- Email: [email protected]
The association between the LRRK2 R1628P variant and the risk of Parkinson’s disease in Asian: a meta-analysis
Neuroscience Letters 623 (2016) 22–27
Contents lists available at ScienceDirect
Neuroscience Letters journal homepage: www.elsevier.com/locate/neulet
Research paper
The association between the LRRK2 R1628P variant and the risk of Parkinson’s disease in Asian: a meta-analysis Xiaoli Wang a , Xiaona Zhang a , Li Xue b , Anmu Xie a,∗ a b
Department of Neurology, The Affiliated Hospital of Qingdao University, Qingdao, China Department of Rehabilitation, The Affiliated Hospital of Qingdao University, Qingdao, China
h i g h l i g h t s • We performed a meta-analysis to assess the R1628P polymorphism of LRRK2 in PD. • LRRK2 R1628P variants were increased risk of PD in allelic and dominant model. • The subgroup analysis also showed LRRK2 R1628P was an increased risk factor in PD among Chinese and non-Chinese Asians.
a r t i c l e
i n f o
Article history: Received 1 March 2016 Received in revised form 7 April 2016 Accepted 26 April 2016 Available online 28 April 2016 Keywords: Parkinson’s disease R1628P Polymorphism Meta-analysis
a b s t r a c t Many published case-control studies have investigated the association between Leucine-rich repeat kinase 2(LRRK2) R1628P and the susceptibility of Parkinson’s disease (PD). However, controversial results were obtained. Herein we performed this meta-analysis to assess the association between the LRRK2 R1628P variants and the risk of PD. Up to January of 2016, 5 databases were searched to identify casecontrol studies involving LRRK2 R1628P variants and PD risk. A total of 5736 PD patients and 4786 controls in 14 case-control studies were included in this meta-analysis. And STATA 12.0 statistics software was used to calculate available data from each study. The pooled odds ratios (OR) and 95% confidence interval (CI) were calculated to assess the genetic association between LRRK2 R1628P polymorphism and the risk of PD. The results indicated that LRRK2 R1628P variants were increased risk of PD when all studies were pooled (C vs. G OR = 1.983, 95% CI 1.640-2.399; GC + CC vs. GG OR = 1.971, 95% CI 1.625–2.391). Moreover, in the subgroup analysis by ethnicity, increased risks were identified among Chinese (GC + CC vs. GG, OR = 1.96, 95% CI 1.60–2.41) as well as in non-Chinese Asian races (GC + CC vs. GG, OR = 2.03, 95% CI 1.13–3.65). Therefore, our results suggest that the C allele, GC and CC genotype of LRRK2 R1628P variants contribute to the susceptibility of PD in Asian. © 2016 Elsevier Ireland Ltd. All rights reserved.
1. Introduction Parkinson’s disease (PD) is a progressive neurodegenerative disorder characterized by Lewy body formation and the loss of dopamine neurons in the substantia nigra (SN). Mechanisms underlying dopamine neuron degeneration in PD remains unknown. It is well known that genetic and environmental factors act together in the disease cascade [1,2]. To date, several candidate genes and susceptibility loci have been identified in both familial and sporadic cases of PD [3]. In particular, mutations within the Leucine-rich repeat kinase 2 (LRRK2) are considered to play a vital role in PD pathogenesis [4].
∗ Corresponding author. E-mail addresses: [email protected], [email protected] (A. Xie). http://dx.doi.org/10.1016/j.neulet.2016.04.056 0304-3940/© 2016 Elsevier Ireland Ltd. All rights reserved.
LRRK2 is mapped to 12p11.2-q13.1 and encodes a rather large (2527 amino acids) peptide containing multiple independent domains, and within these functional domains several pathogenic variations have been reported [5–7]. The most frequently detected were G2019S, G2385R, R1628P, S1647T, P755L, M1646T, and I2020T. Among them, the G2019S mutation in LRRK2 has been found in diverse populations worldwide, and G2385R in this gene but to date have been found exclusively in Asian populations [8]. In 2008, Ross et al. first reported that the R1628P was another common risk variant exclusively in ethic Asian populations [9]. Subsequently, several replication researches in case-control study were performed to investigate whether R1628P variant contributed to PD risk. In Chinese populations, studies in Liaoning, Fujian and Taiwan showed negative results [17,18,20], while the other studies showed the association between R1628P and the risk of PD [10,11,16,19,21]. In non-Chinese Asian populations, only one study
X. Wang et al. / Neuroscience Letters 623 (2016) 22–27
conducted in Thai showed positive results [22], but in the Malay, Singapore and Korean populations, this variant was very rare or absent, and did not appear to be a risk factor [9,12,13]. These conflicting or inconclusive results may be due to the restriction of sample size and ethnic diversity. We therefore carried out the present meta-analysis to provide more useful information about the relationship between LRRK2 R1628P variants and PD susceptibility. 2. Materials and methods 2.1. Literature search A comprehensive literature search was performed electronically on five databases of PubMed, Wed of science, the Chinese National Knowledge Infrastructure,VIP information database and Wanfang Medicine up to January 2016. The following terms and key words (“Parkinson’s disease” or “PD”) and (“LRRK2” or “R1628P”) and/or (“polymorphism” or “variant”) were used in publications searching and all English and Chinese language articles were included. When necessary; the references of the identified records were checked manually to find other potential eligible studies. 2.2. Inclusion and exclusion criteria The following criteria were used to select the eligible studies: (1) case-control studies about the association between LRRK2 R1628P and PD risk, (2) available genotype frequencies for evaluating an odds ratio (OR) and 95% confidence interval (95%CI), (3) fulltext publications written in English or Chinese, (4) the genotypes distribution in the control population were in Hardy-Weinberg equilibrium (HWE), (5) the PD patient diagnosed according to UK Parkinson’s Disease Society Brain Bank clinical diagnostic criteria. Studies were excluded if they were: (1) reviews and meta-analysis (2) lack of genotype data and (3) studies involving PD cell and animal models. 