 |
|
 |
|
 |
 |
|
 |
 |
 |
 |
 |
 |
 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
ARTICLE |
a Departments of Pediatrics and of Human Genetics, University of Michigan, Ann Arbor, Michigan
b Department of Pediatrics, Hacettepe University, Ankara, Turkey
c Department of Human Genetics, Friedrich-Alexander-Universität, Erlangen-Nürnberg, Germany
 |
ABSTRACT |
|---|
 TOP ABSTRACT PATIENTS AND METHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
METHODS. We performed direct exon sequencing of NPHS1, NPHS2, and the relevant exons 8 and 9 of WT1, whereas the LAMB2 gene was screened by enzymatic mismatches cleavage.
RESULTS. We detected disease-causing mutations in 66.3% (53 of 80) families (NPHS1, NPHS2, WT1, and LAMB2: 22.5%, 37.5%, 3.8%, and 2.5%, respectively). As many as 84.8% of families with congenital onset (0–3 months) and 44.1% with infantile onset (4–12 months) of nephrotic syndrome were explained by mutations. NPHS2 mutations were the most frequent cause of nephrotic syndrome among both families with congenital nephrotic syndrome (39.1%) and infantile nephrotic syndrome (35.3%), whereas NPHS1 mutations were solely found in patients with congenital onset. Of 45 children in whom steroid treatment was attempted, only 1 patient achieved a lasting response. Of these 45 treated children, 28 had causative mutations, and none of the 28 responded to treatment.
CONCLUSIONS. First, two thirds of nephrotic syndrome manifesting in the first year of life can be explained by mutations in 4 genes only (NPHS1, NPHS2, WT1, or LAMB2). Second, NPHS1 mutations occur in congenital nephrotic syndrome only. Third, infants with causative mutations in any of the 4 genes do not respond to steroid treatment; therefore, unnecessary treatment attempts can be avoided. Fourth, there are most likely additional unknown genes mutated in early-onset nephrotic syndrome.
Key Words: nephrotic syndrome • LAMB2 • NPHS1 • NPHS2 • WT1
Abbreviations: NS—nephrotic syndrome • NSFL—nephrotic syndrome manifesting in the first year of life • CNS—congenital nephrotic syndrome • INS—infantile nephrotic syndrome • OMIM—Online Mendelian Inheritance in Man • ESRD—end-stage renal disease
Nephrotic syndrome (NS), the association of gross proteinuria, hypoalbuminaemia, edema, and hyperlipidemia, is often a life-threatening condition when manifesting as NS in the first year of life (NSFL). Whereas standardized treatment protocols are widely used for older children,1,2 in whom a response rate to steroid treatment of 
80% can be observed, clinical decision-making in children with NSFL remains challenging. Descriptively, NSFL has been classified as congenital NS (CNS), manifesting in utero or during the first 3 months of life, or infantile NS (INS), with onset between 4 months and 1 year of age.
