Familial nephrotic syndrome: Clinical spectrum and linkage to chromosome 19q13

Transcription

1 Kidney International, Vol. 57 (2000), pp GENETIC DISORDERS DEVELOPMENT Familial nephrotic syndrome: Clinical spectrum and linkage to chromosome 19q13 ABHAY VATS, ALICE NAYAK, DEMETRIUS ELLIS, PARMJEET SINGH RANDHAWA, DAVID N. FINEGOLD, KARA L. LEVINSON, and ROBERT E. FERRELL Division of Nephrology and Endocrinology, Department of Pediatrics, Children s Hospital of Pittsburgh; Division of Transplantation Pathology, Department of Pathology, University of Pittsburgh; and Department of Human Genetics, Graduate School of Public Health, University of Pittsburgh School of Medicine, Pittsburgh, Pennsylvania, USA Familial nephrotic syndrome: Clinical spectrum and linkage to chromosome 19q13. Background. Familial nephrotic syndrome (NS) has both autosomal dominant and recessive forms of inheritance. Recent studies in families with an autosomal dominant form of focal segmental glomerulosclerosis (FSGS) have been at odds concerning linkage to chromosome 19q13 (Mathis et al, Kidney Int 53: , 1998; Winn et al, Kidney Int 55: , 1999), suggesting genetic heterogeneity. This study examines the clini- cal features and confirms linkage to chromosome 19q13 in a family with autosomal dominant NS. Methods. DNA samples were obtained from 16 of 17 family members. Genomic DNA was isolated, and polymerase chain reaction was performed for five markers spanning the area of interest on chromosome 19q13. Data were evaluated using two- and six-point linkage analysis. Results. Clinical features included presentation of NS in childhood, steroid unresponsiveness, and slow progression to renal failure. Renal biopsy in affected family members showed lesions ranging from minimal change to mesangial proliferative glomerulonephritis to FSGS. Linkage was confirmed between the disease state and chromosome 19q13, with a maximum logarithm of odds (LOD) score of Linkage was observed for a 7 cm region on chromosome 19q13, defined by markers D19S425 and D19S220. Conclusions. This study confirms the Mathis et al report of linkage to chromosome 19q13 in a family with autosomal dominant NS. However, there were notable differences in the presenting clinical and histopathologic features of our affected family members compared with those of Mathis et al. This suggests that the gene on chromosome 19q13 may be responsible for considerable phenotypic heterogeneity and variable expression in both clinical presentation and renal histopathology. Massive proteinuria, hypoalbuminemia, and edema characterize idiopathic nephrotic syndrome (NS). Renal Key words: inherited disease, focal segmental glomerulosclerosis, minimal change disease, linkage analysis, progressive renal failure. Received for publication May 13, 1999 and in revised form September 3, 1999 Accepted for publication September 16, by the International Society of Nephrology biopsy in children who respond to steroids usually shows minimal change disease (MCD), which is usually associated with a favorable prognosis [1]. Conversely, steroid nonresponsive NS has a poorer prognosis and is more frequently associated with pathologies such as focal seg- mental glomerulosclerosis (FSGS) and mesangioproliferative glomerulonephritis (GN) [1, 2]. Minimal change nephrotic syndrome (MCNS) and FSGS are thought to be two very distinct pathologic entities with significant differences in clinical presentation and outcome; however, progression from MCNS and mesangioproliferative GN to FSGS has been reported in both steroid-responsive and steroid-nonresponsive NS patients [1 3]. The mechanisms underlying the etiopathogenesis of MCNS/FSGS remain unclear, although genetic factors have been implicated in various studies, even though idiopathic NS is generally regarded as a sporadic disease. Several observations point toward a genetic susceptibil- ity to NS, as familial cases have been reported periodi- cally, and congenital NS of the Finnish type (CNF) is a well-recognized heritable form of NS [1]. Previous studies have also reported human lymphocyte antigen (HLA) linkage to various antigens for both MCD and FSGS [4 6]. However, the genes responsible for these pheno- types have not been elucidated. Recently, both autoso- mal dominant and recessive forms of familial NS have been identified [7 11]. The reported autosomal domi- nant forms of NS are generally less severe than the recessive forms. Although FSGS is the most common histopathologic lesion in patients with familial NS, the clinical features vary widely in age of presentation and rate of progression to end-stage renal disease (ESRD). Hence, elucidation of genetic factors associated with various heritable forms of NS may allow for a greater understanding of etiopathogenesis, and possibly lead to a more precise classification and new therapeutic approaches. Recently, linkage analysis has been reported in two large pedigrees with autosomal dominant NS in which the primary pathology was FSGS. In the first study, Mathis 875

