We have earlier shown that polyethylene glycol (PEG)-precipitated immune complexes (IC) from systemic lupus erythematosus (SLE) patients can induce IL-10 and IL-6 production from peripheral blood mononuclear cells. SLE is associated with IC-induced classical complement activation. Autoantibodies together with autoantigens released from apoptotic cells partly explain the IC formation. We therefore wanted to investigate how complement activation and levels of autoantibodies correlate to the amount of circulating IC and IC-induced cytokine production.
Others have previously shown an inverse correlation between IL-10 and IL-12 in SLE and we have shown that the degree of complement activation affects IC-induced IL-10 and IL-12 in different ways. A second aim was to investigate the effect of SLE IC on the production of IL-10 and IL-12.
In total 195 SLE samples were investigated. Levels of autoantibodies (anti-nuclear antibodies, anti-DNA, SSA, SSB, RNP and Sm) were known for 116/195 samples. In a first study we investigated 19 patients with paired serum samples obtained at different occasions and mostly showing activation of the classical complement pathway. Here we primarily wanted to evaluate the effect of disease activity on IC-induced cytokine production. In a second study 78 samples with normal classical complement function and known autoantibody status were studied. Samples were PEG-precipitated and 10% PEG precipitates were added to serum-free cultures of normal peripheral blood mononuclear cells, and the levels of IL-10, IL-6 and IL-12p40 were measured in the supernatants after 20 hours. Classical complement pathway function, levels of C3, C3d and circulating IC were measured in all samples.
In paired analysis, increased complement activation (lowered classical function and raised C3d/C3 ratios) was associated with increased IL-10 production (P = 0.046). For IL-6 there was a parallel but non-significant trend. Decreased C3 levels were also associated with increased IL-12 production (P = 0.03). When investigating these complement-activating sera in cross-section we found increased IL-10 levels in ENA+ (P = 0.03) and DNA+ samples (P = 0.046) compared with ENA- and DNA- samples with the largest impact of anti-SSA/SSB (P = 0.02). The same pattern was found for IL-6.
In sera with normal complement levels there was no association between antibody status and IL-10 production, but IL-12 levels were increased in ENA+ cultures (P = 0.01) and linked to anti-SSA/SSB (P < 0.0001). There was no association to anti-DNA.
Circulating IC in serum and PEG IC-induced IL-10 production correlated in samples from active SLE (r = 0.55, P = 0.003), and we therefore used circulating IC as a surrogate marker for IC-induced IL-10 production in statistical evaluation of all sera with known antibody status. In analysis of variance with circulating IC as the dependent variable we found a strong interaction between decreased complement function and the occurrence of ENA, especially anti-SSA (complement P < 0.0001, anti-SSA P = 0.0001, interaction P = 0.002). Anti-DNA showed no association.
We have found a link between classical complement activation, levels of circulating IC and induction of the cytokines IL-10, IL-6 and IL-12 with strong interaction between complement activation and presence of autoantibodies, notably anti-SSA. We hypothesize that, in quiescent SLE, autoantibodies circulate in monomeric form. In active SLE apoptotic cells release SSA forming IC with circulating anti-SSA, with subsequent induction of cytokines like IL-10.
PEG-precipitated SLE IC induce the production of IL-10 and IL-12 with parallel patterns depending on complement and autoantibodies. Inverse regulation of IL-10 and IL-12 in SLE therefore does not seem to depend on SLE-specific IC.