Over the last few years, inherited impairments of the glomerular filtration barrier have been identified as important causes of NS. The glomerular filtration barrier separates the bloodstream from the urinary space and consists of 3 layers, fenestrated endothelium, glomerular basement membrane, and podocyte foot processes. The interdigitating podocyte foot processes are connected by the slit-diaphragm. The protein nephrin, encoded by the NPHS1 gene, represents a major component of this slit-diaphragm and connects foot processes in a zipper-like fashion.3 Mutations of NPHS1 were identified by Kestila et al4 as the cause of CNS of the Finnish type (Online Mendelian Inheritance in Man [OMIM] No. 256300). They result in a disruption of the glomerular filtration barrier and consecutive massive protein loss. Detailed studies on NPHS1 were undertaken, and 2 mutations Finmajor (L41fsX90) and Finminor (R1109X) were identified as the predominant cause of CNS in the Finnish population.4 Mutations in the NPHS2 gene, encoding podocin, were then identified by Boute et al5 as a cause of autosomal recessive steroid-resistant NS (OMIM No. 600995) among older children. Podocin represents an essential component of the slit-diaphragm and a close interactor of nephrin. Intracellular retainment of podocin mutants6 and consecutively altered trafficking of nephrin have been demonstrated.7 Mutations of podocin also result in a dysfunction of the glomerular filtration barrier. The preceding studies demonstrated that children with NPHS2 mutations may manifest as NSFL,8–11 but no systematic evaluation focusing on NPHS2 in infants has been undertaken. Furthermore, we recently reported dominant de novo mutations in exons 8 and 9 of the WT1 gene (encoding the transcription factor Wilms tumor suppressor gene 1) as a cause of isolated steroid-resistant NS in infants and children.12 Transcriptional activation of NPHS1 and upregulation of NPHS1 mRNA by WT1 have been shown.13,14 These findings explain why mutations in WT1 result in NS. In addition, WT1 mutations have been detected in patients with Wilms tumor (OMIM No. 194070),15,16 Denys-Drash syndrome (OMIM No. 194080),17 and Frasier syndrome (OMIM No. 136680).18 Finally, laminin-ß2 as a component of the glomerular basement membrane is crucial for podocyte foot process architecture and stability but is also found in other organs, such as the eye.19 Missense mutations of the LAMB2 present with a spectrum of symptoms reaching from isolated early onset NS20 to intermediate phenotypes, whereas patients with truncating LAMB2 mutations present with the full syndromic phenotype of Pierson syndrome with NS, diffuse mesangial sclerosis, distinct eye anomalies, and mental retardation (OMIM No. 609049).19
Mutations in the genes TRPC6,21 ACTN4,22 and CD2AP23 cause adult-onset NS. They follow a dominant pattern of inheritance and lead to focal segmental glomerulosclerosis. Because they have not been reported in children with NSFL, they were not included in this study.
Mutations in NPHS1, NPHS2, WT1, or LAMB2 in NSFL have been described in small patient cohorts only. These studies mainly focused on 1 of these genes respectively. No reliable data on the relative frequencies of mutations in these genes among children with NSFL are available. We, therefore, examined within a large cohort of 975 children with NS ascertained over 10 years, the subgroup of 89 European infants with NSFL for mutations in NPHS1, NPHS2, WT1, and LAMB2. Here we demonstrate that, in two thirds of children with NSFL, a genetic cause can be identified in 1 of these 4 genes and that children with these mutations do not respond to steroid treatment.
|
PATIENTS AND METHODS |
|---|
|
TOPABSTRACTPATIENTS AND METHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
Between 1996 and 2005, 107 children suffering from either sporadic or familial NSFL were enrolled in the study. To better define our group by ethnic origin, we excluded 18 non-European children from Asia and North America. The remaining 89 patients were of non-Scandinavian European descent, including 25 patients of Turkish descent. The 89 European patients studied were from 80 different families and included 71 families with 1 affected child and 9 families with 2 affected children with NSFL (Table 1). Parental consanguinity was documented for 4 sibling pairs and 15 children with sporadic NSFL. Fifty-four children from 46 families were classified as having CNS with first documented presentation in utero or within 90 days from birth (13 patients of Turkish descent). Thirty-five children from 34 families were classified as having INS with first documented presentation between 91 and 365 days (12 patients of Turkish descent). Partial data on 31 patients from 25 families have been published previously.8,9,12,20 In this work, we will refer to the number of patients studied when assessing clinical data and to the number of families studied when assessing genetic data, because affected siblings may follow a different clinical course but should bear the same mutations.
|
CEL I Nuclease Assay for LAMB2 Screening
To screen all 32 exons of the LAMB2 gene in 89 patients, the CEL I nuclease assay was used.26 This assay has a sensitivity of >92% in our hands (E. Otto, PhD, verbal communication, 2006). CEL I nuclease was extracted from celery plants as previously described by Till et al.26 To detect homozygous mutations in addition to heterozygous ones, DNA strands of affected and healthy control individuals were mixed after exon PCR. Samples were denatured at 95°C and cooled to allow heteroduplexes to form. After incubation of heteroduplexes with CEL I nuclease for 5 minutes, the enzymatic cleavage at sites of DNA mismatches was terminated by the addition of 0.5 m of EDTA and immediate incubation on ice. When products were run on a 1% agarose gel, the presence of 
2 bands indicated that cleavage had occurred at the site of a mismatch. Such aberrant samples were then chosen for mutation analysis by direct DNA sequencing as described above.
