Introduction
Systemic lupus erythematosus (SLE) is a complex and heterogeneous autoimmune disease affecting multiple organs that is characterized by circulating antibodies to nuclear antigens. Many studies have demonstrated a strong genetic component to SLE. Several susceptibility loci have recently been identified in genes encoding proteins involved in many immunological pathways [
1], including B-cell signaling and development, cytokine production [
2], the type I interferon pathway [
3,
4], signaling through Toll-like receptors, and neutrophil function [
5].
One of the immune system cascades involved in the etiopathogenesis of SLE is the complement system. Complement is a pivotal part of the innate immunity, protecting the host from infections and participating in many processes that maintain tissue homeostasis [
6]. In active SLE, immune complex deposition and complement activation contribute to tissue inflammation and damage. On the other hand, inherited deficiencies of complement components such as C1, C2 and C4 strongly predispose to the development of SLE [
7]. This predisposition may be because an intact complement system is important for opsonization and clearance of apoptotic and necrotic cells as well as immune complexes, and thus is important for the prevention of autoimmunity. Additionally, complement is involved in B-cell maturation, differentiation and tolerance. Complement is also involved in microbial defense and thus may be related to SLE exacerbations caused by infections.
Complement is a proteolytic cascade that must be tightly regulated by several soluble and membrane-bound inhibitors in order to prevent damage to own tissues. These inhibitors are typically built of complement control protein (CCP) domains and are mainly encoded by the RCA (regulators of complement activation) gene cluster located on the long arm of chromosome 1. The present study was focused on the genes encoding two such proteins: CD46 encoding membrane cofactor protein (MCP), and CFH encoding factor H (FH). MCP is a cell-bound inhibitor, while FH circulates in blood. Nearly all human cell types, with the exception of erythrocytes, express MCP. This protein acts as a cofactor to serine proteinase factor I (FI), which is able to degrade activated complement components C3b and C4b and thereby to inhibit all pathways of complement. MCP is composed of four CCP domains followed by a serine/threonine-rich region, a transmembrane domain and a small intracellular domain. FH is the major soluble inhibitor of the alternative pathway of complement, serving as a cofactor to FI in degradation of C3b. FH is composed of 20 CCP domains, some of which have a high degree of homology with FH-related proteins 1 to 5 (CFHR1 to CFHR5).
Immune complexes generated in SLE can be passively trapped in kidney glomeruli but also directly bound to glomerular structures, causing a wide range of renal lesions including glomerulonephritis, vasculopathy and tubulointerstitial disease [
8]. Defects in adequate inhibition of complement caused by inherited or acquired deficiencies of complement inhibitors could thus be involved in development and exacerbations of SLE nephritis. Importantly, inherited defects in complement inhibitors have already been associated with several kidney diseases. Complete deficiency of FH leads to membranoproliferative glomerulonephritis [
9], complete deficiency of FI results in glomerulonephritis [
10], while heterozygous mutations in genes encoding FH, FI and MCP result in atypical hemolytic uremic syndrome (aHUS).
Because of the well-established role of complement in SLE and the frequent genetic deficiencies of complement inhibitors in kidney diseases, we hypothesized that mutations or polymorphisms in complement inhibitors may be associated with SLE, and in particular with SLE nephritis. Recent genome-wide association studies have been successful in identifying a number of SLE-associated genes [
1,
11‐
13]. However, these studies only investigated the effect of single common variants. In the present study we performed Sanger sequencing of all exons in the
CD46 and
CFH genes in two cohorts of SLE patients collected in southern and mid Sweden to identify rare variants with a potential effect on SLE and SLE nephritis. We also analyzed the effect of haplotypes in
CFH and
CD46 that are known to affect related traits.
