Table of Contents Author Guidelines Submit a Manuscript
Journal of Immunology Research
Volume 2014, Article ID 197548, 10 pages
http://dx.doi.org/10.1155/2014/197548
Review Article

Cellular Signaling and Production of Galactose-Deficient IgA1 in IgA Nephropathy, an Autoimmune Disease

1Department of Microbiology, University of Alabama at Birmingham, 845 19th Street South, BBRB 762, Birmingham, AL 35294, USA
2Department of Medicine, University of Alabama at Birmingham, 845 19th Street South, Birmingham, AL 35294, USA
3Department of Radiation Oncology, University of Alabama at Birmingham, 845 19th Street South, Birmingham, AL 35294, USA

Received 23 April 2014; Accepted 25 June 2014; Published 23 July 2014

Academic Editor: Mario Clerici

Copyright © 2014 Colin Reily et al. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

Abstract

Immunoglobulin A (IgA) nephropathy (IgAN), the leading cause of primary glomerulonephritis, is characterized by IgA1-containing immunodeposits in the glomeruli. IgAN is a chronic disease, with up to 40% of patients progressing to end-stage renal disease, with no disease-specific treatment. Multiple studies of the origin of the glomerular immunodeposits have linked elevated circulating levels of aberrantly glycosylated IgA1 (galactose-deficient in some O-glycans; Gd-IgA1) with formation of nephritogenic Gd-IgA1-containing immune complexes. Gd-IgA1 is recognized as an autoantigen in susceptible individuals by anti-glycan autoantibodies, resulting in immune complexes that may ultimately deposit in the kidney and induce glomerular injury. Genetic studies have revealed that an elevated level of Gd-IgA1 in the circulation of IgAN patients is a hereditable trait. Moreover, recent genome-wide association studies have identified several immunity-related loci that associated with IgAN. Production of Gd-IgA1 by IgA1-secreting cells of IgAN patients has been attributed to abnormal expression and activity of several key glycosyltransferases. Substantial evidence is emerging that abnormal signaling in IgA1-producing cells is related to the production of Gd-IgA1. As Gd-IgA1 is the key autoantigen in IgAN, understanding the genetic, biochemical, and environmental aspects of the abnormal signaling in IgA1-producing cells will provide insight into possible targets for future disease-specific therapy.

1. Introduction

IgA nephropathy (IgAN) is the most common primary glomerulonephritis. IgAN was first described by Berger and Hinglais in 1968 based on observation of “intercapillary deposits of IgA-IgG” in renal tissue of patients with hematuria [1]. The initial manifestation of disease is frequently synpharyngitic hematuria, that is, macroscopic hematuria concurrent with upper respiratory tract infection [1, 2]. Progressive renal damage often culminates in end-stage renal disease (ESRD) that requires dialysis or transplantation as renal replacement therapy [3, 4]. In patients undergoing renal transplantation, the high rate of recurrence, up to 50–60%, suggests that the disease is of extrarenal origin [5]. Diagnosis of IgAN requires a renal biopsy and is based on detection of IgA immunodeposits in the glomerular mesangium [2]. Glomerular pathological changes may include mesangial hypercellularity and matrix expansion, variable degrees of segmental scarring, and crescent formation [68]. Fibrotic changes in the interstitium are observed in patients with more severe disease.

Human IgA has two subclasses, IgA1 and IgA2, with IgA1 constituting about 85% of the total IgA in the circulation. IgA in the glomerular deposits of IgAN patients is exclusively of the IgA1 subclass [9] and predominantly in polymeric form. IgA1 is produced in two distinct molecular forms, monomeric and polymeric (the latter consisting of 2–4 J-chain-linked monomers) [10]. IgA1-producing cells in mucosal tissues secrete predominantly polymeric IgA1; in contrast, the monomeric form accounts for >90% of the IgA1 in the circulation [10].

IgA1 contains both N-and O-linked glycans. In N-glycans, the first sugar added, N-acetylglucosamine (GlcNAc), is linked to a nitrogen atom in a N-X-S/T sequence motif, whereas O-glycans have N-acetylgalactosamine (GalNAc) bound to the oxygen atom of serine or threonine. Both subclasses of IgA are N-glycosylated, but only the IgA1 subclass has O-glycans. IgA1 O-glycosylation sites are located in the hinge region, with 3–6 O-glycans per heavy chain. IgA1 O-glycans are core 1 structures that normally consist of GalNAc with a β1,3-linked galactose (Gal). The final extension of the core 1 structure is an addition of sialic acid to Gal, GalNAc, or both (Figure 1) [1113].

197548.fig.001
Figure 1: IgA1 and hinge-region O-glycosylation in IgAN patients. Figure of monomeric IgA1, with the hinge-region amino-acid sequence below to show the S/T residues that are O-glycosylated (numbered in red). Three to six core 1 O-glycans are attached per hinge region, with N-acetylgalactosamine (white square) being the first sugar added; galactose (blue circle) may be then attached. Both glycans may be sialylated by N-acetylneuraminic acid (pink diamond), attached to N-acetylgalactosamine in an α2,6 linkage and to galactose in an α2,3 linkage. Patients with IgAN have elevated levels of circulatory IgA1 with some galactose-deficient O-glycans, consisting of terminal or sialylated N-acetylgalactosamine.

Patients with IgAN have elevated serum levels of IgA1 that lacks Gal in one or more O-glycans, thus presenting with a terminal GalNAc or sialylated GalNAc (Gal-deficient IgA1; Gd-IgA1) [14]. Upon completion of glycosylation, IgA1 moves out of the Golgi apparatus and the monomers are processed into polymeric forms by addition of joining the (J) chain at the tail end of the heavy chain (Figure 2) [10, 15, 16]. Currently, J chain is not thought to play a role in Gd-IgA1 production, as it does not bind to IgA1 until after O-glycosylation occurs and IgA1 exits the Golgi apparatus [17].

