Immune-complexes (IC) cause inflammation in systemic lupus erythematosus (SLE) in significant part by engagement of FcγR on neutrophils, monocyte/macrophages and plasmacytoid dendritic cells (pDCs). We previously observed that in the absence of C1q, IC bind to FcγRII on pDCs and potently stimulate IFN-α production; whereas IC containing C1q bind predominantly to monocytes leading to a reduction in IFN-α production by pDCs. Heterogeneity of human peripheral blood monocytes is known, but recently 3 subpopulations have been defined based on surface expression of CD14 and CD16. Both the CD14+CD16- and the CD14+CD16+ monocytes were described in mice as having “inflammatory” properties while the CD14dimCD16+ population is thought to be a “patrolling” monocyte that responds to invading pathogens as well as to circulating ICs. Given the differing profiles of these newly defined monocyte subsets, we asked whether SLE-IC show preferential binding to the different monocyte subsets, how this binding is altered in the presence of C1q and what are the functional consequences in regards to inflammatory cytokine signaling.
IC were formed with diluted SLE serum (1:2000) and 0.5% U937 necrotic cell extract as a source of antigen or diluted SLE serum (1:1000) and Alexa Fluor 647 labeled RNP. Monocyte subsets were purified by flow sorting (FACSAria). IC were added to peripheral blood mononuclear cells (PBMCs) or purified monocyte subsets (> 95% pure) prepared from healthy donors in the presence or absence of C1q for 30 minutes on ice (binding studies) or 20 hours (ELISA and flow cytometry). Cytokines in culture supernatants were quantified by ELISA and expression of cell surface markers and the monocyte activation marker CD86 were determined by flow cytometry.
When monocytes subsets were flow sorted according to cell surface markers, CD14 and CD16, and incubated with SLE-IC in the absence of C1q, the IC bound to all monocyte subsets. Similar to one published study, CD14dimCD16+ monocytes (but not the CD14+CD16+ or CD14+CD16- populations) uniquely responded to SLE-IC stimulation by producing TNF-α. In addition, these IC induced strongest up-regulation of CD86 on the CD14dim subset (two-fold increase of MFI). Whereas the addition of C1q enhanced SLE-IC binding to all monocyte subsets as determined by MFI of the labeled RNP antigen, the distribution of binding changed to favor the CD14+CD16+ monocytes (fold increase = 6.13±0.93) and the CD14+CD16- monocytes (fold increase =7.17±0.91) compared to the pro-inflammatory CD14dim (fold increase = 3.00±0.48, p<0.05 compared to the other subsets).
Under steady state conditions SLE immune complexes bind to all of the presently described monocyte subsets, but induce CD86 up-regulation most strongly on the CD14dim population. The presence of C1q diverts the IC away from the TNF-α producing CD14dim "patrolling" monocytes and towards the less inflammatory CD14+CD16+ and CD14+CD16- monocyte subsets. Our findings point to a novel mechanism by which C1q may protect against inflammation in SLE.
*Contributed equally to the work
Disclosure: P. Ghosh*, None; A. Wiedeman*, None; D. M. Santer, None; V. E. Vlamakis, None; K. B. Elkon, None.