J Am Soc Nephrol 14: S195–S201, 2003 Linkage Analysis of Candidate Loci for End-Stage Renal Disease due to Diabetic Nephropathy

SUDHA K. IYENGAR,*† KATHERINE A. FOX,*† MARLENE SCHACHERE,*† FAUZIA MANZOOR,*† MARY E. SLAUGHTER,*† ADRIAN M. COVIC,†‡ S. MOHAMMED ORLOFF,*† PATRICK S. HAYDEN,†‡ JANE M. OLSON,*† JEFFREY R. SCHELLING,†‡ and JOHN R. SEDOR†‡ Departments of *Epidemiology and Biostatistics, and ‡Medicine, Case Western Reserve University, and †Rammelkamp Center for Research and Education, MetroHealth Medical Center, Cleveland, Ohio.

Abstract. Diabetic nephropathy (DN), a major cause of ESRD, ate in the final linkage analysis. To date, we have collected 212 is undoubtedly multifactorial and is caused by environmental sib pairs from 46 CA and 50 AA families. The average age of and genetic factors. To identify a genetic basis for DN suscep- diabetes onset was 46.8 yr versus 36.2 yr for CA and 39.5 yr tibility, we are collecting multiplex DN families in the Cauca- versus 40.2 yr for AA, in males versus females respectively. sian (CA) and African-American (AA) populations for whole Genotyping data were available for 106 sib pairs (43 CA, 63 genome scanning and candidate analysis. A candidate AA) from 27 CA (44% male probands) and 38 AA families gene search of diabetic sibs discordantly affected, concordantly (43% male probands). Average AA and CA sibship size was affected and concordantly unaffected for DN was performed 2.73. Singlepoint and multipoint linkage analyses indicate that with microsatellite markers in genomic regions suspected to marker D10S1654 on 10p is potentially linked to harbor nephropathy susceptibility loci. Regions examined were DN (CA only multipoint P ϭ 4 ϫ 10Ϫ3). Interestingly, the at human chromosome 10p,10q (orthologous to the rat renal majority of the linkage evidence derives from the CA sib pairs. susceptibility Rf-1 locus), and at NPHS1 (), CD2AP, We are now adding sib pairs and increasing marker density on Wilms tumor (WT1), and NPHS2 () loci. Linkage chromosome 10. We have excluded linkage with candidate analyses were conducted using model-free methods (SIBPAL, regions for nephrin, CD2AP, WT1, and podocin in this sample. S.A.G.E.) for AA, CA, and the combined sample. Allele fre- In conjunction with previous reports, our data support evidence quencies and the identity by descent sharing were estimated for a DN susceptibility locus on chromosome 10. separately for AA and CA, and race was included as a covari-

ESRD: a Common Complex Disease With a genesis is mediated by (11–14). Lastly, environmental Genetic Basis factors alone cannot adequately explain the excessive cluster- ESRD is a multifactorial disease with increasing worldwide ing of ESRD in families. incidence, due in part to the aging population (1). Progression Complex segregation of families in which DN was clustered, of diabetic nephropathy (DN) to ESRD can be effectively performed independently by two groups of investigators, dem- slowed by aggressively treating hyperglycemia and hyperten- onstrated a major genetic effect on susceptibility to DN. Im- sion, using either angiotensin converting enzyme inhibitors or peratore et al. (15) defined nephropathy as a urine to angiotensin receptor blockers. Although modifiable risk factors creatinine ratio Ն500 mg/g in 715 nuclear, Pima Indian fam- have been identified, epidemiologic data, primarily from Cau- ilies, and rejected models for no major gene effect and no casian and Pima Indian populations, indicate that DN is genet- transmission of a major effect (P ϭ 0.00001; P ϭ 0.003). In ically determined, with multiple genes regulating the pheno- contrast, they were unable to reject Mendelian transmission type (reviewed in reference 2–5). Several lines of evidence models (P ϭ 0.85), although the specific mode of inheritance suggest that ESRD pathogenesis is mediated by genes. First, could not be determined. Fogarty et al. (16) examined segre- there is significant evidence of familial aggregation for ESRD gation of a quantitative trait, urine albumin to creatinine ratio, (6–10). Second, animal models also suggest that ESRD patho- in 96 large multigenerational pedigrees ascertained for type 2 diabetes and found that a multifactorial mode of inheritance best fit the ratio of albumin to creatinine levels. Together, these Correspondence to Dr. Sudha Iyengar, Department of Epidemiology and segregation analyses support the hypothesis of a major genetic Biostatistics, Case Western Reserve University, R215 Rammelkamp Center for Research, 2500 MetroHealth Drive, Cleveland, OH 44109-4945. Phone: 216- effect on susceptibility to DN. 778-8484; Fax: 216-778-3280; E-mail: [email protected] Strategies that have identified the genes responsible for 1046-6673/1407-0195 monogenic renal diseases such as polycystic kidney disease Journal of the American Society of Nephrology (17–20) and Alport’s syndrome (21,22) are being applied to Copyright © 2003 by the American Society of Nephrology more common causes of progressive renal disease, such as DN, DOI: 10.1097/01.ASN.0000070078.66465.55 hypertensive nephrosclerosis (HN), and glomerulonephritis S196 Journal of the American Society of Nephrology J Am Soc Nephrol 14: S195–S201, 2003

