Mannose-binding lectin and the kidney

The complement system is a key component of innate immunity that plays a major role in host defense against invading pathogens. This is effectuated by a direct attack of pathogens, by mediating inflammation and opsonin-dependent phagocytosis, and by induction and amplification of adaptive immunity. In recent years, it has become increasingly clear that the complement system also plays an important role in immune homeostasis in the healthy situation. In this respect, genetic complement deficiencies are not only associated with infectious diseases, but also with auto-immune and inflammatory diseases. Among the complement deficiencies described in humans, deficiency of the complement factor mannose-binding lectin (MBL) has the highest frequency. Due to its large genetic heterogeneity, MBL is a highly interesting and attractive object of study, and many studies have now revealed the association of a variety of human diseases with the MBL status, being either MBL-deficient or MBL-sufficient. These studies demonstrate that MBL can play both an advantageous role and an adverse role, depending on the context. In the present overview, the role of MBL in diseases of the kidney will be presented and discussed. Activation of the complement system involves three known activation pathways, i.e. the classical pathway, the alternative pathway and the lectin pathway ( Figure 1 ) [ 1 , 2 ]. Recently, a fourth pathway has been described as the C2-bypass pathway [ 3 , 4 ]. Each of these pathways has its own mechanism of target recognition, cleavage of the key complement component C3 and activation of the common terminal complement pathway, finally resulting in formation of the membrane attack complex C5b-9. The three pathways of the complement system. Complement can be activated via the classical pathway, involving C1q as a recognition molecule, via the lectin pathway, involving MBL, L-ficolin, H-ficolin or M-ficolin as recognition molecules, or via the alternative pathway, involving the binding of factor B to spontaneously hydrolysed C3 [C3 (H 2 O)], cleavage of factor B by factor D, and stabilization of the alternative pathway C3 convertase by properdin (P) and a protective surface. Both the classical pathway and the lectin pathway lead to cleavage of C4 and C2 and formation of the C3 convertase C4b2a. Cleavage of C3 by C3 convertase initiates the terminal complement pathway, finally resulting into formation of the membrane attack complex (MAC) C5b-9. Other functional products of complement activation include C3b as a major opsonin and the chemotactic factors C3a and C5a, being important pro-inflammatory mediators. The classical complement pathway involves binding of its recognition factor C1q to, for example, immune complexes, leading to activation of C4 and C2 involving enzymatic cleavage by the serine protease C1s, and generation of the C3 convertase C4b2a. The more recently discovered lectin pathway of complement is activated following an interaction of the plasma lectins mannose-binding lectin (MBL), l -ficolin, H-ficolin or M-ficolin with their carbohydrate ligands [ 5–9 ]. This leads to activation of MASPs (MBL-associated serine proteases) present in a pro-enzymatic complex with these lectins. Activated MASP-2, the functional homologue of C1s, generates C4b2a, followed by C3 cleavage. In contrast to the classical pathway and the lectin pathway, the alternative pathway leads to activation of C3 in a C4-independent way, involving factor B, factor D and properdin. MBL is a multimeric C-type lectin present in human serum [ 10 ]. It consists of collagenous tails similar to those of C1q, and C-type lectin domains that can bind to certain carbohydrates, including mannose and N -acetyl glucosamine, in a calcium-dependent fashion. Three MBL gene ( mbl2 ) polymorphisms (SNPs) have been identified that are associated with functional MBL deficiency. These SNPs are located in codon 54 (B genotype) codon 57 (C genotype), and codon 52 (D genotype) of the first exon, encoding the collagenous region of the MBL molecule [ 11–13 ]. MBL exon 1 SNPs, when present on either one or two alleles, hamper the polymerization of the MBL molecule and affect the structure and function of MBL [ 14 , 15 ]. Furthermore, additional SNPs in the promoter and untranslated region of the mbl2 gene modify the basal serum level of MBL [ 16 , 17 ]. Six SNPs of exon 1 and promoter region are in linkage disequilibrium and define seven common MBL haplotypes [ 17 ]. In the laboratory, MBL deficiency can be detected using a number of different approaches. In the first place, MBL genotyping can be performed, providing clear and basic information on the genetic status. The disadvantage of this approach is that MBL function might still be highly variable within the seven haplotypes that can be defined with the most frequent SNPs, most likely explained by additional genetic variation [ 14 ]. Secondly, MBL serum concentrations can be measured by ELISA. Most of such ELISA's have a strong preference for measurement of high molecular weight (wildtype) MBL and, therefore, provide an excellent correlate to MBL function ( Figure 2 ). Since MBL serum levels are stable and hardly affected by for example, serum storage, and since a recent study also shows that these serum levels are extremely constant over time in the same individual [ 18 ], the measurement of MBL serum concentration is highly useful in clinical studies to determine the MBL status. By ELISA, concentrations of functional MBL