|
RESULTS |
|---|
|
TOPABSTRACTPATIENTS AND METHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
Frequencies of Mutations
Frequencies of detected mutations in NPHS1, NPHS2, WT1, or LAMB2 in 80 European families with NSFL are summarized in Table 1. The correlation of detected mutations with exact age at first clinical presentation for 89 affected individuals is shown in Fig 1.
|
The frequencies of mutations in NPHS1, NPHS2, WT1, and LAMB2 differed between children with congenital and infantile onset (Table 1). In CNS (onset 0–3 months), mutations were found in 1 of the 4 genes in 84.8% (39 of 46) of all families. The distribution was NPHS1, NPHS2, WT1, and LAMB2: 39.1% (18 of 46), 39.1% (18 of 46), 2.2% (1 of 46), and 4.4% (2 of 46). In 15.2% (7 of 46) of families with CNS in whom the disease could not be explained by mutations in these 4 genes, we detected only 1 heterozygous NPHS1 mutation in 3 families. In NPHS1, all of the patients with disease-causing mutations presented as CNS, that is, within the first 3 months of life. Manifestation of children with NPHS1 mutations was, thus, limited to congenital onset. In INS (onset 4–12 months), mutations were found in 44.1% (15 of 34) of families. The distribution was NPHS1, NPHS2, WT1, and LAMB2: 0.0% (0 of 34), 35.2% (12 of 34), 5.9% (2 of 34), and 2.9% (1 of 34; Table 1). The percentage of patients in whom the disease was explained by mutations in these 4 genes, thus, decreased with increasing age at first presentation. Mutations in LAMB2 and WT1 were a rare cause of NSFL between both groups, CNS and INS, and accounted together for NSFL in only 6.3% (5 of 80) of all families.
The frequencies of mutations in NPHS1, NPHS2, WT1, and LAMB2 were different between children of central European and of Turkish descent (Table 1 and Fig 1). In families of central European descent, mutations were found in 75.5% (43 of 57) of families. The distribution was, respectively, NPHS1, NPHS2, WT1, and LAMB2: 21.0% (12 of 57), 47.4% (27 of 57), 5.3% (3 of 57), and 1.8% (1 of 57; Fig 1A). Mutations were found in 43.5% (10 of 23) of families of Turkish descent. The distribution was, respectively, NPHS1, NPHS2, WT1, and LAMB2: 26.1% (6 of 23), 13.1% (3 of 23), 0.0% (0 of 23), and 4.3% (1 of 23; Fig 1B). Although in families of central European descent, mutations in NPHS2 represented the most frequent cause of CNS (18 of 35 [51.4%]), no mutations of this gene were seen in 13 families of Turkish descent with CNS (Fig 1).
Spectrum of Mutations
An exact reference of all of the detected mutations is given in the Appendix.
|
NPHS2
A founder mutation R138Q has been described by Boute et al.5 We confirm R138Q as the most frequent mutation as follows. In 30 families (18 with CNS and 12 with INS) with NPHS2 mutations, the founder mutation R138Q was detected homozygously in 12 families and compound heterozygously in 8 families. In CNS, all of the patients with NPHS2 mutations carried 1 loss of function mutation or R138Q, with the exception of patient F876. In INS, all of the patients with NPHS2 mutations, except for patient F1336, showed either a loss of function mutation or the founder mutation R138Q. The spectrum of detected mutations in NPHS2 was, therefore, similar for patients with CNS and INS.
