Common variants of LRRK2 are not associated with sporadic Parkinson's disease.код для вставкиСкачать
Common Variants of LRRK2 Are Not Associated with Sporadic Parkinson’s Disease Saskia Biskup, MD, PhD,1 Jakob C. Mueller, PhD,1–3 Manu Sharma, BSc,3 Peter Lichtner, PhD,1 Alexander Zimprich, MD,4 Daniela Berg, MD,5 Ullrich Wüllner, MD,6 Thomas Illig, PhD,7 Thomas Meitinger, MD,1,8 and Thomas Gasser, MD 3 Multiple mutations in the gene for the leucine-rich repeat kinase (LRRK2) cause autosomal dominant late-onset parkinsonism (PARK8). The Gly2019Ser mutation appears to be common in different populations. To investigate whether this novel gene influences the non-Mendelian sporadic form of Parkinson’s disease, we genotyped 121 single nucleotide polymorphisms comprehensively covering the entire LRRK2 gene region in a set of 340 Parkinson’s disease patients and 680 matched control subjects from Germany. No association could be demonstrated. We have therefore no evidence for the existence of a common variant in LRRK2 that has a strong influence on Parkinson’s disease risk. Ann Neurol 2005;58:905–908 We and others have recently identified six missense and one putative splice site mutation in the LRRK2 gene in families with autosomal dominant late-onset parkinsonism linked to chromosome 12 (PARK8).1,2 Several other mutations, including an apparently common mutation (Gly2019Ser) in the same gene, were described in sporadic and familial Parkinson’s disease (PD) cases.3–7 The LRRK2 gene consists of 51 exons, spans 144kb of genomic DNA, and encodes a putative protein kinase that belongs to the Roco protein family. Roco proteins include a protein kinase of the mitogenactivated protein kinase kinase kinase (MAPKKK) class, a small RasGTPase domain, and several other protein–protein interaction domains. The function of LRRK2 remains to be determined. Within affected carriers, postmortem diagnoses indicate brainstem dopaminergic degeneration accompanied by strikingly diverse pathologies including Lewy body PD in most cases, but also diffuse Lewy body disease, nigral degeneration without distinctive histopathology, and even progressive supranuclear palsy–like pathology in some. We, therefore, suggested that LRRK2 might be central to the pathogenesis of several major neurodegenerative disorders.1 Although LRRK2 mutations are much more common in clinically typical late-onset PD than mutations in any of the other PD genes identified so far, it is unclear whether common variants in the gene influence the risk for development of sporadic PD. In this study, we investigated the influence of LRRK2 in common sporadic forms of PD. The linkage disequilibrium (LD) structure of LRRK2 and its association with PD was determined in a whole-gene approach, as suggested by several other authors.8 As part of the ENCODE project (Enr123, 12q12, Chr12: 38626477-39126476) a high density of single nucleotide polymorphism (SNP) markers (1SNP/500bp) was available in the selected region (240.3kb in total, 50kb on each side of the LRRK2 gene). Of 447 SNPs, 310 had a minor allele frequency of 1% or greater in 30 parent-offspring trios of European origin Centre d’etude de polymorphisms humaines (CEPH). It has been suggested that populations genotyped in the HapMap project may From the 1Institute of Human Genetics, GSF National Research Center for Environment and Health, Neuherberg; 2Institute for Medical Statistics and Epidemiology and Institute for Psychiatry and Psychotherapy, Technical University Munich; 3Hertie Institute for Clinical Brain Research, Department of Neurodegenerative Disease, Tübingen, Germany; 4Neurological Department, Medical University of Vienna, Vienna, Austria; 5Department for Medical Genetics, Hertie Institute for Clinical Brain Research, Tübingen; 6 Department of Neurology, University of Bonn, Bonn; 7Institute of Epidemiology, GSF National Research Institute, Neuherberg; and 8 Institute of Human Genetics, Technical University Munich, Munich, Germany. Received Aug 1, 2005, and in revised form Aug 18. Accepted for publication Aug 22, 2005. This article includes supplementary materials available via the Internet at http://www.interscience.wiley.com/jpages/0364-5134/suppmat Published online Oct 27, 2005 in Wiley InterScience (www.interscience.wiley.com). DOI: 10.1002/ana.20664 Address correspondence to Dr Gasser, Hertie Institute for Clinical Brain Research, Department of Neurodegenerative Disease, HoppeSeyler Str. 3, Tübingen, Germany. E-mail: email@example.com © 2005 American Neurological Association Published by Wiley-Liss, Inc., through Wiley Subscription Services 905 serve as reference populations for the selection of tagging markers in association studies (The International HapMap Consortium 2003). We genotyped 81 tagging SNPs based on the ENCODE data (including 5 nonsynonymous and 5 synonymous coding SNPs), plus an additional 40 SNPs, filling gaps and enriching the interval around the common kinase domain mutation Gly2019Ser. Altogether, 121 SNPs were analyzed attempting to represent the complete DNA variation of the LRRK2 gene. Subjects and Methods Subjects PD was diagnosed clinically by specialists in movement disorders, and diagnoses were established according to the UK Parkinson’s Disease Society Brain Bank