, St. Louis, Missouri, USA.

4The Edison Family Center for Genome Sciences and Systems Biology, Washington University School of Medicine, St. Louis, Missouri, USA.

5Weill Cornell Medicine, Cornell University, New York, New York, USA.

6Department of Nephrology, University Hospital Regensburg, Regensburg, Germany.

7Department of Inflammation and Immunity, Cleveland Clinic, Cleveland, Ohio, USA.

8Department of Bioengineering and ChEM-H, Stanford University, Stanford, California, USA.

9Department of Molecular Microbiology, Washington University School of Medicine, St. Louis, Missouri, USA.

Address correspondence to: Michael S. Diamond, Department of Medicine, Box 8051, Washington University School of Medicine, 660 S. Euclid Ave, St. Louis, Missouri 63110, USA. Email: mdiamond@wustl.edu.

Find articles by Mack, M. in: PubMed | Google Scholar |

1Department of Medicine and

2Department of Pathology and Immunology, Washington University School of Medicine, St. Louis, Missouri, USA.

3Bioanalytical Chemistry Facility, Donald Danforth Plant Science Center, St. Louis, Missouri, USA.

4The Edison Family Center for Genome Sciences and Systems Biology, Washington University School of Medicine, St. Louis, Missouri, USA.

5Weill Cornell Medicine, Cornell University, New York, New York, USA.

6Department of Nephrology, University Hospital Regensburg, Regensburg, Germany.

7Department of Inflammation and Immunity, Cleveland Clinic, Cleveland, Ohio, USA.

8Department of Bioengineering and ChEM-H, Stanford University, Stanford, California, USA.

9Department of Molecular Microbiology, Washington University School of Medicine, St. Louis, Missouri, USA.

Address correspondence to: Michael S. Diamond, Department of Medicine, Box 8051, Washington University School of Medicine, 660 S. Euclid Ave, St. Louis, Missouri 63110, USA. Email: mdiamond@wustl.edu.

Find articles by Baldridge, M. in: PubMed | Google Scholar |

1Department of Medicine and

2Department of Pathology and Immunology, Washington University School of Medicine, St. Louis, Missouri, USA.

3Bioanalytical Chemistry Facility, Donald Danforth Plant Science Center, St. Louis, Missouri, USA.

4The Edison Family Center for Genome Sciences and Systems Biology, Washington University School of Medicine, St. Louis, Missouri, USA.

5Weill Cornell Medicine, Cornell University, New York, New York, USA.

6Department of Nephrology, University Hospital Regensburg, Regensburg, Germany.

7Department of Inflammation and Immunity, Cleveland Clinic, Cleveland, Ohio, USA.

8Department of Bioengineering and ChEM-H, Stanford University, Stanford, California, USA.

9Department of Molecular Microbiology, Washington University School of Medicine, St. Louis, Missouri, USA.

Address correspondence to: Michael S. Diamond, Department of Medicine, Box 8051, Washington University School of Medicine, 660 S. Euclid Ave, St. Louis, Missouri 63110, USA. Email: mdiamond@wustl.edu.

Find articles by Stappenbeck, T. in: PubMed | Google Scholar

1Department of Medicine and

2Department of Pathology and Immunology, Washington University School of Medicine, St. Louis, Missouri, USA.

3Bioanalytical Chemistry Facility, Donald Danforth Plant Science Center, St. Louis, Missouri, USA.

4The Edison Family Center for Genome Sciences and Systems Biology, Washington University School of Medicine, St. Louis, Missouri, USA.

5Weill Cornell Medicine, Cornell University, New York, New York, USA.

6Department of Nephrology, University Hospital Regensburg, Regensburg, Germany.

7Department of Inflammation and Immunity, Cleveland Clinic, Cleveland, Ohio, USA.

8Department of Bioengineering and ChEM-H, Stanford University, Stanford, California, USA.

9Department of Molecular Microbiology, Washington University School of Medicine, St. Louis, Missouri, USA.

Address correspondence to: Michael S. Diamond, Department of Medicine, Box 8051, Washington University School of Medicine, 660 S. Euclid Ave, St. Louis, Missouri 63110, USA. Email: mdiamond@wustl.edu.

Find articles by Thackray, L. in: PubMed | Google Scholar

1Department of Medicine and

2Department of Pathology and Immunology, Washington University School of Medicine, St. Louis, Missouri, USA.

3Bioanalytical Chemistry Facility, Donald Danforth Plant Science Center, St. Louis, Missouri, USA.