WT1
The 3 mutations detected in WT1 (R366H [A651], D396N [A794], and R395P [A1021]) were de novo changes of exons 8 or 9 manifesting as a dominant disease in female patients.12,16,27,28
LAMB2
The disease-causing mutations detected in the LAMB2 gene of families F1012 (R246Q) and F1234 (C231R and L1393F) were missense mutations as further described by Hasselbacher et al.20
Genotype Versus Treatment Response
Steroid treatment was attempted in 50.6% (45 of 89), not given in 30.3% (27 of 89), and not reported in 19.1% (17 of 89) of children. Detailed clinical data of all 89 individuals with NSFL are provided in the Appendix.
Among children with CNS, steroids were administered less frequently (21 of 54 [38.9%]) than among patients with INS (24 of 35 [68.6%]). Only 6 treated children showed an initial response to steroids. Five of these children showed no mutations in the 4 genes, and 1 (F1313) carried a single heterozygous NPHS1 mutation, P264R, which also does not explain the phenotype. Among the 45 treated children, 5 of 6 initially responsive children relapsed early, whereas 1 child (A90) of Turkish descent achieved a continuous remission through the steroid treatment. This child manifested at 9 months of age, and carried only 1 single NPHS2 R229Q polymorphism. Therefore, 97.8% (44 of 45) of treated children did not achieve a lasting response under steroids, whereas the only child to respond to steroid treatment did not carry disease-causing mutations. In 62.2% (28 of 45) of treated children, disease-causing mutations were detected; none of these children responded to steroid treatment.
Genotype Versus Onset of End-Stage Renal Disease
Evaluation of the clinical follow-up was possible for 88.6% (79 of 89) of patients. At last examination, 46.8% (37 of 79) of children had not reached end-stage renal disease (ESRD), 6.3% (5 of 79) were in ESRD and had not received a kidney transplant, 27.8% (22 of 79) had received a kidney transplant and did not experience transplant failure, 7.6% (6 of 79) had suffered from kidney-transplant failure, and 11.4% (9 of 79) were deceased.
Progression to ESRD was more rapid in children with NPHS1 mutations than in children with NPHS2 mutations (P = .04). Nine children with NPHS1 mutations in ESRD had progressed to ESRD within a mean interval of 4.6 years (range: 1.8–9.3 years) from diagnosis, whereas the equivalent group of 16 children with NPHS2 mutations had developed ESRD after a mean of 6.6 years (range: 2.8–10.0 years). Four of the 19 children with NPHS1 mutations had died of their disease, whereas all 27 children with NPHS2 mutations were alive. Two of 17 children with NPHS2 mutations and kidney transplant (F1221 and A674) showed recurrence of focal segmental glomerulosclerosis in the transplant, and 2 had a rejection of their transplant (A126 and F1030). In 1 patient with disease-causing NPHS1 mutations (F475 II-1), no function was obtained, and in 1 patient with a single NPHS1 mutation (A693), the transplant was rejected.
|
DISCUSSION |
|---|
|
TOPABSTRACTPATIENTS AND METHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
We found a high rate of mutations clarifying the molecular pathogenesis in 66.3% of all 80 families with NSFL and in 84.8% of all families with CNS by screening the 4 known genes NPHS1, NPHS2, WT1, and LAMB2 only. These data show that NSFL is, to a large extent, a monogenic disease, in which genetic testing should be considered as an early diagnostic step. These findings have additional clinical relevance, because data on the relative frequency of genetically caused NSFL has been missing, and data on the prevalence of steroid responsiveness among children with NSFL is sparse. The efficacy of steroid treatment in infants with NS is considered to be low.29,30 However, the documented initiation of steroid treatment in more than half of all patients (45 of 89) reported here illustrates that this treatment remains in clinical practice in children with NSFL.
For 45 children in this study who were treated with steroids, an initial response to treatment was reported in 6 children only. Just 1 of these 6 initially responsive children (A90) achieved a remission but did not carry a disease-causing mutation in 1 of the 4 analyzed genes. In 28 (62.2%) of 45 treated children, we found disease-causing mutations in NPHS1, NPHS2, WT1, or LAMB2. None of the children who were treated with steroids and who carried a mutation showed a response. We, therefore, conclude that mutation analysis of NPHS1, NPHS2, WT1, or LAMB2 is warranted in children with NSFL. If DNA analysis reveals pathogenic mutations, the children should be spared the adverse effects of unwarranted steroid treatment attempts.