may vary up to 1000-fold in the human population. Finally, MBL status can be measured functionally, using e.g. MBL complex assays to determine the activity of the MBL–MASP complex, or MBL pathway assays to measure activation of the complement pathway initiated by MBL [ 19 , 20 ]. These assays are the first choice in a clinical diagnostic setting, but in a research setting their use might be hampered, since they require high-quality and well-stored material. MBL concentration and function in relation to genotype. Paired serum and DNA samples of 70 healthy non-selected donors were used [ 14 ]. MBL genotyping revealed 46 donors with MBL exon 1 wildtype alleles (A/A), 21 donors with one SNP in exon 1 (A/O) and three donors with two SNPs in exon 1 (O/O). MBL serum concentration [ 59 ] and MBL complex activity [ 14 ] in serum were assessed as described. MBL complex activity was measured in an assay originally described by Petersen et al . [ 19 ], and measures the functional ability of the MBL-MASP-2-complex to bind to mannan and to activate exogenously added C4. Depending on ethnicity, the total allele frequency of the B, C and D alleles in exon 1 of the MBL gene may be above 40% [ 17 ]. Since these polymorphisms apparently are not subject to a high negative selection pressure, it has been proposed that the polymorphisms are associated with host protection in certain situations [ 21 ]. Furthermore, the clinical effect of MBL deficiency is strongly dependent on the immune status and the pathogen pressure of the individuals tested, and most MBL-deficient individuals are apparently healthy. Therefore, although MBL gene polymorphisms do have important functional consequences for activation of the lectin pathway of complement, most affected persons have alternative mechanisms for target recognition to reach a sufficient level of anti-microbial protection [ 14 ]. However, MBL deficiency may confer a strongly increased risk for infectious disease in immune-compromised individuals [ 22 , 23 ]. IgA nephropathy (IgAN) is a common renal disease, characterized by mesangial deposition of IgA and complement. In vivo and in vitro evidence has been presented for involvement of the alternative pathway, whereas the classical pathway is unlikely to contribute. However, several studies have shown glomerular deposition of C4, as well as circulating C4 activation products in a subpopulation of IgAN patients [ 24 , 25 ], suggesting a contribution of the lectin pathway of complement. Indeed, deposition of MBL in association with IgA in glomeruli was reported in a subpopulation of IgAN patients by several authors [ 26–28 ]. Whereas Matsuda et al . [ 26 ] reported MBL deposition in association with deposition of C2 and C4, proteinuria, subnormal creatinine clearance and mesangial proliferation, these findings could not be reproduced by Lhotta et al . [ 27 ] and by Endo et al . [ 28 ]. Interestingly, MBL deposition was also observed in glomeruli from patients with Henoch Schönlein purpura nephritis [ 29 , 30 ]. Although in the first publication [ 29 ], a relation with complement activation or clinical disease characteristics could not be observed, Hisano et al . [ 30 ] presented a relation between MBL deposition and clinical and histological parameters of disease, including haematuria, proteinuria, mesangial matrix expansion and glomerular sclerosis in children with Henoch Schönlein purpura nephritis. On the basis of in vitro experiments with purified molecules, our group provided evidence for a calcium-dependent interaction between human polymeric but not monomeric IgA and the lectin domain of MBL, suggesting that MBL does bind to IgA-associated carbohydrates [ 31 ]. These studies strongly suggest that human IgA exposes carbohydrates that serve as a ligand for MBL. We hypothesize that such an interaction takes place in glomeruli from IgAN patients. In line with this hypothesis, we recently confirmed, in an Italian–Dutch collaboration, the presence of IgA in the mesangial area of glomeruli from IgAN patients [ 32 ]. We divided our population of IgAN patients in two groups based on the complement deposition profile. About 75% of IgAN patients showed negative glomerular staining for MBL, L-ficolin, MASP, C4d and C4-binding protein, indicating that C3 and C5b-9 activation in these patients most likely occurs via the alternative pathway. In contrast, 25% of IgAN patients show glomerular deposition of MBL, L-ficolin, MASP, C4d but not C1q, which is strongly indicative for activation of complement via the lectin pathway of complement. Importantly, the clinical and histological data clearly indicate that activation of the lectin pathway of complement is associated with more severe renal damage, as demonstrated by proteinuria, the presence of renal failure, and histological findings of mesangial proliferation, extracapillary proliferation, glomerular sclerosis and interstitial fibrosis [ 32 ]. Taken together, the available data suggest that activation of the lectin pathway of complement in glomeruli from IgAN patients, involving not only MBL but also L-ficolin [ 32 ], plays an adverse role in this disease. Increased disease progression in association with glomerular lectin pathway activation may possibly be ascribed to the increased production of pro-inflammatory activation products of the complement system. Although the pathogenesis of type I and type II diabetes is different, both diseases may show similar complications upon disease progression, as a consequence of elevated blood glucose