197548.fig.002
Figure 2: Pathways of IgA1 O-glycosylation. The hinge region of IgA1 is O-glycosylated in the Golgi apparatus, a structure composed of multiple layers compacted together, with three designated regions: cis, medial, and trans. These structures support differential localization of various glycosyltransferases, an arrangement that confers a level of specificity and order of addition of sugar moieties. Path I shows normal (in healthy controls; HC) sequential addition of N-acetylgalactosamine (GalNAc), galactose (Gal), and N-acetylneuraminic acid (NeuAc) by their respective enzymes, GalNAc-transferase, core 1 galactosyltransferase (C1GalT1), and sialyltransferases, ST6GalNAc-II (adds NeuAc to GalNAc) and/or ST3Gal (adds NeuAc to Gal). There are multiple GalNAc-transferases that can initiate IgA1 O-glycosylation. Path II shows a possible deviation in the localization of ST6GalNAc-II that could lead to a galactose deficiency due to premature sialylation of GalNAc. The addition of sialic acid blocks a later addition of galactose. Path III shows a scenario in which a reduced expression of C1GalT1 in IgA1-secreting cells from IgAN patients would decrease the addition of galactose to IgA1. Pathways II and III represent possible deviations from normal glycosylation that could lead to production of galactose-deficient IgA1. The polymeric form of IgA1 is formed after exit of IgA1 from the Golgi apparatus through the addition of J chain that binds covalently to the tail pieces of the heavy chains. The resultant polymeric IgA1 may be a dimer or higher oligomer.

A major difference between IgA1 and IgA2 is the extended hinge region in IgA1 heavy chain with nine potential O-glycosylation sites, of which up to six are occupied. An early study reported decreased binding of IgA1 from sera of patients with IgAN to jacalin (a lectin specific for GalNAc-Gal disaccharide) compared to that in sera of healthy controls [18]. These observations were later explained by the discovery of elevated serum levels of Gd-IgA1 in IgAN patients [14, 19, 20]. This finding prompted the hypothesis that a lower degree of O-glycan galactosylation on this immunoglobulin plays a role in the pathogenesis of IgAN. Indeed, the glomerular deposits of IgA1 of IgAN patients are enriched with Gd-IgA1 [21, 22]. Gd-IgA1 in the blood is primarily in the polymeric form and bound in circulating immune complexes (CIC). Investigators interpreted this finding to indicate that either systemic IgA1-producing cells altered the molecular form of secreted IgA1 or that mucosa-based IgA1-secreting cells had migrated to a location more proximal to the systemic blood flow [23].

In most IgAN patients the serum Gd-IgA1 levels are elevated compared to those of the normal population. In multiplex families with two or more IgAN patients (familial IgAN), almost half of the first-degree relatives have elevated Gd-IgA1 levels, although most exhibit no urinary abnormality [24]. Similarly, for patients with apparently sporadic IgAN (no relative with known kidney disease), about a quarter of asymptomatic blood relatives have an elevated serum Gd-IgA1 level. In contrast, about 5% of the general population and married-in relatives of patients with IgAN have a high serum level of Gd-IgA1. These studies are consistent with a hereditary factor affecting serum Gd-IgA1 levels. Segregation analysis indicated a likely influence of a major dominant gene, on a polygenic background, on the serum Gd-IgA1 level [24]. As most blood relatives with a high serum Gd-IgA1 level are apparently unaffected, elevated amounts of Gd-IgA1 alone are not sufficient for development of clinically apparent IgAN [24].

Previous research has shown that normal metabolism of IgA1 can be slowed by terminal Gal deficiency. IgA1 with enzymatically removed terminal Gal exhibits a longer clearance interval after injection in mice [25]. In human as well as murine hepatocytes, IgA1 is taken up by the asialoglycoprotein receptor that recognizes terminal GalNAc, but not if it has undergone sialylation [26, 27]. It should also be considered that, if Gd-IgA1 is bound by anti-glycan antibody, the aggregate complex is too large to pass into the space of Disse to reach the hepatocytes [28]. Together, these studies indicate that increased serum Gd-IgA1 levels could be related to lower hepatic catabolism rather than simply an increased production.

Because elevated serum Gd-IgA1 levels are associated with IgAN but likely not sufficient to induce clinically manifested disease, postulated models for the mechanisms of disease evolved to a proposal that multiple hits are required for renal injury. Several hypotheses tried to explain CIC formation and renal deposition, including Gd-IgA1 “aggregation,” Gd-IgA1 forming complexes with CD89 (IgA Fc-specific receptor), formation of complexes with environmental or microbial antigens, circulating antibodies that recognized Gal-deficient moieties on Gd-IgA1, and Gd-IgA1 directly binding to extracellular matrix proteins in the glomerular mesangium due to a novel sugar motif [2932]. It is now thought that circulating Gd-IgA1 is recognized by anti-glycan autoantibodies of the IgA1 and/or IgG isotype, resulting in formation of CIC [20, 33]. Some of these pathogenic (nephritogenic) CIC deposit in the glomerular mesangium and activate resident mesangial cells, thus initiating glomerular injury [9, 34].

Recent genome-wide association studies (GWAS) of Han Chinese, European, and American Caucasian populations found several loci that were associated with IgAN [3538]. These loci comprise genes encoding multiple components of the immune system, including the complement pathway (complement factor H-related genes 1 and 3, CFHR1,3); adaptive immunity (major histocompatibility complex, MHC); innate immunity, growth factors, and cytokines (LIF and OSM in the HORMAD2 locus; human αdefensin, DEFA; and tumor necrosis factor superfamily 13, TNFSF13) (Table 1) [36, 37]. The distribution of risk alleles in the loci differed between ethnic groups. Specifically, East Asian populations contained the most risk alleles, followed by Europeans, then Africans with the fewest. The risk differences among the groups were also reflected in the US, such that prevalence rates of IgAN are highest among Asian populations, with progressively lower rates in Caucasians and African-Americans. When effect size of each allele was taken into account, the differences in risk association between these groups were further enhanced [39, 40].

tab1
Table 1: Susceptibility loci for IgAN from multiple GWAS [3540].