Discussion
Because complement is intricately involved in SLE and mutations/polymorphisms in complement inhibitors are associated with a number of kidney diseases, we hypothesized that SLE patients, particularly those with nephritis, may also carry such genetic defects. Kidney involvement is common in SLE, occurring in up to 25 to 30% of affected Caucasian adults at some stage during the course of their disease [
19]. In most cases, renal disease develops within the first 3 years following diagnosis. Lupus nephritis still remains a main morbidity and mortality determinant for patients with SLE and the current treatment is not satisfactory considering the rate of success and adverse side effects. If inadequate control of the complement system caused by genetic predisposition is involved in SLE nephritis, it could become an additional indication for use of the emerging complement inhibitors such as eculizumab, which is a monoclonal antibody inhibiting cleavage and activation of C5.
To test our hypothesis we sequenced all exons of the
CD46 and
CFH genes in two Swedish cohorts of SLE patients. Importantly, we could relate the obtained genetic data to a carefully characterized clinical history of the patients. Interestingly, even though we did not find significant association between the presence of mutations in
CD46 and
CFH and SLE or SLE nephritis, we observed that these mutations are associated with a younger age at onset of glomerulonephritis. The treatment of all patients with SLE was determined by the clinical manifestations and was similar according to clinical praxis in the rheumatology clinics in southern and mid Sweden, in accordance with EULAR recommendations [
20]. There is therefore no reason to believe that different treatment regimens could have affected the time of nephritis onset.
The observed earlier onset of nephritis in SLE patients carrying mutations in
CD46 and
CFH is consistent with a recently published study showing that FH deficiency accelerates development of lupus nephritis in MRL-lpr mice, which share many features of human SLE including production of autoantibodies and consumptive hypocomplementemia [
21]. Histopathologic findings in these animals included marked deposition of immune complexes containing C3 in glomerular subendothelial, mesangial and subepithelial locations and glomerular inflammation with infiltrated neutrophils and macrophages [
21]. Complement has complicated and paradoxical roles in SLE. Although deficiencies of early components of the complement cascade are associated with development of SLE due to their role in clearance of dying cells and tissue debris [
7] and their importance in development of tolerance [
22], clearance of immune complexes [
23] and cytokine regulation [
24], activation of complement at later stages of C3 and beyond also contributes to the pathogenesis of SLE. Studies using experimental models showed that development of lupus nephritis was dependent on generation of C5a in glomeruli [
25] and on the presence of iC3b in glomerular immune complexes [
26]. Such observed complement activation in glomerulus was triggered by immune complexes. MCP and FH are both localized to glomerular capillary walls, where they attenuate complement activation under normal conditions. However, this level of protection may be compromised by inherited or acquired deficiency of these proteins.
The A353V polymorphism in MCP has previously been shown to affect the ability of MCP to control the alternative pathway activation [
27]. This conservative amino acid substitution in the transmembrane domain did not affect the ability of recombinant MCP to bind C3b/C4b and to act as a cofactor to FI in the fluid phase. The mutant was defective in its regulatory activity, however, when embedded in the membrane - although the mechanism underlying this impairment of function was not defined. The A353V polymorphism has been identified in several patients with renal pathology such as aHUS, glomerulonephritis with C3 deposits and HELLP syndrome (hemolysis, elevated liver enzymes, and low platelets) [
27]. We did not detect association of this polymorphism with SLE in general or SLE nephritis in the current study, however, and we found the polymorphism in 5.8% healthy controls analyzed in the present study and in 3.1% of 192 healthy controls of Caucasian origin included in the study of spontaneous pregnancy loss (manuscript in preparation, Mohlin F, Mercier E, Fremeaux-Bacchi V, Liszewski K, Atkinson JP, Gris JC, Blom AM). The NCBI database estimates 2% frequency for this SNP in European populations [
28]. Taken together, these frequencies suggests that the A353V polymorphism in MCP cannot be a strong causative factor for SLE/SLE nephritis, and most probably not aHUS either, but could be a modifying factor in the presence of additional defects in complement regulation often observed in these patients.