2.3. Data extraction Two investigators independently reviewed all publications. And disagreements were resolved by group discussion. The following information was retrieved: first author, year of publication, country and ethnicity of the study population, number of cases and controls, allele and genotype frequency, number of the early-onset PD (EOPD) and late-onset PD (LOPD) patients, sex ratio, age of onset, genotyping method, and p-value for HWE. Notably, We defined EOPD as age of onset at <50 years and LOPD as age at ≥50 years. 2.4. Statistical analysis The pooled OR with its 95% CI was used to evaluate the strength of association between the LRRK2 R1628P variants and PD risk. Owing to low frequency for these variations, we then used allele model (C vs. G) and dominant model (GC + CC vs. GG when C is the risk allele) to estimate the risks of the genotypes on PD. Heterogeneity across individual studies was determined by I2 statistical. We used a fixed effects rather than a random effects model when taking into account heterogeneity between multi studies (if I2 was above 50% the random model was chosen) [14]. Publication bias was assessed by Begg’s test and visual inspection of a funnel plot [15]. Moreover, subgroup analysis was performed according to ethnicity. To assess the stability of the results, sensitivity analysis was performed by omitting each individual study in turn from the total and reanalyzing the remainder. All statistical analyses were con-
23
ducted with STATA 12.0 software (Stata Corp, College Station, TX, USA) and p-values were two-tailed. 3. Results 3.1. Characteristics of eligible studies As is shown in flow diagram of study selection (Fig. 1), 14 casecontrol studies from 12 articles with a total of 5736 PD patients and 4786 controls were included in this meta-analysis [9–13,16–22]. To note, the paper by Ross et al. consisted of three study groups and each group was treated separately [9]. The characteristics of these studies are listed in Table 1. Of these studies, 5 were carried out in China mainland [10,11,16–18], 5 on Taiwan [9,19–21], 2 were conducted on Singapore [9,12], 1 on Korean [13], and 1 on Thailand [22]. In terms of the ethnicity, 10 studies were Chinese, 4 studies were non-Chinese. The genotype and allele distributions for each case-control study were listed in Table 2. 3.2. Quantitative synthesis Previous studies showed that the minor allele C frequency was associated with the increased risk of PD, so we first investigated the minor allele C distribution of LRRK2 R1628P in PD and control groups. There was tiny heterogeneity for the analysis of allele model and the fixed effects model was chosen (I2 = 0.9%). When all studies were pooled into the meta-analysis, the minor allele frequency C was detected as an increased risk of PD (Fig. 2). The pooled OR was 1.983 (C vs. G OR = 1.983 95% CI 1.640–2.399; p < 0.0001). Since the minor allele C was associated with the increased susceptibility of PD, we next investigated the dominant genotype (GC + CC) distribution in both groups (Fig. 3). A significant association with the increased risk of PD was also detected in the dominant model(GC + CC vs. GG OR = 1.971; 95% CI 1.625–2.391; p < 0.0001). 3.3. Stratified analyses Additionally, we further conducted a subgroup analysis by ethnicity. Considering conclusion may be impaired by a limited number of studies, only those ethnicities with more than three case-control studies were analyzed in the stratified analysis. Subgroup analysis showed that LRRK2 R1628P polymorphism was significantly associated with increased PD risk in both ethnicities, among Chinese (C vs. G OR = 1.974 95% CI 1.614-2.414; GC + CC vs. GG, OR = 1.96, 95% CI1.60-2.41, Figs. 2 and 3) as well as in non-Chinese(C vs. G OR = 2.064 95% CI 1.159-3.675;GC +CC vs. GG, OR = 2.03, 95% CI 1.13–3.65, Figs. 2 and 3). Table 3 summarizes the evidence of an association between LRRK2 R1628P polymorphism and the risk of PD. The table includes the minor allele C frequency and dominant genetic model (GC + CC vs. GG) by ethnicity. 3.4. Sensitivity analysis and publication bias By omitting one study each time, results showed no significantly alterations in pooled OR and 95% CI in both allelic and dominant model, which suggests the stability of this meta-analysis(Fig. 4). Funnel plots were conducted to evaluate the publication bias of the studies. The shape of the funnel plots seemed nearly symmetric by the visual inspection and P for publication bias = 0.976, indicating no publication bias in this meta-analysis (Fig. 5). 4. Discussion Parkinson’s disease has been the second common neurodegenerative disorder in the worldwide with the prevalence of 1% in the
24
X. Wang et al. / Neuroscience Letters 623 (2016) 22–27
Fig. 1. Flow diagram of study selection.
Table 1 Characteristics of studies. Author
Country
Genotype method
No. of case/control
No. of EOPD/LOPD
Age distribution in case/control
Male of case/control
HWE
Lu et al. [19] Ross (a) et al. [9] Ross (b) et al. [9] Ross (c) et al. [9] Tan et al. [12] Pulkes et al. [22] Yu et al. [22] Zhang et al. [10] Kim et al. [13] Zhou et al. [17] Cai et al. [18] Fu et al. [11] Wu et al. [20] Wuzhou et al. [21]
Taiwan Taiwan Taiwan Singapore Malay Thai China China Korea China China China Taiwan Taiwan
TaqMan RFLP RFLP TaqMan PCR RFLP PCR-RFLP PCR SNaPshot PCR-RFLP PCR-RFLP PCR PCR PCR
834/543 484/341 345/316 250/250 132/160 154/156 328/300 600/459 384/384 202/212 257/298 446/403 573/503 747/461
NA NA NA NA NA NA 138/190 150/450 NA NA 57/200 NA NA NA
65.7+11.8/51.9+18.4 62/57 NA NA NA 61.2+9.8/NA NA 55.9+11.16/53.38+13.39 62.7+9.5/63.3+9.0 62.68+10.29/62.87+10.44 58.1+11/58.2+11.8 60.82+11.20/59.21+9.3 62.1+11.5/59.4+12.9 63.2+7.8/70+6.8
491/224 NA NA NA NA NA 178/173 354/257 174/167 106/125 153/173 263/252 317/255 NA
YES YES YES YES YES YES YES YES YES YES YES YES YES YES
NA, not available; LOPD, late-onset PD; EOPD, early-onset PD; HWE, Hardy-Weinberg equilibrium.