criteria.9 After obtaining informed consent, blood samples were drawn for DNA extraction. The PD patients were from two independent recruitments referred to as sample sets I and II. The 340 PD patients from clinical centers in Munich and Tübingen (sample set I) had a male:female ratio of 1.38 and showed a median age at onset of 52 years. These patients were screened for the published LRRK2 mutations.1–7 One of the patients was found to be heterozygous for the Gly2019Ser mutation (0.14%). Six hundred eighty healthy age- and sex-matched individuals from a population-based cohort panel (Cooperative Health Research in the Region of Augsburg, Germany, KORA S2000 project) were used as control subjects.10 A second sample (sample set II) mainly from clinical centers in Bonn and Rostock (322 PD patients) was used to attempt replication of initial results. Three hundred twenty-two additional healthy sex- and age-matched subjects of KORA were used as control subjects for the second set of PD patients. Tagging Single Nucleotide Polymorphism Selection The high-density genotype information from the HapMap/ ENCODE project was used to define bins of highly intercorrelated markers (r2 ⬎ 0.8). One to two tagging markers of each bin were selected to represent all markers in the bin with high confidence (r2 ⬎ 0.8).11 Genotyping A total of 121 SNPs (81 tagSNPs plus 40 additional SNPs) with an average distance between SNPs of approximately 2kb were identified. All information for the selected SNPs was extracted from the public dbSNP database. Genotyping of SNPs was done by primer extension of multiplex polymerase chain reaction products with the detection of the allele-specific extension products by matrixassociated laser desorption/ionization time of flight (MALDITOF; Sequenom, San Diego, CA) mass spectrometry. The frequencies of genotypes from successfully typed SNPs (average call rate, 98%) were in Hardy–Weinberg equilibrium. Statistical Analysis LD was estimated by D⬘ and r2 calculation as implemented in the program Haploview.12 Genotypic association of SNPs with PD phenotype was tested under different assumptions: 906 Annals of Neurology Vol 58 No 6 December 2005 no specific risk model (independent genotypic effects), recessive model with each of the alleles as risk alleles, and trend model assuming a dose effect of alleles. The association of all 7,260 two-locus combinations was tested in logistic regression models including all main and interaction effects.13 Multiple testing within each risk model was corrected for by a randomization procedure (1,000 permutations). In addition, a sliding window approach was performed to test for two- and three-locus haplotype associations.14 Results One large (approximately 130kb) LD block with high D⬘ values spans the whole 3⬘ region of LRRK2 from intron 11 (marker rs1907633) up to 18kb downstream of exon 51 (marker rs2098963) (Fig). This block defines a region where whole haplotypes are inherited with a low probability of being fractionated by recombination events. Ancient extended haplotypes are therefore most likely to be found within this region. Several marginal and overlapping LD blocks are distributed across the rest of the gene. As expected, pairwise r2 values are low among the tagSNPs. High r2 values appear only next to the marker-enriched region around exon 41. No overall significant genotype–phenotype association could be found in our sample set I (see Supplementary Table 1 available via the Internet at http:// www.interscience.wiley.com/jpages/0364-5134/ suppmat). The lowest single-marker p values were 0.0042 and 0.0048 for rs1388596 and rs1427273, respectively, but were shown to be insignificant after adjustment for multiple testing. Furthermore, the low p values could not be repeated in the replication set II (altogether, six SNPs have been tested in the replication sample set II: rs1388596, rs1427273, and four adjacent SNPs; see Supplementary Table 1). The lowest p value for the two-locus interaction model tests was 0.0002, which is also not significant given the high number of performed tests (see Supplementary Table 2 available via the Internet at http://www.interscience.wiley.com/jpages/0364-5134/suppmat). There was also no tablewise haplotype association. In addition, we could not detect consistent significant results with any SNP after stratification according to sex or age. There was also no age-of-onset effect with any SNP in the patient group. Discussion LRRK2 mutations are a common cause of autosomal dominant late-onset PD. Screening in familial and sporadic PD patients showed frequencies of the Gly2019Ser mutation up to 7% in familial and almost 1% in sporadic PD cases.3–7 However, even if other relatively common mutations remain to be identified in this large gene, these mutations still account for only a minority of all PD cases. Fig. Schematic representation of LRRK2 gene. (a) Genomic localization according to hg17. (b) Mutations and domains. (c) Exon/ intron structure. (d) Linkage disequilibrium structure: black cells indicate high pairwise r2 values; white cells indicate low r2 values. This study was designed to identify a potential role of LRRK2 in the cause of the more common sporadic form of PD. Rather than examining individual coding or noncoding