3.015 PLA2G6 POLYMORPHISMS AND HAPLOTYPES IN PREDICTING PARKINSON'S DISEASE RISK IN CHINESE POPULATION

S172

Wednesday, 14 December 2011 / Parkinsonism and Related Disorders 18S2 (2012) S161–S234

47,000 oligonucleotide probes. Results show that gene expression data from a selected set of informative genes can be used to classify PD from the neurologically healthy subjects with high agreement to clinical diagnosis. In conclusion, PD affects gene expression in blood, suggesting the potential for the development of a bloodbased gene expression test. 3.011 GENETIC VARIATION OF BDNF IN PARKINSON’S DISEASE K.M. Prakash, Y. Zhao, E.K. Tan. Department of Neurology, National Neuroscience Institute (SGH Campus), Singapore, Singapore Introduction: Experimental evidence suggests that brain-derived neurotrophic factor (BDNF) can increase neuronal growth and protect dopaminergic neurons in Parkinson disease (PD) animal models. Objective: We conduct a case control study to evaluate if genetic variation of BDNF can modulate risk of PD. Methods: PD cases diagnosed according to standardized criteria were recruited from a tertiary center. Age and gender matched controls were also included. Genotyping was carried out using the allelic discrimination analysis and sequencing was used to confirm the variant. Results: A total of 800 subjects comprising of 400 PD and 400 controls were included. 60% of the subjects were men and the median age of PD and controls was 65 years and 60 years. BDNF (rs11030104) (A/G) was found in 0.53 vs 0.48 of PD and controls (p = 0.045, OR 1.2, 95%CI 0.9–1.5). Conclusion: We found a significant association of the BDNF (rs11030104) variant with PD. How genetic variation in BDNF modulates neuronal growth and protect against dopaminergic neurons need to be addressed in further studies. 3.012 POLYGENIC DETERMINANTS OF PARKINSON’S DISEASE J. Guo, L. Wang, X. Yan, B. Tang. Department of Neurology, Xiangya Hospital, Central South University, Changsha, China Objective: A systemic evaluation of the independent and combined effects of Rep1, rs11931074, rs356165in SNCA gene; G2385R, R1628P in LRRK2 gene; rs242562, rs2435207 in MAPT gene and L444P in GBA gene on the sporadic PD patients and normal controls in Chinese population was done to discuss that multiple genes may interact with one another. Methods: Gene mutation analysis of SNCA, LRRK2, MAPT and GBA gene were carried out by polymerase chain reaction (PCR) combined with DNA direct sequencing in 1011 sporadic PD patients and 1016 normal controls to discuss the interactions of multiple genes. Results: Rep1, rs356165 in SNCA gene, G2385R in LRRK2 gene, L444P in GBA gene were associated with PD with adjustment of sex and age (P < 0.05) in the analysis of eight variants. PD risk was increased when Rep1 and rs11931074, rs356165, Rep1 and G2385R, rs356165 and G2385R were combined for association analysis. In addition, PD risk increased cumulatively with the increasing number of variants (OR for carrying 5 variants: 8.456). Conclusion: In this study, we first confirm that Rep1, rs356165 in SNCA gene and G2385R in LRRK2 gene have an independent and combined significant association with PD. L444P in GBA gene was speculated as an independent risk factor. SNPs in four genes have a cumulative effect with PD. 3.013 POPULATION STUDY OF (CAG/CAA)N ALLELE VARIATION IN THE TATA BOX-BINDING PROTEIN AND ATXN2-TBP MUTATIONS IN A FAMILIY WITH TYPICAL PARKINSON J. Laffita1 , L. Laguna Salvia2 , Y. Vazquez ´ Mojena2 , M. Verdecia 1 Ram´ırez2 , L.C. Velazquez-P ´ erez ´ , J.A. Valdevila Figueira3 , N. Canales Ochoa1 , T. Cruz Marino ˜ 2 . 1 NeuroBiology, 2 Molecular Neuro Biology, CIRAH, 3 Neurology, Clinical and Surgical Hospital Luc´ıa ´I˜ niquez Land´ın, Holguin, Cuba Objective: (1) To determine the variation of TBP gene in the healthy and SCA population. (2) To determine the frequency of inherited