population older than 60 years old [23]. However, the pathogenesis of PD has not yet been fully understood. Environmental risk factors and gene interactions are involved in the disease cascade [2]. To date, several candidate genes including LRRK2 have been reported conferring risk for PD. At least 10 established mutations in LRRK2 that have been found in multi-generation PD families including G2019S, G2385R,and R1628P. The G2019S variant was commonly found in North African Arab and Caucasian PD patients, but this locus was rare in Asians. While the G2385R and R1628P variants are probably risk factors associated with the Asian population [8]. In 2015, Xie et al. conducted a comprehensive meta-analysis based on 23 case-control studies and concluded that the LRRK2 G2385R variants contributed to the susceptibility of PD especially in Chi-
nese PD [24]. But the relationship between LRRK2 R1628P and PD risk is still contradictory. Thus, we performed a meta-analysis to investigate whether LRRK2 R1628P variants are associated with an increased risk of PD in Chinese ethnicity. LRRK2, located on chromosome 12q12, encodes 2527 amino acids and contains Armadillo (ARM), Ankyrin repeat (ANK), leucine rich repeat (LRR), Ras of complex proteins; GTPase (ROC), Cterminal of ROC (COR), mitogen activated kinase kinase kinase (MAP-KKK) and WD40 domains [25]. The LRRK2 protein is likely to contribute to the pathogenesis of PD in general. LRRK2 R1628P is located in the COR domain,in which a highly basic polar arginine (R) is substituted with a neutral nonpolar proline (P). Proline is considered an a-helix breaker that introduces a b-hairpin turn.
X. Wang et al. / Neuroscience Letters 623 (2016) 22–27
25
Table 2 Genotype and allele distributions for each case-control study. Case
Control
author
GG
GC
CC
G
C
GG
GC
CC
Lu et al. [19] Ross (a) et al. [9] Ross (b) et al. [9] Ross (c) et al. [9] Tan et al. [12] Pulkes et al. [22] Yu et al. [16] Zhang et al. [10] Kim et al. [13] Cai et al. [18] Zhou et al. [17] Fu et al. [11] Wu et al. [20] Wuzhou et al. [21]
772 452 324 237 129 139 311 557 381 241 200 399 539 689
60 31 21 13 3 14 17 40 3 16 2 47 34 56
2 1 0 0 0 1 0 3 0 0 0 0 0 2
1604 935 669 487 261 292 639 1154 765 498 402 845 1112 1434
64 33 21 13 3 16 17 46 3 16 2 47 34 60
523 330 302 244 154 151 294 448 383 288 207 382 481 443
20 11 14 6 6 5 6 11 1 10 5 21 22 18
0 0 0 0 0 0 0 0 0 0 0 0 0 0
G
C
1066 671 618 494 314 307 594 907 767 586 419 785 984 904
20 11 14 6 6 5 6 11 1 10 5 21 22 18
Fig. 2. Forest plot in fixed-effect model for the relationship of R1628P polymorphism and the risk of Parkinson’s disease under the allele model (C vs. G).
Table 3 Pooled odds ratios by ethnicity for allelic and genotypic model. Ethnicity
Contrast
Studies, n
OR(95%CI),p value
Heterogeneity p value,I2 (%)
Chinese
C vs. G GC + CC vs. GG
10 10
1.97(1.61-2.41), P < 0.00001 1.96(1.60-2.41), P < 0.00001
0.429,1.0 0.488, 0.0
NonChinese
C vs. G GC + CC vs. GG
4 4
2.06(1.16-3.67), P < 0.05 2.03(1.13–3.65), P < 0.05
0.261, 25.1 0.276, 22.4
CI = confidence interval, OR = odds ratio.
It is postulated that this substitution affects the dynamic interaction among the Roc, COR, and mitogen-activated protein kinase kinase kinase domains critical for activity, and may disrupt LRRK2 dimerization [9]. Therefore, the substitution causes a conformational change in the protein secondary structure, thus altering the function of the protein. The R1628P variant was first identified by Mata et al. [26]. And Ross et al. presented the first evidence to sup-
port LRRK2 R1628P as the second common genetic risk factor for PD in Taiwan and Singapore populations [9]. Subsequently, several replication researches in case-control study also showed R1628P contributed to PD risk in Chinese population [10,11]. But this association was not found in Indian, Malay, Japanese or Caucasian ethnic groups [4].
26
X. Wang et al. / Neuroscience Letters 623 (2016) 22–27
Fig. 3. Forest plot in fixed-effect model for the relationship of R1628P polymorphism and the risk of Parkinson disease under the dominant model (GC + CC vs. GG).
Fig. 4. Sensitivity analysis of the summary OR coefficients under the dominant model (GC + CC vs. GG).
In current meta-analysis, our results provided evidence supporting R1628P variant was associated with an increased risk of PD when all studies were pooled (C vs. G OR = 1.983 95% CI 1.6402.399; GC + CC vs. GG OR = 1.971 95% CI 1.625–2.391), which is obviously inconsistent with evidence published on Lancet by Ross OA et al. in 2011. In this article, 1376 Asian patients and 962 controls were assessed in the case-control study and the frequency of the minor allele C was 1.2%. In addition, participants in this study were from Taiwan, South Korean and Japan. Notably, a nonsignificant protective effect was investigated for this variant in the Taiwanese series (minor allele frequency3.8%, OR = 0.56, 95% CI 0.32–1.01) and the predicted risk effect for R1628P was noted in the South Korean series, particularly at the Seoul site (minor allele frequency 0.2%, OR 2.47, 95% CI 0.28–22.15). R1628P was not noted in the Japanese series. So we think the discrepancy may be
Fig. 5. Funnel plot for publication bias in selection of studies on the R1628P gene polymorphism (GC + CC vs. GG).
