Immune-complex glomerulonephritis with a membranoproliferative pattern in Frasier syndrome: a case report and review of the literature

Abstract

Background

Mutations in the Wilms tumor 1 gene cause a spectrum of podocytopathy ranging from diffuse mesangial sclerosis to focal segmental glomerulosclerosis. In a considerable fraction of patients with Wilms tumor 1 mutations, the distinctive histology of immune-complex-type glomerulonephritis has been reported. However, the clinical relevance and etiologic mechanisms remain unknown.

Case presentation

A 5-year-old child presented with steroid-resistant nephrotic range proteinuria. Initial renal biopsy revealed predominant diffuse mesangial proliferation with a double-contour and coexisting milder changes of focal segmental glomerulosclerosis. Immunofluorescence and electron microscopy revealed a full-house-pattern deposition of immune complexes in the subendothelial and paramesangial areas. Serial biopsies at 6 and 8 years of age revealed that more remarkable changes of focal segmental glomerulosclerosis had developed on top of the initial proliferative glomerulonephritis. Identification of a de novo Wilms tumor 1 splice donor-site mutation in intron 9 (NM_024426.6:c.1447 + 4C > T) and 46,XY-gonadal dysgenesis led to the diagnosis of Frasier syndrome.

Conclusions

Our findings, together with those of others, point to the importance of heterogeneity in clinicopathological phenotypes caused by Wilms tumor 1 mutations and suggest that immune-complex-mediated membranoproliferative glomerulopathy should be considered as a histological variant.

Background

Mutations in Wilms tumor 1 (WT1) gene cause several diseases characterized by renal and /or genital anomalies, such as Denys–Drash syndrome (DDS), Frasier syndrome (FS), and isolated focal segmental glomerulosclerosis (FSGS). DDS patients typically present early-onset diffuse mesangial sclerosis (DMS), a 46,XY disorder of sex differentiation, and Wilms tumor (WT). FS patients tend to exhibit milder phenotypes with an onset at adolescence, including FSGS, male-to-female sex reversal, and gonadoblastoma, but usually lack WT [1]. Given the high incidence of WT and gonadoblastoma in DDS and FS, prophylactic gonadectomy and nephrectomies are recommended [1]. Over 95% of DDS patients carry missense mutations in exons 8 and 9, whereas FS is commonly caused by a splice-donor site mutation in intron 9 [1]. WT1-related nephropathy is generally ascribed to developmental defects in glomerular podocytes [2, 3]. Several patients with DDS or FS display membranoproliferative glomerulonephritis (MPGN) that is mainly characterized by subendothelial immune deposits [3,4,5,6,7,8,9], suggesting that renal pathologies resulting from WT1 mutations are complex and affected by multiple factors. Possible pathogenic mechanisms are discussed by reviewing the current literature. Here, we present a case of a child with FS. To the best of our knowledge, this is the first case report where renal histological changes have been followed up for 4 years before and after immunosuppressive therapy. Our case report should alert clinicians to consider the possible existence or WT1 mutations behind seemingly immunologic forms of MPGN.

Case presentation

A child was first diagnosed with proteinuria during a regular check-up at 3 years of age but remained untreated until the age of 5 years old, when proteinuria reached the nephrotic range. Since the proteinuria did not resolve after 4 weeks of oral prednisolone (2 mg/kg/day), the child was referred to our hospital for further evaluation.

Pregnancy and perinatal periods were uneventful, with no family history of kidney disease. Physical examination revealed no edema, rash, or arthralgia, and the child had normal female external genitalia. Laboratory studies indicated nephrotic syndrome: hypoalbuminemia (2.3 g/dL), hyperlipidemia (total cholesterol, 342 mg/dL), and massive proteinuria (11 g/g creatinine) without hematuria. Blood-cell count, renal function, and serum complement C3 and C4 levels were all normal. Serological tests for hepatitis B and C and anti-nuclear antibodies were negative. In the first renal biopsy at age 5 (See Additional files 1, 2, 3 and 4), ~ 50% of glomeruli displayed global mild-to-severe mesangial proliferation, whereas others showed FSGS (Fig. 1a). In some glomerular tufts, the capillary wall was irregularly thickened with a double-contour configuration (Fig. 1b). Foam cells had infiltrated focally around the tubular interstitium. Immunofluorescence revealed a coarsely granular, full-house deposition pattern of 1+, positive for IgG, IgM, IgA, C3, and C1q along with trace C4+/− labeling in the mesangial and peripheral capillary loops. Electron microscopy revealed electron-dense deposits in the subendothelial and paramesangial areas (Fig. 1c). The glomerular basement membranes (GBM) exhibited normal thickness and contour, while focally showing subendothelial widening. These histologic features were consistent with endocapillary immune-complex (IC) glomerulonephritis with partial MPGN pattern, whereas minor FSGS changes were occasionally observed. Based on pathological assessment, intravenous methylprednisolone pulse, intravenous cyclophosphamide, cyclosporine A, and mycophenolate mofetil were administered. However, the proteinuria was unresponsive.