4The Edison Family Center for Genome Sciences and Systems Biology, Washington University School of Medicine, St. Louis, Missouri, USA.

5Weill Cornell Medicine, Cornell University, New York, New York, USA.

6Department of Nephrology, University Hospital Regensburg, Regensburg, Germany.

7Department of Inflammation and Immunity, Cleveland Clinic, Cleveland, Ohio, USA.

8Department of Bioengineering and ChEM-H, Stanford University, Stanford, California, USA.

9Department of Molecular Microbiology, Washington University School of Medicine, St. Louis, Missouri, USA.

Address correspondence to: Michael S. Diamond, Department of Medicine, Box 8051, Washington University School of Medicine, 660 S. Euclid Ave, St. Louis, Missouri 63110, USA. Email: mdiamond@wustl.edu.

Find articles by Hsieh, C. in: PubMed | Google Scholar

1Department of Medicine and

2Department of Pathology and Immunology, Washington University School of Medicine, St. Louis, Missouri, USA.

3Bioanalytical Chemistry Facility, Donald Danforth Plant Science Center, St. Louis, Missouri, USA.

4The Edison Family Center for Genome Sciences and Systems Biology, Washington University School of Medicine, St. Louis, Missouri, USA.

5Weill Cornell Medicine, Cornell University, New York, New York, USA.

6Department of Nephrology, University Hospital Regensburg, Regensburg, Germany.

7Department of Inflammation and Immunity, Cleveland Clinic, Cleveland, Ohio, USA.

8Department of Bioengineering and ChEM-H, Stanford University, Stanford, California, USA.

9Department of Molecular Microbiology, Washington University School of Medicine, St. Louis, Missouri, USA.

Address correspondence to: Michael S. Diamond, Department of Medicine, Box 8051, Washington University School of Medicine, 660 S. Euclid Ave, St. Louis, Missouri 63110, USA. Email: mdiamond@wustl.edu.

Find articles by Handley, S. in: PubMed | Google Scholar

1Department of Medicine and

2Department of Pathology and Immunology, Washington University School of Medicine, St. Louis, Missouri, USA.

3Bioanalytical Chemistry Facility, Donald Danforth Plant Science Center, St. Louis, Missouri, USA.

4The Edison Family Center for Genome Sciences and Systems Biology, Washington University School of Medicine, St. Louis, Missouri, USA.

5Weill Cornell Medicine, Cornell University, New York, New York, USA.

6Department of Nephrology, University Hospital Regensburg, Regensburg, Germany.

7Department of Inflammation and Immunity, Cleveland Clinic, Cleveland, Ohio, USA.

8Department of Bioengineering and ChEM-H, Stanford University, Stanford, California, USA.

9Department of Molecular Microbiology, Washington University School of Medicine, St. Louis, Missouri, USA.

Address correspondence to: Michael S. Diamond, Department of Medicine, Box 8051, Washington University School of Medicine, 660 S. Euclid Ave, St. Louis, Missouri 63110, USA. Email: mdiamond@wustl.edu.

Find articles by Guo, C. in: PubMed | Google Scholar

1Department of Medicine and

2Department of Pathology and Immunology, Washington University School of Medicine, St. Louis, Missouri, USA.

3Bioanalytical Chemistry Facility, Donald Danforth Plant Science Center, St. Louis, Missouri, USA.

4The Edison Family Center for Genome Sciences and Systems Biology, Washington University School of Medicine, St. Louis, Missouri, USA.

5Weill Cornell Medicine, Cornell University, New York, New York, USA.

6Department of Nephrology, University Hospital Regensburg, Regensburg, Germany.

7Department of Inflammation and Immunity, Cleveland Clinic, Cleveland, Ohio, USA.

8Department of Bioengineering and ChEM-H, Stanford University, Stanford, California, USA.

9Department of Molecular Microbiology, Washington University School of Medicine, St. Louis, Missouri, USA.

Address correspondence to: Michael S. Diamond, Department of Medicine, Box 8051, Washington University School of Medicine, 660 S. Euclid Ave, St. Louis, Missouri 63110, USA. Email: mdiamond@wustl.edu.

Find articles by Fischbach, M. in: PubMed | Google Scholar

1Department of Medicine and

2Department of Pathology and Immunology, Washington University School of Medicine, St. Louis, Missouri, USA.