due to a combination of the low frequency of the C variant, natural sampling variation, and population heterogeneity. In fact, we have tried to include this paper into our meta-analysis to conduct a more precise and comprehensive result, but the data of the allele and genotype frequencies were insufficient and we failed to contact authors to inquire for further information. Hence, our result should be interpreted with caution and more eligible studies should also be included in our meta-analysis to validate this association. Additionally, subgroup analysis by ethnicity showed that the association between R1628P variant and PD susceptibility was significant in both Chinese and non-Chinese populations. But the variant C allele carriers had a more increasing risk of PD in non-Chinese than Chinese. This may be because that the number of non-Chinese case-control studies is dramatically less than Chinese
X. Wang et al. / Neuroscience Letters 623 (2016) 22–27
case-control studies. What’s more, immigration and assimilation of various regional ethnicities and tribes over the last thousand years may lead to genetic diversity and similarity in Asia. To the best of our knowledge, this is the most comprehensive meta-analysis evaluating the association between the LRRK2 R1628P variants and the risk of PD. Compared with previous metaanalysis [27], we added 4 more case-control studies to the current study conducted here. But several limitations of our investigation merit comment. Firstly, due to insufficient data, we failed to perform subgroup analysis based on age of onset in both LOPD and EOPD cases. Secondly, only published studies with genotype frequency were analyzed in our meta-analysis, thus, publication bias may have occurred. Finally, there is a great deal of genetic and populations heterogeneity among Chinese and non-Chinese, and the results should be cautiously interpreted. In conclusion, our meta-analysis showed that LRRK2 R1628P variants might be associated with increased risk of PD susceptibility in Asian. Considering the limitations above, further well-designed studies with larger sample sizes and ethnicities are warranted to validate this association. Conflict of interest The authors declare no conflicts of interest. Acknowledgments We would like to acknowledge the investigators for their helpful comments on this paper. This work was supported by grants from National Natural Science Foundation of China (81571225). References [1] K. Steece-Collier, E. Maries, Kordower JH,Etiology of Parkinson’s disease: genetics and environment revisited, Proc. Natl. Acad. Sci. U. S. A. 99 (2002) 13972–13974. [2] J. Gao, M.A. Nalls, M. Shi, et al., An exploratory analysis on gene-environment interactions for Parkinson disease, Neurobiol. Aging 33 (2012), 2528 e2521–2526. [3] A.B. Singleton, M.J. Farrer, V. Bonifati, The genetics of Parkinson’s disease: progress and therapeutic implications, Mov. Disord. 28 (2013) 14–23. [4] T. Peeraully, E.K. Tan, Genetic variants in sporadic Parkinson’s disease: east vs West, Parkinsonism Relat. Disord. 18 (Suppl. 1) (2012) S63–65. [5] M. Funayama, K. Hasegawa, H. Kowa, M. Saito, S. Tsuji, F. Obata, A new locus for Parkinson’s disease (PARK8) maps to chromosome 12p11.2-q13, Ann. Neurol. 51 (2002) 296–301. [6] O.A. Ross, A.I. Soto-Ortolaza, M.G. Heckman, et al., Association of LRRK2 exonic variants with susceptibility to Parkinson’s disease: a case-control study, Lancet Neurol. 10 (2011) 898–908.
27
[7] J.C. Dachsel, M.J. Farrer, LRRK2 Parkinson disease, Arch. Neurol. 67 (2010) 542–547. [8] S. Bardien, S. Lesage, A. Brice, J. Carr, Genetic characteristics of leucine-rich repeat kinase 2 (LRRK2) associated Parkinson’s disease, Parkinsonism Relat. Disord. 17 (2011) 501–508. [9] O.A. Ross, Y.R. Wu, M.C. Lee, et al., Analysis of Lrrk2 R1628 P as a risk factor for Parkinson’s disease, Ann. Neurol. 64 (2008) 88–92. [10] Z.J. Zhang, J.M. Burgunder, X.K. An, et al., LRRK2 R1628 P variant is a risk factor of parkinson’s disease among Han-Chinese from mainland China, Mov. Disord. 24 (2009) 1902–1905. [11] X.L. Fu, Y.F. Zheng, H. Hong, et al., LRRK2 G2385R and LRRK2 R1628 P increase risk of parkinson’s disease in a Han Chinese population from southern mainland China, Parkinsonism Relat. Disord. 19 (2013) 397–398. [12] E.K. Tan, M. Tang, L.C. Tan, et al., Lrrk2 R1628 P in non-Chinese Asian races, Ann. Neurol. 64 (2008) 472–473. [13] J.M. Kim, J.Y. Lee, H.J. Kim, et al., The LRRK2 G2385R variant is a risk factor for sporadic Parkinson’s disease in the Korean population, Parkinsonism Relat. Disord. 16 (2010) 85–88. [14] C.L. Ma, L. Su, J.J. Xie, J.X. Long, P. Wu, L. Gu, The prevalence and incidence of Parkinson’s disease in China: a systematic review and meta-analysis, J. Neural Transm. 121 (2014) 123–134. [15] J.L. Peters, A.J. Sutton, D.R. Jones, K.R. Abrams, L. Rushton, Comparison of two methods to detect publication bias in meta-analysis, JAMA 295 (2006) 676–680. [16] L. Yu, F. Hu, X. Zou, et al., LRRK2 R1628P contributes to Parkinson’s disease susceptibility in Chinese Han populations from mainland China, Brain Res. 1296 (2009) 113–116. [17] Y.S. Zhou, X.G. Luo, F.R. Li, et al., Association of Parkinson’s disease with six single nucleotide polymorphisms located in four PARK genes in the northern Han Chinese population, J. Clin. Neurosci. 