1. Introduction
Postinfectious glomerulonephritis (PIGN) is one of the representative diseases presenting with acute glomerulonephritis (AGN)
[1,2,3][1][2][3]. The typical presentation of PIGN includes hematuria, mild to moderate proteinuria, edema, and hypertension with a latency period following infection. Rapidly progressive GN also occurs rarely, and some patients progress to chronic kidney disease (CKD)
[3]. During the past century, most cases of PIGN occurred in childhood and followed streptococcal infections, leading to it being called post-streptococcal acute glomerulonephritis (PSAGN)
[2,3][2][3]. However, in the past three decades, the incidence of PSAGN has decreased, particularly in advanced countries, probably due to the improvement in the living environment and the appropriate administration of antibiotics
[1,2,3][1][2][3]. Currently, the estimated annual incidence of PSAGN is 90–280 per 1,000,000 population
[4,5][4][5]. On the other hand, in advanced countries in the modern era, the number of adult patients with AGN has been increasing. A significant proportion of adult AGN cases is associated with non-streptococcal infection, especially
Staphylococcus infection
[1,3][1][3]. At present, adult-onset AGN caused by
Staphylococcus infection is as common as PSAGN, and such cases of AGN occur in elderly or immunocompromised patients, in whom it is a formidable issue because of their poor renal prognosis
[1,2,3][1][2][3]. Moreover, non-streptococcal infections in adult AGN have heterogeneous infection sites and tend to be ongoing at the time of definitive diagnosis by renal biopsy (RB). Therefore, the term “infection-related glomerulonephritis (IRGN)” has been proposed
[6] as the comprehensive concept including PSAGN and atypical types of adult-onset of AGN related to various infections (e.g., staphylococci, Gram-negative bacteria, various viruses, fungi, or protozoa)
[7].
2. Complement Activation in IRGN and C3G
2.1. Complement Cascade
The complement system consists of three pathways: the CP, the lectin pathway (LP), and the AP. The complement cascade is briefly summarized as follows: the amplification phase follows the initiation phase and segues into the terminal phase. Generally, the CP and LP need to be activated via recognition of pathway-specific triggers, such as pathogen-associated molecular patterns, antigen-antibody complexes, or microbial polysaccharides. Meanwhile, the AP is constitutively active through the spontaneous hydrolysis of a thioester bond on C3 to produce C3(H
2O). This process is termed tick-over and occurs at a rate of around 1% of total C3 per hour
[42,43][8][9]. Subsequently, activated C3 in the AP leads to the formation of C3 convertase, C3bBb, by combining with FB through cleavage by factor D. Once C3 is activated, a feedback loop amplifies the initial complement response and activation of the terminal pathway by the generation of C5 convertase (C3bBbC3b). This C5 convertase cleaves C5 into C5a and C5b, starting the generation of C5b-9 or membrane attack complex, which induces cell lysis
[13,16,44][10][11][12]. In addition, C5a, like C3a, promotes inflammation as a potent anaphylatoxin
[16,44][11][12].
The above-mentioned complement activation system plays a pivotal role in the progression of both IRGN and C3G. In IRGN, activation of CP, LP, and AP might be involved, which depends on the clinical stage or pathophysiological state
[19][13]. In C3G, acquired or genetic alterations of the regulatory proteins in the AP are essential for pathogenesis
[11,12,35][14][15][16]. In both diseases, hypocomplementemia, mainly a reduction in the serum C3 level, is frequently seen, and the level of serum C3 usually returns to the normal range within a month or so in IRGN, especially in cases of PSAGN. However, some patients with IRGN also demonstrated a persistent reduction in the serum C3 level, which is proposed to be due to the effect of AP dysregulation
[3,19][3][13].