3Bioanalytical Chemistry Facility, Donald Danforth Plant Science Center, St. Louis, Missouri, USA.

4The Edison Family Center for Genome Sciences and Systems Biology, Washington University School of Medicine, St. Louis, Missouri, USA.

5Weill Cornell Medicine, Cornell University, New York, New York, USA.

6Department of Nephrology, University Hospital Regensburg, Regensburg, Germany.

7Department of Inflammation and Immunity, Cleveland Clinic, Cleveland, Ohio, USA.

8Department of Bioengineering and ChEM-H, Stanford University, Stanford, California, USA.

9Department of Molecular Microbiology, Washington University School of Medicine, St. Louis, Missouri, USA.

Address correspondence to: Michael S. Diamond, Department of Medicine, Box 8051, Washington University School of Medicine, 660 S. Euclid Ave, St. Louis, Missouri 63110, USA. Email: mdiamond@wustl.edu.

Find articles by Artyomov, M. in: PubMed | Google Scholar

1Department of Medicine and

2Department of Pathology and Immunology, Washington University School of Medicine, St. Louis, Missouri, USA.

3Bioanalytical Chemistry Facility, Donald Danforth Plant Science Center, St. Louis, Missouri, USA.

4The Edison Family Center for Genome Sciences and Systems Biology, Washington University School of Medicine, St. Louis, Missouri, USA.

5Weill Cornell Medicine, Cornell University, New York, New York, USA.

6Department of Nephrology, University Hospital Regensburg, Regensburg, Germany.

7Department of Inflammation and Immunity, Cleveland Clinic, Cleveland, Ohio, USA.

8Department of Bioengineering and ChEM-H, Stanford University, Stanford, California, USA.

9Department of Molecular Microbiology, Washington University School of Medicine, St. Louis, Missouri, USA.

Address correspondence to: Michael S. Diamond, Department of Medicine, Box 8051, Washington University School of Medicine, 660 S. Euclid Ave, St. Louis, Missouri 63110, USA. Email: mdiamond@wustl.edu.

Find articles by Diamond, M. in: PubMed | Google Scholar |

Published July 15, 2026 - More info

Published in Volume 136, Issue 14 on July 15, 2026
J Clin Invest. 2026;136(14):e202262. https://doi.org/10.1172/JCI202262.
© 2026 Zhao et al. This work is licensed under the Creative Commons Attribution 4.0 International License. To view a copy of this license, visit http://creativecommons.org/licenses/by/4.0/. Published July 15, 2026 - Version history
Received: November 10, 2025; Accepted: June 2, 2026 View PDF Abstract

Oral antibiotics can predispose to joint inflammation, but this phenomenon remains poorly understood. Here, we leverage mouse models of alphavirus-induced arthritis to investigate the roles of gut commensals, metabolites, and host immune mechanisms in promoting musculoskeletal inflammation. Mice treated with a short course of oral antibiotics exhibited worsened arthritis after chikungunya (CHIKV) or Mayaro virus infections. This phenotype was associated with loss of short-chain fatty acids (SCFAs), greater intestinal permeability, and activation of gut-associated immune cells and required TLR4 signaling, MyD88 expression, monocytes, antigen-specific and bystander CD4+ T cells, and proinflammatory cytokines. Administration of exogenous SCFAs or colonization of mice with bacterial species that generate SCFAs mitigated CHIKV-induced joint inflammation. scRNA-seq revealed that gut-derived SCFAs ameliorate the inflammatory phenotype of synovial CD4+ T cells, infiltrating monocytes, and resident osteoclast-like cells. Thus, antibiotic-triggered gut dysbiosis exacerbates alphavirus arthritis by shaping the inflammatory profile of both infiltrating and resident immune cells in joint tissues.

Graphical Abstract Introduction

Antibiotic usage is linked to an increased risk for onset or relapse of inflammatory arthritis, including rheumatoid (RA) and juvenile idiopathic arthritis (1–6). Although a gut-joint linkage has been described (7–9), the basis for this increased risk with antibiotics use is not well understood, particularly how joint inflammation is regulated by interactions between gut microbes, mucosal barriers, and immune cells. Less is known about the effects of antibiotics and the commensal intestinal microbiota on inflammation in the context of pathogens that infect joint tissues.

