Comparison between siblings and twins supports a role for modifier genes in ADPKD

Transcription

1 Kidney International, Vol. 66 (2004), pp GENETIC DISORDERS DEVELOPMENT Comparison between siblings and twins supports a role for modifier genes in ADPKD ALEXANDRE PERSU, MICHEL DUYME, YVES PIRSON, XOSÉ M. LENS, THIERRY MESSIAEN, MARTIJN H. BREUNING, DOMINIQUE CHAUVEAU, MICHELINE LEVY, JEAN-PIERRE GRÜNFELD, and OLIVIER DEVUYST Division of Nephrology, Université Catholique de Louvain Medical School, Brussels, Belgium; Laboratory of Statistics, Medical University, Montpellier, France; Division of Nephrology, Complexo Hospitalario Universitario de Santiago, Santiago de Compostela, Spain; Division of Nephrology, KUL Medical School, Leuven, Belgium; Department of Clinical Genetics, Leiden University Medical Center, Leiden, The Netherlands; Division of Nephrology and INSERM Unit 508, Hôpital Necker AP-HP, Paris, France; and INSERM Unit 535, Paris, France Comparison between siblings and twins supports a role for modifier genes in ADPKD. Background. Autosomal-dominant polycystic kidney disease (ADPKD) is characterized by intrafamilial variability in renal disease progression, which could result from a combination of environmental and genetic factors. Although a role for modifier genes has been evidenced in mouse models, direct evidence in ADPKD patients is lacking. The analysis of variability in affected siblings and monozygotic (MZ) twins would help evaluate the relative contribution of environment and genetic factors on renal disease progression in ADPKD. Methods. The difference in the age at end-stage renal disease (ESRD) and the intraclass correlation coefficient (ICC) were quantified in a large series of ADPKD siblings from western Europe and compared with the values obtained in a series of MZ ADPKD twins from the same geographic area. Results. Fifty-six sibships (including 129 patients) and nine pairs of MZ twins were included. The difference in the age at ESRD was significantly higher in siblings (6.9 ± 6.0 years, range 2 months to 23 years) than in MZ twins (2.1 ± 1.9 years, range 1 month to 6 years; P = 0.02). Furthermore, the intraclass correlation coefficient was significantly lower in siblings than in MZ twins (0.49 vs. 0.92, respectively; P = 0.003). The intrafamilial difference in the age at ESRD was not influenced by gender. Conclusion. These data substantiate the existence of a large intrafamilial variability in renal disease progression in ADPKD siblings. The fact that the variability in siblings is in a significant excess of that found in MZ twins strongly suggests that modifier genes account for a significant part of this variability. Autosomal-dominant polycystic kidney disease (ADPKD) is the most common inherited renal disease, Keywords: modifier genes, twins, siblings, renal disease progression. Received for publication March 16, 2004 and in revised form May 25, 2004 Accepted for publication June 8, 2004 C 2004 by the International Society of Nephrology accounting for 5% to 10% of patients requiring renal replacement therapy. ADPKD is characterized by the development of multiple cysts in both kidneys, leading to end-stage renal disease (ESRD) in about half of the patients by the age of 50 years [1]. The disease is due to mutations in two genes, PKD1, responsible for about 85% of cases, and PKD2, which accounts for the vast majority of the remaining cases [2]. The proteins encoded by PKD1 and PKD2 (polycystin-1 and polycystin-2, respectively) probably interact in the primary cilia of renal epithelial cells to participate in complex signal transduction pathways involved in mechanosensory function and renal tubular cell maturation and/or proliferation [2]. One of the most striking aspects of ADPKD is the existence of an important variability in renal disease progression [3]. Interfamilial variability in the age at ESRD is well established. It has been attributed to the distinct effects of mutations in PKD1 and PKD2 [4], the position of the mutation itself [5], and also environmental factors [3]. Intrafamilial variability in the age at ESRD has been observed in large ADPKD families [6], dizygotic (DZ) twins [7], affected parent-offspring pairs [8] and, most recently, families characterized for PKD1 [5] and PKD2 [9] mutations. Such variability could result from a combination of environmental and genetic factors, influencing all the steps leading to the disease [3]. Data obtained both in patients and mouse models suggest that cyst formation in ADPKD epithelia is triggered by a second hit (i.e., the occurrence of a somatic mutation in the allele unaffected by germline mutation) [10]. Micro-environmental or genetic factors determining the rate of second hit could thus modulate cystogenesis and, thereby, renal disease progression. Alternatively, modifier genes in ADPKD patients could affect the polycystins-mediated signal transduction pathways, play a role in cyst fluid 2132