Parkinson disease caused by expansion in the TBP gene. (3) To report a familiy with typical Parkinsonism with CAG expansions in TBP and ATXN2. Background: Parkinson disease is very frequent in Holguin province, altogether with expansion causing classical SCA2; however, the etiology is unknown. Methods: CAG expansions were determined in the TBP and ATXN2 gene of 288 chromosomes (48 unrelated controls and 240SCA). Results: CAG in the TBP gene was higher in SCA population than in control group (t-test, p < 0.05). When SCA sample was distributed by CAG expansions in the major SCA causing genes, expansions in the ATXN2 gene was associated (Chi2: 4.41, p < 0.05) with large normal TBP alleles (>38CAG/CAA). A typical PD familiy with anomalous CAG expansions in the TBP (38/42 repeats) and ATXN2 (36 repeats) was found. Somatic mosaicism was higher in affected cases in this typical PD familiy than in pure SCA2 population with the same CAG. A genetic interaction of both mutations caused an ALS-like phenotype in carriers of such mutations. Detailed clinical characterization in this familiy and electrophysiological studies are provided delineating differences between typical PD with imperfect mutations in the ATXN2 gene respecting pure SCA2. Conclusions: TBP and ATXN2 CAG expansions interact genetically. Somatic mosaicism is a rationale modifier of the PD onset and the phenotype of carriers of ATXN2 CAG expansions. 3.014 INTEGRATION OF CLINICAL AND MICROARRAY DATA WITH CANONICAL PARTIAL LEAST SQUARES FOR PREDICTION OF PARKINSON’S DISEASE 1 M.K. Karlsson1,2 , H.-M. Andersen1 , A. Lonneborg ¨ , S. Sæbø2 . 1 DiaGenic ASA, Oslo, 2 Department of Chemistry, Biotechnology, and ˚s, Norway Food Science, Norwegian University of Life Sciences, A The clinical diagnosis of Parkinson’s disease rests on the identification of the characteristics related to dopamine deficiency that are a consequence of progressive neuronal loss of the substantia nigra and other brain structures. However, nondopaminergic and non-motor symptoms are sometimes present before diagnosis and almost inevitably emerge with disease progression. Clinical data, such as gender, age, patient history, laboratory factors – which are the basis of day-to-day clinical decision support – are often underused to guide the diagnosis of Parkinson’s disease in the presence of microarray data. We apply a method called CPLS to incorporate clinical data as additional response variables to stabilize the extraction of latent components in Partial Least Squares (PLS) regression. Four different PLS regression methods for predicting Parkinson’s disease status were compared. A repeated random 10-fold CV routine was used to evaluate their classification performance. The CV-results show that including clinical information in the model building improves predictions and produces simpler and more stable models. We found that CPLS and the closely related CPPLS extract more information in the first few latent components than regular PLS does. 3.015 PLA2G6 POLYMORPHISMS AND HAPLOTYPES IN PREDICTING PARKINSON’S DISEASE RISK IN CHINESE POPULATION L. Zhanyun1 , G. Jifeng2 , T. Beisha2 . 1 Department of Neurology, Xiangya Hospital, 2 Department of Neurology, Xiangya Hospital, Central South University, Changsha, China Genetic variants of many genes have been reported to be associated with Parkinson’s disease (PD) in several populations except PLA2G6 gene. So we conducted a case-control study and genotyped four SNPs of PLA2G6 gene: rs4375, rs2267369, rs132985 and rs2284063 in 536 mainland Chinese PD patients and 567 healthy controls. Rs132985 and rs2284063 were selected from a genome-wide