19 (2012) 1011–1015. [18] J.P. Cai, Y. Lin, W.J. Chen, et al., Association between G2385R and R1628 P polymorphism of LRRK2 gene and sporadic Parkinson’s disease in a Han-Chinese population in south-eastern China, Neurol. Sci. 34 (2013) 2001–2006. [19] C.S. Lu, Y.H. Wu-Chou, M. van Doeselaar, et al., The LRRK2 Arg1628Pro variant is a risk factor for Parkinson’s disease in the Chinese population, Neurogenetics 9 (2008) 271–276. [20] Y.R. Wu, K.H. Chang, W.T. Chang, et al., Genetic variants of LRRK2 in taiwanese parkinson’s disease, PLoS One 8 (2013). [21] Y.H. Wu-Chou, Y.T. Chen, T.H. Yeh, et al., Genetic variants of SNCA and LRRK2 genes are associated with sporadic PD susceptibility: a replication study in a Taiwanese cohort, Parkinsonism Relat. Disord. 19 (2013) 251–255. [22] T. Pulkes, C. Papsing, S. Mahasirimongkol, M. Busabaratana, K. Kulkantrakorn, S. Tiamkao, Frequencies of LRRK2 variants in Thai patients with Parkinson’s disease: evidence for an R1628P founder, J. Neurol. Neurosur. Psychiatry 82 (2011) 1179–1180. [23] B.S. Connolly, A.E. Lang, Pharmacological treatment of Parkinson disease: a review, JAMA 11 (2014) 1670–1683. [24] C.L. Xie, J.L. Pan, W.W. Wang, et al., The association between the LRRK2 G2385R variant and the risk of Parkinson’s disease: a meta-analysis based on 23 case-control studies, Neurol. Sci. 35 (2014) 1495–1504. [25] C. Schulte, T. Gasser, Genetic basis of Parkinson’s disease: inheritance, penetrance, and expression, Appl. Clin. Genet. 4 (2011) 67–80. [26] I.F. Mata, J.M. Kachergus, J.P. Taylor, et al., Lrrk2 pathogenic substitutions in Parkinson’s disease, Neurogenetics 6 (2005) 171–177. [27] X. Wu, K.F. Tang, Y. Li, et al., Quantitative assessment of the effect of LRRK2 exonic variants on the risk of Parkinson’s disease: a meta-analysis, Parkinsonism Relat. Disord. 18 (July (6)) (2012) 722–730.
Contents lists available at ScienceDirect
Neuroscience Letters journal homepage: www.elsevier.com/locate/neulet
Research paper
The association between the LRRK2 R1628P variant and the risk of Parkinson’s disease in Asian: a meta-analysis Xiaoli Wang a , Xiaona Zhang a , Li Xue b , Anmu Xie a,∗ a b
Department of Neurology, The Affiliated Hospital of Qingdao University, Qingdao, China Department of Rehabilitation, The Affiliated Hospital of Qingdao University, Qingdao, China
h i g h l i g h t s • We performed a meta-analysis to assess the R1628P polymorphism of LRRK2 in PD. • LRRK2 R1628P variants were increased risk of PD in allelic and dominant model. • The subgroup analysis also showed LRRK2 R1628P was an increased risk factor in PD among Chinese and non-Chinese Asians.
a r t i c l e
i n f o
Article history: Received 1 March 2016 Received in revised form 7 April 2016 Accepted 26 April 2016 Available online 28 April 2016 Keywords: Parkinson’s disease R1628P Polymorphism Meta-analysis
a b s t r a c t Many published case-control studies have investigated the association between Leucine-rich repeat kinase 2(LRRK2) R1628P and the susceptibility of Parkinson’s disease (PD). However, controversial results were obtained. Herein we performed this meta-analysis to assess the association between the LRRK2 R1628P variants and the risk of PD. Up to January of 2016, 5 databases were searched to identify casecontrol studies involving LRRK2 R1628P variants and PD risk. A total of 5736 PD patients and 4786 controls in 14 case-control studies were included in this meta-analysis. And STATA 12.0 statistics software was used to calculate available data from each study. The pooled odds ratios (OR) and 95% confidence interval (CI) were calculated to assess the genetic association between LRRK2 R1628P polymorphism and the risk of PD. The results indicated that LRRK2 R1628P variants were increased risk of PD when all studies were pooled (C vs. G OR = 1.983, 95% CI 1.640-2.399; GC + CC vs. GG OR = 1.971, 95% CI 1.625–2.391). Moreover, in the subgroup analysis by ethnicity, increased risks were identified among Chinese (GC + CC vs. GG, OR = 1.96, 95% CI 1.60–2.41) as well as in non-Chinese Asian races (GC + CC vs. GG, OR = 2.03, 95% CI 1.13–3.65). Therefore, our results suggest that the C allele, GC and CC genotype of LRRK2 R1628P variants contribute to the susceptibility of PD in Asian. © 2016 Elsevier Ireland Ltd. All rights reserved.
1. Introduction Parkinson’s disease (PD) is a progressive neurodegenerative disorder characterized by Lewy body formation and the loss of dopamine neurons in the substantia nigra (SN). Mechanisms underlying dopamine neuron degeneration in PD remains unknown. It is well known that genetic and environmental factors act together in the disease cascade [1,2]. To date, several candidate genes and susceptibility loci have been identified in both familial and sporadic cases of PD [3]. In particular, mutations within the Leucine-rich repeat kinase 2 (LRRK2) are considered to play a vital role in PD pathogenesis [4].
∗ Corresponding author. E-mail addresses: [email protected], [email protected] (A. Xie). http://dx.doi.org/10.1016/j.neulet.2016.04.056 0304-3940/© 2016 Elsevier Ireland Ltd. All rights reserved.
