What determines arthritogenicity of bacterial cell wall? A study on Eubacterium cell wall-induced arthritis

Citation
X. Zhang et al., What determines arthritogenicity of bacterial cell wall? A study on Eubacterium cell wall-induced arthritis, RHEUMATOLOG, 39(3), 2000, pp. 274-282
Citations number
41
Categorie Soggetti
Rheumatology
Journal title
RHEUMATOLOGY
ISSN journal
14620324 → ACNP
Volume
39
Issue
3
Year of publication
2000
Pages
274 - 282
Database
ISI
SICI code
1462-0324(200003)39:3<274:WDAOBC>2.0.ZU;2-R
Abstract
Objective. To study what determines the arthritogenicity of the bacterial c ell wall (CW) using Eubacterium CW-induced arthritis in the rat. Methods. Eubacterium aerofaciens, previously reported as arthritogenic, and E. limosum and E. alactolyticum, known as non-arthritogenic, were used. Ga s chromatography-mass spectrometry (GC-MS) was applied to analyse the chemi cal composition of the bacterial cell wall. Cellular immune response was me asured by concanavalin A (Con A) stimulation and FACScan analysis. Also, se rum antibodies against the injected cell wall were determined. Results. Unexpectedly, from the two strains of E. aerofaciens used only one proved to be arthritogenic (with a CW inducing chronic arthritis after a s ingle intraperitoneal injection), even though these two strains were 100% i dentical by 16S rDNA analysis. CW of the other E. aerofaciens strain induce d only transient acute arthritis. CW of E. limosum and E. alactolyticum ind uced weak signs of acute arthritis. Based on the CC-MS analysis and on the results published previously, putative structures of peptidoglycan (PG) in the four CW preparations are presented. It is apparent that the presence of lysine in position 3 of the PG stem peptide contributes to arthritogenicit y but is alone not decisive. Both strains of E. aerofaciens were immunosupp ressive, when tested by Con A response at 2 weeks after CW injection. Such an immunosuppression was not observed after injection of CW from E. limosum or E. alactolyticum. FACScan analysis for six T cell markers and studies o n serum antibody responses did not reveal any differences in the effect of the four bacterial strains used. Conclusions. The results obtained suggest that the chemical structure of PG present in the bacterial CW is decisive in determining arthritogenicity/no n-arthritogenicity. Therefore, from two bacterial strains belonging to norm al human intestinal flora and 100% identical by 16S rDNA analysis, one prov ed to be arthritogenic and the other non-arthritogenic.