Two of the IgAN-associated loci contain genes for cytokines that may affect the circulating levels of IgA1. One locus, the Chr. 22q12 HORMAD2 locus, encompasses two cytokines, leukemia inhibitory factor (LIF) and oncostatin M (OSM) (Table 1). The IgAN risk allele in this locus is associated with elevated serum levels of IgA [36]. LIF and OSM are IL-6-related cytokines that signal through gp130 coupled to their respective cytokine-specific receptor(s) [41]. This family of cytokines has received some attention in the IgAN field, as the level of IL-6 is increased in the circulation of some IgAN patients [42]. Notably, LIF and OSM modulate mucosal immune responses. Recent in vitro studies, using IgA1-producing cells from IgAN patients stimulated with IL-6 family of cytokines, have demonstrated that these cytokines enhanced Gd-IgA1 production in IgA1-secreting cells from IgAN patients [17]. The effects on glycosylation of IgA1 produced by IgA1-secreting cells from healthy controls were minimal, suggesting differential responses of the cells from IgAN patients. A second IgAN-associated locus encompasses the TNFSF13 gene that encodes APRIL (a proliferation-inducing ligand), a cytokine involved in augmenting IgA production in B cells independently of T-cell activation (Table 1) [43]. In Han Chinese with IgAN, a risk allele in this locus is associated with higher levels of serum IgA [37]. The effect of IgAN risk alleles on the expression of the encoded proteins, such as APRIL, is not known, nor is the mechanism by which APRIL may affect production of Gd-IgA1 well understood. How signaling by growth factors and cytokines in IgA1-producing cells from IgAN patients differs from that in healthy individuals is still being investigated, but there is evidence for involvement of enhanced activation of the JAK/STAT pathway [4446].

Currently, there is no disease-specific treatment of IgAN and up to 40% of patients progress to ESRD within 20 years of diagnosis by biopsy [47, 48]; new targets for future disease-specific treatments are needed [48]. A reduction in serum Gd-IgA1 levels would decrease formation of nephritogenic CIC leading to less renal injury. We hypothesize that targeting a specific signaling pathway(s) would normalize expression, activity, and/or localization of specific glycosyltransferase(s) and thereby reduce Gd-IgA1 production. Manipulation of some of the unique signaling pathways promoting production of Gd-IgA1 in IgAN may constitute targets for therapeutic intervention.

2. Gd-IgA1 in IgAN

The role of circulating Gd-IgA1 in the formation of immune complexes was not fully appreciated until IgA1-IgG CIC were isolated and characterized. The Gd-IgA1 was found to be bound to IgG directed against hinge-region O-glycans, without Gal [20, 32, 33, 49]. Subsequent studies revealed that serum levels of Gd-IgA1 and the anti-glycan antibodies were associated with progression to ESRD [50, 51]. Thus, Gd-IgA1 and the corresponding autoantibodies are important in the pathogenesis of IgAN.

A unique resource has been developed for studies of the mechanisms of aberrant O-glycosylation of IgA1: EBV-immortalized IgA1-producing cells derived from peripheral blood mononuclear cells of patients with IgAN and healthy and disease controls. Notably, the degree of Gal deficiency of IgA1 secreted by the cell lines mirrored that of IgA1 in serum of the donors [52]. Thus, EBV immortalization and cloning of IgA1-producing cells did not significantly change the O-glycosylation pattern of the secreted IgA1. Using IgA1-producing cell lines, we have identified changes in some signaling pathways, such as STAT3, and activities of key glycosylation enzymes for the O-glycosylation of IgA1 [45, 46, 52]. These cell lines therefore represent a unique resource for in-depth studies of the glycosylation of IgA1.

3. O-glycosylation Pathways and Glycosyltransferases Involved in Gd-IgA1 Production

O-glycosylation takes place primarily within the Golgi apparatus that consists of multiple vesicles compacted into cis, medial, and trans compartments. These compartments function to delineate specific posttranslational modifications that help to control the order of addition of sugars to proteins. Glycosyltransferase enzymes reside in specific compartments of the Golgi apparatus, allowing the cell to control the glycosylation process at multiple points through localization, expression, and activity of specific enzymes [53, 54].

The initiation of O-glycosylation on glycoproteins by the attachment of GalNAc to a serine or threonine is catalyzed by GalNAc-transferases (GalNAc-Ts) [55]. For IgA1, this process is thought to be mediated primarily by GalNAc-T2, but other GalNAc-Ts are abundantly expressed in IgA1-producing cells and may contribute as well [56]. Core 1 β1,3-galactosyltransferase (C1GalT1) adds Gal to GalNAc, generating a GalNAc-Gal disaccharide. Expression of a C1GalT1-specific chaperone (Cosmc) is necessary for stability of the nascent C1GalT1 protein. Sialyltransferases can then add sialic acid to either saccharide or both. In IgA1-producing cells, α-N-acetylgalactosaminide α-2,6-sialyltransferase 2 (ST6GalNAc-II) adds sialic acid to GalNAc and galactosyl-α-N-acetylgalactosaminide α-2,3-sialyltransferase (ST3Gal) adds sialic acid to Gal [5760].

Analysis of the expression and activity of key enzymes in IgA1-producing cells revealed important differences between cells from patients with IgAN and those from healthy and disease controls, such as patients with other glomerular diseases (including membranous nephropathy, minimal change disease, or immune-complex-associated lupus nephritis) [52]. Specifically, expression of C1GalT1 and Cosmc was lower and, conversely, expression of ST6GalNAc-II was higher in the cells from IgAN patients versus controls. All core 1 O-glycosylation structures have GalNAc as the first sugar in the series and, as depicted in Figure 2, tend to be localized within the cis region of the Golgi apparatus. C1GalT1 is expressed at lower levels in the cells from IgAN patients compared to controls. The decrease in C1GalT1 levels in IgAN cells may also result from the decreased levels of C1GalT1-specific chaperone, Cosmc [45, 61]. The attachment of Gal to GalNAc likely occurs in the medial section of the Golgi apparatus due to the predominant localization of C1GalT1 there (Figure 2, pathway I). ST6GalNAc-II is overexpressed in IgA1-producing cells from IgAN patients. As sialylation of GalNAc in the disaccharide occurs late in the glycosylation process, ST6GalNAc-II is likely compartmentalized in the trans section of the Golgi apparatus. Migration of this enzyme to locations more proximal than that of GalNAc-Ts could lead to premature sialylation of GalNAc, before the addition of Gal (Figure 2, pathway III) [56], that inhibits subsequent attachment of Gal [46]. Thus, changes in the expression and/or localization of key enzymes may explain the differences in the pattern of O-glycosylation of IgA1 between IgAN patients and controls.