(GN). Classical linkage analysis of extended families has met bility to progressive renal failure is independent of the etiology with moderate success in mapping genes for familial forms of of ESRD (8,33), suggesting that genes or loci linked to non- focal segmental glomerulosclerosis (FSGS) and IgA nephrop- diabetic causes of kidney disease should be assessed as candi- athy (IgAN). Familial FSGS has been linked to markers on date pathogenesis genes for DN. Whole genome linkage anal- chromosome 19q13 (23,24), 11q21–q22 (25), and 1q25–31 ysis in the fawn-hooded rat, a model of hypertension and (26). Mutations in the gene encoding ␣-actinin-4, an actin nephrosclerosis, identified two renal failure susceptibility filament cross-linking protein, were identified in three families genes, Rf-1 and Rf-2 (11). The region of human homology for with an autosomal dominant form of FSGS linked to the 19q13 Rf-1 was identified as the distal portion of chromosome 10q. locus (27), identifying yet another molecular basis for a com- Yu et al. (34) investigated if markers on human chromosome mon histologic phenotype. Gharavi et al. (28) used genome- 10 were linked with chronic renal failure in 129 African- wide linkage analysis of 30 multiplex IgAN kindreds and American nondiabetic sibling pairs concordant for ESRD. The identified a locus for IgAN on 6q22–23, with an autosomal investigators observed weak evidence for linkage with markers dominant model of transmission with incomplete penetrance. on 10p, but not in the Rf-1 region. The investigators obtained a lod score of 5.6, and observed that In vitro, animal and human data suggest that podocyte 60% of kindreds were linked to markers on . dysregulation can initiate and/or perpetuate progressive glo- Collection of extended family data are problematic for the merular scarring (35). Both transgenic animal models and most common causes of nephropathy, such as DN and HN, familial glomerulosclerosis have been linked to mutations in because the age of disease in probands is later in life and two novel and two previously known genes, which encode relatives are often deceased or unavailable for collection. Fur- podocyte that comprise components of the slit dia- thermore, basic assumptions of traditional linkage analysis for phragm. Although mesangial expansion is the classic lesion of Mendelian traits often cannot be applied to multifactorial dis- DN, recent work suggests that podocyte injury occurs early in eases. The nature of complex diseases precludes the use of the course of DN and filtration barrier dysfunction is a hall- specific genetic models in the linkage analysis, because the mark of the disease. NPHS1 (nephrin) and NPHS2 (podocin) mode of inheritance is often unknown. Locus heterogeneity, as were identified by positional cloning in families with congen- observed for FSGS above, as well as allelic (mutation) heter- ital nephrotic syndrome of the Finnish type (36), and families ogeneity, further constrains the usefulness of traditional link- with steroid-resistant nephrotic syndrome (37), respectively. age methods. New linkage methods have been developed that Nephrin, an Ig superfamily member, is thought to be a major use nuclear family structures, such as affected and discordant transmembrane component of the slit diaphragm (38–40). sibling pairs, which make no assumptions about the mode of Podocin, which is similar to stomatin family scaffold mole- inheritance of the disease (for reviews see references 29 and cules, has a single hairpin-like structure with both the N- 30). These novel, model-free methods have formed the basis of terminal and C-terminal domains in the cytosol (37). ACTN4, genetic investigations in common forms of nephropathy. In which encodes the actin cross-linking protein ␣-actinin-4 (41), aggregate, these data suggest that genes mediating ESRD was identified as a candidate gene on the 19q13 locus mapped pathogenesis are complex in etiology. from three families with FSGS. CD2 associated protein (CD2AP) was originally identified as an adapter protein that Candidate Loci for Diabetic Nephropathy interacts with the cytoplasmic domain of CD2, a T cell adhe- In an effort to reduce genetic heterogeneity, our group has sion protein but subsequently was found to cause progressive limited recruitment to families with DN. Association analyses glomerulosclerosis in mice lacking CD2AP (42). of candidate DN pathogenesis genes using case-control designs are common, but results often cannot be replicated. Only two Materials and Methods linkage analyses of families in which DN is clustered have Sib Pair Study Design been reported. Moczulski et al. (31) performed a linkage study In our ongoing project we are collecting sibling pairs who are using 66 type 1 diabetic Caucasian sib pairs who were discor- concordant for diabetes and renal disease (affected sib pairs, ASP), dant for DN. Chromosomal regions containing genes for ACE, based upon the rationale that genes predisposing to ESRD are more angiotensinogen, and angiotensin II type 1 receptor (AT1) likely to segregate in families with more than one affected sibling. To were examined. The investigators observed significant evi- distinguish between diabetes susceptibility loci and DN loci, we are dence of linkage at microsatellites near AT1. Follow-up anal- also collecting sibling pairs who are concordant for diabetes and yses demonstrated that DN was linked with a major nephrop- discordant for DN (discordant sib pairs, DSP). In diseases such as DN athy susceptibility locus in a 20-cM region, which included where the recurrence risk for diabetes among siblings is high, DSP AT1 (P ϭ 7.7 ϫ 10Ϫ5). Imperatore et al. (32) undertook a may provide as much information as ASP (43). Our selection criteria more comprehensive genome-wide survey in 98 diabetic sib- for ASP and DSP with type 2 diabetes and DN, and the strategy for data collection are described in a previously published methods paper ling pairs concordant for diabetes and nephropathy to identify (44). Briefly, the clinical characteristics (phenotype) of index cases DN susceptibility loci among the Pima Indians. They observed from dialysis clinics and family members (predominantly sibs) are suggestive evidence for linkage on 3, 7, 9 and ascertained with a questionnaire, from medical record review, and 20. from measurement of proteinuria, serum creatinine and HgbA1C. The Although our goal is to identify genes that regulate DN, general classification scheme for sib pairs with varying degrees of several epidemiologic studies indicate that inherited suscepti- albumin excretion is described in Figure 1. Diabetes is defined by J Am Soc Nephrol 14: S195–S201, 2003 Genetic Susceptibility to Diabetic Nephropathy S197