levels. Vascular complications of type I and type II diabetes are common and a leading cause for renal failure, frequently leading to dialysis and transplantation. A number of studies have, up to now, investigated the potential role of MBL in diabetes. Type I diabetic patients have increased plasma levels of MBL [ 33 , 34 ]. This is also found at disease onset [ 35 ]. The increased levels of circulating MBL in type I diabetics as compared to healthy aged-matched controls are most apparent in carriers of alleles that promote a high production of MBL [ 35 ]. However, the difference could not be explained by genetic differences, indicating that the increased levels of MBL are probably not directly related to susceptibility to the disease. It has been suggested that the high production of MBL in type I diabetics is related to impaired insulin production and subnormal concentrations of insulin in the portal vein [ 34 ]. The latter fact would also be apparent during insulin substitution therapy, due to systemic administration. In line with this hypothesis, treatment of severely ill patients with insulin resulted in less production of MBL and suppressed the acute phase response as assessed by CRP production [ 36 ]. Alternatively, MBL production could be increased as a result of hyperglycaemia, as can be deduced from the reported correlation between MBL levels and HbA1c [ 37 ]. Interestingly, type II diabetic patients were reported to have normal levels of circulating MBL [ 38 ], possibly supporting a role for insulin in MBL regulation. In a population of albuminuric type I diabetic patients without overt albuminuria, MBL serum levels were correlated with the urinary albumin excretion [ 33 ]. In a follow-up study, these authors showed that type I diabetic patients with diabetic nephropathy showed higher levels of circulating MBL than diabetic controls without proteinuria [ 34 ]. Furthermore, this association could be confirmed by MBL genotyping: diabetic nephropathy was clearly associated with the presence of MBL alleles that promote high MBL production [ 34 ]. MBL levels measured early in the disease were demonstrated to be an independent risk factor for the later development of persistent micro- or macro-albuminuria [ 39 ]. In a recent analysis of type II diabetic patients, high plasma levels of MBL were shown to predict albuminuria only when present in combination with high levels of CRP [ 38 ]. Importantly, in type II diabetics, a high MBL level is a significant and independent risk factor for death, probably mainly explained by cardiovascular disease [ 38 ]. Recently, the role of MBL in diabetic injury has been addressed experimentally, using streptozotocin-induced diabetes in mice. In this experimental model, a role for MBL in development of renal injury could not yet be unequivocally demonstrated [ 40 ]. Taken together, MBL plays an unfavourable role in diabetic patients, most likely involving a pro-inflammatory role of MBL and complement activation at the vascular level. At present, to our knowledge, no data are available concerning the binding of MBL to proteins modified by high glucose. Basic studies in this direction would contribute significantly to the understanding of the role of MBL in diabetic vascular injury. Renal ischaemia/reperfusion (I/R) injury is a major issue in nephrology, both in the situation of renal transplantation and in other diseases characterized by acute renal failure. A large number of experimental studies have provided evidence that complement is involved in I/R injury (reviewed in [ 41 ]). The complement activation pathway involved could be the classical pathway, the alternative pathway, and/or the lectin pathway, depending on the model systems studied. For experimental renal I/R injury, it has been shown that the terminal complement pathway and the membrane attack complex are involved, whereas C4 is not required [ 42–45 ], suggesting a role for the alternative pathway. A number of studies provide evidence that activation of the lectin pathway can be responsible for complement activation and related inflammation in I/R injury [ 46 ]. Cultured endothelial cells that were subjected to oxidative stress bind MBL and show deposition of C3 upon their exposure to human serum [ 47 ]. In an in vivo model for myocardial infarction [ 48 ], treatment of rats with a blocking antibody directed against rat MBL showed significantly less myocardial damage upon occlusion of a coronary artery than controls. Inhibition of MBL resulted also in reduced inflammation and expression of pro-inflammatory genes [ 48 ]. Whereas humans have one functional MBL gene, mice and rats produce two MBL species, named MBL-A and MBL-C. Recently, experimental studies became available, using mice that are genetically deficient for both MBL-A and MBL-C, showing that these mice were protected against intestinal and myocardial I/R injury [ 49 , 50 ]. In renal I/R injury in the mouse, clear deposition of MBL was demonstrated in tubular and endothelial cells [ 51 , 52 ]. Furthermore, mice deficient for MBL-A and MBL-C were partially protected against renal I/R injury [ 52 ]. Together, these studies confirm that MBL and the lectin pathway of complement are involved in I/R injury. With respect to renal I/R injury, several pathways are probably involved, including MBL-mediated complement activation. A recent study suggests that MBL may bind to natural IgM antibodies that recognize auto-antigens in ischaemic tissue, thereby activating the lectin pathway [ 53 ]. Further studies are required