Our data show that children with mutations in NPHS1 always manifest within the first 3 months of life. Only small numbers of cases with NPHS2 mutations and manifestation during these first 3 months of life have been described.8,10 Our findings now identify NPHS2 mutations as a frequent cause of CNS among patients of central European descent (51.4%). This finding emphasizes the necessity to include in the mutation analysis of CNS not only NPHS1 but also NPHS2. If no mutations are found in these 2 genes, the screening of WT1 and LAMB2 may be indicated, because mutations were found in 3.8% and 2.5% of CNS, respectively.
The relative frequency of mutations in NPHS1, NPHS2, WT1, or LAMB2 in patients with NSFL should be seen in the context of the patients' ethnic origin. Turkish children in our cohort had a higher percentage of absence of mutations in any of the 4 genes (56.5%) compared with central European children (24.5%). In contrast to central European children, no NPHS2 mutations were found in Turkish children with CNS. The spectrum of mutations in NSFL has been reported to vary by ethnicity for NPHS1 and NPHS2. Mutations in NPHS1 show a high incidence in the Finnish population and are almost the exclusive cause of CNS there.4,31 Two truncating mutations, Finmajor and Finminor, of NPHS1 are predominantly seen in the Finnish population.4 Outside Finland, however, this Finnish-type CNS is less frequent, and a diversity of different mutations are found in NPHS1.32 Both findings are further supported by our data on mutations, so that further studies will be necessary to evaluate the frequency of mutations in these genes in NSFL among children of different ethnic origin.
Our results reveal a specific spectrum of "severe" NPHS2 mutations resulting in NSFL. Twenty-eight (93.3%) of 30 families with NPHS2 mutations carried 1 truncating mutation or the founder mutation R138Q, a finding also supported by data from Weber et al.10 Published functional data may provide the first answers as to why these specific alleles cause NS with early onset. It was shown that the NPHS2 variant R138Q, which we frequently detected in our patients, is retained in the endoplasmatic reticulum disrupting the targeting of NPHS1 to lipid raft microdomains.7,33,34 Based on these experimental findings, Nishibori et al34 speculated that this altered nephrin positioning resulting from the R138Q mutation in NPHS2 could resemble CNS of the Finnish type not only pathophysiologically but also clinically.
|
CONCLUSIONS |
|---|
|
TOPABSTRACTPATIENTS AND METHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
|
ACKNOWLEDGMENTS |
|---|
Members of the Study Group of the Arbeitsgemeinschaft für Pädiatrische Nephrologie include A. Noyan (Adana, Turkey); A. Bakkaloglu (Ankara, Turkey); S. Spranger (Bremen, Germany); S. Briese, D. Müller, U. Querfeld (Berlin, Germany); G. Reusz (Budapest, Hungary); R. Bogdanovic (Belgrad, Serbia); B. Beck, B. Hoppe, C. Licht, M.T.F. Wolf (Cologne, Germany); J. Dötsch, C. Plank, W. Rascher (Erlangen, Germany); P. Hoyer (Essen, Germany); M. Schröder (Frankfurt, Germany); M. Brandis, A. Fuchshuber, M. Pohl, C.v. Schnakenburg (Freiburg, Germany); C. Mache (Graz, Austria); F. Schäfer, T. Knüppel, O. Mehls, B. Tönshoff, D. Wenning (Heidelberg, Germany); M. Kemper, D.E. Müller-Wiefel (Hamburg, Germany); J.H.H. Ehrich, G. Offner (Hannover, Germany); M. Barenbrock (Hamm, Germany); T. Jungraithmayr, B. Zimmerhackl (Innsbruck, Austria); J. Misselwitz (Jena, Germany); S. Wygoda (Leipzig, Germany); D. Böckenhauer (London, United Kingdom); M. Schuhmacher (Luebeck, Germany); M. Benz, M. Griebel, J. Höfele, L. Weber (Munich, Germany); H. Fehrenbach (Memmingen, Germany); M. Bulla, E. Kuwertz-Bröcking, A. Schulze Everding (Muenster, Germany); M. Shenoy (Newcastle, United Kingdom); L. Patzer (Halle, Germany); T. Seeman (Prag, Czeck Republic); A. Gianviti, G. Rizzoni (Rome, Italy); O. Amon (Tuebingen, Germany); J. Mühleder (Wels, Austria); T. Neuhaus (Zuerich, Switzerland); and T. Stuckert (Zwickau, Germany).