2 876 Vats et al: NS linked to chromosome 19q13 Fig. 1. Children s Hospital of Pittsburg (CHP) family 101 pedigree. The pedigree has been altered to preserve anonymity of the family. The three-digit numbers located under the individual number refer to chromosomal marker used in the study. Symbols are: ( ) male; ( ) female; ( ) affected; ( ) deceased. / et al reported a kindred from Oklahoma with linkage to a narrow region on chromosome 19q13 [8, 9]. This region also includes the gene for CNF [12 14]. However, Winn et al excluded linkage to the same locus in a large New Zealand family with a similar disease and inheritance pattern [11], suggesting genetic heterogeneity. In this study, we report a family from Pittsburgh with autosomal dominant NS with significantly different clinical and histopathological features in which the trait maps to chromosome 19q13. METHODS The study family (#101) has been followed at the Children s Hospital of Pittsburgh (CHP) since The family is a 17-member kindred spanning three generations (Fig. 1). One of the family members is deceased (I:2). Six affected family members have undergone renal biopsy over the years (Table 1). The biopsy on I:2 was not available for review. One member (III:4) was biopsied at Texas Children s Hospital, and the four remaining family members were biopsied at CHP. Two of the patients were biopsied twice (II:2 and II:5). P.S.R. reviewed the histopathology independently on all available renal biopsies. An evaluation of the affected family members included a complete history, physical examination, and laboratory investigations, including assays of serum cre- atinine and urinary protein. Asymptomatic individuals were screened for proteinuria by semiquantitative uri- nalysis (dipstick). Family members were considered affected if they had a documented proteinuria of greater than 3 on urinalysis, a urinary protein to creatinine ratio of greater than 2.0, or a 24-hour urine protein excre- tion of greater than 2 g. One child (III:6) was considered affected, although she had a proteinuria of 2 because her serum albumin was 2.1 g/dl. Of the 17 family members, 10 were considered affected. Peripheral blood was obtained on 15 family members, and an oral mucosal brushing was acquired from one family member (III:7) for DNA extraction. Studies were performed after obtaining informed consent in accor- dance with a protocol approved by the Human Rights Committee at CHP. DNA isolation and genotyping DNA was isolated from ethylenediaminetetraacetic acid (EDTA) anticoagulated whole blood and cheek

3 Vats et al: NS linked to chromosome 19q Table 1. Selected clinical characteristics of Children s Hospital of Pittsburgh (CHP) family 101 Laboratory Values at Presentation Current Age at Urine age presentation 24 Hour protein/ S Cr S Alb UA protein Cr ratio Kidney biopsy Individual years mg/dl g/dl (dipstick) g mg/mg (age at biopsy) Clinical course I:2 Deceased (Exact age of onset of symptoms and details of early presentation unknown) ESRD at age 50. Transplant at 51. II: NA 2.2 FSGS (8 y) Steroid unresponsive. PE and DVT Focal global sclerosis as adult. (11 y) II: Focal global sclerosis No steroid trial documented. (9 y) Asymptomatic as an adult. II: NA MCD (4 y) Steroid, mercaptopurine, and cyclophosphamide unresponsive. Intermittent edema. III: NA MCD (6 1/2 y) Ongoing proteinuria. No steroid trial recorded. III: Mesangial Prolifera- Steroid, cyclosporine and chloramtive GN (19 mo) bucil unresponsive. ESRD at age 15. Hemodialysis dependent. III: Focal global sclerosis Steroid unresponsive (15 mo) III: No steroid trial. Continued proteinuria. III: NA NA No steroid trial. Continued proteinuria. III: NA NA 5.2 No steroid trial. Abbreviations are: S Cr, serum creatinine; S Alb, serum albumin; UA, urine analysis; FSGS, focal segmental glomerular sclerosis; MCD, minimal change disease; DVT, deep venous thrombosis; PE, pulmonary embolism; NA, not available. cells using the Qiagen kit (Qiagen, Inc., Valencia, CA, generation denied renal disease in prior generations. USA). Fluorescently labeled primers for chromosome Segregation of the disease in CHP family 101 followed 19 markers D19S425, D19S208, D19S191, D19S224, and an autosomal dominant pattern. The family