Functional consequences of anti-C1q autoantibodies from systemic lupus erythematosus patients

The complement system is a key component of the innate immune system. It consists of several plasma and cell-associated proteins and acts as an enzymatic-driven protein cascade. The complement system can be activated via at least three different pathways: the classical pathway (CP), the lectin pathway (LP), and the alternative pathway (AP). C1q, the recognition and starter molecule of the CP of complement, is considered to be involved in the pathogenesis of systemic lupus erythematosus (SLE). This observation is based on the fact that homozygous C1q deficiency is the strongest known genetic disease susceptibility for the development of SLE in humans. In addition, C1q-deficient mice develop a SLE-like disease. However, most SLE patients do not suffer from primary C1q deficiency, moreover, aberrant complement activation is accounted for secondary hypocomplementemia. Another reason for low C1q levels are autoantibodies directed against C1q (anti-C1q) which are present in 20-50% of SLE patients. The occurrence of anti-C1q not only correlates with hypocomplementemia, but additionally, also with the occurrence of severe lupus nephritis. The association of anti-C1q with lupus nephritis suggests a pathogenic role of these autoantibodies in this inflammatory kidney disease. Even though, anti-C1q correlate with renal involvement, the direct evidence how these autoantibodies contribute to the pathogenesis of lupus nephritis is not yet available. Animal studies demonstrated that anti-C1q were only pathogenic in combination with predeposited glomerular C1q-containing immune complexes. Trouw et al. proposed that extensive anti-C1q-mediated complement activation might damage the kidney and lead to the infiltration of immune cells resulting in glomerular injury by Fcgamma receptor (FcγR)-mediated mechanisms suggesting a role for both, the complement system as well as phagocytes, in lupus nephritis. Nevertheless, little knowledge is available on the pathogenic properties of these autoantibodies and their biological function is not well defined. The aim of my thesis was to analyze the interaction of anti-C1q with the complement system and the down-stream effects of anti-C1q on macrophages. The following questions have been addressed during my thesis: Part 1: Do anti-C1q trigger the activation of the complement system? Anti-C1q strongly correlate with the occurrence of lupus nephritis and hypocomplementemia in SLE. Although a direct pathogenic role of anti-C1q has been suggested, the assumed complement-activating capacity remains to be elucidated. Using an ELISA-based assay, we found that anti-C1q activate the CP and LP depending on the anti-C1q immunoglobulin-class repertoire present in the patient’s serum. IgG anti-C1q resulted in the activation of the CP as reflected by C4b deposition in the presence of purified C1 and C4 in a dose-dependent manner. The extent of C4b deposition correlated with anti-C1q levels in SLE patients but not in healthy controls. Our data indicate that SLE patient-derived anti-C1q can activate the CP and the LP but not the alternative pathway of complement. These findings are of importance for the understanding of the role of anti-C1q in SLE suggesting a direct link to hypocomplementemia. Part 2: Do anti-C1q induce a proinflammatory phenotype in macrophages? Anti-C1q are frequently found in patients with SLE and correlate with the occurrence of proliferative lupus nephritis. A previous study of anti-C1q in experimental lupus nephritis demonstrated an important role of FcγR in the pathogenesis of lupus nephritis suggesting a direct effect on phagocytes. Therefore, we developed an in vitro model to study the effect of SLE patient-derived anti-C1q bound to immobilized C1q (imC1q) on human monocyte-derived macrophages (HMDMs) obtained from healthy donors and SLE patients. HMDMs were investigated by analyzing the cell morphology, LPS-induced cytokine profile, surface marker expression, and the phagocytosis rate of apoptotic Jurkat cells. Morphologically, bound anti-C1q induced cell aggregations of HMDMs when compared to imC1q or IgG alone. In addition, anti-C1q reversed the effect of imC1q alone shifting the LPS-induced cytokine release towards a proinflammatory response. By FcγR-blocking experiments, the secretion of proinflammatory cytokines was found to be mediated via FcγRII. The anti-C1q-induced inflammatory cytokine profile was accompanied by a downregulation of CD163 and an upregulation of LPS-induced CD80, CD274, and MHC class II. Finally, HMDMs primed on bound anti-C1q versus imC1q alone displayed a significantly lower phagocytosis rate of early and late apoptotic cells accompanied by a reduced MerTK expression. Interestingly, anti-C1q-dependent secretion of proinflammatory cytokines was similar in SLE patient-derived cells with the exception of IL-10 being slightly increased. In conclusion, anti-C1q induced a proinflammatory phenotype in HMDMs reversing the effects of imC1q alone. This effect might exacerbate underlying pathogenic mechanisms in lupus nephritis. Part 3: Do anti-C1q influence C1q secretion by macrophages? Anti-C1q are frequently found in patients with SLE. They strongly correlate with the occurrence of lupus nephritis and low circulating C1q levels. Previous studies demonstrated that myeloid cells, i.e., dendritic cells and macrophages, are a major source of C1q. However, a direct effect of anti-C1q on C1q secretion by macrophages has not yet been established. In the present study, we investigated the C1q secretion profile of HMDMs obtained from healthy donors and SLE patients in vitro. The effect of SLE patient-derived anti-C1q bound to imC1q and imC1q alone on HMDMs was investigated by C1q secretion levels, the expression of membrane-bound and intracellular C1q using flow cytometry and ImageStreamX technology, and testing the ability of secreted C1q to activate the CP of complement. Bound anti-C1q induced significantly higher C1q secretion levels as compared to imC1q alone or healthy donor IgG. The extent of C1q secretion by HMDMs correlated with IgG anti-C1q levels of SLE patients but not of healthy controls. Furthermore, bound autoantibodies and imC1q induced continuous and de novo C1q synthesis as evident by the intracellular C1q content, which correlated with C1q secretion levels. Finally, secreted C1q was able to activate the CP as reflected by C4b deposition. Interestingly, anti-C1q-dependent C1q secretion could also be observed in SLE patient-derived cells. In conclusion, our data indicate that imC1q-bound anti-C1q strongly stimulate the C1q production by HMDMs. Anti-C1q-induced C1q secretion might be an important immune-modulatory factor in SLE. Taken together, I can demonstrate that anti-C1q activate the complement system via the CP and the LP but not via the AP dependent of the anti-C1q Ig-class present in the patient’s sera. Furthermore, I show that anti-C1q induce a proinflammatory phenotype in HMDMs via an FcγRII-dependent pathway. In addition, I can also demonstrate that anti-C1q trigger C1q secretion by HMDMs