LRRK2 is mapped to 12p11.2-q13.1 and encodes a rather large (2527 amino acids) peptide containing multiple independent domains, and within these functional domains several pathogenic variations have been reported [5–7]. The most frequently detected were G2019S, G2385R, R1628P, S1647T, P755L, M1646T, and I2020T. Among them, the G2019S mutation in LRRK2 has been found in diverse populations worldwide, and G2385R in this gene but to date have been found exclusively in Asian populations [8]. In 2008, Ross et al. first reported that the R1628P was another common risk variant exclusively in ethic Asian populations [9]. Subsequently, several replication researches in case-control study were performed to investigate whether R1628P variant contributed to PD risk. In Chinese populations, studies in Liaoning, Fujian and Taiwan showed negative results [17,18,20], while the other studies showed the association between R1628P and the risk of PD [10,11,16,19,21]. In non-Chinese Asian populations, only one study
X. Wang et al. / Neuroscience Letters 623 (2016) 22–27
conducted in Thai showed positive results [22], but in the Malay, Singapore and Korean populations, this variant was very rare or absent, and did not appear to be a risk factor [9,12,13]. These conflicting or inconclusive results may be due to the restriction of sample size and ethnic diversity. We therefore carried out the present meta-analysis to provide more useful information about the relationship between LRRK2 R1628P variants and PD susceptibility. 2. Materials and methods 2.1. Literature search A comprehensive literature search was performed electronically on five databases of PubMed, Wed of science, the Chinese National Knowledge Infrastructure,VIP information database and Wanfang Medicine up to January 2016. The following terms and key words (“Parkinson’s disease” or “PD”) and (“LRRK2” or “R1628P”) and/or (“polymorphism” or “variant”) were used in publications searching and all English and Chinese language articles were included. When necessary; the references of the identified records were checked manually to find other potential eligible studies. 2.2. Inclusion and exclusion criteria The following criteria were used to select the eligible studies: (1) case-control studies about the association between LRRK2 R1628P and PD risk, (2) available genotype frequencies for evaluating an odds ratio (OR) and 95% confidence interval (95%CI), (3) fulltext publications written in English or Chinese, (4) the genotypes distribution in the control population were in Hardy-Weinberg equilibrium (HWE), (5) the PD patient diagnosed according to UK Parkinson’s Disease Society Brain Bank clinical diagnostic criteria. Studies were excluded if they were: (1) reviews and meta-analysis (2) lack of genotype data and (3) studies involving PD cell and animal models. 2.3. Data extraction Two investigators independently reviewed all publications. And disagreements were resolved by group discussion. The following information was retrieved: first author, year of publication, country and ethnicity of the study population, number of cases and controls, allele and genotype frequency, number of the early-onset PD (EOPD) and late-onset PD (LOPD) patients, sex ratio, age of onset, genotyping method, and p-value for HWE. Notably, We defined EOPD as age of onset at <50 years and LOPD as age at ≥50 years. 2.4. Statistical analysis The pooled OR with its 95% CI was used to evaluate the strength of association between the LRRK2 R1628P variants and PD risk. Owing to low frequency for these variations, we then used allele model (C vs. G) and dominant model (GC + CC vs. GG when C is the risk allele) to estimate the risks of the genotypes on PD. Heterogeneity across individual studies was determined by I2 statistical. We used a fixed effects rather than a random effects model when taking into account heterogeneity between multi studies (if I2 was above 50% the random model was chosen) [14]. Publication bias was assessed by Begg’s test and visual inspection of a funnel plot [15]. Moreover, subgroup analysis was performed according to ethnicity. To assess the stability of the results, sensitivity analysis was performed by omitting each individual study in turn from the total and reanalyzing the remainder. All statistical analyses were con-
23
ducted with STATA 12.0 software (Stata Corp, College Station, TX, USA) and p-values were two-tailed. 3. Results 3.1. Characteristics of eligible studies As is shown in flow diagram of study selection (Fig. 1), 14 casecontrol studies from 12 articles with a total of 5736 PD patients and 4786 controls were included in this meta-analysis [9–13,16–22]. To note, the paper by Ross et al. consisted of three study groups and each group was treated separately [9]. The characteristics of these studies are listed in Table 1. Of these studies, 5 were carried out in China mainland [10,11,16–18], 5 on Taiwan [9,19–21], 2 were conducted on Singapore [9,12], 1 on Korean [13], and 1 on Thailand [22]. In terms of the ethnicity, 10 studies were Chinese, 4 studies were non-Chinese. The genotype and allele distributions for each case-control study were listed in Table 2. 3.2. Quantitative synthesis Previous studies showed that the minor allele C frequency was associated with the increased risk of PD, so we first investigated the minor allele C distribution of LRRK2 R1628P in PD and control groups. There was tiny heterogeneity for the analysis of allele model and the fixed effects model was chosen (I2 = 0.9%). When all studies were pooled into the meta-analysis, the minor allele frequency C was detected as an increased risk of PD (Fig. 2). The pooled OR was 1.983 (C vs. G OR = 1.983 95% CI 1.640–2.399; p < 0.0001). Since the minor allele C was associated with the increased susceptibility of PD, we next investigated the dominant genotype (GC + CC) distribution in both groups (Fig. 3). A significant association with the increased risk of PD was also detected in the dominant model(GC + CC vs. GG OR = 1.971; 95% CI 1.625–2.391; p < 0.0001). 3.3. Stratified analyses Additionally, we further conducted a subgroup analysis by ethnicity. Considering conclusion may be impaired by a limited number of studies, only those ethnicities with more than three case-control studies were analyzed in the stratified analysis. Subgroup analysis showed that LRRK2 R1628P polymorphism was significantly associated with increased PD risk in both ethnicities, among Chinese (C vs. G OR = 1.974 95% CI 1.614-2.414; GC + CC vs. GG, OR = 1.96, 95% CI1.60-2.41, Figs. 2 and 3) as well as in non-Chinese(C vs. G OR = 2.064 95% CI 1.159-3.675;GC +CC vs. GG, OR = 2.03, 95% CI 1.13–3.65, Figs. 2 and 3). Table 3 summarizes the evidence of an association between LRRK2 R1628P polymorphism and the risk of PD. The table includes the minor allele C frequency and dominant genetic model (GC + CC vs. GG) by ethnicity. 3.4. Sensitivity analysis and publication bias By omitting one study each time, results showed no significantly alterations in pooled OR and 95% CI in both allelic and dominant model, which suggests the stability of this meta-analysis(Fig. 4). Funnel plots were conducted to evaluate the publication bias of the studies. The shape of the funnel plots seemed nearly symmetric by the visual inspection and P for publication bias = 0.976, indicating no publication bias in this meta-analysis (Fig. 5). 4. Discussion Parkinson’s disease has been the second common neurodegenerative disorder in the worldwide with the prevalence of 1% in the
24
X. Wang et al. / Neuroscience Letters 623 (2016) 22–27
Fig. 1. Flow diagram of study selection.