Chikungunya virus (CHIKV) is a mosquito-transmitted alphavirus that causes epidemics of inflammatory arthritis. Recent resurgence in the Indian Ocean, Africa, and Asia and autochthonous cases in New York have triggered concern about another widespread epidemic. Three related alphaviruses (Mayaro [MAYV], Ross River, and O’nyong-nyong viruses) also cause musculoskeletal disease in the Americas, Oceania, and Africa, respectively (10). CHIKV infection causes an acute febrile syndrome in humans, with viremia, rash, myalgia, severe joint inflammation, and elevated systemic cytokines and chemokines. CHIKV arthritis clinically mimics RA, can persist for months to years after initial infection, and can cause bone erosions and joint damage (11–13).

In mice, subcutaneous inoculation of CHIKV leads to arthritis and myositis, like in humans (14–16). Inflammation is associated with tissue infiltration of monocytes, neutrophils, and T cells. However, there often is a discordance between viral burden and arthritis severity, suggesting that inflammation and pathology are not directly related to the viral burden in joint-associated tissues. Indeed, CHIKV infection of Rag1–/– mice lacking mature T and B cells leads to persistent local infection yet less joint swelling than infected WT mice (17). Mice lacking CD4+ T cells also have attenuated joint swelling after CHIKV infection without substantive effects on viral infection (18, 19). These data highlight the differential contributions of immune cells in restricting CHIKV infection and causing musculoskeletal disease.

Here, we use mouse models of alphavirus arthritis to investigate how gut commensals, their metabolites, and host immune responses promote joint inflammation in the setting of antibiotic-induced dysbiosis. Intestinal dysbiosis resulting from a short course of oral antibiotics results in increased gastrointestinal (GI) permeability due to loss of specific microbiota-derived short-chain fatty acids (SCFAs) and enhanced foot swelling and joint infiltration of monocytes and CD4+ T cells after CHIKV infection. The increased inflammation was dependent on gut-associated immune cells and required TLR4 and MyD88 signaling, with contributions from intestinal epithelial cells. The inflammatory phenotypes in antibiotic-treated mice were attenuated by neutralizing IL-17A, TNF, IL-18, or IFN-γ and not observed in alphavirus-infected germ-free (GF) mice. Reconstitution with SCFA-producing bacterial species or oral SCFA supplementation restored intestinal barrier integrity and reversed CHIKV-induced arthritis phenotypes in antibiotic-treated mice. Finally, SCFA supplementation modulated the inflammatory phenotypes of CD4+ T cells, monocytes, and osteoclast-like cells triggered by antibiotic treatment. Our data suggest that antibiotic-induced disruption of the intestinal microbiota results in loss of microbe-derived SCFAs, which drives inflammation in joint-associated tissues by shaping the functions of infiltrating and resident immune cells.

Results

Perturbation of the intestinal microbiota exacerbates CHIKV-induced arthritis. To assess the role of the microbiota on disease severity, we treated C57BL6/J mice with a short, 3-day course of minimally absorbable oral antibiotics, ampicillin, and vancomycin (Figure 1A). This treatment, which was chosen to limit systemic exposure, alters the bacterial community structure in the GI tract by depleting bacteria primarily from the Bacteroidetes phylum, leading to overrepresentation of Proteobacteria and Firmicutes (Figure 1, B–D, and Supplemental Figure 1, A–C; supplemental material available online with this article; https://doi.org/10.1172/JCI202262DS1). While treatment with either vancomycin or ampicillin alone increased CHIKV-induced foot swelling, the 2 antibiotics together resulted in a more pronounced phenotype (Figure 1, E–G); consequently, we used the ampicillin and vancomycin (AV) combination for the remainder of our studies. Notably, the enhanced inflammatory phenotype persisted for at least 8 weeks after antibiotic cessation (Supplemental Figure 1, D–G). Similar increases in joint swelling were also observed in AV-treated mice after infection with MAYV, a related arthritogenic alphavirus (Supplemental Figure 1H).