We thank the patients and their families for taking part in this study. We particularly thank Bethan Hoskins, PhD, for very helpful advice and critical reading of the article.
|
FOOTNOTES |
|---|
Address correspondence to Friedhelm Hildebrandt, MD, University of Michigan Health System, 8220C MSRB III, 1150 W Medical Center Dr, Ann Arbor, MI 48109-0646. E-mail: fhilde{at}umich.edu
The authors have indicated they have no financial relationships relevant to this article to disclose.
|
REFERENCES |
|---|
|
TOPABSTRACTPATIENTS AND METHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
This article has been cited by other articles:
 |
|
 |
|
  A. Philippe, F. Nevo, E. L. Esquivel, D. Reklaityte, O. Gribouval, M.-J. Tete, C. Loirat, J. Dantal, M. Fischbach, C. Pouteil-Noble, et al. Nephrin Mutations Can Cause Childhood-Onset Steroid-Resistant Nephrotic Syndrome J. Am. Soc. Nephrol., October 1, 2008; 19(10): 1871 - 1878. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. V. Lemley Yet More Ways to Skin a Cat: Nephrin Mutations outside the Neonatal Period J. Am. Soc. Nephrol., October 1, 2008; 19(10): 1837 - 1838. [Full Text] [PDF]
|
|
|
|
|
|
J. Ratelade, T. A. Lavin, A. O. Muda, L. Morisset, G. Mollet, O. Boyer, D. S. Chen, A. Henger, M. Kretzler, N. Hubner, et al. Maternal Environment Interacts with Modifier Genes to Influence Progression of Nephrotic Syndrome J. Am. Soc. Nephrol., August 1, 2008; 19(8): 1491 - 1499. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. F. Heeringa, C. N. Vlangos, G. Chernin, B. Hinkes, R. Gbadegesin, J. Liu, B. E. Hoskins, F. Ozaltin, F. Hildebrandt, and Members of the APN Study Group Thirteen novel NPHS1 mutations in a large cohort of children with congenital nephrotic syndrome Nephrol. Dial. Transplant., May 23, 2008; (2008) gfn271v1. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. Gbadegesin, B. G. Hinkes, B. E. Hoskins, C. N. Vlangos, S. F. Heeringa, J. Liu, C. Loirat, F. Ozaltin, S. Hashmi, F. Ulmer, et al. Mutations in PLCE1 are a major cause of isolated diffuse mesangial sclerosis (IDMS) Nephrol. Dial. Transplant., April 1, 2008; 23(4): 1291 - 1297. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
B. Hinkes, C. Vlangos, S. Heeringa, B. Mucha, R. Gbadegesin, J. Liu, K. Hasselbacher, F. Ozaltin, F. Hildebrandt, and and Members of the APN Study Group Specific Podocin Mutations Correlate with Age of Onset in Steroid-Resistant Nephrotic Syndrome J. Am. Soc. Nephrol., February 1, 2008; 19(2): 365 - 371. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. R. Pollak, M. P. Alexander, and J. M. Henderson A Case of Familial Kidney Disease Clin. J. Am. Soc. Nephrol., November 1, 2007; 2(6): 1367 - 1374. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
 |
||
|
||
Home | My Pediatrics | Journal Information | Current Issue | Past Issues | Subscriptions & Services | Contact Us Click here for faster international access © 2007 American Academy of Pediatrics. All rights reserved. |
||
|
||