2 Persu et al: Renal disease progression in ADPKD siblings and twins 2133 accumulation, or influence other mechanisms involved in the progression of the disease [3, 11]. Although a role for modifier genes has been evidenced in mouse models of ADPKD [12], studies in ADPKD patients are required to evaluate the importance of genetic versus environmental factors in renal disease progression. The analysis of intrafamilial variability in renal disease progression in ADPKD in multigenerational families [5, 6] and parent-offspring pairs [8] is not optimal to identify the possible contribution of modifier genes because differences in environment and way of life, and also distinctive medical care, may influence renal disease progression [13, 14]. The quantification of the variability in the age at ESRD in affected ADPKD siblings, who theoretically share a more similar environment and medical care than patients from multiple generations, and its comparison with the variability observed in genetically identical monozygotic (MZ) ADPKD twins, would be a strong element to evaluate the contribution of modifier genes influencing renal disease progression. In order to uncover the contribution of modifier genes, we compared the variability in the age at ESRD in a large series of ADPKD sibships with that observed in MZ ADPKD twins. METHODS ADPKD siblings During the three-year recruitment period (September 1998 September 2001), we looked for patients with ADPKD at ESRD (dialysis or transplantation) in four academic institutions: Saint-Luc Academic Hospital, Brussels (Belgium), U.Z. Gasthuisberg, Leuven (Belgium), Necker Hospital, Paris (France), and Complexo Hospitalario Universitario, Santiago (Spain). If these patients (index cases) had at least one affected sibling also at ESRD at that time, they were included in the analysis. Within sibships, all affected siblings at ESRD at the time of recruitment were taken into account (i.e., two or more within each family). Thus, all siblings included had reached ESRD before September 2001, and many of them before September The diagnosis of ADPKD was established on the basis of well-established criteria, including bilateral enlarged cystic kidneys and a family history suggestive of autosomal-dominant inheritance [1]. The age at ESRD was defined as the age at starting renal replacement therapy. Linkage analysis for PKD1 and PKD2 was performed in informative families as described previously [15], and known PKD2 families (N = 2) were excluded from the analysis. The study was approved by the ethics committee of each center; informed consent was obtained from all patients. Table 1. Intrafamilial variability in renal disease progression in ADPKD: Comparison between families of siblings and monozygotic twins Families of MZ twins siblings pairs P value N 56 9 (N of ADPKD patients) (129) (18) Mean difference of age at 5.2 ± birth years Mean difference in age at 6.9 ± 6.0 a 2.1 ± b ESRD years Intraclass correlation coefficient c 0.49 d 0.92 d e Abbreviations are: ADPKD, autosomal-dominant polycystic kidney disease; MZ, monozygotic; ESRD, end-stage renal disease. a After adjustment for difference of age for date of birth. b Mann-Whitney U test. c One-way ANOVA. d TheICC was significant (P < ) within each group. e F test. ADPKD monozygotic twins MZ twins affected with ADPKD were identified in the series of 32 twins recruited through a European Concerted Action [16]. Determination of zygosity using questionnaires, comparison of blood groups, HLA antigens and, when necessary, hypervariable DNA markers, allowed the identification of 20 pairs of MZ twins. Among these 20 MZ pairs, nine pairs had reached ESRD and were included in the study. The analysis of variability was thus