The
CD46
ggaac
haplotype has been previously shown to be overrepresented in aHUS patients compared with controls (odds ratio = 2.68). Interestingly, this association was mainly due to the aHUS patients with mutations in
CFH,
CD46 or
CFI (odds ratio = 5.25), whereas aHUS patients without identified mutations in
CFH,
CD46 or
CFI showed no differences with the control group [
17]. The functional consequences of the five synonymous SNPs in the
CD46
ggaac
haplotype have not been fully elucidated, but it has been shown that the rs2796267 and the rs2796268 SNPs located in the promoter region are involved in the transcriptional activity potentially due to disruption of the CBF-1/RBP-Jk binding site [
17]. Two other SNPs in the
CD46
ggaac
haplotype block are intronic, while the third lies in the 3' UTR. In the current study we found weak indication that the
CD46
ggaac
haplotype could perhaps be protective for SLE, and in particular SLE with nephritis, whereas the
CD46
agaac
could represent a risk haplotype. These indications must now be assessed in a larger patient material before final conclusions are made.
Several nonsynonymous polymorphisms in
CFH have been linked to age-related macular degeneration [
29,
30] and susceptibility to meningococcal disease [
31], but not to rheumatoid arthritis [
32] or coronary heart disease [
33]. A large number of studies have shown that I62V polymorphism in CCP1 of FH affects its FI cofactor activity and ability to accelerate decay of convertases [
34,
35] and that the Y402H polymorphism in CCP7 affects binding of FH to several ligands such as C-reactive protein, DNA, dying cells and heparin [
36,
37]. The effect of the D936E polymorphism in CCP16, however, has not yet been analyzed. In the current study we did not detect any significant associations with these three SNPs in
CFH nor any of the haplotypes thereof. This finding is consistent with a recent publication assessing association of genetic variants in
CFH with SLE susceptibility [
38]. The authors studied 60 SNPs covering
CFH and FH-related genes for association with SLE in over 15,000 case-control subjects from four ethnic groups. They found significant allelic associations with SLE in European Americans and African Americans, which could be attributed to an intronic
CFH SNP and intergenic SNP between
CFHR1 and
CFHR4 rather than the exonic SNPs we studied.
Acknowledgements
The authors would like to acknowledge the financial support of the Söderberg Foundation, the Swedish Research Council (K2009-68X-14928-06-3, K2011-52X-12672-14-3 and 2008-2201), the Swedish Foundation for Strategic Research, the Swedish Rheumatism Association, Swedish Society of Medicine, the National Board of Health and Welfare and Skåne University Hospital, and the Swedish Heart-Lung Foundation, as well as Österlund, Greta and Johan Kock, King Gustaf V's 80th Birthday, Knut and Alice Wallenberg, Inga-Britt and Arne Lundberg, Professor Nanna Svartz and Pharmacist Hedberg Foundations.
Competing interests
The authors declare that they have no competing interests.
Authors' contributions
AMB initiated the collaborative project, designed the study, monitored data collection, designed the statistical analysis plan, and drafted and revised the paper. AJ collected clinical data and samples for patients in Lund, contributed to design of the study and revised the draft paper. SCN cleaned and assembled the data, performed statistical analyses, contributed to the design of the study and revised the draft paper. EA performed statistical analyses and revised the draft paper. ES collected clinical data and samples for patients at Karolinska Hospital, contributed to design of the study and revised the draft paper. IG collected clinical data and samples for patients at Karolinska Hospital, contributed to design of the study and revised the draft paper. KGE collected clinical data and samples for patients in Uppsala and revised the draft paper. AB collected clinical data and samples for patients in Lund and revised the draft paper. AZ collected clinical data and samples for patients at Karolinska Hospital and revised the draft paper. M-LE collected healthy control samples from Uppsala and revised the draft paper. LT provided samples of healthy controls from Lund and revised the draft paper. LR collected clinical data and samples for patients in Uppsala and revised the draft paper. GN collected clinical data and samples for patients in Uppsala, contributed to design of the study and revised the draft paper. GS collected clinical data and samples for patients in Lund, contributed to design of the study and revised the draft paper. All authors read and approved the final manuscript.