pedigree is D19S220 were purchased from Research Genetics (Huntsville, shown in Figure 1. Not all family members from generaprotocols AL, USA). Markers were amplified using standard tion one are shown. The selected clinical features of and the products resolved on an ABI 377 automatic affected family members are described in Table 1. Of sequencer (Applied Biosystems, Foster City, CA, the 10 affected members, the age of presentation ranged USA) and were analyzed using the GENOTYPER program from three months to four years, with a mean age of package [15]. onset of 18.4 months. Eight of 10 affected patients pre- sented with periorbital and extremity edema. The re- Linkage analysis maining two patients (III:6, III:7) were evaluated for Prior to recruitment of the family, a simulation analysis proteinuria prior to onset of symptoms. Twenty-four using the computer program SLINK (abstract; Weeks et hour protein excretion ranged from 0.87 to 7.76 g, with al, Am J Human Genet 47:A204, 1990), gave an estimated a mean of 2.32 g. Serum creatinine was in the normal probability of observing a logarithm of odds (LOD) score range for age and weight of the patient at the time of of 2.0 of 93%, assuming a fully penetrant autosomal diagnosis (Table 1). Microscopic hematuria was noted dominant disease model and a disease allele frequency of in one member at presentation (III:2) and in two other Two-point linkage analysis was performed using family members (II:2, III:3) during their clinical course. VITESSE with the same parameters as the simulation Hematuria was noted in family member II:5 after the [16]. A locus order cen-s425-s208-s191-s224-s220-tel administration of cytotoxic agents and was diagnosed as was taken from Mathis et al [9]. hemorrhagic cystitis. One of the affected family members (III:6) was born with congenital corneal opacity (Peters anomaly) and required corneal transplantation. None of RESULTS these individuals displayed any evidence of other causes Clinical evaluation and family data of renal dysfunction, such as diabetes or collagen vascular Children s Hospital of Pittsburgh family #101 is a 17- disease. member kindred of mixed African American and Caucasian heritage spanning three generations. One member (II:6) is of African origin. Family members in the first Of the six renal biopsies, five were available for review. On light microscopy, focal global sclerosis (Fig. 2a) was seen in two patients (II:4 and III:4). Classic lesions of

4 878 Vats et al: NS linked to chromosome 19q13 Fig. 2. Spectrum of pathological lesions. (a) A globally sclerotic glomerulus in a biopsy from case II:4 ( 700). (b) Focal segmental sclerosis seen in a renal biopsy from patient II:2 ( 700). (c) A glomerulus from case III:3 showing focal mesangial proliferation ( 350). (d) An essentially normal glomerulus illustrating the histologic appearance classified as minimal change disease ( 350). (e) Electron micrograph of III:4 showing extensive foot process fusion. onstrated a moderate amount (2 ) of mesangial deposits of IgM, whereas electron microscopy documented mes- angial and paramesangial electron-dense deposits. Ultrastructural examination of the remaining cases revealed subendothelial granular translucent material in the glo- FSGS (Fig. 2b) were seen in one case (II:2). Focal mesangial proliferation (Fig. 2c) was seen in three cases (II:4, III:3, III:4), and one of these could be classified as mesangial proliferative GN (IgM nephropathy variant; III:3). Immunofluorescence examination in this latter case dem-

5 Vats et al: NS linked to chromosome 19q merular capillary loops and variable fusion of the podocyte foot processes (Fig. 2e). One case (II:5) was considered to represent MCD because all of the 24 glomeruli in two biopsies appeared to be morphologically normal (Fig. 2d). Two family members (I:2 and III:3) progressed to renal failure. Family member I:2 progressed to renal failure in the fifth decade, was transplanted at age 51, and died at age 57 of myocardial infarction and respiratory complications. Family member III:3 abruptly progressed to renal failure recently at age 15 after maintaining stable renal function for the first 14 years of life. His serum creatinine at 12 years of age was 0.5 mg/dl and was 0.9 to 1.0 mg/dl in the 14th year of life. He is currently hemodialysis dependent. This individual had numerous infectious episodes as an infant, including two episodes of Hemophilus influenzae meningitis, leading to deafness. He required repeated infusions of intravenous immunoglobulin for hypogammaglobulinemia. He also has severe pulmonary dysfunction secondary to presumed undiagnosed