Table 1 Characteristics of studies. Author
Country
Genotype method
No. of case/control
No. of EOPD/LOPD
Age distribution in case/control
Male of case/control
HWE
Lu et al. [19] Ross (a) et al. [9] Ross (b) et al. [9] Ross (c) et al. [9] Tan et al. [12] Pulkes et al. [22] Yu et al. [22] Zhang et al. [10] Kim et al. [13] Zhou et al. [17] Cai et al. [18] Fu et al. [11] Wu et al. [20] Wuzhou et al. [21]
Taiwan Taiwan Taiwan Singapore Malay Thai China China Korea China China China Taiwan Taiwan
TaqMan RFLP RFLP TaqMan PCR RFLP PCR-RFLP PCR SNaPshot PCR-RFLP PCR-RFLP PCR PCR PCR
834/543 484/341 345/316 250/250 132/160 154/156 328/300 600/459 384/384 202/212 257/298 446/403 573/503 747/461
NA NA NA NA NA NA 138/190 150/450 NA NA 57/200 NA NA NA
65.7+11.8/51.9+18.4 62/57 NA NA NA 61.2+9.8/NA NA 55.9+11.16/53.38+13.39 62.7+9.5/63.3+9.0 62.68+10.29/62.87+10.44 58.1+11/58.2+11.8 60.82+11.20/59.21+9.3 62.1+11.5/59.4+12.9 63.2+7.8/70+6.8
491/224 NA NA NA NA NA 178/173 354/257 174/167 106/125 153/173 263/252 317/255 NA
YES YES YES YES YES YES YES YES YES YES YES YES YES YES
NA, not available; LOPD, late-onset PD; EOPD, early-onset PD; HWE, Hardy-Weinberg equilibrium.
population older than 60 years old [23]. However, the pathogenesis of PD has not yet been fully understood. Environmental risk factors and gene interactions are involved in the disease cascade [2]. To date, several candidate genes including LRRK2 have been reported conferring risk for PD. At least 10 established mutations in LRRK2 that have been found in multi-generation PD families including G2019S, G2385R,and R1628P. The G2019S variant was commonly found in North African Arab and Caucasian PD patients, but this locus was rare in Asians. While the G2385R and R1628P variants are probably risk factors associated with the Asian population [8]. In 2015, Xie et al. conducted a comprehensive meta-analysis based on 23 case-control studies and concluded that the LRRK2 G2385R variants contributed to the susceptibility of PD especially in Chi-
nese PD [24]. But the relationship between LRRK2 R1628P and PD risk is still contradictory. Thus, we performed a meta-analysis to investigate whether LRRK2 R1628P variants are associated with an increased risk of PD in Chinese ethnicity. LRRK2, located on chromosome 12q12, encodes 2527 amino acids and contains Armadillo (ARM), Ankyrin repeat (ANK), leucine rich repeat (LRR), Ras of complex proteins; GTPase (ROC), Cterminal of ROC (COR), mitogen activated kinase kinase kinase (MAP-KKK) and WD40 domains [25]. The LRRK2 protein is likely to contribute to the pathogenesis of PD in general. LRRK2 R1628P is located in the COR domain,in which a highly basic polar arginine (R) is substituted with a neutral nonpolar proline (P). Proline is considered an a-helix breaker that introduces a b-hairpin turn.
X. Wang et al. / Neuroscience Letters 623 (2016) 22–27
25
Table 2 Genotype and allele distributions for each case-control study. Case
Control
author
GG
GC
CC
G
C
GG
GC
CC
Lu et al. [19] Ross (a) et al. [9] Ross (b) et al. [9] Ross (c) et al. [9] Tan et al. [12] Pulkes et al. [22] Yu et al. [16] Zhang et al. [10] Kim et al. [13] Cai et al. [18] Zhou et al. [17] Fu et al. [11] Wu et al. [20] Wuzhou et al. [21]
772 452 324 237 129 139 311 557 381 241 200 399 539 689
60 31 21 13 3 14 17 40 3 16 2 47 34 56
2 1 0 0 0 1 0 3 0 0 0 0 0 2
1604 935 669 487 261 292 639 1154 765 498 402 845 1112 1434
64 33 21 13 3 16 17 46 3 16 2 47 34 60
523 330 302 244 154 151 294 448 383 288 207 382 481 443
20 11 14 6 6 5 6 11 1 10 5 21 22 18
0 0 0 0 0 0 0 0 0 0 0 0 0 0
G
C
1066 671 618 494 314 307 594 907 767 586 419 785 984 904
20 11 14 6 6 5 6 11 1 10 5 21 22 18
Fig. 2. Forest plot in fixed-effect model for the relationship of R1628P polymorphism and the risk of Parkinson’s disease under the allele model (C vs. G).
Table 3 Pooled odds ratios by ethnicity for allelic and genotypic model. Ethnicity
Contrast
Studies, n
OR(95%CI),p value
Heterogeneity p value,I2 (%)
Chinese
C vs. G GC + CC vs. GG
10 10
1.97(1.61-2.41), P < 0.00001 1.96(1.60-2.41), P < 0.00001
0.429,1.0 0.488, 0.0
NonChinese
C vs. G GC + CC vs. GG
4 4
2.06(1.16-3.67), P < 0.05 2.03(1.13–3.65), P < 0.05
0.261, 25.1 0.276, 22.4
CI = confidence interval, OR = odds ratio.
It is postulated that this substitution affects the dynamic interaction among the Roc, COR, and mitogen-activated protein kinase kinase kinase domains critical for activity, and may disrupt LRRK2 dimerization [9]. Therefore, the substitution causes a conformational change in the protein secondary structure, thus altering the function of the protein. The R1628P variant was first identified by Mata et al. [26]. And Ross et al. presented the first evidence to sup-
port LRRK2 R1628P as the second common genetic risk factor for PD in Taiwan and Singapore populations [9]. Subsequently, several replication researches in case-control study also showed R1628P contributed to PD risk in Chinese population [10,11]. But this association was not found in Indian, Malay, Japanese or Caucasian ethnic groups [4].