polymorphisms, we chose a comprehensive approach by examining a large number of SNPs evenly distributed over the entire length of the gene including adjacent genomic regions and establishing its LD structure. This comprehensive study clearly argues against a significant contribution of common LRRK2 variants to the sporadic form of PD. No significant association was found with any of the tagSNPs, which capture practically all of the genetic diversity of the LRRK2 gene and its adjacent regulatory regions. The ENCODE genotype data supported the saturated tagSNP selection. Our sample size would have a power of 98% to detect a risk allele of 10% frequency with a relative risk of 1.7 (dose-effect model of alleles). These are reasonable assumptions for general complex diseases. In summary, we provide strong evidence against a common genetic variation in the LRRK2 gene with a strong effect on disease susceptibility. However, it does not exclude the overall involvement of the LRRK2 pathway in the pathogenesis of non-Mendelian PD. Genetic heterogeneity based on multiple rare variants within the LRRK2 would make it difficult to assess the contribution of LRRK2 for sporadic PD. LRRK2 includes one MAPKKK domain. MAPKKKs are part of MAPK cascades, which are involved in signal transduction, gene expression, and modification of diverse cytoskeletal proteins. MAPKKKs themselves are often activated by small GTPases and phosphorylate MAPKKs, which, in turn, perform phosphorylation of MAPKs. MAPKs are universal signal mediators of diverse extra- Biskup et al: Common LRRK2-Variants and PD 907 cellular signals. Scaffold proteins, as LRRK2 is suggested to be, bring specificity into MAPK signaling pathways. Interaction partners and substrates of LRRK2 remain to be determined to assign a specific role for LRRK2 in one of the pathways mentioned earlier. It is known from the literature that the different pathways are highly interconnected with each other.15 We can therefore speculate that members of the same or of other cross-linking pathways, or both, might have effects on the susceptibility to PD. The work was funded by the German National Genome Network (NGFN; German Ministry for Education and Research, 01GS0116, T.G., T.M.) and the Competence Network Parkinson’s disease (German Ministry for Education and Research, 01Gl0201, T.G., U.W.). The KORA group (Cooperative Health Research in the Region of Augsburg, Germany) consists of H.E. Wichmann, H. Löwel, C. Meisinger, T. Illig, R. Holle, and C. John and their coworkers who are responsible for the design and the conduct of the KORA studies. We gratefully acknowledge the participation of all patients and family members. References 1. Zimprich A, Biskup S, Leitner P, et al. Mutations in LRRK2 cause autosomal dominant Parkinsonism with pleomorphic pathology (Park8). Neuron 2004;44:601– 607. 2. Paisan-Ruiz C, Jain S, Ewans EW, et al. Cloning of the gene containing mutations that cause PARK8-linked Parkinson’s disease. Neuron 2004;44:595– 600. 3. Di Fonzo A, Rohe CF, Ferreira J, et al. A frequent LRRK2 gene mutation associated with autosomal dominant Parkinson’s disease. Lancet 2005;365:412– 415. 4. Gilks WP, Abou-Sleiman PM, Gandhi S, et al. A common LRRK2 mutation in idiopathic Parkinson’s disease. Lancet 2005;365:415– 416. 908 Annals of Neurology Vol 58 No 6 December 2005 5. Hernandez DG, Paisan-Ruiz C, McInerney-Leo A, et al. Clinical and positron emission tomography of Parkinson’s disease caused by LRRK2. Ann Neurol 2005;57:453– 456. 6. Kachergus J, Mata IF, Hulihan M, et al. Identification of a novel LRRK2 mutation linked to autosomal dominant Parkinsonism: evidence of a common founder across European populations. Am J Hum Genet 2005;76:672– 680. 7. Nichols WC, Pankratz N, Hernandez D, et al. Genetic screening for a single common LRRK2 mutation in familial Parkinson’s disease. Lancet 2005;365:410 – 412. 8. Neale BM, Sham PC. The future of association studies: genebased analysis and replication. Am J Hum Genet 2004;75: 353–362. 9. Hughes AJ, Daniel SE, Kilford L, Lees AJ. Accuracy of clinical diagnosis of idiopathic Parkinson’s disease: a clinicopathological study of 100 cases. J Neurol Neurosurg Psychiatry 1992;55:181–184. 10. Heid IM, Vollmert C, Hinney A, et al. Association of the 103I MC4R allele with decreased body mass in 7937 participants of two population-based surveys, 2005. J Med Genet 2005;42: e21. 11. Carlson CS, Eberle MA, Rieder MJ, et al. Selecting a maximally informative set of single-nucleotide polymorphisms for association analyses using linkage disequilibrium. Am J Hum Genet 2004;74:106 –120. 12. Barrett JC, Fry B, Maller J, Daly MJ. Haploview: analysis and visualization of LD and haplotype maps. Bioinformatics 2005; 15:263–265. 13. Cordell HJ, Clayton DG. A unified stepwise regression procedure for evaluating the relative effects of polymorphisms within a gene using case/control or family data: application to HLA in type 1 diabetes. Am J Hum Genet 2002;70:124 –141. 14. Zaykin DV, Westfall PH, Young SS, et al. Testing association of statistically inferred haplotypes with discrete and continuous traits in samples of unrelated individuals. Hum Hered 2002;53: 79 –91. 15. Chang L, Karin M. Mammalian MAP kinase signalling cascades. Nature 2001;410:37– 40.