performed in a series of nine MZ ADPKD twin pairs at ESRD. Statistical analyses The Gaussian distribution of age at ESRD was verified by the Shapiro-Wilk s W test. Similarity for renal disease progression in ADPKD siblings and MZ twins was assessed by the mean difference in the age at ESRD (adjusted for the difference of date of birth for siblings) and the intraclass coefficient (ICC) [17]. The Mann-Whitney U test was used to compare the intrasibship differences in the age at ESRD. The intraclass correlation coefficient (ICC) is a classic mean of evaluating the heritability of quantitative traits in twins or siblings. It is a measure of the similarity of observations (here, the age at ESRD) within sibships compared to that between sibships. As a ratio, it might vary from 0 to 1. The higher the similarity within sibships, the higher the ICC. The ICC was computed by a one-way analysis of variance (ANOVA) in which observations were grouped by family to provide estimates of resemblance based on the variability within and between families [17]. The degrees of similarity in MZ twin pairs and sibships were compared using the F test for comparison of ICCs [17]. The values are reported as mean ± SD (Tables 1 and 2). RESULTS Fifty-six consecutive families including two (N = 43), three (N = 10), four (N = 2), or five (N = 1) siblings at

3 2134 Persu et al: Renal disease progression in ADPKD siblings and twins Table 2. Variability in the age at ESRD within same-gender and opposite-gender pairs of siblings affected with ADPKD Same Opposite gender gender P value N of pairs 60 a 35 a Mean difference of age at 5.5 ± ± birth years Mean age at ESRD years 51.2 ± ± Mean difference of age at 7.3 ± 6 b 6.1 ± 6 b 0.35 ESRD years Intraclass correlation coefficient c a All the possible pairs within the 56 families with siblings were included in the analysis. b After adjustment for difference of age for date of birth. c The ICC was significant for the same gender (P = ) and opposite gender (P = 0.002) groups. ESRD (total: 129 patients), and nine pairs of MZ twins at ESRD were recruited (Table 1). The year of birth was similar in siblings and twins (median year: 1940 in both groups). The distribution of the age at ESRD was not significantly different from a Gaussian distribution in both groups. The mean age at ESRD of the 56 index patients was 51.2 ± 8.5 years (range years), similar to that reported in PKD1-linked patients in general [4, 5] and those linked to PKD1 in this study (51.0 ± 7.7 years, N = 23). The mean age at ESRD was 46.8 ± 7.4 years in the MZ twins (range years). The age at ESRD was not influenced by center. The differences in the age at ESRD within the 56 families of siblings were highly variable, ranging from two months to 23 years. The intrafamilial difference in the age at ESRD averaged 6.9 ± 6.0 years in the 56 families and 6.1 ± 4.4 in the 23 PKD1-linked sibships. The variance within siblings estimated by the ICC was 0.49 (95% CI ; P < ). The mean difference in the age at ESRD among the nine pairs of MZ twins was 2.1 ± 1.9 years (range 1 month to 6 years), whereas the ICC within this group was 0.92 (95% CI ; P < ). The difference in the age at ESRD in siblings (6.9 ± 6.0 years) was significantly higher than that observed in MZ twins (2.1 ± 1.9 years) (Mann-Whitney U test, P = 0.02). Furthermore, the variability within sibships, computed with the ANOVA method, was also significantly higher in siblings (N = 56, within variance: 37.99) than in MZ twins (N = 9, within variance: 4.98) (F = 8.67, P = 0.003), thus reflecting the increased ICC in the latter (Table 1). There was no significant effect of gender on the age at ESRD either in siblings (males 50.6 ± 8.8 years, N = 71 vs. females 52.0 ± 8.1 years, N = 58; P = 0.34) or MZ twins (males 45.0 ± 7.8 years, N = 12 vs. females 50.5 ± 5.4 years, N = 6; P = 0.14). These data were confirmed by a systematic analysis of the variability in the age at ESRD within same- and opposite-gender pairs, taking into account all the possible pairs within the 56 ADPKD families with siblings (Table 2). Considering that renal failure may progress more slowly in females than males with ADPKD, a selection bias in gender recruitment was not excluded. However, the elimination of the non-esrd patients was not reflected by a preferential loss of females or a