Statistics from Altmetric.com
The pathogenesis of two important inflammatory rheumatic diseases, rheumatoid arthritis (RA) and ankylosing spondylitis (AS), is not clear. In both diseases an immune response against an unknown autoantigen may have a crucial role.1,2 It has been repeatedly suggested that heat shock proteins (HSP) play a part in various autoimmune diseases such as RA, diabetes, and multiple sclerosis,3 based on high interspecies sequence homologies, inducible tissue expression, and a strong immunogenicity. On the other hand, some studies indicate that HSP may even be protective in arthritis.4
The 19 kDa urease β subunit of yersinia is regarded as a major immunodominant protein for both T cell and antibody responses in patients with yersinia induced reactive arthritis.5 Because 20–30% of HLA-B27+ patients with yersinia induced reactive arthritis develop the full picture of AS after 10–20 years,6 it is an obvious question to ask whether a T cell response against the yersinia-specific 19 kDa protein is also detectable in patients with AS. Until now, there has been insufficient evidence to support a relationship between yersinia infection and RA.
PATIENTS AND METHODS
In this study 45 patients with active AS and 22 with active RA were recruited from the outpatient clinic of the University Hospital Benjamin Franklin, Berlin. In vitro stimulation of CD4+ T cells by protein antigen was carried out for six hours, as described previously.7,8 Brefeldin A (10 μg/ml; Sigma) was added for the last four hours of the stimulation for accumulation of intracellular cytokine. After fixation with 2% formaldehyde, the cells were quadruply stained for CD4, CD69 surface marker, and for the intracellular cytokines interferon γ (IFNγ) and tumour necrosis factor α (TNFα).7,8 After gating on CD4+ T cells, only cytokine positive T cells which were also positive for the early activation antigen CD69 were counted.
After stimulation with staphylococcal enterotoxin B the percentage of positive cells was always >8%, whereas after stimulation with no antigen the percentage was always <0.01%. Samples with more than 0.05% CD69/cytokine double positive cells out of the CD4+ T cell population were regarded as positive.7,8 Figure 1 shows examples of T cell responses to human (h)-HSP60 and Y-19 kDa in one patient with AS, one with RA and a healthy control (HC).
There was no difference between the patients with AS, those with RA, and healthy controls (fig 2) in the peripheral blood T cell response to the h-HSP60 as judged by the percentage of CD69/IFNγ or CD69/TNFα double positive CD4+ T cells.
For synovial fluid (SF), 5/7 (71%) patients with AS and 3/4 (75%) patients with RA showed a T cell response to h-HSP60 by IFNγ secretion. Six of seven (86%) patients with AS and 3/4 (75%) with RA showed a TNFα secretion in response to h-HSP60. In comparison with peripheral blood, the h-HSP60-specific IFNγ/TNFα responses were stronger in SF (mean (SD) 0.18 (0.3) in SF v 0.09 (0.08) in peripheral blood for IFNγ; 0.35 (0.64) in SF v 0.15 (0.05) in peripheral blood for TNFα), but this difference was not significant (p>0.05 for both cytokines).
As judged by the percentage of CD69/IFNγ double positive CD4+ T cells, a similar T cell response to the Y-19 kDa was also found in peripheral blood in all three groups (fig 2). Although a higher percentage of patients with AS (3/7 (43%) patients for IFNγ; 5/7 (71%) patients for TNFα) than with RA (1/4 (25%) patients for IFNγ; 2/4 (50%) patients for TNFα) showed a synovial T cell response to the 19 kDa protein; this difference was non-significant, possibly owing to the small number of patients.
These results suggest that T cell responses to h-HSP60 and to Y-19 kDa are present in all three groups and seem therefore not to be pathogenic. We cannot exclude the possibility that T cells specific for one of the two antigens are present in synovium and just are not detected in peripheral blood. However, the absence of clear differences between antigen-specific T cells in the SF of patients with RA and AS argues against this, unless a similar pathogenesis for these two diseases is assumed, which is unlikely.