podocin (NPHS2) loci. Linkage analyses were conducted using mod- el-free methods (SIBPAL2, S.A.G.E.) for AA, CA, and the combined sample. Allele frequencies and the identity by descent sharing were estimated separately for AA and CA. Race was included as a covariate in the final linkage analysis. The protocol has been approved by Institutional Review Boards at MetroHealth Medical Center and University Hospitals of Cleveland.

Results Data Collection To date, we have obtained information on 1577 type 2 diabetic index cases (see Figure 2 and Table 1). Comparison of family history records from subjects with all causes of ESRD (n ϭ 2804) versus DN only (n ϭ 1577) revealed several trends (Table 1). First, there is a significant excess of females with a Figure 1. Classification of sib pairs for linkage analysis on the basis family history among all-cause ESRD and DN (P ϭ 8.5 ϫ Ϫ of albumin excretion. Model-free linkage analysis used three types of 10 9), although this gender difference is great in DN (P ϭ 3.4 sib pairs; concordantly affected, concordantly unaffected, and discor- ϫ 10Ϫ13). Second, there is no difference in the percentage of dant. Unaffected sibs were long-standing diabetics (diabetes mellitus African Americans (AA) and Caucasians (CA) with and with- duration Ͼ10 yr) who did not demonstrate evidence of incipient out a family history for all cause ESRD versus DN (with family nephropathy. In contrast, affected sibs were individuals who pro- history P ϭ 0.968, without family history P ϭ 0.091). We also gressed to overt nephropathy. compared the DN probands enrolled in the genetic portion of the study (n ϭ 229) to those who participated in the initial questionnaire (n ϭ 964), but did not meet study criteria (Table HgbA1C Ն7.0 mg/dl, fasting serum glucose Ն126 mg/dl, random 2). There were no significant differences in the average ages of glucose Ն200 mg/dl, or prevalent treatment with insulin or oral the two groups. Comparing index cases with a living diabetic hypoglycemic agents. DN is defined by diabetes duration Ն10 yr, sib versus the unenrolled population, we did observe an in- urine protein Ն500 mg/g creatinine, and background or proliferative crease in the percentage of AA compared with CA (P ϭ retinopathy is defined by ophthalmology records, or history of laser 0.023). The source of this difference is likely to be the decrease surgery. Index cases and affected sibs must meet diabetes and DN in the percentage of male CA with living diabetic sibs. criteria. Discordant sibs must have diabetes Ն10 yr, but no proteinuria We have completed the administration of questionnaires in (Ͻ30 ␮g albumin/g creatinine). 399 families and their 287 relatives (see Figure 2). Of these Database families, 96 have questionnaire data confirmed by medical record review, and have provided necessary blood and urine All phenotyping information generated from questionnaires, med- ical record reviews, and laboratory data are entered into a centralized specimens. Average AA and CA sibship size was 2.73. These ESRD database. The database design uses client-server architecture with a Microsoft Access server acting as a graphical front-end, and a high quality UNIX-based SQL server to store data at the back-end. Currently, data can be entered from four different forms that are linked to each other. To maintain patient confidentiality, all personal information, including the name and address of each study subject, is segregated from the rest of the phenotypic information in the initial data entry form. After entering a subject’s name and address, a unique identifier and bar code is assigned to all records pertaining to that individual. All subsequent data entry for that patient or relative(s) uses only the unique identifier, thereby ensuring continued confidentiality, as well as linking family members related to the proband. All pheno- typic data are entered via graphical dialog boxes with drop-down selection boxes. Each screen has integrated logic to perform internal data encoding and checks to detect common coding and consistency errors.

Candidate Gene and Linkage Analyses A candidate gene search of diabetic sibs discordantly affected, concordantly affected, and concordantly unaffected for DN was per- formed with microsatellite markers in regions suspected to harbor nephropathy susceptibility loci. Regions examined were at chromo- some 10p and 10q, orthologous to the rat renal susceptibility Rf-1 Figure 2. Recruitment update for the Cleveland ESRD population locus, NPHS1 (nephrin, 19q), Wilm’s tumor (WT1), CD2AP, and (updated from 44). S198 Journal of the American Society of Nephrology J Am Soc Nephrol 14: S195–S201, 2003

Table 1. Comparison of the family history in all ESRD versus diabetic nephropathy only

All Cause ESRD (n ϭ 2638) DN Only (n ϭ 1577)

Family history Yes No Yes No ESRD (n ϭ 802) (n ϭ 1729) (n ϭ 380) (n ϭ 921) Male (%) 45.6 52.9 40.8 49.9 Female (%) 54.5 47.1 59.2 50.1 Caucasian (%) 38.3 53.9 38.2 57.3 African American (%) 61.7 46.1 61.8 42.7

Table 2. DN probands enrolled in the study versus those not enrolleda

Index Cases with Living Diabetic Sib Diabetic Population Not (n ϭ 229) Enrolled (n ϭ 964)