episodes of pneumonia during his infancy and childhood, leading to bilateral bronchiectasis. Most of the other individuals had mild clinical disease with intermittent edema, which appeared to be very slowly progressive. All of the affected children in the third generation have also been diagnosed with severe reactive airway disease. One individual (II:4) had remission of clinical symptoms at age 17 but continues to have pro- teinuria. All patients were nonresponsive to steroids and cytotoxic agents. DNA samples were collected from a total of 16 individuals, including 9 affected and 7 unaf- fected, which included spouses. Linkage analysis The family pedigree with five locus haplotypes of each available family member is shown in Figure 1. All af- fected family members inherited the identical five locus haplotype and two point linkage analysis using this haplotype gave a maximum LOD score of 2.25 at 0. Six point linkage analysis using five locus genotypes gave a maximum LOD score of 2.41 at 0. These results confirm the estimated maximum LOD score obtained in the SLINK simulation analysis and provide evidence for a gene-causing autosomal dominant NS within the 7 cm region defined by markers D19S425 and D19S220. DISCUSSION We have reported a kindred with familial NS, which shows linkage to a region on chromosome 19q13, confirming the findings of Mathis et al [9]. Phenotypic similarities between CHP family 101 and the family of Mathis et al include an autosomal dominant inheritance pattern, as well as the presence of significant proteinuria. However, there were significant differences between the two families. The Mathis et al family was Caucasian and much larger (100 vs. 17 in our family), yet had a smaller proportion of affected individuals (29 affected vs. 10 in ours), indicating variable expressivisity and penetrance of the trait. Patients in the Mathis et al family usually presented with NS in adulthood (mean age at the fifth decade) and progressed to ESRD in a mean of nine years. In contrast, CHP family 101 demonstrated com- plete penetrance. Furthermore, the affected members presented early in life (mean age 18.4 months, range 3 months to 4 years) and maintained normal or near normal renal function for several decades. Two family mem- bers in CHP family 101 developed ESRD, one in the fifth decade of life and the other abruptly at age 15. The clinical course of the affected individuals in CHP family 101 is somewhat unusual in that they had severe steroid unresponsive NS with an onset at a young age, and, except for the two individuals mentioned, did not prog- ress to ESRD up to 30 years later. We speculate that iatrogenic factors, including intravenous immunoglobu- lin, may have contributed to the rapid decline in renal function in the 15-year-old member [17, 18]. Interestingly, the pathological abnormality seen in our family varied from minimal change and mesangial prolif- erative GN to FSGS at the time of initial presentation (Fig. 2), in contrast to mostly FSGS seen in the family reported by Mathis et al [8, 9]. A biopsy of one of the affected members in our family (II:2) demonstrated classic focal segmental sclerosis (Fig. 2b), whereas two others (II:4, III:4) were noted to have focal global glomerulo- sclerosis with no segmental involvement (Fig. 2a). It is possible that the biopsies of these two family members and the member diagnosed with MCD (II:5) may have demonstrated FSGS lesions if serial sections of the biopsy tissue were examined [19]. It has been postulated that MCD and FSGS could be variants of the same disease [1, 2]. As all three different histologic lesions were linked to the same localized chromosomal region in our family, it is possible that the same locus or allele is re- sponsible for the differing histopathologic manifesta- tions. This supports the notion that in a subset of patients with NS, MCD and FSGS may represent two ends of the histologic spectrum of the same disease. However, it is possible that modifying genes not yet identified could account for the differing histopathologic findings among family members. The different rates of progression and outcome variability in the CHP 101 family and that of Mathis et al may be partly due to the different histologic manifestations at presentation in the two families, as FSGS is typically associated with a faster rate of progres- sion to ESRD in comparison to MCD [1]. It is likely that significant differences in the expression, penetrance, and progression of the disease in the Mathis et al and CHP families may be due to allele or locus heterogeneity or epistatic and/or environmental factors.

