Systemic lupus erythematosus (SLE) is a chronic autoimmune disease characterized by loss of self-tolerance, aberrant autoantibody production and multiple organ damage. Treating the disease symptoms (flares) has remarkably improved the survival of SLE patients, but thrombosis remains a major negative prognostic factor. Currently available tests do not predict thrombotic events; hence there is a need for identification of novel biomarkers of thrombosis associated with SLE and characterization of the relevant pathophysiologic mechanisms. We have shown that inhibition of cell-surface protein disulfide isomerase (PDI) can regulate the thrombotic properties of the endothelium in vitro, through exposure of phosphatidylserine (PS). We therefore hypothesized that inhibitory autoantibodies to PDI may promote or enhance the prothrombotic phenotype in SLE.
We developed a quantitative ELISA to measure the titers of anti-PDI autoantibodies present in the sera and performed a cross-sectional analysis of sera from 217 SLE patients and 99 healthy controls enrolled in the Oklahoma Lupus Cohort. Low levels of PDI autoreactivity were detected in 54 out of 99 healthy controls tested, with a median of 4.185 AU and an interquartile range (IQR) of 2.25–4.18. Anti-PDI autoantibodies were detected in 107 out of 217 SLE sera tested, with significantly higher PDI autoreactivity (median 5.982, IQR 2.43–11.23, p = 0.042 Mann-Whitney test). We defined elevated anti-PDI titers as those to be higher than the 95th percentile of serum anti-PDI levels in the control group. Thirty-five out of the 217 SLE patients (16%) exhibited anti-PDI titers above the 95th percentile cut-off, an incidence significantly higher (p=0.0057, Fisher's exact test) than the control group (5/99, 5% by cut-off definition). Interestingly, the incidence of patients with elevated anti-PDI autoreactivity was doubled in SLE patients with secondary antiphospholipid syndrome (APS), 20 out of 89 (22%), as compared to SLE patients without APS (15 out of 128, 11%, p = 0.04 Fisher's exact test). Accordingly, we found a significant correlation between the titers of anti-PDI autoantibodies and the anti-cardiolipin IgG levels (Spearman r= 0.212, P = 0.003), indicating a link with the presence of antiphospholipid antibodies. Since the APS diagnosis involves both the detection of antiphospholipid antibodies and the presence of clinical thrombotic events, this might be an indication that anti-PDI autoantibodies are associated with thrombosis in SLE.
Clinical history revealed that SLE patients with elevated PDI autoreactivity had higher SLE Disease Activity Index (SLEDAI) scores than those with normal anti-PDI levels (P = 0.0256, Mann-Whitney test), indicating a correlation between anti-PDI levels and disease activity. We observed significant inverse correlations between anti-PDI titers and complement levels as measured by the total hemolytic complement assay (Spearman r = −0.134, P = 0.049), C3 levels (Spearman r = −0.169, P = 0.013) and C4 levels (Spearman r = −0.215, P < 0.001). While the biological significance of this correlation is under investigation, activation of the classical complement pathway on the surface of PS-exposing cells has been documented in the literature.
We have shown that PDI inhibition on endothelial surface promotes PS exposure and a thrombotic phenotype. We tested the inhibitory properties of anti-PDI autoantibodies and found that 50% of the SLE sera with elevated anti-PDI titers inhibited PDI reductase activity in vitro, thus indicating that these autoantibodies can modulate PDI function. Furthermore, affinity-purified anti-PDI autoantibodies from SLE sera were able to induce a prothrombotic phenotype on endothelial cells in vitro as evidenced by tissue factor decryption and enhanced prothrombinase activity, a direct indicator that anti-PDI autoantibodies induce anionic phospholipid exposure in vascular cells.
In conclusion, we found an increased incidence of elevated anti-PDI titers in SLE, especially in patients with antiphospholipid syndrome, that correlate with disease activity. These autoantibodies inhibit PDI enzymatic activity and are prothrombotic in vitro. To our knowledge this is the first report of autoantibodies directly inducing the exposure of anionic phospholipids on vascular cells with mechanistic implications for the pathophysiology of SLE and APS.
No relevant conflicts of interest to declare.
Asterisk with author names denotes non-ASH members.