Age (yr) 64.5 Ϯ 8.73 65.9 Ϯ 12.8 AA/CA/others (%) 52.4/44.9/2.6 45.1/52.2/2.7 Age AA 63.4 Ϯ 8.7 65.9 Ϯ 11.9 Age CA 65.7 Ϯ 8.6 66.1 Ϯ 13.4 % male 41.9 48.1 % male AA 40 43.6 % male CA 43.7 51.8

a All diabetic index cases (n ϭ 1,577) were not included in these estimates. AA, African American; CA, caucasian. families contain 212 sibs pairs, of which 59 and 72 meet ASP from the CA sib pairs. A second peak, with weaker evidence and DSP criteria, respectively. The remaining sibs pairs (n ϭ for linkage, is identified at the Rf-1 locus on 10q D10S1230). 81) fail to meet phenotype requirements, due primarily to sibs In addition, we have excluded linkage with candidate regions with microalbuminuria or diabetes duration Ͻ10 yr. Sixty for CD2AP, WT1, podocin, nephrin, and ACTN4 (Table 4). additional families are under evaluation. We also compared the age-at-diabetes-onset in probands and Discussion their diabetic sibs (Table 3). When analyzed as a group, ages of The linkage results demonstrate strong evidence for a DN diabetes onset are similar in sibs and probands. After stratifi- susceptibility locus on chromosome 10p in CA families. Yu et cation of the sibs by phenotype, DSP sibs are significantly al. (34) also showed linkage of two adjacent markers on 10p, older than the probands at the onset of diabetes, despite having D10S1435 and D10S249, to ESRD in sibling pairs with non- diabetes duration as long as the proband. diabetic causes of ESRD (P ϭ 0.035 pairwise, P ϭ 0.082 multipoint for D10S1435; P ϭ 0.074 pairwise, P ϭ 0.063 Candidate Gene Analysis multipoint for D10S249). In contrast to Yu et al. (34), we also Genotyping data are available for 106 sib pairs (43 CA, 63 observed a second smaller linkage peak at the Rf-1 locus. Our AA) from 25 CA (44% male probands) and 37 AA families results indicate the existence of one DN susceptibility locus of (43% male probands). Singlepoint and multipoint linkage anal- strong effect and a second locus of weak/moderate effect on yses indicate that markers on chromosome 10p demonstrate chromosome 10, consistent with emerging evidence from ge- suggestive evidence for linkage with a putative nephropathy netic linkage studies performed in other populations (46,47). susceptibility locus (Figure 3, P ϭ 0.03 for multipoint analysis) We have previously demonstrated that the sibling recurrence (45). Interestingly, the majority of the linkage evidence derives risk and sibling recurrence risk ratio are greater in CA versus

Table 3. Comparison of DM onset and duration in probands versus sibsa

n Age at DM Onset (yr) DM Duration (yr)

Proband 395 46.6 Ϯ 14.8 20.5 All diabetic sibs 141 50.1 Ϯ 12.7 15.8 Sibs with ASP 51 45.6 Ϯ 10.5 18.8 Sibs with DSP 54 56.6 Ϯ 12.7 18.1 Unclassified sibs 34 55.1 Ϯ 13.5 8.0

a ASP, affected sib pairs; DSP, discordant sib pairs. J Am Soc Nephrol 14: S195–S201, 2003 Genetic Susceptibility to Diabetic Nephropathy S199

Figure 3. Significant evidence of linkage to ESRD with markers on chromosome 10p in the CA population. The asymptotic P values corresponding to the linkage results obtained using the SIBPAL2 program in the Statistical Analysis for Genetic Epidemiology (S.A.G.E.) linkage package were plotted for Caucasian American (CA), African American (AA), and both races combined. Distance in cM on chromosome Ϫ 10 was plotted along the x axis. A function of P values obtained from the linkage analysis [ log10 (P value)], was plotted along the y axis.

