Atypical post-infectious glomerulonephritis is associated with abnormalities in the alternative pathway of complement

INTRODUCTION

Post-infectious glomerulonephritis is a type of glomerulonephritis seen in both children and adults following an infection. The glomerulonephritis is manifested by development of a nephritic syndrome within 1–3 weeks of an infectious episode.^1–3 The infection is often mild and typically resolves by the time the glomerulonephritis is diagnosed. The pathogenesis of post-infectious glomerulonephritis is thought to be due to (1) glomerular in-situ deposition of bacterial antigens with subsequent antibody build-up and formation of in-situ immune complexes and/or (2) glomerular deposition of circulating immune-complexes.^2–6

The characteristic features of post-infectious glomerulonephritis on kidney biopsy are a proliferative glomerulonephritis on light microscopy (LM), bright C3 staining with or without immunoglobulins on immunofluorescence (IF) microscopy, and subepithelial deposits called ‘humps’ on electron microscopy (EM).^{3, 7, 8} It should be pointed out that in some cases the diagnosis of post-infectious glomerulonephritis is made based on these biopsy findings even in the absence of any clinical, bacterial or serological evidence of a preceding infection.

The majority of patients achieve complete remission of the nephritic syndrome following post-infectious glomerulonephritis. However, in a small percentage the disease takes longer to resolve or persists with renal dysfunction evidenced by continuing hematuria and proteinuria. A smaller percentage still progresses to end-stage kidney disease. ^{1, 4, 7–10} The terms ‘persistent’, ‘resolving’ or ‘chronic’ post-infectious glomerulonephritis have been used to describe these entities.

During an infection that triggers development of post-infectious glomerulonephritis, there is activation of the AP of complement.^{4, 11, 12} Following elimination of the infection, immune-complexes are cleared, AP activity is controlled, and glomerulonephritis resolves.^{11, 13} We hypothesized that in patients who take longer to resolve, develop persistent proteinuria and hematuria, or progress to end stage kidney disease there is a defect in the regulating mechanisms of the AP of complement that prevents its down-regulation following resolution of the infection. The consequence is excessive deposition of complement proteins and breakdown products in the glomeruli seen as the pathognomonic bright C3 staining on IF microscopy on kidney biopsy. An inflammatory response ensues with the development of persistent proliferative glomerulonephritis. We define such cases as ‘atypical’ post-infectious glomerulonephritis, to highlight the presence of an abnormality in the AP of complement. To test this hypothesis, we evaluated the AP of complement in 11 patients with persistent hematuria and proteinuria diagnosed with post-infectious glomerulonephritis on kidney biopsy to determine whether an underlying abnormality of AP of complement could be identified. The study included 6 cases with a kidney biopsy diagnosis of post-infectious glomerulonephritis even though there was no evidence of a preceding infection.

RESULTS

DISCUSSION

Post-infectious glomerulonephritis is a relatively common glomerulonephritis that affects both children and adults.^{1, 8–10} Patients typically present with nephritic syndrome or acute renal failure, which rapidly resolve. In a minority of patients, however, hematuria and proteinuria persist or there is progression to end stage kidney disease.^{10, 14} The cause of persistent hematuria and proteinuria is not known. Kidney biopsy done in such cases shows features of post-infectious glomerulonephritis, i.e. proliferative glomerulonephritis on LM, bright glomerular C3 staining with or without immunoglobulins on IF, and subepithelial humps on EM. We also include cases of persistent hematuria and proteinuria that show features of post-infectious glomerulonephritis on kidney biopsy, but have no evidence of a preceding infection. We classify these patients as ‘atypical’ post-infectious glomerulonephritis and they form the basis of our study. Thus, patients with persistent hematuria and proteinuria, and kidney biopsy findings of a post-infectious glomerulonephritis, with or without a preceding infection, are included in the study. In this study, we show that such patients have an underlying abnormality of the AP of complement.

Most patients that are clinically diagnosed with post-infectious glomerulonephritis recover full renal function and are not biopsied, and hence as a group are difficult to study. However, persistent hematuria and proteinuria, in the setting of a prior clinical diagnosis of post-infectious glomerulonephritis, is an indication for kidney biopsy. We propose that when an infection occurs it triggers the AP, which under normal circumstances is quickly brought under control once the infection abates. However, in patients with a defect in AP regulation, there is continual AP activation with deposition of complement proteins and their breakdown products in the glomeruli, even after resolution of the infection, that leads to a development of ‘atypical’ proliferative glomerulonephritis. If the defect is mild, AP control occurs albeit at much slower rate than normal and in these patients, the glomerulonephritis eventually resolves. If the defect in AP regulation is more severe, hematuria and proteinuria persist, often exacerbated by recurrent bouts of infection (Fig 2). Thus, in ‘atypical’ post-infectious glomerulonephritis, an infection unmasks the underlying AP abnormality. In our study, we had documentation of infection in only five of 11 patients raising the possibility that sub-clinical infections may have occurred in the remaining patients. These are the cases that showed biopsy findings of post-infectious glomerulonephritis, yet had no clinical evidence of a preceding infection.