Fig. 1

Representative light and electron microscopy images of the first and second biopsies. a Mesangial proliferation with segmental sclerosis (arrowheads; periodic acid-Schiff staining; original magnification, 400×). b Double-contour formation (arrows; periodic acid methenamine silver staining; original magnification, 400×) (c) Subendothelial (asterisks) and paramesangial (arrowheads) deposits (electron micrograph; original magnification, 4000×). (a–c) Images from the first biopsy at age 5. d Irregularities of the GBM (asterisks) with mesangial interposition (asterisks; electron micrograph; original magnification, 8000×). d Image from the second biopsy at age 6. GBM: glomerular basement membrane

The second biopsy at age 6 (See Additional files 5, 6 and 7), following intravenous methylprednisolone pulse and cyclophosphamide, showed remarkable attenuation of IgA, C3, and C1q depositions in mesangiocapillary areas relative to the first biopsy. Irregular GBM thickening was more apparent with a double-contour pattern due to mesangial interposition (Fig. 1d). Along with a stepwise increase in cyclosporine A dosage, proteinuria gradually declined (3 g/g creatinine), thereby achieving partial remission. The third biopsy at age 8 (See Additional files 8 and 9) revealed coarse granular deposits of IgM (2+) and C3(1+) at the capillary periphery, suggesting macromolecule entrapment in sclerosing glomeruli, and we observed significantly fewer foam cells around the tubular interstitium. Throughout the clinical course, IC depositions had ameliorated in response to immunosuppressive therapy. However, as the child aged, glomerular capillary remodeling and podocyte injuries had progressed, thereby causing the development of more remarkable FSGS features onto the initial proliferative glomerulopathy.

Because of the steroid-resistant nephrotic syndrome, genetic testing was recommended. Targeted sequencing for 88 known renal disease genes (Additional files 10) detected a splice-donor site mutation in WT1 intron 9 (NM_024426.6: c.1447 + 4C > T). Segregation analysis of family members confirmed a de novo mutation (Fig. 2). Subsequent G-band analysis revealed a 46,XY karyotype. Bilateral streak gonads were observed by magnetic resonance imaging, confirming FS diagnosis. The patient underwent gonadectomy at age 10, was diagnosed with gonadoblastoma, and is currently treated with cyclosporine A and an angiotensin II-receptor blocker. The proteinuria is now in the nephrotic range, but renal function remains normal.

Fig. 2

Pedigree and sequencing analyses. Sanger sequencing of the WT1 exon 9 and intron 9 boundary in the affected individual and family members. The affected child (II-1, proband shown by arrow) harbored a single nucleotide substitution in the canonic donor KTS splice site of WT1 intron 9 (IVS9; c.1447 + 4C > T; RefSeq NM_024426.6:WT1 isoform D; ClinVar:000003500, dbSNP:rs587776577), which was absent in family members, indicating a de novo mutation. This variant has been reported elsewhere under alternate variant designations (e.g., 1432 + 4C > T or IVS9 + 4C > T). WT1: Wilms tumor suppressor gene

Discussion and conclusions

Herein, we presented a case of a child with FS with steroid-resistant nephrotic syndrome, whose renal histology initially showed a predominant proliferative glomerulonephritis that later progressed into FSGS. The first renal histology at age 5 was consistent with mesangial proliferative glomerulonephritis with MPGN pattern diagnosis based on mesangial proliferation, double-contour GBM, and “full-house” granular IC deposits along the glomerular capillary loops, as well as the paramesangial region. The pathologic findings were indistinguishable from those commonly seen in lupus nephritis, although our patient lacked serological abnormalities. The second and third biopsies, following immunosuppressive therapy, at ages 6 and 8, respectively, revealed that FSGS features (i.e., focal segmental capillary obsolescence and tubule interstitial foam cell infiltration) were superimposed on the IC glomerulonephritis and had become more apparent. Despite a partial response, persistent steroid-resistant nephrotic proteinuria incentivized us to conduct genetic testing, thereby allowing the diagnosis of FS caused by a typical splice-donor mutation in intron 9 of WT1.