Table 4. Lack of evidence of linkage for ESRD with Acknowledgments markers on chromosomes 1, 6, and 19 This work was supported by grants from the National Institutes of Health (DK54644, DK54178, DK38558, DK51472, DK02281, Marshfield a DK57329), Northeast Ohio chapter of the American Heart Associa- Characteristics Marker Map (cM) P Value tion, Central Ohio Diabetes Association, Kidney Foundation of Ohio, 1 Podocin (D1S215) 194.9 0.631 Leonard Rosenberg Foundation, Juvenile Diabetes Foundation, and 6 CD2AP (D6S1699) 73.1 0.868 Baxter Extramural Grant Program. 19 Nephrin, ACTN4 63.1 0.854 b (D19S417 ) References a P-values from Haseman-Elston linkage (regression) analysis in 1. Cooper L: USRDS 2001 Annual data report. Nephrol News SIBPAL2 corresponding to the marker given in parentheses next to Issues 15: 31, 34–35, 38, 2001 the candidate locus. 2. Iyengar SK, Schelling JR, Sedor JR: Approaches to understand- b Both nephrin and ACTN4 lie within 5 cM of D19S417. ing susceptibility to nephropathy: From genetics to genomics. Kidney Int 61[Suppl 1]: 61–67, 2002 3. Schelling JR, Zarif L, Sehgal A, Iyengar S, Sedor JR: Genetic AA (SK Iyengar et al., submitted). The results of our linkage susceptibility to end-stage renal disease. Curr Opin Nephrol Hypertens 8: 465–472, 1999 analyses confirm that CA sib pairs are a powerful tool to map 4. Krolewski AS, Ng DP, Canani LH, Warram JH: Genetics of susceptibility genes for DN. We anticipate that fine mapping of diabetic nephropathy: How far are we from finding susceptibility chromosome 10p and 10q in a CA subpopulation that includes genes? Adv Nephrol Necker Hosp 31: 295–315, 2001 additional sibs may yield specific candidate genes, which reg- 5. Freedman BI, Bowden DW, Rich SS, Appel RG: Genetic initi- ulate susceptibility for the development of DN and progression ation of hypertensive and diabetic nephropathy. Am J Hypertens to ESRD. 11: 251–257, 1998 S200 Journal of the American Society of Nephrology J Am Soc Nephrol 14: S195–S201, 2003