The role of AP activity in post-infectious glomerulonephritis is supported by other studies.¹⁵ As early as 1985, Meri and colleagues showed that sera from four patients with post-streptococcal glomerulonephritis contained factors other than immune-complexes that activated the complement system.¹⁶ Hisano et al., by immunohistochemical staining methods showed that glomeruli in post-infectious glomerulonephritis contained components of both the lectin and AP of complement.¹⁷ We have also completed laser microdissection and mass spectrometry of glomeruli of atypical post-infectious glomerulonephritis and verified the presence of AP complement proteins and their breakdown products in diseased glomeruli.¹⁸ Another indirect evidence of the role of AP is persistently low C3 levels noted in a small percentage of patients with post-infectious glomerulonephritis, even though C3 titers typically normalize within 8 weeks of resolution of the infection.^{19, 20}

In this study of ‘atypical’ post-infectious glomerulonephritis patients, we show that such patients have an underlying defect in the AP of complement. In 10 of the 11 patients, we demonstrated abnormalities in AP regulation. Five patients had C3Nefs detected by C3CSAP. C3Nefs are autoantibodies to C3 convertase that impair its control by protein regulators of complement activity, effectively prolonging the half-life of C3 convertase from a few seconds to 4–60 minutes.²¹ In two patients, IFE was positive. This assay does NOT detect antibodies but rather measures the presence of C3c in the serum as an indication of C3 convertase dysregulation. Four patients also had novel variants in complement-regulatory genes. CFH, which carried variants in three patients, is the most important fluid-phase regulator of the AP. It competes with factor B for binding to C3b to control formation of C3 convertase, promotes decay of formed C3 convertase, and serves as a cofactor for factor I to degrade C3b. We also detected a novel variant in CFHR5, a member of the factor H gene family, which is also a regulator of the C3 convertase. None of these variants is reported in NHLBI Exome Sequencing Project, a database of over 10,000 chromosomes, and current data suggest that rare variants such as those reported here make an important contribution to human phenotypic variation and disease susceptibility.²² Consistent with this possible functional effect, when we applied three commonly used algorithms (SIFT, PolyPhen2 and BLOSUM) to predict the consequence of the non-synonymous variants we identified, in each case at least two of these three programs classified these variants as damaging (predicted to have an effect on function).^22–24 Our results are also supported by a case report showing persistent glomerulonephritis following streptococcal infection in a patient with CFHR5 deficiency.²⁵ Nevertheless, functional testing of the isolated mutant proteins was not performed and therefore the pathological significance of these variants remains speculative.

Recent studies have shown that dysregulation of the AP also results in a proliferative glomerulonephritis called C3 glomerulonephritis (C3GN). C3GN is characterized by glomerular C3 deposition and the presence of numerous deposits in the mesangium and capillary walls, including subepithelial deposits.^{18, 26–29} Thus, there is considerable overlap in the biopsy findings of patients with ‘atypical’ post-infectious glomerulonephritis and C3GN.²⁶ This overlap is not surprising since both ‘atypical’ post-infectious glomerulonephritis and C3GN are due to abnormalities of AP of complement. However differences include the diffuse proliferative pattern of glomerulonephritis with ‘atypical’ post-infectious glomerulonephritis, as opposed to the membranoproliferative glomerulonephritis most commonly seen with C3GN, and the possible presence of Ig’s in ‘atypical’ post-infectious glomerulonephritis, which are typically absent in C3GN. On electron microscopy, ‘atypical’ post-infectious glomerulonephritis shows numerous subepithelial humps with mesangial and subendothelial deposits, while in C3GN there are numerous mesangial and subendothelial deposits and only occasional subepithelial deposits (Table 4). It should be pointed out that subepithelial humps can be present occasionally in Henoch-Schonlein purpura and Dense Deposit Disease.

Finally, we speculate that post-infectious and ‘atypical’ post-infectious glomerulonephritis are conceptually similar to hemolytic uremic syndrome (HUS) and atypical hemolytic uremic syndrome (aHUS) in that both HUS and post-infectious glomerulonephritis are caused by an infection, while in aHUS and ‘atypical’ post-infectious glomerulonephritis, although there may be an infectious trigger, it is the underlying abnormality of the AP that drives the disease process.

To summarize, we define the entity of ‘atypical’ post-infectious glomerulonephritis that mimics post-infectious glomerulonephritis on kidney biopsy, yet behaves differently in that these patients have persistent proteinuria and hematuria and may progress to end stage kidney disease. We show that patients with ‘atypical’ post-infectious glomerulonephritis have an underlying abnormality in the AP of complement. Based on our data, we recommend testing for abnormalities in the AP of complement in all patients with ‘atypical’ post-infectious glomerulonephritis. Novel treatment strategies that target the underlying complement defect should also be considered in patients with progressive renal failure.

METHODS

Renal biopsies from 11 Mayo Clinic patients were evaluated. In all cases, routine work-up including light, immunofluorescence and electron microscopy was performed. Clinical information was obtained from the charts. The Institutional Review Boards at the Mayo Clinic and University of Iowa approved the study.