4. Signaling in IgA1-Producing Cells

Even with a genetic predisposition to elevated serum Gd-IgA1 levels in patients with IgAN, the onset of the disease requires a trigger resulting in the production of pathogenic CIC containing Gd-IgA1. The molecular nature of the trigger(s) is currently unknown, but it is well recognized clinically that upper respiratory tract infections are frequently associated with the first apparent manifestation of disease, “synpharyngitic hematuria,” that, in some patients, may be recurrent [62, 63]. It has been shown that these episodes of disease activity are associated with elevated serum levels of Gd-IgA1-containing CIC [51, 6466]. This clinical association suggests a mucosal origin of an altered immunity in IgAN. This assumption is supported by findings from GWAS. One of the GWAS IgAN-related susceptibility loci, chromosome 17p13, contains the TNFSF13 gene encoding APRIL [36]. APRIL is involved in T-cell-independent generation of IgA-secreting plasma cells and IgA class switching [43]. Serum levels of APRIL are elevated in some patients with IgAN and the 17p23 risk variant is associated with elevated levels of serum IgA [67]. Moreover, overexpression of a related factor, B-cell activating factor of the TNF family (BAFF), results in autoimmune disease with commensal microbiotoa-dependent glomerular IgA deposits in mice [67]. Furthermore, another IgAN-associated locus, encompassing several genes including those encoding LIF and OSM, also influences serum IgA levels. Risk alleles in this locus are associated with elevated serum IgA levels [36]. LIF and OSM belong to the IL-6 family of cytokines and are expressed in mucosal tissues where they exert immunoregulatory effects [68]. Together, these data implicate growth factors APRIL and BAFF and IL-6 family cytokines in the pathogenesis of IgAN.

Based on the accumulated clinical observations and biochemical and genetic data, we hypothesize that the onset and/or severity and progression of the disease depend on the activation of the immune system and that cytokines and B-cell growth factors play important roles in that aspect. Namely, these factors regulate Gd-IgA1 production and B-cell and plasma-cell maintenance and proliferation. This hypothesis offers several possible scenarios: (1) Gd-IgA1-producing B-cell-specific proliferation and survival are enhanced during mucosal infections, (2) B-cell signaling is altered and directly regulates expression and activity of specific glycosyltransferases, and (3) there is a direct signaling effect on the function of the Golgi apparatus in IgA1-producing cells that increases synthesis of Gd-IgA1.

Currently, there is no robust animal model mimicking the pathogenetic pathways of human IgAN, mainly because the IgA1 isotype is present in only humans and hominoid primates. Serendipitously, an IgA-associated glomerulonephritis developed in mice overexpressing BAFF [67]. The manifestations included hematuria and albuminuria and were associated with IgA-containing CIC deposited in glomeruli [67]. BAFF, encoded by the TNFSF13B gene, is related to APRIL, sharing multiple receptors and functions. BAFF ligand can bind to several receptors, including BAFF-R (transmembrane activator), B-cell maturation antigen (BCMA), and the calcium modulator and cyclophilin ligand interactor (TACI), to activate the NF-κB pathway (Figure 3). BAFF is a B-cell-survival cytokine and, in a mouse model of nephrotic syndrome, induces B-cell hyperplasia and higher serum levels of IgA. Notably, murine IgA resembles human IgA2 and lacks a hinge region for attachment of clustered O-glycans. Alterations in IgA production in humans could be due to class-switching recombination (CSR), changing from IgM or IgG to IgA upon stimulation with BAFF, which can affect B cells at mucosal interfaces [69, 70]. BAFF and/or APRIL levels are elevated in the circulation of some IgAN patients, suggesting a possible effect on circulatory B cells [67, 71]. Future studies will need to characterize mechanisms by which B-cell growth factors affect production of Gd-IgA1 in patients with IgAN and to answer many important questions. For example, do these factors affect IgA1 O-glycosylation in IgA1-producing cells? Do these factors enhance survival of plasma cells already differentiated into Gd-IgA1-secreting cells that would not survive in healthy individuals? Is abnormal signaling in IgA1-producing cells from patients with IgAN induced by these growth factors?

197548.fig.003
Figure 3: Examples of potential signaling mechanisms involved in the O-glycosylation of IgA1. Several signaling mechanisms likely regulate O-glycosylation patterns of IgA1 in the Golgi apparatus. In this scheme, we have highlighted examples of three types of signaling systems that are potential players in the production of galactose-deficient IgA1 (Gd-IgA1). BAFF signaling through the BAFF receptor(s) (TACI, BAFF-R, and BCMA/TRAF) prevents degradation of NF-κB and enhances its subsequent translocation to the nucleus. This mechanism represents an example of a signaling pathway related to survival and proliferation of cells producing Gd-IgA1. Cytokine LIF, that binds the LIF-R/gp130 receptor complex, signals in B cells through STAT1 phosphorylation by JAK, followed by dimerization and translocation of the active STAT dimers to the nucleus. IL-6 activates the STAT3 pathway in B cells via JAK activation at the IL-6/gp130 complex. LIF and IL-6 signaling pathways are examples of a mechanism by which environmental factors may alter expression of specific glycosyltransferases. Activation of the MAPK pathways through various intraextracellular mechanisms can lead directly to the Golgi apparatus. MAPK signaling in B cells is another example of a pathway for an environment-mediated influence to change expression of specific glycosyltransferases, to signal directly to the Golgi apparatus, or both.

Another class of proteins affecting IgA1 production includes IL-6 and related cytokines. IL-6 is a plasma-cell differentiation factor and canonical inducer of IgA1 production; its serum level is elevated in some IgAN patients [46]. IL-6 binds the IL-6R/gp130 complex and activates the JAK/STAT3 pathway, leading to nuclear translocation of STAT3 (Figure 3) [41, 68]. In IgA1-producing cells from IgAN patients, exposure to IL-6 prolongs and significantly increases STAT3 activation compared to cells from controls [72]. IL-6 signaling increases IgA1 production in IgA1-secreting cells from healthy controls and IgAN patients. Moreover, IL-6 enhances production of Gd-IgA1, but only in cells from IgAN patients, further supporting the role of immune activation in exacerbating the underlying pathology [73]. A possible mechanism for STAT3 to increase Gd-IgA1 production is through altered expression of specific glycosyltransferases: increased for ST6GalNAc-II and decreased for C1GalT1 [17, 46].

Activities of the corresponding enzymes mirror the changes in gene expression. These observations suggest a possible mechanism for IL-6/STAT3-mediated Gd-IgA1 production, one that involves regulation of expression and activity of key glycosyltransferases. Thus, changes in O-glycosylation patterns of IgA1 secreted by cells from IgAN patients reflect differences in signaling pathways that ultimately affect the Golgi apparatus.