significant change in the gender ratio in our series (data not shown). DISCUSSION This study is the first detailed assessment of intrafamilial variability in renal survival comparing siblings and MZ twins with ADPKD. Intrafamilial variability in renal disease progression is commonly observed in ADPKD, as recently shown in multigeneration PKD1 [5] and PKD2 [9] families. In particular, coexistence of young patients with ESRD and elderly patients with moderate renal failure was noted within such families [5, 9]. However, the potential impact of enhanced awareness about risk factors for renal progression (e.g., smoking, hypertension) and more efficient antihypertensive treatment could hamper the analysis of renal progression in multigenerational families [3, 13, 14]. The present study, which is restricted to siblings, who arguably share a similar environment and way of life at least until adulthood, and benefit from similar medical care [3, 14], thus confirms and extends the findings obtained in multigenerational families [4, 5, 6, 8, 9]. Most importantly, our study demonstrates a significant excess of variability in renal disease progression in ADPKD siblings versus MZ twins. The difference between the two groups is reflected both by the age at ESRD and by the intrasibship variability, as shown by the ICC (Table 1). Consistent with findings in PKD1 and PKD2 families [5, 9], the discordant variability among siblings and MZ twins represents a strong argument for modifier genes influencing renal disease progression in ADPKD. The difference in the age at ESRD within sibships theoretically results from the interplay between genetic and environmental factors. By contrast, in genetically identical MZ twins, the difference can only be explained by environmental factors and/or the rate of somatic mutations in PKD1 or PKD2, ifthe latter is a stochastic event that plays an early role in cystogenesis [10]. It must be said, however, that this reasoning could have limitations, since MZ twins are probably not genetically identical in all tissues and may have experienced differences in placental vascular supply and/or cell number allocation during very early post-zygotic events [18]. Nevertheless, since ADPKD siblings and MZ twins studied here share a roughly similar environment (date of birth, geographic origin), the excess of variability found within siblings should be mainly explained by genetic factors [19].

4 Persu et al: Renal disease progression in ADPKD siblings and twins 2135 It is increasingly recognized that the phenotype of Mendelian disorders is influenced by modifier genes (i.e., inherited genetic variations distinct from the disease locus) [20]. Such modifier genes have been evidenced in murine models of polycystic kidney disease [12]. Furthermore, recent studies have demonstrated a greater-thanadditive cystic phenotype in trans-heterozygous Pkd1 +/ : Pkd2 +/ mice versus singly heterozygous Pkd1 +/ or Pkd2 +/ mice that is best explained by a modifier effect exerted by the polycystins genes themselves [21]. A few studies have investigated candidate genes in ADPKD patients, including the polymorphisms of the angiotensinconverting enzyme (ACE) and endothelial nitric oxide synthase (ENOS or NOS3) genes, or mutations in the CF gene that encodes the camp-regulated CFTR chloride channel [5, 15, 22 26]. These studies yielded divergent results, which can be explained by small sample size, population admixture, and lack of consideration of genetic and environmental factors [27]. Two aspects of this study should be further commented. The first is the relative rarity of twins recruited. It is generally assumed that about 1 in 40 babies born in Europe will be a twin [28]. Considering the prevalence of ADPKD (1:400 to 1:1000), one should expect a prevalence of twins with ADPKD ranging approximately from 25 to 60 per million living births. Despite a multicentric effort involving academic centers centralizing the recruitment of ADPKD patients, only 32 pairs could be recruited and analyzed. Furthermore, although DZ twins should be twice as numerous as MZ twins in Caucasians, only 12 DZ twins versus 20 MZ twins were recruited. One explanation is that twins were not recruited just because they were not identified as such by the referring physicians. This potential bias is illustrated by the small size of