Methods
Paired samples of serum and synovial fluid from 131 RA patients and serum samples from 34 healthy volunteers from Goteborg, Sweden were assayed for survivin and anti-survivin antibodies. RA patients were characterized by the presence/absence of erosive disease (defined as at least one erosion on hand and foot radiographs). Survivin and anti-survivin antibody levels were compared between control and RA patients, between serum and synovial fluid levels, and within the RA group, stratified by presence or absence of erosive disease and by status of treatment with DMARDs.
In a series of in vitro experiments, peripheral blood mononuclear cells (PBMCs) from RA patients and non-RA controls were stimulated with phytohemagglutinine (PHA), a T-cell mitogen, and survivin expression was assessed. In a commercial human mononuclear cell line, THP-1, cells were transfected with inhibitory and non-inhibitory oligonucleotides targeting survivin mRNA. Cell proliferation, apoptosis, and expression of survivin and IL-6 were measured after stimulation with PHA.
Results
Survivin levels: Mean synovial and serum survivin levels were roughly equivalent and were significantly higher in RA patients compared to non-RA controls (330 pg/ml vs. 121 pg/ml, respectively; p=0.002). Mean survivin levels were also significantly higher in RA patients classified as having erosive disease compared to those with non-erosive RA (430 pg/ml vs. 127 pg/ml, respectively; p=0.002). There was no correlation between survivin levels and serum CRP, IL-6, WBC, or synovial WBC levels.
In comparisons of RA patients stratified according to high survivin level (>300 pg/ml) and low survivin level (<300 pg/ml), 96% of RA patients with high survivin levels had erosive RA while only 60% of RA patients with low survivin levels demonstrated erosive manifestations on radiographs; corresponding to an 18.76 greater odds of erosive disease in RA patients with high survivin levels compared to RA patients with low survivin levels (95% CI 2.45 143.65; p=0.0048). This relationship remained significant after adjustment for serum rheumatoid factor, RA disease duration, and gender (OR 16.02 (95% CI 2.02 127.19: p=0.013)). To account for the effect of treatment with DMARDs, patients were stratified according to DMARD therapy at the time of sampling. RA patients treated with any DMARD, including TNF inhibitors, demonstrated significantly lower mean survivin levels than RA patients not receiving any DMARDs (approximately 300-400 pg/ml vs. 666 pg/ml, respectively: p= 0.02-0.028).
Autoantibodies to Survivin: Mean synovial and serum IgG anti-survivin antibody levels were equivalent and were significantly higher in RA patients compared to non-RA controls. Mean serum IgM anti-survivin antibody levels were significantly higher than synovial anti-survivin antibody levels, and were also significantly higher in RA patients compared to non-RA controls. RA patients with non-erosive disease had significantly higher mean blood and synovial fluid levels of IgG and IgM anti-survivin antibodies than RA patients with erosive disease or non-RA controls. Although RA patients with erosive disease and the highest levels of survivin were noted to have high levels of anti-survivin antibodies; in general, there was no correlation between survivin and anti-survivin antibody levels (r = 0.05).
In vitro transfection experiments: After PHA stimulation of THP-1 cells transfected with anti-sense oligonucleotides targeting survivin mRNA, both survivin and IL-6 expression were downregulated. Transfection with inhibitory oligonucleotides did not affect measured proliferation or apoptosis of THP-1 cells. Transfection of THP-1 cells with non-sense oligonucleotides to survivin mRNA did not alter survivin or IL-6 expression, or affect cells proliferation or apoptosis.
Conclusion
High serum and synovial levels of survivin were associated with RA. Erosive RA was more strongly associated with higher survivin levels than non-erosive RA. In addition, patients with non-erosive RA tended to have higher levels of anti-survivin antibodies than patients with erosive RA. A consequence of survivin expression in monocytes may be the increased expression of the pro-inflammatory cytokine IL-6.
Editorial Comments
These are interesting preliminary results suggesting a role for the apoptosis inhibitor survivin in the pathogenesis of RA. As this is a cross-sectional study, no causal relationships can be inferred between survivin and RA disease activity or the effect of DMARDs on survivin levels. However, the associations presented here are compelling and warrant further investigation, preferably involving a longitudinally followed cohort. Survivin-directed therapies are currently under development for the treatment of a variety of malignancies. If feasible for clinical application, these therapies may prove effective in RA and other autoimmune disorders, as well. Importantly, these therapies, if they were to prove safe and effective in RA, are many years from being available for general use.