Other IL-6-related cytokines, LIF and OSM, also activate B cells through similar receptors and associated STAT pathways [68]. The potential role of these cytokines in the pathogenesis of IgAN has been outlined in a GWAS report that found a significant correlation between SNPs in the HORMAD2  locus and prevalence of the renal disease. SNPs in this interval, rs2412971, were associated with decreased serum IgA1 levels in patients and span a region that contains LIF and OSM [40]. LIF and OSM cytokines have been tested on IgA1-producing cells from IgAN patients and controls; both increased production of Gd-IgA1, but only in the cells from IgAN patients [74].

The signaling mechanism involves binding of the ligands to the LIF receptor or OSM receptor that recruits gp130 and activates the JAK/STAT pathway (Figure 3). In IgA1-producing cells, STAT1 is phosphorylated and can induce Gd-IgA1 production in cells from IgAN patients, similar to the activity promoted by IL-6/STAT3-mediated signaling [41]. Notably, in most cases of JAK recruitment through IL-6 family members, downstream kinases are also phosphorylated, such as the mitogen-activated protein kinase (MAPK) pathway. MAPK pathways are known to modulate Golgi morphology and glycosylation functions, as outlined below.

A recent study by Chia et al. screened many signaling genes by using siRNA knockdown and assessed the individual effects on glycosylation patterns and morphology of the Golgi apparatus of HeLa cells [75]. Glycosylation was assessed for changes in terminal sugars using a panel of lectins that recognize specific glycans. These experiments showed that knockdown of dual-specificity phosphatases (DUSP), proteins that negatively regulate MAPK pathways, enhanced MAPK phosphorylation. DUSP siRNA knockdown altered the morphology of the Golgi apparatus and decreased Gal content of O-glycans on newly synthesized glycoproteins. Using this data set, it is possible to search for genes and pathways that may produce the same or similar O-glycosylation pattern as in the hinge region of Gd-IgA1 from patients with IgAN.

As signaling by IL-6/LIF/OSM family members affects IgA1 production in all cells but alters IgA1 O-glycosylation only in cells from IgAN patients, we speculate that increased activation of a signaling pathway(s) plays a direct role in influencing the functions of the Golgi apparatus [46, 74]. Our preliminary data revealed that several signaling pathways, including MAPK, are enhanced in IgA1-producing cells from IgAN patients versus controls (unpublished data). Thus, it is conceivable that increased activation of the MAPK pathway, possibly through environmental factors, alters IgA1 O-glycosylation through direct signaling into the Golgi apparatus in IgA1-producing cells of IgAN patients (Figure 3). These observations link signaling pathways with IgA1 aberrant glycosylation, a model that offers potential for testing new targets for manipulating O-glycosylation through downregulation of “inappropriate” signaling in IgAN.

5. Conclusions

Progressive deterioration of renal function in IgAN patients remains a significant concern, as current therapeutic options do not include a disease-specific approach. Treatment with ACEI/ARB to reduce proteinuria and control of hypertension and, for some patients, immunosuppressants can slow the loss of renal-clearance function [2, 76]. These therapeutic approaches are not tailored to IgAN. The ongoing renal injury would be better controlled with molecular therapies targeted for the disease. The current multihit hypothesis of the pathogenesis of IgAN, wherein Gd-IgA1 production in conjunction with anti-glycan antibodies creates CIC that drive the disease, suggests that limiting either of the two initial hits would provide a long-term benefit [77]. Recent studies have shown that several cytokine- and B-cell-specific survival/signaling pathways apparently play a role in driving aberrant glycosylation of IgA1. Further research is required to more clearly define how these pathways influence IgA1 glycosylation in B cells. Focusing on pathways to reduce production of Gd-IgA1 will likely provide opportunities for development of future disease-targeted treatment of patients with IgAN.

Conflict of Interests

The authors declare that there is no conflict of interests regarding the publication of this paper.

Acknowledgments

This study was supported in part by the NIH Grants T32 DK079337, DK078244, GM098539, and DK082753 and by a gift from the IGA Nephropathy Foundation of America. The authors express their gratitude to all colleagues and collaborators who participated in various aspects of these studies.