many twin series, and by the fact that discordant DZ twins may classically escape inclusion [16, 19]. The point must be noted since DZ twins would be even more adequate control than siblings to evaluate the influence of environment [19]. Alternatively, the low twinning rate could reflect a discrete fertility problem in the ADPKD population. In contrast with the rate of MZ twinning, which is stable in humans, DZ twinning occurs at different rates that probably reflect geographic, seasonal, and familial variations in polyovulation. The spontaneous DZ twinning rate is an index of the combination of male plus female fertility [28, 29], and a lagged correlation between decreased DZ twinning rates and sperm counts has been suggested [30]. There is no evidence suggesting fertility problems in females with ADPKD [1]. In contrast, infertility associated with necrospermia [31] or axonemal 9+0 defects in the sperm flagella [32] has been documented in ADPKD patients. A role of polycystins in male fertility is further suggested by studies in model organisms, which showed that the Drosophila homolog of polycystin- 2 is required for sperm movement and male fertility [33], whereas the association of homologs of polycystin-2 and polycystin-1 could mediate the sperm acrosomal reaction in the sea urchin [34]. Another hypothesis arises from the cardiac and vascular abnormalities observed in several mouse models of ADPKD with targeted mutations in Pkd1 and Pkd2 [12]. Abnormalities in placentation or arteriovenous malformations have a critical influence on fetal development and neonatal outcome in multiple gestations [35] and, by extension, may also participate in the low prevalence of twinning in ADPKD. The second aspect is the lack of influence of gender on renal disease progression, as shown by a similar difference in the age at ESRD in same and opposite gender pairs (Table 2). Male gender has been associated with an earlier age at ESRD in the whole ADPKD population [36, 37]. However, recent studies suggest that this gender effect, which has not been found in PKD1 families [4, 5, 15], could be restricted to PKD2 families [9]. CONCLUSION We have characterized the intrafamilial variability in renal disease progression in a large series of ADPKD siblings and MZ twins. The sibling analysis and the excess variability compared to MZ twins support the existence of modifier genes influencing renal survival in ADPKD. These data provide a strong rationale for the design of a multicentric sib-pair study looking for modifier loci in human ADPKD. ACKNOWLEDGMENTS We are indebted to the ADPKD patients, their families, and the staff of the nephrology units involved. We are grateful to Profs. X. Jeunemaitre and A. Sessa for useful advice, and to Mrs. N. Lannoy and Prof. C Verellen for performing part of the linkage analyses. We thank Drs. J. Bagon, M. De Meyer, J. Vanwalleghem, as well as the members of the U.C.L. Nephrology group, Mr A. Herelixka and the Leuvense Samenwerking Groep voor Transplantatie, Drs. P. Serbelloni and F. Conte, and members of the Multicentric Twins European Study for information and DNA samples. These studies were supported in part by the Belgian agencies FNRS and FRSM, the ARC 00/05 260, and grants from the Fondation Saint-Luc and the Société denéphrologie. Reprint requests to Olivier Devuyst, Division of Nephrology, UCL Medical School, 10 Avenue Hippocrate, B-1200 Brussels, Belgium. devuyst@nefr.ucl.ac.be REFERENCES 1. PIRSON Y, CHAUVEAU D, GRÜNFELD JP: Autosomal dominant polycystic kidney disease, in Oxford Textbook of Clinical Nephrology, edited by Cameron JS, Davison AM, Grünfeld JP, Kerr DNS, Ritz E, Oxford, Oxford University Press, Oxford, 1998, pp IGARASHI P, SOMLO S: Genetics and pathogenesis of polycystic kidney disease. JAmSoc Nephrol 13: , PETERS DJM, BREUNING MH: Autosomal dominant polycystic kidney disease: Modification of disease progression. Lancet 358: , HATEBOER N, VAN DIJK MA, BOGDANOVA N, et al: Comparison of phenotypes of polycystic kidney disease types 1 and 2. Lancet 353: , 1999