6. Buckalew VMJ, Freedman BI: Candidate gene analysis in Afri- losclerosis maps to chromosome 19q13. Kidney Int 53: 282–286, can-Americans with renal failure. Ren Fail 21: 403–407, 1999 1998 7. Ferguson R, Grim CE, Opgenorth TJ: A familial risk of chronic 24. Vats A, Nayak A, Ellis D, Randhawa PS, Finegold DN, Levinson renal failure among blacks on dialysis? J Clin Epidemiol 41: KL, Ferrell RE: Familial nephrotic syndrome: clinical spectrum and 1189–1196, 1988 linkage to chromosome 19q13. Kidney Int 57: 875–881, 2000 8. Freedman BI, Soucie JM, Stone SM, Pegram S: Familial clus- 25. Winn MP, Conlon PJ, Lynn KL, Howell DN, Gross DA, Rogala tering of end-stage renal disease in blacks with HIV-associated AR, Smith AH, Graham FL, Bembe M, Quarles LD, Pericak- nephropathy. Am J Kidney Dis 34: 254–258, 1999 Vance MA, Vance JM: Clinical and genetic heterogeneity in 9. Freedman BI: Familial aggregation of end-stage renal failure: familial focal segmental glomerulosclerosis. International Col- aetiological implications. Nephrol Dial Transplant 14: 295–297, laborative Group for the Study of Familial Focal Segmental 1999 Glomerulosclerosis. Kidney Int 55: 1241–1246, 1999 10. Seaquist ER, Goetz FC, Rich S, Barbosa J: Familial clustering of 26. Fuchshuber A, Jean G, Gribouval O, Gubler MC, Broyer M, diabetic kidney disease. Evidence for genetic susceptibility to Beckmann JS, Niaudet P, Antignac C: Mapping a gene (SRN1) diabetic nephropathy. N Engl J Med 320: 1161–1165, 1989 to chromosome 1q25-q31 in idiopathic nephrotic syndrome con- 11. Brown DM, Provoost AP, Daly MJ, Lander ES, Jacob HJ: Renal firms a distinct entity of autosomal recessive nephrosis. Hum Mol disease susceptibility and hypertension are under independent Genet 4: 2155–2158, 1995 genetic control in the fawn-hooded rat. Nat Genet 12: 44–51, 27. Kaplan JM, Kim SH, North KN, Rennke H, Correia LA, Tong 1996 HQ, Mathis BJ, Rodriguez-Perez JC, Allen PG, Beggs AH, 12. Provoost AP, Shiozawa M, Van Dokkum RP, Jacob HJ: Transfer Pollak MR: Mutations in ACTN4, encoding alpha-actinin-4, of the Rf-1 region from FHH onto the ACI background increases cause familial focal segmental glomerulosclerosis. Nat Genet 24: susceptibility to renal impairment. Physiol Genomics 8: 123– 251–256, 2000 129, 2002 28. Gharavi AG, Yan Y, Scolari F, Schena FP, Frasca GM, Ghiggeri ϩ 13. Zalups RK: The Os/ mouse: A genetic animal model of re- GM, Cooper K, Amoroso A, Viola BF, Battini G, Caridi G, duced renal mass. Am J Physiol 264: F53–F60, 1993 Canova C, Farhi A, Subramanian V, Nelson-Williams C, Wood- 14. Zheng F, Striker GE, Esposito C, Lupia E, Striker LJ: Strain ford S, Julian BA, Wyatt RJ, Lifton RP: IgA nephropathy, the differences rather than hyperglycemia determine the severity of most common cause of glomerulonephritis, is linked to 6q22–23. glomerulosclerosis in mice. Kidney Int 54: 1999–2007, 1998 Nat Genet 26: 354–357, 2000 15. Imperatore G, Knowler WC, Pettitt DJ, Kobes S, Bennett PH, 29. Lander ES, Schork NJ: Genetic dissection of complex traits Hanson RL: Segregation analysis of diabetic nephropathy in [published erratum appears in Science 1994 266(5184):353]. Pima Indians. Diabetes 49: 1049–1056, 2000 Science 265: 2037–2048, 1994 16. Fogarty DG, Hanna LS, Wantman M, Warram JH, Krolewski 30. Elston