Figure 1

Depletion of the intestinal microbiota by oral antibiotics exacerbates musculoskeletal tissue inflammation after CHIKV infection. (A) Schematic of experimental setup. (B–D) Colonic contents were collected from water- or AV-treated C57BL/6J mice at 0 or 7 dpi. Relative abundance of bacterial phyla (B), number of bacterial taxa (richness) (C), and beta diversity (weighted UniFrac distance) (D) in colonic contents (2 experiments, n = 8–9 per group). (E–G) Foot swelling after CHIKV infection in mice treated with ampicillin (E, n = 10), vancomycin (F, n = 10), AV (G, n = 15), or water (E–G, n = 15). (H–J) H&E staining of foot tissues from water- or AV-treated mice at 0 (H), 3 (I), and 6 (J) dpi (n = 7–10 per group). Original magnification, ×2.5; scale bars: 1 mm. D, dermis; B, bone; BM, bone marrow; M, muscle. Double-headed arrows indicate edema. Arrowheads indicate periosteal inflammation. (K) H&E staining showing synovitis (arrows) in the mid-foot of water- or AV-treated mice at 6 dpi. Original magnification, ×5; scale bars: 100 μm. (L) Scoring of joint inflammation and tissue damage. (M) Toluidine blue staining of foot tissues from water- or AV-treated mice at 0 and 6 dpi (representative of n = 2 uninfected and n = 8 infected mice per group). Original magnification, ×40; scale bars: 100 μm. SC, superficial noncalcified cartilage; DC, deeper calcified cartilage. Arrowheads indicate destaining (proteoglycan loss) of superficial hyaline cartilage. Statistical analysis: C, Wilcoxon’s test; D, permutational multivariate ANOVA (Adonis); E–G, mean ± SEM, 2-way ANOVA with Šidák’s post test or 1-way ANOVA with Dunnett’s post test for AUC; L, unpaired 2-tailed t test. ****P < 0.0001; ***P < 0.001; *P < 0.05.

Histological examination of ipsilateral foot tissues at 3 days postinfection (dpi) showed greater soft tissue edema in AV-treated CHIKV-infected mice than water-treated, infected, or uninfected controls (Figure 1, H and I). Mononuclear cell infiltration into the joint and adjacent muscle was seen in both water- and AV-treated CHIKV-infected mice at this time point (Figure 1I). By 6 dpi, extensive soft tissue inflammation, myositis with muscle degeneration and necrosis, and periostitis were evident in tissues from all CHIKV-infected mice (Figure 1J). However, synovitis and erosion of synovial membranes were more severe in AV-treated mice (Figure 1, K and L).

Given the association between inflammation and cartilage damage, we evaluated articular cartilage integrity using toluidine blue staining. At 6 dpi, AV-treated mice showed reduced proteoglycan content in the superficial noncalcified cartilage layer compared with water-treated controls (Figure 1M). We also performed tartrate resistant acid phosphatase (TRAP) staining to visualize osteoclasts. In both water- and AV-treated CHIKV-infected mice, we observed transcortical vascular channels close to the epiphysis and subchondral bone, structures that expand during inflammation and lead to osteoclast-induced bone erosions (20–22), with increased cellularity in AV-treated mice (Figure 2A). AV treatment was also associated with increased density of subchondral TRAP+ osteoclasts at 6 dpi (Figure 2, A–C).

Figure 2

Oral antibiotics increase subchondral osteoclasts without altering viral infection during CHIKV infection. (A and B) TRAP staining of foot tissues from water- or AV-treated mice at 0 and 6 dpi (representative of n = 8 per group); red staining shows TRAP+ cells. Original magnification, ×10; scale bars: 200 μm. B, bone; BM, bone marrow; S, synovium; C, cartilage. In the higher magnification views, double-headed arrows indicate the width of the vascular channels. Dashed ovals in B encircle subchondral TRAP+ osteoclasts (red, nucleated). (C) Quantitation of subchondral TRAP+ osteoclasts in water- and AV-treated mice at 0 and 6 dpi. (D) CHIKV infection in ipsilateral feet of water- and AV-treated mice at 1, 3, and 6 dpi was determined by focus-forming assay (2 experiments, n = 5–8 per group). FFU, focus-forming units. (E) CHIKV RNA in situ hybridization of foot tissue sections from Zika virus–infected mice (negative control, top panel) and CHIKV-infected, water- or AV-treated mice at 3 dpi (middle panels) or 6 dpi (bottom panels) (representative of n = 3 per group). Original magnification, ×2.5; scale bars: 1 mm. Statistical analysis: C and D, Mann-Whitney test. *P < 0.05.

To determine whether antibiotic-mediated exacerbation of CHIKV arthritis was related to increased viral burden, we quantified viral load and visualized viral RNA within joint tissues using in situ hybridization. Notably, viral titers in the ipsilateral foot were comparable between water- and AV-treated mice at both 3 and 6 dpi, and intense viral RNA staining was detected in the periosteum, articular cartilage, synovium, and muscles in both groups (Figure 2, D and E). These data suggest that enhanced musculoskeletal inflammation in AV-treated mice is not due to increased viral replication or altered viral distribution in joint tissues.