5 2136 Persu et al: Renal disease progression in ADPKD siblings and twins 5. ROSSETTI S, BURTON S, STRMECKI L, et al: The position of the polycystic kidney disease 1 (PKD1) gene mutation correlates with the severity of renal disease. J Am Soc Nephrol 13: , MILUTINOVIC J, RUST PF, FIALKOW PJ, et al: Intrafamilial phenotypic expression of autosomal dominant polycystic kidney disease. Am J Kidney Dis 19: , PERAL B, ONG AC, SAN MILLAN JL, et al:astable, nonsense mutation associated with a case of infantile onset polycystic kidney disease 1 (PKD1). Hum Mol Genet 5: , GEBERTH S, RITZ E, ZEIER M, STIER E: Anticipation of age at renal death in autosomal dominant polycystic kidney disease (ADPKD)? Nephrol Dial Transplant 10: , MAGISTRONI R, HE N, WANG K, et al: Genotype-renal function correlation in type 2 autosomal dominant polycystic kidney disease. J Am Soc Nephrol 14: , PEI Y: A two-hit model of cystogenesis in autosomal dominant polycystic kidney disease? Trends Mol Med 7: , PERSU A, DEVUYST O: Transepithelial chloride secretion and cystogenesis in autosomal dominant polycystic kidney disease. Nephrol Dial Transplant 15: , GUAY-WOODFORD LM: Murine models of polycystic kidney disease: Molecular and therapeutic insights. Am J Physiol Renal Physiol 285:F1034 F1049, ROSCOE JM, BRISSENDEN JE, WILLIAMS EA, et al: Autosomal dominant polycystic kidney disease in Toronto. Kidney Int 44: , SCHRIER RW, MCFANN KK, JOHNSON AM: Epidemiological study of kidney survival in autosomal dominant polycystic kidney disease. Kidney Int 63: , PERSU A, STOENOIU MS, MESSIAEN T, et al:modifier effect of ENOS in autosomal dominant polycystic kidney disease. Hum Mol Genet 11: , LEVY M, DUYME M, SERBELLONI P, et al: Is progression of renal involvement similar in twins with ADPKD? Contrib Nephrol 115:65 71, DONNER A, ELIASZIW M: Methodology for inferences concerning familial correlations: A review. J Clin Epidemiol 44: , HALL JG: Twinning: Mechanisms and genetic implications. Curr Opin Genet Dev 6: , BOOMSMA D, BUSJAHN A, PELTONEN L: Classical twin studies and beyond. Nat Rev Genet 3: , DIPPLE KM, MCCABE ER: Modifier genes convert simple Mendelian disorders to complex traits. Mol Genet Metab 71:43 50, WU G, TIAN X, NISHIMURA S, et al: Trans-heterozygous Pkd1 and Pkd2 mutations modify expression of polycystic kidney disease. Hum Mol Genet 11: , BABOOLAL K, RAVINE D, DANIELS J, et al: Association of the angiotensin I converting enzyme gene deletion polymorphism with early onset of ESRF in PKD1 adult polycystic kidney disease. Kidney Int 52: , PERSU A, EL-KHATTABI O, MESSIAEN T, et al: Influence of ACE (I/D) G460W polymorphism of a-adducin in autosomal dominant polycystic kidney disease. Nephrol Dial Transplant 18: , O SULLIVAN DA, TORRES VE, GABOW PA, et al: Cystic fibrosis and the phenotypic expression of autosomal dominant polycystic kidney disease. Am J Kidney Dis 32: , PERSU A, DEVUYST O, LANNOY N, et al: CF gene and cystic fibrosis transmembrane conductance regulator expression in autosomal dominant polycystic kidney disease. J Am Soc Nephrol 11: , WALKER D, CONSUGAR M, SLEZAK J, et al: The ENOS polymorphism is not associated with severity of renal disease in polycystic kidney disease 1. Am J Kidney Dis 41:90 94, DEVUYST O, PERSU A, VO-CONG MT: Autosomal dominant polycystic kidney disease: Modifier genes and endothelial dysfunction. Nephrol Dial Transplant 18: , HALL JG: Twinning. Lancet 362: , TONG S, SHORT RV: Dizygotic twinning as a measure of human fertility. Hum Reprod 13:95 98, JAMES WH: Secular trends in monitors of reproductive hazard. Hum Reprod 12: , FANG S, BAKER HW: Male infertility and adult polycystic kidney disease are associated with necrospermia. Fertil Steril 79: , OKADA H, FUJIOKA H, TATSUMI N, et al: Assited reproduction for infertile patients with 9+0 immotile spermatozoa associated with autosomal dominant polycystic kidney disease. Hum Reprod 14: , WATNICK TJ, JIN Y, MATUNIS E, et al: A flagellar polycystin-2 homolog required for male fertility in Drosophila. Curr Biol 13: , NEILL AT, MOY GW, VACQUIER VD: Polycystin-2 associates with the polycystin-1 homolog, surej3, and localizes to the acrosomal region of the sea urchin spermatozoa. Mol Reprod Dev 67: , BENISCHKE K: The biology of the twinning process: How placentation influences outcome. Semin Perinatol 19: , GRETZ N, ZEIER M, GEBERTH S, et al: Isgender a determinant for evolution of renal failure? A study in autosomal dominant polycystic kidney disease. Am J Kidney Dis 14: , NEUGARTEN J, ACHARYA A, SILBIGER SR: Effect of gender on the progression of non-diabetic renal disease: A meta-analysis. JAm Soc Nephrol 11: , 2000