RC: The genetic dissection of multifactorial traits. Clin AS, Rich SS: Segregation analysis of urinary albumin excretion Exp Allergy 25[Suppl 2]: 103–106, 1995 in families with type 2 diabetes. Diabetes 49: 1057–1063, 2000 31. Moczulski DK, Rogus JJ, Antonellis A, Warram JH, Krolewski 17. Ariza M, Alvarez V, Marin R, Aguado S, Lopez-Larrea C, AS: Major susceptibility locus for nephropathy in type 1 diabetes Alvarez J, Menendez MJ, Coto E: A family with a milder form on chromosome 3q: Results of novel discordant sib-pair analysis. of adult dominant polycystic kidney disease not linked to the PKD1 (16p) or PKD2 (4q) genes. J Med Genet 34: 587–589, Diabetes 47: 1164–1169, 1998 1997 32. Imperatore G, Hanson RL, Pettitt DJ, Kobes S, Bennett PH, 18. Connors TD, Van Raay TJ, Petry LR, Klinger KW, Landes GM, Knowler WC: Sib-pair linkage analysis for susceptibility Burn TC: The cloning of a human ABC gene (ABC3) mapping genes for microvascular complications among Pima Indians to chromosome 16p13.3. Genomics 39: 231–234, 1997 with type 2 diabetes. Pima Diabetes Genes Group. Diabetes 19. Mochizuki T, Wu G, Hayashi T, Xenophontos SL, Veldhuisen B, 47: 821– 830, 1998 Saris JJ, Reynolds DM, Cai Y, Gabow PA, Pierides A, Kimber- 33. Freedman BI, Wilson CH, Spray BJ, Tuttle AB, Olorenshaw ling WJ, Breuning MH, Deltas CC, Peters DJ, Somlo S: PKD2, IM, Kammer GM: Familial clustering of end-stage renal dis- a gene for polycystic kidney disease that encodes an integral ease in blacks with lupus nephritis. Am J Kidney Dis 29: membrane protein. Science 272: 1339–1342, 1996 729–732, 1997 20. Germino GG, Weinstat-Saslow D, Himmelbauer H, Gillespie 34. Yu H, Sale M, Rich SS, Spray BJ, Roh BH, Bowden DW, GA, Somlo S, Wirth B, Barton N, Harris KL, Frischauf AM, Freedman BI: Evaluation of markers on human chromosome 10, Reeders ST: The gene for autosomal dominant polycystic kidney including the homologue of the rodent Rf-1 gene, for linkage to disease lies in a 750- kb CpG-rich region. Genomics 13: 144– ESRD in black patients. Am J Kidney Dis 33: 294–300, 1999 151, 1992 35. Kriz W: Podocyte is the major culprit accounting for the pro- 21. Martin P, Heiskari N, Pajari H, Gronhagen-Riska C, Kaariainen gression of chronic renal disease. Microsc Res Tech 57: 189, H, Koskimies O, Tryggvason K: Spectrum of COL4A5 muta- 195., 2002 tions in Finnish Alport syndrome patients. Hum Mutat 15: 579, 36. Kestila M, Lenkkeri U, Mannikko M, Lamerdin J, McCready P, 2000 Putaala H, Ruotsalainen V, Morita T, Nissinen M, Herva R, 22. Barker DF, Hostikka SL, Zhou J, Chow LT, Oliphant AR, Kashtan CE, Peltonen L, Holmberg C, Olsen A, Tryggvason K: Gerken SC, Gregory MC, Skolnick MH, Atkin CL, Tryggvason Positionally cloned gene for a novel glomerular protein–nephri- K: Identification of mutations in the COL4A5 collagen gene in n–is mutated in congenital nephrotic syndrome. Mol Cell 1: Alport syndrome. Science 248: 1224–1227, 1990 575–582, 1998 23. Mathis BJ, Kim SH, Calabrese K, Haas M, Seidman JG, Seidman 37. Boute N, Gribouval O, Roselli S, Benessy F, Lee H, Fuchshuber A, CE, Pollak MR: A locus for inherited focal segmental glomeru- Dahan K, Gubler MC, Niaudet P, Antignac C: NPHS2, encoding J Am Soc Nephrol 14: S195–S201, 2003 Genetic Susceptibility to Diabetic Nephropathy S201