6 880 Vats et al: NS linked to chromosome 19q13 The region 19q13 defined by the markers used in this 003 nephrnav.htm) has been developed to provide more study spans a genetic distance of 7 cm. A mutation in information about this study. the nephrin gene (NPHS1, MIM ), which encodes a unique protein expressed predominantly in glomeruli ACKNOWLEDGMENTS [14], has been associated with CNF and is located in the We are indebted to the members of the study family for their same region [12 14]. Besides the Finnish population, the cooperation and assistance. We thank Texas Children s Hospital for CNF gene is also implicated in congenital NS in non- allowing review of pathological slides, Ms. Nancy Petro and Ms. Meg Alexandra for DNA preparation and genotype analysis, and Ms. Bobbi Finnish families [20]. Similar to children with CNF (who Wierioch for secretarial assistance. often have proteinuria detectable at birth and are resistant to steroids), two of the affected members of the Reprint requests to Abhay Vats, M.D., Children s Hospital of Pitts- burgh, Division of Pediatric Nephrology, 3705 Fifth Avenue, Pittsburgh, CHP family 101 had detectable proteinuria within the Pennsylvania 15213, USA. first four months of life, and six of the seven remaining vatsa@chplink.chp.edu affected members presented with the disease by two years of age. None of the affected persons responded to REFERENCES steroids. Despite these similarities, CHP family 101 had 1. Barrat TM, Clark G: Minimal change nephrotic syndrome and an autosomal dominant inheritance pattern and slow focal segmental glomerulosclerosis, in Pediatric Nephrology (3rd progression to ESRD, whereas CNF is an autosomal ed), edited by Holliday MA, Barratt TM, Avner ED, Baltimore, Williams & Wilkins, 1994, pp recessive disease and patients show a rapid decline in 2. Korbet SM: Primary focal segmental glomeruosclerosis. J Am Soc renal function within the first three to four years of life. Nephrol 9: , 1998 It is possible that either nephrin or a gene(s) closely 3. Nash MA, Bakare MA, D Agati V, Pirani CL: Late development of chronic renal failure in steroid responsive nephrotic syndrome. linked or related to nephrin may be involved in the J Pediatr 101: , 1982 etiopathogenesis of this disorder. 4. Goodman DJ, Clarke B, Hope RN, Miach PJ, Dawborn JK: Although idiopathic NS is usually sporadic, various Familial focal glomerulosclerosis: A genetic linkage to HLA locus? Am J Nephrol 15: , 1995 types of familial NS have been reported [1] and lately 5. McEnery PT, Walsh TM: Major histocompatibilty complex antigens in steroid sensitive nephrotic syndrome. Pediatr Nephrol 3:33 have been shown to be linked to various genes in addition to the locus on chromosome 19q13 such as SRN-1 (chro- 36, Moncreiff MW, White RHR, Glasgow EF, Winterborn MH, mosome 1q25-31) and WT-1 [7, 20]. The gene SRN-1 is Cameron JS, Ogg CS: The familial nephrotic syndrome. II. A linked to an autosomal recessive NS and is associated clinopathological study. Clin Nephrol 1: , 1973 with rapid progression to ESRD in families of European 7. Fuchshuber A, Jean G, Gribouval O, Gubler MC, Broyer M, Beckmann JS, Niaudet P, Antignac C: Mapping a gene (SRN1) and African ancestry [7]. The histopathologic lesion seen to chromosome 1q25-31 in idiopathic nephrotic syndrome confirms in the majority (18 out of 20) of the affected patients in a distinct entity of autosomal recessive nephrosis. Hum Mol Genet that study was of FSGS. Other studies have reported 4: , Mathis BJ, Calabrese KE, Slick GL: Familial glomerular disease that mutations in the gene WT-1 are associated with with asymptomatic proteinuria and nephrotic syndrome: A new diffuse mesangial sclerosis both with and without Denys clinical entity. J Am Osteopath Assoc 92: , 1992 Drash syndrome [21, 22]. Thus, the linkage of various 9. Mathis BJ, Kim SH, Calabreses