2.2. Complement Profile and Activation in IRGN, Especially in PIGN
According to a recent review
[30][17], activation of the CP in PSAGN is suppressed by chemokine-binding evasins secreted by
Streptococcus and by proteins on the streptococcal surface that bind a C4b-binding protein
[45,46,47][18][19][20]. Evasins are a family of salivary proteins produced in parasitic ticks that are capable of turning off the first steps of an immune response brought about by chemokines
[47][20]. In fact, normal C1q levels are almost universal in PIGN, and there has been research that indicated the suppressive state of CP in PIGN, but glomerular C1q deposition in PIGN is sometimes detected in RB specimens. Thus, it might be important to keep in mind the partial effect of CP on the development of PIGN. Regarding the LP in PIGN, the association of activated LP with the occurrence of PSAGN has been reported
[30][17]. Polymorphisms of the genes in the LP were demonstrated to aggravate the immune reaction
[48][21]. Indeed, a low serum C4 level in patients with PIGN is occasionally seen, which is thought to be the result of activation of the LP. Taken together, activation of the LP, as well as the CP, is not negligible in PIGN. Nevertheless, there is no doubt that the AP is the most important pathway in the pathogenesis of PIGN. Atypical cases of PIGN in which dysregulation of the AP-induced CKD with persistent proteinuria and hematuria were highlighted in observational studies
[2,20,31,49][2][22][23][24].
As a cause of the dysregulation of the AP in PIGN, the focus has been on the effect of autoimmune mechanisms for decades. The aspect of abnormal complement gene variants in PIGN was not emphasized, although several previous studies performed genetic analyses
[50,51][25][26]. So far, several characteristic autoantibodies have been demonstrated in patients with PIGN. Anti-C1q antibodies were anticipated as one of the key effectors of the activation of the AP in PIGN. Kozyro et al. reported that 8 of 24 children with PIGN had anti-C1q antibodies
[52][27]. C3NeF, known as the autoantibody targeting C3Bb, was also considered the driver for AP dysregulation in PIGN. A recent review article suggested that transient C3NeF generation was the cause of AP activation in PIGN
[30][17]. In addition, Sethi et al. assessed 11 patients who were diagnosed as having PIGN with persistent activation of the AP, and 7 of them were found to have positive activity for C3NeF
[49][24]. However, the presence of the above-mentioned autoantibodies has not provided the crucial breakthrough for understanding the mechanism of the activated AP in PIGN. Recently, Chauvet et al. showed the presence of anti-FB antibodies (mainly IgG1 subclass) in pediatric patients with PIGN as a useful diagnostic marker
[29][28]. In the acute phase of GN, anti-FB autoantibodies were identified in 31 of 34 (91%) children with PIGN and in four of 28 (14%) children with hypocomplementemic C3G. The sensitivity and specificity of the detection of anti-FB antibodies for diagnosing PIGN were 95% and 82%, respectively. Furthermore, the anti-FB autoantibodies were transient and correlated inversely with plasma levels of C3 and correlated directly with levels of C5b-9. Furthermore, the anti-FB autoantibodies amplified the activity of the C3 convertase in AP. The anti-FB autoantibodies did not stabilize the C3 convertase formed on red blood cells, indicating that anti-FB autoantibody is different from C3NeF in pediatric patients with C3G. Although the mechanism of the emergence of anti-FB autoantibodies remains to be elucidated, a genetic predisposition is posited. Collectively, identifying anti-FB autoantibodies provided critical insight into the pathophysiologic mechanism of PIGN. Screening for anti-FB antibodies might be helpful to avoid misdiagnosing IRGN as C3G.
2.3. Complement Profile and Activation in C3G
The involvement of genetic drivers is inevitable when discussing the AP activation in C3G. It was reported that approximately 25% of C3G patients carry variants of complement-related genes, including
C3,
CFB,
CFH,
CFI, and
CFHR5, which encode C3, FB, FH, Factor I, and complement FHR protein 5, respectively
[11,12,35][14][15][16]. According to a recent review, the most commonly identified genetic variant in C3G is rearrangement at the
CFH locus, creating CFHR fusion genes
[12][15]. These genetic variants drive the continuous activation of AP, resulting in the development of C3G. Thus, etiopathogenetic diagnosis based on genetic analysis is desirable for C3G patients, although performing such analysis might be a high hurdle in resource-constrained settings.