Wednesday, 14 December 2011 / Parkinsonism and Related Disorders 18S2 (2012) S161–S234

association study, and the two SNPs were found associated strongly with nevus number, and a high melanocytic nevi count was the strongest known risk factor for cutaneous melanoma. Our findings showed that the PLA2G6 gene rs2267369 polymorphism was associated with PD (p = 0.022). There were no statistical differences in genotype or allele frequencies of rs4375, rs132985 and rs2284063 variants in PLA2G6 gene between sporadic PD (SPD) group and healthy control group in our study. Moreover, no significant LD was observed among the PLA2G6 SNPs, the sub-haplotypes C-C-T(1 1 0 1), C-C(1 0 0 1) and A-C(0 0 1 1) may increase the susceptibility of PD in Chinese population, and the sub-haplotypes A-C (0 0 1 1) may be informative of the relationship between PD and melanoma. 3.016 ASSOCIATION OF GWAS SINGLE-NUCLEOTIDE POLYMORPHISMS WITH PD IN CHINESE PATIENTS P.Y. Chen, F.H. Shang, W. Song, K. Chen, L.P. Pan, R. Huang, P.J. Li, Z.Z. Zheng. West China Hospital/Sichuan University, Chengdu Sichuan, China Background: Numerous single-nucleotide polymorphisms (SNPs) were reported to be associated with the risk of Parkinson disease (PD) through genome-wide association studies (GWAS). Ethnic specific effect is a very important factor for analyzing the GWAS association. The strength of these associations has not to be proved in Mainland China. Patients and Methods: Three hundred and sixty-two patients (M/F: 1.45/1) with PD from Department of Neurology of West China hospital, Sichuan University were included in the study. All medical data for each patient was recorded in our database. Three hundred unrelated healthy controls from the same region were included as control group. Ten SNPs at 8 loci (GAK rs1564282, DGKQ rs11248060, MAPT rs1052553 rs242557, GLIS1 rs797906, PARK16 rs11240572, LRP8 rs3820198, SNCA rs5174, PCGF3 2242235 and PLEKHM1 rs11012) were genotyped by GWAS. We analyzed the association between PD and above SNPs by Fisher’s exact test and Odds ratio (OR) together with 95%CI methods. A case-control methodology was used and 300 unrelated matched healthy controls were included. Results: The frequencies of minor alleles in GAK rs1564282 (P = 0.01, OR = 0.63, 95% CI = 0.45, 0.89) and DGKQ rs11248060 (P = 0.02, OR = 0.65, 95% CI = 0.46, 0.93) were found significantly lower in patients than that in controls. No significant differences in other above SNPs between patients with PD and controls were observed. In addition, no association of age, gender, initial symptoms and these SNPs were found. Conclusion: Our study suggests that GAK rs1564282 and DGKQ rs11248060 are associated with lower risk of PD in Mainland China. 3.017 ASSOCIATION OF MUTATIONS IN THE GLUCOCEREBROSIDASE GENE WITH PARKINSON DISEASE IN A KOREAN POPULATION Y.J. Kim1,2 , J.M. Choi1 , W.C. Kim3 , C.H. Lyoo4 , S.Y. Kang2 , P.H. Lee4 , J.S. Baik5 , S.B. Koh6 , H.-I. Ma2 , Y.H. Sohn4 , M.S. Lee4 . 1 Ilsong Institute of Life Science, Hallym University, Anyang-si, 2 Dept. of Neurology, Hallym University College of Medicine, Chooncheon, 3 Dept.of Neurology, CHA Bundang Medical Center, CHA University, Seongnam-si, 4 Dept. of Neurology, Yonsei University College of Medicine, 5 Dept. of Neurology, Inje University Ilsan-Paik Hospital, 6 Dept. of Neurology, Korea University College of Medicine, Seoul, Republic of Korea Background: Recent studies have shown an association between Parkinson disease (PD) and mutations in the gene encoding the lysosomal enzyme glucocerebrosidase (GBA), which is deficient in Gaucher disease. In Asian populations, 2 studies have been performed in all exons of GBA; one study in a Japanese population showed the highest odds ratio among all ethnic groups, whereas the other study in Chinese observed a trend of a higher frequency of GBA mutation in PD patients without statistical significance.