MH, Seidman JG, Seidman CE, Pollack MR: A locus for inherited focal segmental glomeruloscleforms of heritable NS with different chromosomal re- rosis maps to chromosome 19q13. Kidney Int 53: , 1998 gions underscores genetic and phenotypic heterogeneity 10. Conlon PJ, Butterly D, Albers F, Rodby R, Gunnells JC, that will be more fully understood once various associsegmental glomerulosclerosis. Am J Kidney Dis 26:34 40, 1995 Howell DN: Clinical and pathologic features of familial focal ated genes are identified. The identification of various 11. Winn MP, Conlon PJ, Lynn KL, Howell DN, Gross DA: Clinical nephropathy susceptibility genes would aid in early desclerosis. Kidney Int 55: , 1999 and genetic heterogeneity in familial focal segmental glomerulo- tection and more precise characterization of high-risk 12. Ruotsalainen V, Ljungberg P, Wartiovaara J, Lenkkeri U, Kespatients, which can be useful for prognostication of clini- tila M, Ualanko H, Holmberg C, Tryggvason K: Nephrin is cal course in affected individuals and could possibly lead specifically located at the slit diaphragm of glomerular podocytes. to more focused therapeutic approaches. Proc Natl Acad Sci USA 96: , Lenkkeri U, Mannikko M, McCready P, Lamerdin J, Gribouval O, Niaudet P, Antignac C, Kashtan CE, Holmberg C, Olsen Postscript A, Kestila M, Tryggvason K: Structure of the gene for congenital We have identified several families with inherited NS/ nephrotic syndrome of the Finnish type (NPHS1) and characterization of mutatioins. Am J Hum Genet 64:51 61, 1999 FSGS from the Pittsburgh/Western Pennsylvania region 14. Kestila M, Lenkkeri U, Manniko M, Lamerdin J, McCready P, and have expanded our search to other geographic re- Putaala H, Routsalainen V, Morita T, Nissinen M, Herva R, gions. Families or physicians interested in participating Kashtan CE, Peltonen L, Holmberg C, Olsen A, Tryggvason K: Positionally cloned gene for a novel glomerular protein linkage analysis and gene localization studies for inher- nephrin is mutated in congenital nephrotic syndrome. Mol Cell ited NS/FSGS are requested to contact Abhay Vats (vats@ 1: , 1998 chplink.chp.edu) or Alice Nayak (anayak2@yahoo.com) 15. Ferrell RE, Levinson KL, Esman JH, Kimak MA, Lawrence EC, Barmada MM, Finegold DN: Hereditary lymphedema: Evi- dence for linkage and genetic heterogeneity. Hum Mol Genet 7: , 1998 by or telephone ( ). A webpage (

7 Vats et al: NS linked to chromosome 19q O Connell JR, Weeks DE: The VITESSE algorithm for rapid drome of the Finnish type in non-finnish families. JAmSoc exact multilocus linkage analysis via genotype set-recording and Nephrol 7: , 1996 fuzzy inheritance. Nat Genet 11: , Schumacher V, Schver K, Wulh E, Altrogge H, Bonzel KE, 17. Cayco AV, Perazella MA, Hayslett JP: Renal insufficiency after Guschmann M, Neuhaus TJ, Pollastro RM, Kuwertz-Broking intravenous immune globulin therapy: A report of two cases and E, Bulla M, Tondera AM, Mundel P, Helmchen U, Waldherr an analysis of the literature. J Am Soc Nephrol 8: , 1997 R, Weirich A, Royer-Pokora B: Spectrum of early onset nephrotic 18. Koboski J, Nicol P: Renal toxicity of intravenous immunoglobulin. syndrome associated with WT1 missense mutations. Kidney Int 53: , 1998 Clin Nephrol 37: , Jeanpierre C, Denamur E, Henry I, Cabanis MO, Luce S, Cecille 19. Remuzzi A, Pergolizzi R, Mauer MS, Bertaini T: Three-dimen- A, Elion J, Peuchmaur M, Loirat C, Niaudet P, Gubler MC, sional morphometric analysis of segmental glomerulosclerosis in Junien C: Identification of constitutional WT1 mutations, in pathe rat. Kidney Int 38: , 1990 tients with isolated diffuse mesangial sclerosis, and analysis of 20. Manniko M, Lenkkeri U, Kashtan C, Kestila M, Holmberg C, genotype/phenotype correlations by use of a computerized mutation Trygguason K: Haplotype analysis of congenital nephrotic syn- database. Am J Hum Genet 62: , 1998