Oral antibiotic treatment increases immune cell infiltration and cytokine levels in musculoskeletal tissues after CHIKV infection. We next evaluated the effects of AV treatment on inflammation in CHIKV-infected joint-associated tissues. At 4 dpi, AV-treated mice had higher levels of IL-4, IL-5, CCL11, LIF, IL-6, IFN-γ, TNF, M-CSF, and G-CSF, as well as the IFN-induced chemokines, CCL2, CCL3, CCL4, and CXCL10, in the ipsilateral foot (Figure 3A and Supplemental Figure 1, I and J) compared with water-treated controls. To characterize immune cell infiltrates, we performed flow cytometry analyses of cell composition in the ipsilateral foot after CHIKV infection (Supplemental Figure 2A). AV-treated mice exhibited increased numbers of monocytes and neutrophils at 4 dpi and increased neutrophils at 7 dpi (Figure 3, B and C). No significant differences were observed in numbers of NK, T, or B cell populations (Figure 3, B and C).

Figure 3

Oral antibiotics increase joint inflammation during CHIKV infection. (A) Heatmap of cytokine and chemokine levels in foot-associated musculoskeletal tissue homogenates from water- or AV-treated mice at 0 and 4 dpi (2 experiments, n = 5–8 per group, log10 fold differences in protein levels are relative to water-treated animals at 0 dpi; see also Supplemental Figure 1J and Supplemental Tables 1 and 2). (B and C) Numbers of immune cells in joint-associated tissues in the ipsilateral feet of water- or AV-treated mice at 4 (B) or 7 (C) dpi (2 experiments, n = 6–9 per group). (D) Percentages of IFN-γ–, TNF-, IL-17–, and IL-22–producing CD4+ T cells in the ipsilateral feet of water- and AV-treated mice at 4 dpi (2–3 experiments, n = 6–11 per group). (E) Percentages of IFN-γ– and TNF-producing, perforin-expressing, or CD107a-cycling CD8+ T cells in the ipsilateral feet of control and AV-treated mice at 4 dpi (3 experiments, n = 8–17 per group). (F–I) Percentages and numbers CD4+CD25+FoxP3+ Tregs in the draining popliteal lymph nodes (F and G) and spleens (H and I) of water- and AV-treated mice at 4 dpi. (J) IL-17A/F levels in joint-associated tissue homogenates from ipsilateral feet (2 experiments, n = 5–8 per group). (K–M) Foot swelling of water- and AV-treated mice that were administered isotype control or neutralizing mAb against TNF (K), IL-17A (L), or IFN-γ (M) (2 experiments, n = 8–9 per group). Statistical analysis: B–J, unpaired t test; K–M, 2-way ANOVA with Šidák’s post test or 1-way ANOVA with Dunnett’s post test for AUC; mean values ± SEM are shown. ****P < 0.0001; ***P < 0.001; **P < 0.01; *P < 0.05.

We also assessed whether AV treatment altered qualitative T cell responses in joint-associated tissues after CHIKV infection. At 4 dpi, AV-treated mice showed increased proportions and numbers of TNF-, IFN-γ–, IL-17A–, and IL-22–producing CD4+ T cells in the joint (Figure 3D and Supplemental Figure 2B). In contrast, CD8+ T cells displayed similar percentages of degranulation and IFN-γ production in both groups, although AV-treated mice showed a small increase in the proportion of perforin- and TNF-expressing CD8+ T cells (Figure 3E). Significant differences in the percentages or numbers of FoxP3+ Tregs were not detected between water- and AV-treated mice in the draining popliteal lymph node (Figure 3, F and G) or spleen (Figure 3, H and I) at 4 dpi.

Using enzyme-linked immunosorbent assays, we confirmed that AV-treated mice had elevated levels of IL-17A, TNF, and IFN-γ in joint-associated tissues at 4 dpi compared with water-treated mice (Figure 3J and Supplemental Figure 1I). To determine their functional contribution, we administered neutralizing mAbs to water- and AV-treated mice. Treatment with anti-TNF or anti–IL-17A mAb improved foot swelling in antibiotic-treated mice following CHIKV infection (Figure 3, K and L). Neutralization of IFN-γ in AV-treated mice led to comparatively smaller reductions in inflammation at 3–4 dpi but a marked reduction in swelling at and after 7 dpi (Figure 3M). Thus, a