References

  1. J. Berger and N. Hinglais, “Intercapillary deposits of IgA-IgG,” Journal d'Urologie et de Néphrologie, vol. 74, pp. 694–695, 1968. View at Google Scholar
  2. R. J. Wyatt and B. A. Julian, “IgA nephropathy,” New England Journal of Medicine, vol. 368, pp. 2402–2414, 2013. View at Google Scholar
  3. S. J. Barbour and H. N. Reich, “Risk stratification of patients with IgA nephropathy,” American Journal of Kidney Diseases, vol. 59, pp. 865–873, 2012. View at Publisher · View at Google Scholar
  4. G. D'Amico, “Natural history of idiopathic IgA nephropathy: role of clinical and histological prognostic factors,” American Journal of Kidney Diseases, vol. 36, no. 2, pp. 227–237, 2000. View at Google Scholar
  5. G. Canaud, V. Audard, T. Kofman, P. Lang, C. Legendre, and P. Grimbert, “Recurrence from primary and secondary glomerulopathy after renal transplant,” Transplant International, vol. 25, no. 8, pp. 812–824, 2012. View at Publisher · View at Google Scholar
  6. A. M. Herzenberg, A. B. Fogo, H. N. Reich et al., “Validation of the Oxford classification of IgA nephropathy,” Kidney International, vol. 80, pp. 310–317, 2011. View at Google Scholar
  7. R. Coppo, S. Troyanov, R. Camilla et al., “The Oxford IgA nephropathy clinicopathological classification is valid for children as well as adults,” Kidney International, vol. 77, pp. 921–927, 2010. View at Publisher · View at Google Scholar
  8. D. C. Cattran, R. Coppo, H. T. Cook et al., “The Oxford classification of IgA nephropathy: rationale, clinicopathological correlations, and classification,” Kidney International, vol. 76, pp. 534–545, 2009. View at Publisher · View at Google Scholar
  9. M. E. Conley, M. D. Cooper, and A. F. Michael, “Selective deposition of immunoglobulin A1 in immunoglobulin A nephropathy, anaphylactoid purpura nephritis, and systemic lupus erythematosus,” Journal of Clinical Investigation, vol. 66, pp. 1432–1436, 1980. View at Google Scholar
  10. J. Mestecky, I. Moro, and M. Kerr, “Mucosal immunoglobulins,” in Mucosal Immunology, J. Mestecky, M. Lamm, J. McGhee, J. Bienenstock, L. Mayer, and W. Strober, Eds., pp. 153–181, Elsevier, 2009. View at Google Scholar
  11. T. S. Mattu, R. J. Pleass, A. C. Willis et al., “The glycosylation and structure of human serum IgA1, Fab, and Fc regions and the role of N-glycosylation on Fcα receptor interactions,” Journal of Biological Chemistry, vol. 273, pp. 2260–2272, 1998. View at Google Scholar
  12. M. C. Field, R. A. Dwek, C. J. Edge, and T. W. Rademacher, “O-linked oligosaccharides from human serum immunoglobulin A1,” Biochemical Society Transactions, vol. 17, pp. 1034–1035, 1989. View at Google Scholar
  13. M. Tomana, W. Niedermeier, J. Mestecky, and F. Skvaril, “The differences in carbohydrate composition between the subclasses of IgA immunoglobulins,” Immunochemistry, vol. 13, pp. 325–328, 1976. View at Publisher · View at Google Scholar
  14. J. Mestecky, M. Tomana, P. A. Crowley-Nowick, Z. Moldoveanu, B. A. Julian, and S. Jackson, “Defective galactosylation and clearance of IgA1 molecules as a possible etiopathogenic factor in IgA nephropathy,” Contributions to Nephrology, vol. 104, pp. 172–182, 1993. View at Google Scholar
  15. J. M. Woof and J. Mestecky, “Mucosal immunoglobulins,” in Mucosal Immunology, J. Mestecky, W. Strober, M. W. Russell, B. Kelsall, H. Cheroutre, and B. Lambrecht, Eds., 4th edition, 2014, In press. View at Google Scholar
  16. M. E. Koshland, “The coming of age of the immunoglobulin J chain,” Annual Review of Immunology, vol. 3, pp. 425–453, 1985. View at Google Scholar
  17. H. Suzuki, M. Raska, K. Yamada et al., “Cytokine-mediated dysregulation of specific glycosyltransferases contributes to aberrant glycosylation of IgA1 in patients with IgA nephropathy,” Journal of the American Society of Nephrology, vol. 23, p. 854A, 2012. View at Google Scholar
  18. P. M. Andre, P. Le Pogamp, and D. Chevet, “Impairment of jacalin binding to serum IgA in IgA nephropathy,” Journal of Clinical Laboratory Analysis, vol. 4, pp. 115–119, 1990. View at Google Scholar
  19. A. C. Allen, E. M. Bailey, J. Barratt, K. S. Buck, and J. Feehally, “Analysis of IgA1 O-glycans in IgA nephropathy by fluorophore-assisted carbohydrate electrophoresis,” Journal of the American Society of Nephrology, vol. 10, pp. 1763–1771, 1999. View at Google Scholar
  20. M. Tomana, K. Matousovic, B. A. Julian, J. Radl, K. Konecny, and J. Mestecky, “Galactose-deficient IgA1 in sera of IgA nephropathy patients is present in complexes with IgG,” Kidney International, vol. 52, pp. 509–516, 1997. View at Google Scholar
  21. Y. Hiki, H. Odani, M. Takahashi et al., “Mass spectrometry proves under-O-glycosylation of glomerular IgA1 in IgA nephropathy,” Kidney International, vol. 59, pp. 1077–1085, 2001. View at Publisher · View at Google Scholar
  22. A. C. Allen, E. M. Bailey, P. E. Brenchley, K. S. Buck, J. Barratt, and J. Feehally, “Mesangial IgA1 in IgA nephropathy exhibits aberrant O-glycosylation: observations in three patients,” Kidney International, vol. 60, pp. 969–973, 2001. View at Google Scholar
  23. J. Feehally and A. C. Allen, “Structural features of IgA molecules which contribute to IgA nephropathy,” Journal of Nephrology, vol. 12, pp. 59–65, 1999. View at Google Scholar
  24. A. G. Gharavi, Z. Moldoveanu, R. J. Wyatt et al., “Aberrant IgA1 glycosylation is inherited in familial and sporadic IgA nephropathy,” Journal of the American Society of Nephrology, vol. 19, pp. 1008–1014, 2008. View at Google Scholar
  25. J. Novak, W. J. Cook, B. A. Julian, J. Mestecky, and M. Tomana, “IgA nephropathy (IgAN): a similarity in the mechanism of IgA1 deposition in the kidney in humans and mice,” Journal of the American Society of Nephrology, vol. 11, pp. 478–479A, 2000. View at Google Scholar
  26. M. Tomana, R. Kulhavy, and J. Mestecky, “Receptor-mediated binding and uptake of immunoglobulin A by human liver,” Gastroenterology, vol. 94, pp. 762–770, 1988. View at Google Scholar
  27. M. Tomana, J. O. Phillips, R. Kulhavy, and J. Mestecky, “Carbohydrate-mediated clearance of secretory IgA from the circulation,” Molecular Immunology, vol. 22, pp. 887–892, 1985. View at Google Scholar