for the definition of the mechanisms involved. Renal transplantation involves early immunological and non-immunological processes including ischaemia and reperfusion, leading to tissue damage. Lessons from patients and experimental models have indicated that early graft damage determines the long-term outcome of a graft. Therefore, we hypothesize that complement contributes to early and late graft damage upon renal transplantation. Several studies point to a role for complement in renal transplantation. Inhibition of complement by soluble complement receptor-1 prolonged graft survival and decreased complement activation and tissue injury in a model of renal allotransplantation in the rat [ 54 , 55 ]. A more recent study showed a critical role for local complement production in the graft, using an animal model for allotransplantation. Kidneys from C3-deficient mice showed long-term survival when transplanted in MHC-incompatible C3 +/+ without immune whereas C3 +/+ from the same were subject to acute [ ]. in a recent study in human renal transplantation patients, the C3 was shown to be a risk factor for renal [ 57 ]. These studies provide evidence that C3 contributes to renal but they do not provide into the mechanism of C3 activation. A recent study in a renal transplantation model in contrast to studies in C3-deficient mice [ ], deficiency of C4, a key complement factor of both the classical and the lectin pathway of complement, does not modify the survival of renal [ ]. These studies suggest that but not C4, is involved in acute in this model, and may suggest a role for the alternative pathway in C3 activation. the of our knowledge, studies directly the role of of the lectin pathway in experimental transplantation have not yet been on the adverse role of MBL in inflammatory diseases, and on the contribution of MBL to I/R injury, we that MBL might play an unfavourable role in human renal transplantation via its contribution to complement activation. this hypothesis, we the MBL status of of renal [ 59 ]. were divided on basis of their serum MBL concentration into two i.e. high MBL and MBL patients with high MBL levels to have a significantly graft survival than patients with a MBL level. We could that the graft in the group was due to ( Figure 3 ). for renal in relation to MBL status. In of renal from graft later than 3 following transplantation was assessed 10 of were divided into two groups based on their MBL serum MBL ( and MBL ( The of graft was significantly higher in the group which could be ascribed to graft from [ 59 ]. In follow-up to these we a similar analysis in diabetic patients with transplantation. In this study, we were to confirm the protective effect of MBL not only with respect to renal but for MBL to be a major in survival of these patients, and high levels of MBL were associated with a of for The adverse effect of MBL was explained by cardiovascular in this population [ ]. In to the mechanisms and in the local of a renal early analysis of complement deposition in human is both following early I/R injury as during In a study, MBL deposition was shown in early of from donors [ 51 ]. Renal deposition of a for acute and could not be associated with MBL deposition [ , ], but recently an association with deposition of recognition molecule of the lectin pathway, has been presented [ ]. Together, these studies indicate a role for complement and for the lectin pathway in human renal injury. The mechanisms of complement activation and the contribution of complement to the are and experimental in vivo and in vitro studies are However, it is likely that early graft injury, either by the transplantation I/R or by early may the binding of MBL and other innate immune to and cause complement activation and production of inflammatory mediators. These processes may lead to inflammation and as well as amplification of graft recognition by the adaptive immune system. Furthermore, local production of complement including C3 contributes to early and injury [ ]. The kidney is an that is to immunological injury, and complement plays a role in many renal disease processes [ 59 , ]. The present that MBL, a recognition molecule of a more recently discovered complement activation pathway, plays an adverse role in a number of renal inflammatory diseases diabetic renal I/R injury, renal [ ]). This adverse role of MBL is in contrast to the role of MBL in protection against and infectious disease. In this we recently reported that a status is strongly protective against severe in of [ 23 ]. We hypothesize that the contribution of MBL to renal injury is due to the deposition of circulating MBL in the followed by activation of the complement and production of pro-inflammatory data demonstrate that circulating MBL is by the whereas production of circulating MBL could not be [ 23 ]. The kidney does not show for MBL [ ]. However, we do not that MBL might be in the kidney inflammatory for by cells in response to inflammatory or by inflammatory The latter is by data suggesting level production of MBL in the and by cells [ ]. local MBL, as has now been proposed for may contribute to the inflammatory Further studies are required to the different mechanisms involved in MBL-mediated renal injury. The recent of mice will of the role of MBL and the lectin pathway in experimental renal disease. It is that into the role of the lectin pathway of complement in renal inflammatory disease will lead to development of and We the for the for our of . A on this subject was in in the of the