the glomerular protein podocin, is mutated in autosomal recessive lacking CD2-associated protein. Science 286: 312–315, 1999 steroid-resistant nephrotic syndrome [published erratum appears in 43. Rogus JJ, Krolewski AS: Using discordant sib pairs to map loci Nat Genet 25: 125, 2000]. Nat Genet 24: 349–354, 2000 for qualitative traits with high sibling recurrence risk. Am J Hum 38. Tryggvason K: Unraveling the mechanisms of glomerular ultra- Genet 59: 1376–1381, 1996 filtration: Nephrin, a key component of the slit diaphragm. JAm 44. Covic AM, Iyengar SK, Olson JM, Sehgal AR, Constantiner M, Soc Nephrol 10: 2440–2445, 1999 Jedrey C, Kara M, Sabbagh E, Sedor JR, Schelling JR: A 39. Holthofer H, Ahola H, Solin ML, Wang S, Palmen T, Luimula P, family-based strategy to identify genes for diabetic nephropathy. Miettinen A, Kerjaschki D: Nephrin localizes at the podocyte Am J Kidney Dis 37: 638–647, 2001 filtration slit area and is characteristically spliced in the human 45. Lander E, Kruglyak L: Genetic dissection of complex traits: kidney. Am J Pathol 155: 1681–1687, 1999 guidelines for interpreting and reporting linkage results. Nat. 40. Holzman LB, St. John PL, Kovari IA, Verma R, Holthofer H, Genet 11: 241–247, 1995 Abrahamson DR: Nephrin localizes to the slit pore of the glo- 46. Freedman BI, Rich SS, Yu H, Roh BH, Bowden DW: Linkage merular epithelial cell. Kidney Int 56: 1481–1491, 1999 heterogeneity of end-stage renal disease on human chromosome 41. Kaplan JM, Kim SH, North KN, Rennke H, Correia LA, Tong HQ, 10. Kidney Int 62: 770–774, 2002 Mathis BJ, Rodriguez-Perez JC, Allen PG, Beggs AH, Pollak MR: 47. Hunt SC, Hasstedt SJ, Coon H, Camp NJ, Cawthon RM, Wu LL, Mutations in ACTN4, encoding alpha-actinin-4, cause familial focal Hopkins PN: Linkage of creatinine clearance to chromosome 10 segmental glomerulosclerosis. Nat Genet 24: 251–256, 2000 in Utah pedigrees replicates a locus for end-stage renal disease in 42. Shih NY, Li J, Karpitskii V, Nguyen A, Dustin ML, Kanagawa humans and renal failure in the fawn-hooded rat. Kidney Int 62: O, Miner JH, Shaw AS: Congenital nephrotic syndrome in mice 1143–1148, 2002