WT1 mutations can cause a broad spectrum of clinical diseases affecting urogenital development and sexual differentiation at variable severity and combinations. Mutational survey of WT1 in steroid-resistant nephrotic syndrome cohorts [2, 3] revealed that an intron 9 splice mutation typically causes FSGS with a gonadal tumor, whereas missense and truncating mutations result in DMS with nephroblastoma. However, morphologic abnormalities considerably vary in histologic appearance among individuals with DDS/FS [3,4,5,6,7,8,9]. Detailed analysis of the renal histology of DDS individuals revealed complex glomerular changes, including endotheliosis-like endothelial injuries, foot-process fusion, and GBM alterations [6]. Previous studies report a significant fraction of DDS/FS individuals, including original and some familial cases, display MPGN with IC deposition in addition to FSGS or DMS, a histopathology commonly seen in WT1-related glomerulopathy [3,4,5,6,7,8,9,10,11,12,13,14,15,16,17] (Tables 1 and 2). Out of six DDS cases, four patients harbored a p.Arg467Trp (NM_024426.6:c.1399C > T) variant [4,5,6], the most common substitution (present in 40% of DDS patients), and two harbored a nonsense p.Arg463Ter (NM_024426.6:c.1387C > T) variant [3, 7] manifesting in an MPGN pattern. Moreover, nine FS cases have been reported, including two monozygous twins harboring a donor splice site mutation in intron 9 and initially presenting with MPGN [8, 9]. In these cases, ICs, comprised of either full-house or combined IgG + C3 patterns, were deposited along glomerular capillaries. Glomerulopathy in DDS usually manifests earlier and progresses faster into end-stage renal disease, relative to FS. Notably, there are DDS cases initially presenting with thrombotic microangiopathy or atypical hemolytic uremia syndrome (HUS) [18,19,20] (Table 3). It is not clear how the WT 1 variant could facilitate HUS-like severe endothelial injuries. Co-occurrence of atypical HUS with other glomerular diseases (e.g. FSGS) has been reported [21], suggesting that complement activation and podocyte dysfunction may be related, mechanistically. However, to address this hypothesis, we need the description of further case reports as well as additional data collection from experimental studies. As Noris et al. suggested [21], we should also take into account the possibility that another genetic abnormality or triggering environmental factor may be the main MPGN etiology in this case, over the WT1 glomerulopathy background.

Table 1 Summary of WT1-related glomerulopathy with MPGN pattern

Table 2 Summary of renal manifestations in Denys-Drash syndrome with Wilms tumor

Table 3 Summary of WT1-related glomerulopathy presenting an atypical hemolytic uremic syndrome or thrombotic microangiopathy phenotype

The mechanisms by which WT1 mutations cause MPGN have not been defined. Three factors might be implicated in the pathogenesis of IC-mediated glomerulonephritis involving WT1 mutations: 1) the WT-derived precipitating antigen promoting IC formation, 2) altered immune responses, and 3) increased vulnerability to endothelial injuries in structurally maldeveloped glomerular capillaries. First, MPGN is occasionally associated with malignancy, typically in lymphoproliferative disorders, but also in solid tumors (i.e., lung, colon, and renal carcinoma) [22, 23]. In this context, it is plausible that ICs might form through aberrant immune responses against oncofetal and/or non-autologous tumor antigens and trigger endothelial injuries of glomerular capillaries in WT patients [23]. Clinically, in most DDS cases, glomerulonephritis precedes or manifests simultaneously with WT diagnosis [3,4,5,6,7, 12,13,14,15,16] (Tables 1 and 2). Moreover, nephrotic syndrome can persist, even after complete excision of tumors with no evidence of recurrence and metastasis [4, 7]. However, the existence of WT-specific circulating antibodies has not been well defined. These observations indicate no convincing biological evidence linking WT to MPGN, thereby warranting further study.

Second, dysfunctional WT1 might be associated with aberrant immune responses, leading to IC formation [22], based on its role in the transcriptional regulation of multiple genes implicated in the differentiation of hematopoietic stem cells and apoptosis [24]. Consistent with our case, clinical studies report the effectiveness of cyclosporin in WT1 glomerulopathy [25], with no report of MPGN recurrence after renal transplant in patients bearing WT1 mutations, except for one DDS patient [4], which represents an unusual case of MPGN recurrence in the allografted kidney, even after WT resection and subsequent thorough immunosuppressive therapy.

An MPGN pattern resembling a glomerular morphology distinctive from classic podocytopathy (FSGS/DMS) is recognizable in some DDS/FSGS cases by the disrupted glomerular capillary integrity and, as of yet, undetermined predisposing factors for IC deposition. In-depth evaluation of glomerular histology in DDS patients and mouse models harboring p.Arg467Trp (NM_024426.6:c.1399C > T) demonstrates complex disturbances in podocytes and endothelial cells, as well as GBM maturation [6, 26]. Several studies of FS patients suggest GBM alterations as the first histological changes that precede overt features of MPGN (IC deposits) and FSGS (interstitial foam cells) [8, 9].

This case adds to the evidence of endothelial injuries as essential components in the pathogenesis of WT1-related glomerulopathy. In addition to DMS or FSGS pathologies, IC-mediated MPGN should be considered as a histological variant in patients harboring WT1 mutations. Early recognition of WT1 mutations allows for personalized choices of immunosuppressive reagents and prevention of tumorigenesis.