  28. J. Novak, B. A. Julian, M. Tomana, and J. Mestecky, “Progress in molecular and genetic studies of IgA nephropathy,” Journal of Clinical Immunology, vol. 21, no. 5, pp. 310–327, 2001. View at Publisher · View at Google Scholar
  29. P. J. van der Boog, C. van Kooten, J. W. de Fijter, and M. R. Daha, “Role of macromolecular IgA in IgA nephropathy,” Kidney International, vol. 67, pp. 813–821, 2005. View at Publisher · View at Google Scholar
  30. J. K. Boyd and J. Barratt, “Immune complex formation in IgA nephropathy: CD89 a “saint” or a “sinner”?” Kidney International, vol. 78, pp. 1211–1213, 2010. View at Publisher · View at Google Scholar
  31. T. Kokubo, Y. Hiki, H. Iwase et al., “Protective role of IgA1 glycans against IgA1 self-aggregation and adhesion to extracellular matrix proteins,” Journal of the American Society of Nephrology, vol. 9, pp. 2048–2054, 1998. View at Google Scholar
  32. H. Suzuki, R. Fan, Z. Zhang et al., “Aberrantly glycosylated IgA1 in IgA nephropathy patients is recognized by IgG antibodies with restricted heterogeneity,” Journal of Clinical Investigation, vol. 119, pp. 1668–1677, 2009. View at Google Scholar
  33. M. Tomana, J. Novak, B. A. Julian, K. Matousovic, K. Konecny, and J. Mestecky, “Circulating immune complexes in IgA nephropathy consist of IgA1 with galactose-deficient hinge region and antiglycan antibodies,” Journal of Clinical Investigation, vol. 104, pp. 73–81, 1999. View at Google Scholar
  34. J. Novak, Z. Moldoveanu, M. B. Renfrow et al., “IgA nephropathy and Henoch-Schoenlein purpura nephritis: aberrant glycosylation of IgA1, formation of IgA1-containing immune complexes, and activation of mesangial cells,” Contributions to Nephrology, vol. 157, pp. 134–138, 2007. View at Google Scholar
  35. J. Feehally, M. Farrall, A. Boland et al., “HLA has strongest association with IgA nephropathy in genome-wide analysis,” Journal of the American Society of Nephrology, vol. 21, pp. 1791–1797, 2010. View at Google Scholar
  36. A. G. Gharavi, K. Kiryluk, M. Choi et al., “Genome-wide association study identifies susceptibility loci for IgA nephropathy,” Nature Genetics, vol. 43, pp. 321–327, 2011. View at Google Scholar
  37. X. Q. Yu, M. Li, H. Zhang et al., “A genome-wide association study in Han Chinese identifies multiple susceptibility loci for IgA nephropathy,” Nature Genetics, vol. 44, no. 2, pp. 178–182, 2012. View at Publisher · View at Google Scholar
  38. M. Fennessy, G. A. Hitman, R. H. Moore et al., “HLA-DQ gene polymorphism in primary IgA nephropathy in three European populations,” Kidney International, vol. 49, pp. 477–480, 1996. View at Publisher · View at Google Scholar
  39. K. Kiryluk, Y. Li, S. Sanna-Cherchi et al., “Geographic differences in genetic susceptibility to IgA nephropathy: GWAS replication study and geospatial risk analysis,” PLOS Genetics, vol. 8, Article ID e1002765, 2012. View at Google Scholar
  40. K. Kiryluk, J. Novak, and A. G. Gharavi, “Pathogenesis of immunoglobulin A nephropathy: recent insight from genetic studies,” Annual Review of Medicine, vol. 64, pp. 339–356, 2013. View at Google Scholar
  41. P. C. Heinrich, I. Behrmann, S. Haan, H. M. Hermanns, G. Muller-Newen, and F. Schaper, “Principles of interleukin (IL)-6-type cytokine signalling and its regulation,” Biochemical Journal, vol. 374, pp. 1–20, 2003. View at Google Scholar
  42. G. Rostoker, J. C. Rymer, G. Bagnard, M. Petit-Phar, M. Griuncelli, and Y. Pilatte, “Imbalances in serum proinflammatory cytokines and their soluble receptors: a putative role in the progression of idiopathic IgA nephropathy (IgAN) and Henoch-Schőnlein purpura nephritis, and a potential target of immunoglobulin therapy?” Clinical & Experimental Immunology, vol. 114, pp. 468–476, 1998. View at Google Scholar
  43. J. V. Stein, M. Lopez-Fraga, F. A. Elustondo et al., “APRIL modulates B and T cell immunity,” Journal of Clinical Investigation, vol. 109, pp. 1587–1598, 2002. View at Google Scholar
  44. K. Harada, Y. Akai, N. Kurumatani, M. Iwano, and Y. Saito, “Prognostic value of urinary interleukin 6 in patients with IgA nephropathy: an 8-year follow-up study,” Nephron, vol. 92, no. 4, pp. 824–826, 2002. View at Publisher · View at Google Scholar
  45. J. Novak, M. Raska, M. Tomana et al., “Transcriptional regulation by cytokines of O-glycan-specific glycosyltransferases in human IgA1-producing cells: implications for IgA nephropathy (IgAN),” Journal of the American Society of Nephrology A, vol. 14, p. 632A, 2003. View at Google Scholar
  46. H. Suzuki, M. Raska, K. Yamada et al., “Cytokines alter IgA1 O-glycosylation by dysregulating C1GalT1 and ST6GalNAc-II enzymes,” Journal of Biological Chemistry, vol. 289, pp. 5330–5339, 2014. View at Google Scholar
  47. T. Knoop, B. E. Vikse, E. Svarstad, S. Leh, A. V. Reisæter, and R. Bjørneklett, “Mortality in patients with IgA nephropathy,” American Journal of Kidney Diseases, vol. 62, no. 5, pp. 883–890, 2013. View at Publisher · View at Google Scholar
  48. J. Lv, D. Xu, V. Perkovic et al., “Corticosteroid therapy in IgA nephropathy,” Journal of the American Society of Nephrology, vol. 23, pp. 1108–1116, 2012. View at Google Scholar
  49. Y. Hiki, “O-linked oligosaccharides of the IgA1 hinge region: roles of its aberrant structure in the occurrence and/or progression of IgA nephropathy,” Clinical and Experimental Nephrology, vol. 13, pp. 415–423, 2009. View at Publisher · View at Google Scholar
  50. F. Berthoux, H. Suzuki, L. Thibaudin et al., “Autoantibodies targeting galactose-deficient IgA1 associate with progression of IgA nephropathy,” Journal of the American Society of Nephrology, vol. 23, pp. 1579–1587, 2012. View at Google Scholar
  51. N. Zhao, P. Hou, J. Lv et al., “The level of galactose-deficient IgA1 in the sera of patients with IgA nephropathy is associated with disease progression,” Kidney International, vol. 82, no. 7, pp. 790–796, 2012. View at Publisher · View at Google Scholar
  52. H. Suzuki, Z. Moldoveanu, S. Hall et al., “IgA1-secreting cell lines from patients with IgA nephropathy produce aberrantly glycosylated IgA1,” Journal of Clinical Investigation, vol. 118, pp. 629–639, 2008. View at Google Scholar