Acquired drivers, that is, autoantibodies to the complement system, could also affect the activation of the AP in C3G. The majority of patients with C3G were positive for some autoantibodies against complement convertase or specific complement proteins that impair normal convertase or protein function
[12,35][15][16]. C3NeF against C3bBb is the most common, reported in 50–80% of DDD patients and 44–50% of C3GN patients
[27,28][29][30]. C3NeF stabilizes and prolongs the half-life of this convertase in the amplification phase by protecting C3bBb from FH-mediated decay
[55][31]. Similarly, C5 nephritic factor (C5NeF) is also common. This autoantibody binds to and stabilizes C3bBbC3b, increasing the half-life of the C5 convertase and generation of sC5b–9
[35][16]. In the previous analysis, C5NeF, rather than DDD, was deeply involved in the pathogenesis of C3GN
[54][32]. C4 nephritic factor (C4NeF), which binds to and stabilizes C4b2a, was also identified in patients with C3G
[53][33]. Furthermore, not only these nephritic factors, but also autoantibodies against FH and FB, were occasionally detected
[29,54][28][32]. In the previous reports, anti-FH autoantibodies were detected in approximately 10–20% of C3G patients
[29,30,56][17][28][34]. It was demonstrated that these autoantibodies ultimately stabilize C3 convertase by impairing FH-mediated decay
[56,57][34][35]. Therefore, physicians need to make an effort to evaluate autoantibodies to the complement system when diagnosing C3G. Further studies are required to establish a reliable, reproducible method for detecting such autoantibodies in C3G.
3. Glomerular NAPlr Deposition and Plasmin Activity in IRGN and C3G
3.1. NAPlr as a Key Diagnostic Marker for IRGN
NAPlr
[19,20,21,22,23,24][13][22][36][37][38][39] and streptococcal pyrogenic exotoxin B (SPEB)
[58][40] have been identified as proteins causing IRGN. NAPlr is a nephritogenic protein isolated from Group A streptococcus that is homologous to streptococcal glyceraldehyde-3-phosphate dehydrogenase (GAPDH). Although GAPDH is known as a housekeeping gene, bacterial GAPDH has pleiotropic functions, such as energy production, regulation of gene expression, and plasmin-binding capacity
[59,60,61][41][42][43]. Plasmin bound with NAPlr is suspected to be protected from its physiological inhibitors in vivo and keep its enzymatic activity. Accordingly, NAPlr deposited in glomeruli binds with plasmin and induces plasmin-dependent glomerular damage in IRGN
[19,20,21][13][22][36]. The proteolytic activity of plasmin in glomeruli would induce glomerular injury directly by degrading extracellular matrix proteins and indirectly by activating pro–matrix metalloproteases
[20,23][22][38]. Moreover, the plasmin is recognized to have a proinflammatory function by activating and accumulating inflammatory cells
[20,23][22][38].
With regard to SPEB, it was also isolated from Group A streptococci and was regarded as another nephritogenic protein in PSAGN
[58][40]. Since SPEB was reported to localize within the subepithelial hump by immunoelectron microscopy, according to the paper of Batsford et al.
[62][44], it was suspected to mainly induce glomerular damage as the component of IC. However, SPEB was also found to have plasmin-binding capacity and might induce glomerular injury via plasmin activity
[2].
Methodologically, glomerular NAPlr deposition is detected by immunostaining, and plasmin activity can be evaluated by zymography using a plasmin-sensitive synthetic substrate
[23][38]. The distributions of NAPlr deposition and plasmin activity in glomeruli, which approximately merge into each other, are seen in the early phase of PSAGN. A previous report demonstrated that RB specimens of PSAGN showed glomerular deposition of NAPlr and plasmin activity in all patients within 2 weeks of disease onset
[63][45]. Furthermore, such glomerular staining or expression patterns have been demonstrated in patients with IRGN caused by infections other than streptococci, such as
Streptococcus pneumoniae [64][46],
Aggregatibacter actinomycetemcomitans [65][47],
Mycoplasma pneumoniae [66][48], and
Staphylococcus aureus [19,21][13][36]. In fact, researchers also saw a patient with IRGN after the onset of a streptococcal urinary tract infection and sexually transmitted infections by
Neisseria gonorrhoeae and
Chlamydia (manuscript in preparation), in which RB specimens clearly demonstrated glomerular NAPlr deposition and plasmin activity in a similar fashion. As mentioned earlier, NAPlr is the same substance as the GAPDH of
Streptococcus, and GAPDH is universally expressed and may have high homology among various bacteria. Consequently, anti-NAPlr antibodies are likely to cross-react with the GAPDH of bacteria other than
Streptococcus pyogenes. Therefore, positive glomerular staining for NAPlr and plasmin activity could provide critical histological evidence for the substantial involvement of bacterial infection in the development of GN.
