Background/Purpose: To test the hypothesis that an altered gut microbiota (dysbiosis) plays a causal role in the obese OA phenotype (obesity with both hand and knee OA).

Methods: Stool and blood samples were collected from 92 participants with BMI ≥ 30 and age ≥ 55 yrs recruited from the Johnston County Osteoarthritis Project. OA cases (n=50) had clinical and/or radiographic hand OA, defined as involvement of at least 3 joints across both hands, and had Kellgren-Lawrence (KL) grade 2-4 knee OA (or TKR) in at least one knee on preliminary reading. Controls (N=42) had no hand OA and KL grade 0-1 knees. Exclusions included consumption of antibiotics/probiotics or intestinal surgery within the past 6 weeks, known bowel disease or prior fecal microbiota transplantation. Compositional analysis of the gut microbiome was carried out by 16S rRNA bacterial amplicon sequencing.  Data was analyzed using QIIME 1.9.0. Shannon index, Chao1, and Phylogenetic Diversity (PD) were estimated from sequencing data. Beta diversity estimates were calculated using weighted and unweighted Unifrac approaches. Permutation Multivariate Analysis of Variance (PERMANOVA) and Analysis of Similarities (ANOSIM) were used to test the null hypothesis that case and control microbial communities share the same distribution. The Wilcoxon tests adjusted for multiple comparisons were used to determine differences in relative genera abundances between case and control groupsl. Blood samples were used for a multiplex cytokine analysis (n=73) and measures of LPS and LPS binding protein (n=78). To test for causality, pooled fecal transplant from 5 cases and 5 controls were gavaged every 4 weeks to 31 germ-free mice (n=16 cases; n=15 controls). Mice were on a Western Diet (40% fat, high sucrose) for 40 weeks. Fecal samples were collected 2 weeks after each gavage. Knee OA was evaluated histologically (grading of cartilage damage, osteophytes and synovitis) at week 40.

Results: OA cases were slightly older with more females, higher BMI, greater WOMAC pain and KL grades (Table 1). Nine participants recruited as controls had KL grade 2 in one knee after final readings by the study radiologist. A sensitivity analysis was performed assigning these participants to the case group and this did not change any of the results. There were no significant differences in diversity between cases and controls. Likewise, there were no significant compositional differences between groups at the genus level. Cases had higher levels of osteopontin (a pro-inflammatory cytokine originally isolated from bone) and LPS (a sign of a leaky gut) (Fig 1).

Mice transplanted with case or control microbiota gained similar weight but exhibited a significant difference in microbial diversity (p=0.02, Fig.2). Both groups developed histologic findings of OA without any differences in OA severity between the groups.      

Conclusion: The lack of differences in the gut microbiota yet increased serum LPS levels in obese humans with hand and knee OA suggest that a leaky gut allowing for greater absorption of LPS (and potentially other bacterial products), rather than a dysbiotic microbiota, may contribute to development of OA. This was supported by the failure to promote more severe OA in mice by microbial transplant from humans with OA.

« Back to ACR Convergence 2020

ACR Meeting Abstracts - https://acrabstracts.org/abstract/association-of-a-leaky-gut-but-not-microbial-dysbiosis-with-obesity-related-oa-a-translational-study/