  53. E. Losev, C. A. Reinke, J. Jellen, D. E. Strongin, B. J. Bevis, and B. S. Glick, “Golgi maturation visualized in living yeast,” Nature, vol. 441, pp. 1002–1006, 2006. View at Google Scholar
  54. H. Schachter, “Coordination between enzyme specificity and intracellular compartmentation in the control of protein-bound oligosaccharide biosynthesis,” Biology of the Cell, vol. 51, pp. 133–145, 1984. View at Publisher · View at Google Scholar
  55. B. C. O'Connell and L. A. Tabak, “A comparison of serine and threonine O-glycosylation by UDP-GalNAc:polypeptide N-acetylgalactosaminyltransferase,” Journal of Dental Research, vol. 72, pp. 1554–1558, 1993. View at Google Scholar
  56. M. Raska, K. Yamada, M. Horynova et al., “Role of GalNAc-transferases in the synthesis of aberrant IgA1 O-glycans in IgA nephropathy,” Journal of the American Society of Nephrology, vol. 22, p. 625A, 2011. View at Google Scholar
  57. D. H. van den Eijunden, P. Stoffyn, A. Stoffyn, and W. E. Schiphorst, “Specificity of sialytransferase: structure of α1-acid glycoprotein sialylated in vitro. A new methodology for the determination of the structure of the linkage formed by glycosyltransferase action on galactosyl-oligosaccharide-protein acceptors,” European Journal of Biochemistry, vol. 81, pp. 1–7, 1977. View at Google Scholar
  58. T. Kudo, T. Iwai, T. Kubota et al., “Molecular cloning and characterization of a novel UDP-Gal:GalNAc(α) peptide β 1, 3-galactosyltransferase (C1Gal-T2, an enzyme synthesizing a core 1 structure of O-glycan,” Journal of Biological Chemistry, vol. 277, pp. 47724–47731, 2002. View at Google Scholar
  59. R. P. Aryal, T. Ju, and R. D. Cummings, “Tight complex formation between Cosmc chaperone and its specific client non-native T-synthase leads to enzyme activity and client-driven dissociation,” Journal of Biological Chemistry, vol. 287, pp. 15317–15329, 2012. View at Google Scholar
  60. J. E. Sadler, J. I. Rearick, and R. L. Hill, “Purification to homogeneity and enzymatic characterization of an α-N-acetylgalactosaminide α2→6 sialyltransferase from porcine submaxillary glands,” Journal of Biological Chemistry, vol. 254, pp. 5934–5941, 1979. View at Google Scholar
  61. T. Ju, V. I. Otto, and R. D. Cummings, “The Tn antigen-structural simplicity and biological complexity,” Angewandte Chemie International Edition in English, vol. 50, no. 8, pp. 1770–1791, 2011. View at Publisher · View at Google Scholar
  62. J. Novak, Z. Moldoveanu, B. A. Julian et al., “Aberrant glycosylation of IgA1 and anti-glycan antibodies in IgA nephropathy: role of mucosal immune system,” Advances in Otolaryngology, vol. 72, pp. 60–63, 2011. View at Google Scholar
  63. K. Yamaguchi, Y. Ozono, T. Harada, and K. Hara, “Changes in circulating immune complex and charge distribution with upper respiratory tract inflammation in IgA nephropathy,” Nephron, vol. 69, no. 4, pp. 384–390, 1995. View at Google Scholar
  64. J. Novak, M. B. Renfrow, A. G. Gharavi, and B. A. Julian, “Pathogenesis of immunoglobulin A nephropathy,” Current Opinion in Nephrology and Hypertension, vol. 22, pp. 287–294, 2013. View at Publisher · View at Google Scholar
  65. R. Coppo, A. Amore, B. Gianoglio et al., “Macromolecular IgA and abnormal IgA reactivity in sera from children with IgA nephropathy. Italian Collaborative Paediatric IgA Nephropathy Study,” Clinical Nephrology, vol. 43, no. 1, pp. 1–13, 1995. View at Google Scholar
  66. J. Novak, B. A. Julian, M. Tomana, and J. Mestecky, “IgA glycosylation and IgA immune complexes in the pathogenesis of IgA nephropathy,” Seminars in Nephrology, vol. 28, pp. 78–87, 2008. View at Publisher · View at Google Scholar
  67. D. D. McCarthy, J. Kujawa, C. Wilson et al., “Mice overexpressing BAFF develop a commensal flora-dependent, IgA-associated nephropathy,” Journal of Clinical Investigation, vol. 121, pp. 3991–4002, 2011. View at Google Scholar
  68. J. S. Silver and C. A. Hunter, “gp130 at the nexus of inflammation, autoimmunity, and cancer,” Journal of Leukocyte Biology, vol. 88, no. 6, pp. 1145–1156, 2010. View at Publisher · View at Google Scholar
  69. A. Davidson, “The rationale for BAFF inhibition in systemic lupus erythematosus,” Current Rheumatology Reports, vol. 14, pp. 295–302, 2012. View at Google Scholar
  70. A. Cerutti, I. Puga, and M. Cols, “Innate control of B cell responses,” Trends in Immunology, vol. 32, no. 5, pp. 202–211, 2011. View at Publisher · View at Google Scholar
  71. G. Xin, W. Shi, L. X. Xu, Y. Su, L. J. Yan, and K. S. Li, “Serum BAFF is elevated in patients with IgA nephropathy and associated with clinical and histopathological features,” Journal of Nephrology, vol. 26, pp. 683–690, 2013. View at Publisher · View at Google Scholar
  72. C. Reily, K. Yamada, Z. Q. Huang et al., “Abnormal STAT3 signaling enhances production of autoantigen in an autoimmune disease, IgA nephropathy,” Journal of the American Society of Nephrology, vol. 24, p. 20A, 2013. View at Google Scholar
  73. H. Suzuki, Z. Moldoveanu, S. Hall et al., “Cytokines regulate aberrant glycosylation of IgA1 in cell lines from patients with IgA nephropathy,” Journal of the American Society of Nephrology, vol. 18, p. 188A, 2007. View at Google Scholar
  74. K. Yamada, Z. Q. Huang, M. Raska et al., “Effects of leukemia inhibitory factor and oncostatin M on IgA1-producing cells from patients with IgA nephropathy,” Journal of the American Society of Nephrology, vol. 23, p. 912A, 2012. View at Google Scholar
  75. J. Chia, G. Goh, V. Racine, S. Ng, P. Kumar, and F. Bard, “RNAi screening reveals a large signaling network controlling the Golgi apparatus in human cells,” Molecular Systems Biology, vol. 8, article 629, 2012. View at Publisher · View at Google Scholar
  76. Y. H. Zhou, L. G. Tang, S. L. Guo et al., “Steroids in the treatment of IgA nephropathy to the improvement of renal survival: a systematic review and meta-analysis,” PLoS One, vol. 6, article e8, 2011. View at Google Scholar
  77. H. Suzuki, K. Kiryluk, J. Novak et al., “The pathophysiology of IgA nephropathy,” Journal of the American Society of Nephrology, vol. 22, no. 10, pp. 1795–1803, 2011. View at Google Scholar