3.2. Activation of the AP by NAPlr
The mechanism of overactivation of the AP in IRGN remains to be completely elucidated. Persistent AP activation with a reduction in the serum C3 level, which is apparent in IRGN, is considered to be a factor in the progression to CKD in patients with IRGN. As a possible cause of its activation, the involvement of the aforementioned autoantibodies, such as anti-FB autoantibody
[29][28], has been reported. Furthermore, according to the latest reviews and a previous report
[19[13][17][22][36][49],
20,21,30,67], there is a subpopulation of patients with IRGN having a mild genetic or acquired deficiency in AP regulatory proteins, which is regarded as a trigger for uncontrolled activation of the AP. Yoshizawa et al. demonstrated that NAPlr converted C3 to C3b and induced the formation of iC3b in a dose-dependent manner in human serum samples in vitro
[19,24][13][39]. Because the glomerular distributions of NAPlr deposition and C3 deposition are essentially different, the complement AP activation by NAPlr might occur mainly in the circulation (liquid phase) rather than in situ in glomeruli. Direct activation of the AP by NAPlr in the liquid phase may release complement-related anaphylatoxins, including C3a and C5a, resulting in glomerular accumulation of macrophages and neutrophils. However, the evidence for the direct activation of AP by NAPlr in vitro is limited. Further basic research will be indispensable, because this approach might get to the core of the pathogenesis of C3-dominant IRGN.
3.3. C3G with Glomerular NAPlr Deposition
Fundamentally, positive key diagnostic markers for IRGN are glomerular staining for NAPlr and for plasmin activity. However, positive glomerular staining for these markers was also detected in various glomerular diseases other than IRGN
[22][37]. These cases include C3G
[68[50][51][52][53],
69,70,71], MPGN type I
[72][54], antineutrophil cytoplasmic antibody-associated vasculitis
[73][55], and IgA vasculitis
[74,75][56][57]. No significant difference in staining patterns of NAPlr and plasmin activity among those glomerular diseases and IRGN was detected, which made it hard to differentiate these diseases from IRGN. Moreover, it might be difficult to interpret cases with C3G exhibiting positive glomerular staining for NAPlr and for plasmin activity, despite no apparent history of infection, as recently reported
[70][52].
4. Causal Relationship between IRGN and Development of C3G
The overlapping clinicopathological features of C3G and IRGN, mainly the PIGN pattern, have been an intriguing matter for discussion. In a recent review, Yoshizawa et al. mentioned that pathogenic mechanisms responsible for the development of C3G are essentially the same as the mechanism for C3-dominant PSAGN
[19][13]. As described, several characteristic factors might be useful for the differentiation of these diseases, but there have been several reports that let people presume that C3G and C3-dominant IRGN could just be subtypes of one category of glomerular disease.
Meanwhile, the causal relationship between C3G and IRGN has been considered as follows. First, preceding infection is suggested. As previously reported
[76][58], a preceding streptococcal infection initially induced PSAGN, but the patient later developed chronic C3G. According to the analysis of Al-Ghaithi et al., 25 (75.8%) of 33 children with PIGN who underwent RB because of an unusual clinical course were ultimately diagnosed as having IRGN, but the remaining 8 (24.2%) were eventually diagnosed as having C3G
[77][59]. Second, persistent causative infection could modify the condition of the host’s complement activation state. Oda et al. mentioned the possibility of alteration of host AP activation due to persistent infection
[20][22]. RB findings in case 3 showed C3G with positive glomerular staining for NAPlr and for plasmin activity, despite the absence of apparent infection
[70][52], which may be due to the modulation of the AP by subclinical infection. Thus, thoroughly assessing the hidden infection by not only bacterial cultures, but also histological staining for NAPlr and for plasmin activity, is critical in cases in which IF findings indicate C3-dominant glomerular deposition. Third, researchers propose the exacerbation of latent C3G with mild AP dysregulation after the onset of IRGN. Generally, C3G patients show severe clinical presentations such as NRP, but cases of C3G with milder renal clinical manifestations are not rare
[15,78][60][61]. Environmental triggers of C3G have not been sufficiently documented, but several previous reports suggested that infection might be a trigger for the development of C3G
[68,69,76][50][51][58].