S173

To investigate whether there is an association between PD and mutations of GBA in a Korean population, we analyzed mutations of GBA and compared mutation frequencies between Korean PD patients and a control population. Methods: We analyzed mutations in GBA by sequencing exons of GBA in 277 Korean PD patients and 291 control subjects. All exons of GBA were sequenced in all PD cases and 100 control subjects. Exon 2 and exons 5–10, where mutations of GBA were found in our PD patients, were analyzed in an additional 191 control subjects. Results: Five different pathogenic heterozygous GBA mutations, including N188S, P201H, R257Q, S271G, and L444P, were identified in 9 PD cases (3.2%), whereas there were no GBA mutations found in control subjects (p < 0.01, OR 20.6, 95% CI 1.2–356.4). Age-at-onset of heterozygous GBA variants carriers were younger than that of noncarriers (48.6±11.9 versus 57.9±13.5, p < 0.05, Mann-Whitney test). Conclusions: Our results suggest that heterozygous mutations of GBA represent a risk factor for PD in Koreans. 3.018 PREVALENCE OF PARKIN (PARK2) MUTATIONS AND DELETIONS IN PARKINSON’S DISEASE PATIENTS FROM EXTREMADURE (SOUTHERN SPAIN) R. Ronco-Barrantes1 , I. Casado-Naranjo2 , F. Castellanos3 , C. Duran4 , 1 R. Gonzalez-Polo1 , M. Niso-Santano1 , R. Gomez-Sanchez ´ , 2 J.M. Bravo-San Pedro1 , E. Pizarro-Estrella1 , M. Gomez-Guti ´ errez ´ , 2 1 G. Gamez-Leyva ´ , J. Martin-Zurdo3 , R. Rodriguez-Lopez ´ , J.M. Fuentes1 . 1 Ciberned, Extremadura, 2 Neurology, Complejo Hospitalario de Caceres, Caceres, 3 Neurology, Hospital Virgen del Puerto, Plasencia, 4 Hospital Infanta Cristina, Badajoz, Spain Objective: To determine parkin (D394N and V380L) mutations and exons 2–6 deletions frequency in a well characterized cohort of familial and sporadic Parkinson’s disease patients from Extremadura (southern Spain). Methods: We have genotyped two previously reported parkin mutation (D394N and V380L) in 108 cases with PD and 191 controls enrolled in this study. Cases and controls were recruited without knowledge of family history of PD. Results: Parkin D394N mutation was present in nine of 108 cases (8.3%) and in three of 191 controls (1.6%). Parkin V380L mutation was heterozygous in 34 of 108 cases (31.5%) and in 42 of 191 controls (22%) and homozygous in 7 of 108 cases (6.5%) and none in controls. Our results indicate that there is an association between D394N but not with V380L parkin mutation in the development of PD in this poblation (p < 0.05). Deletions in exons 2 to 6 were found neither cases nor controls. Conclusions: Our results indicate that there might be an association between Parkinson’s disease and parkin’s D394N and V380L mutation. The true prevalence of these mutations in idiopathic and familial disease, their penetrance, and the phenotypic heterogeneity of associated cases has important implications for genetic screening in the clinical field. The funding was from Junta de Extremadura (GR10054), FUNDESALUD (PRIS10013). Instituto de Salud Carlos III (CP080010), Ministerio de Ciencia e Innovacion ´ (SAF2010–14993) and Fondo Social Europeo (FSE). 3.019 ASSOCIATION OF PARKINSON’S DISEASE WITH SIX SNPS LOCATED ON FOUR PARK S GENES IN NORTHERN HAN CHINESE POPULATION Y. Zhou1 , F. Li1 , X. Tian1 , L. Zhu1 , Y. Yang2 , X. Luo3 , Y. Ren3 , H. Pang1 . 1 Department of Forensic Serology, School of Forensic Medicine, China Medical University, Shenyang, 2 Department of Neurosurgery, Affiliated Xinhua Hospital of Dalian University, Dalian, 3 Department of Neurology, First Affiliated Hospital of China Medical University, Shenyang, China Parkinson’s disease (PD) has widely been reported to be associated with mutations in PARKs genes. To investigate potential genetic risk