26
X. Wang et al. / Neuroscience Letters 623 (2016) 22–27
Fig. 3. Forest plot in fixed-effect model for the relationship of R1628P polymorphism and the risk of Parkinson disease under the dominant model (GC + CC vs. GG).
Fig. 4. Sensitivity analysis of the summary OR coefficients under the dominant model (GC + CC vs. GG).
In current meta-analysis, our results provided evidence supporting R1628P variant was associated with an increased risk of PD when all studies were pooled (C vs. G OR = 1.983 95% CI 1.6402.399; GC + CC vs. GG OR = 1.971 95% CI 1.625–2.391), which is obviously inconsistent with evidence published on Lancet by Ross OA et al. in 2011. In this article, 1376 Asian patients and 962 controls were assessed in the case-control study and the frequency of the minor allele C was 1.2%. In addition, participants in this study were from Taiwan, South Korean and Japan. Notably, a nonsignificant protective effect was investigated for this variant in the Taiwanese series (minor allele frequency3.8%, OR = 0.56, 95% CI 0.32–1.01) and the predicted risk effect for R1628P was noted in the South Korean series, particularly at the Seoul site (minor allele frequency 0.2%, OR 2.47, 95% CI 0.28–22.15). R1628P was not noted in the Japanese series. So we think the discrepancy may be
Fig. 5. Funnel plot for publication bias in selection of studies on the R1628P gene polymorphism (GC + CC vs. GG).
due to a combination of the low frequency of the C variant, natural sampling variation, and population heterogeneity. In fact, we have tried to include this paper into our meta-analysis to conduct a more precise and comprehensive result, but the data of the allele and genotype frequencies were insufficient and we failed to contact authors to inquire for further information. Hence, our result should be interpreted with caution and more eligible studies should also be included in our meta-analysis to validate this association. Additionally, subgroup analysis by ethnicity showed that the association between R1628P variant and PD susceptibility was significant in both Chinese and non-Chinese populations. But the variant C allele carriers had a more increasing risk of PD in non-Chinese than Chinese. This may be because that the number of non-Chinese case-control studies is dramatically less than Chinese
X. Wang et al. / Neuroscience Letters 623 (2016) 22–27
case-control studies. What’s more, immigration and assimilation of various regional ethnicities and tribes over the last thousand years may lead to genetic diversity and similarity in Asia. To the best of our knowledge, this is the most comprehensive meta-analysis evaluating the association between the LRRK2 R1628P variants and the risk of PD. Compared with previous metaanalysis [27], we added 4 more case-control studies to the current study conducted here. But several limitations of our investigation merit comment. Firstly, due to insufficient data, we failed to perform subgroup analysis based on age of onset in both LOPD and EOPD cases. Secondly, only published studies with genotype frequency were analyzed in our meta-analysis, thus, publication bias may have occurred. Finally, there is a great deal of genetic and populations heterogeneity among Chinese and non-Chinese, and the results should be cautiously interpreted. In conclusion, our meta-analysis showed that LRRK2 R1628P variants might be associated with increased risk of PD susceptibility in Asian. Considering the limitations above, further well-designed studies with larger sample sizes and ethnicities are warranted to validate this association. Conflict of interest The authors declare no conflicts of interest. Acknowledgments We would like to acknowledge the investigators for their helpful comments on this paper. This work was supported by grants from National Natural Science Foundation of China (81571225). References [1] K. Steece-Collier, E. Maries, Kordower JH,Etiology of Parkinson’s disease: genetics and environment revisited, Proc. Natl. Acad. Sci. U. S. A. 99 (2002) 13972–13974. [2] J. Gao, M.A. Nalls, M. Shi, et al., An exploratory analysis on gene-environment interactions for Parkinson disease, Neurobiol. Aging 33 (2012), 2528 e2521–2526. [3] A.B. Singleton, M.J. Farrer, V. Bonifati, The genetics of Parkinson’s disease: progress and therapeutic implications, Mov. Disord. 28 (2013) 14–23. [4] T. Peeraully, E.K. Tan, Genetic variants in sporadic Parkinson’s disease: east vs West, Parkinsonism Relat. Disord. 18 (Suppl. 1) (2012) S63–65. [5] M. Funayama, K. Hasegawa, H. Kowa, M. Saito, S. Tsuji, F. Obata, A new locus for Parkinson’s disease (PARK8) maps to chromosome 12p11.2-q13, Ann. Neurol. 51 (2002) 296–301. [6] O.A. Ross, A.I. Soto-Ortolaza, M.G. Heckman, et al., Association of LRRK2 exonic variants with susceptibility to Parkinson’s disease: a case-control study, Lancet Neurol. 10 (2011) 898–908.
27
[7] J.C. Dachsel, M.J. Farrer, LRRK2 Parkinson disease, Arch. Neurol. 67 (2010) 542–547. [8] S. Bardien, S. Lesage, A. Brice, J. Carr, Genetic characteristics of leucine-rich repeat kinase 2 (LRRK2) associated Parkinson’s disease, Parkinsonism Relat. Disord. 17 (2011) 501–508. [9] O.A. Ross, Y.R. Wu, M.C. Lee, et al., Analysis of Lrrk2 R1628 P as a risk factor for Parkinson’s disease, Ann. Neurol. 64 (2008) 88–92. [10] Z.J. Zhang, J.M. Burgunder, X.K. An, et al., LRRK2 R1628 P variant is a risk factor of parkinson’s disease among Han-Chinese from mainland China, Mov. Disord. 24 (2009) 1902–1905. [11] X.L. Fu, Y.F. Zheng, H. Hong, et al., LRRK2 G2385R and LRRK2 R1628 P increase risk of parkinson’s disease in a Han Chinese population from southern mainland China, Parkinsonism Relat. Disord. 19 (2013) 397–398. [12] E.K. Tan, M. Tang, L.C. Tan, et al., Lrrk2 R1628 P in non-Chinese Asian races, Ann. Neurol. 64 (2008) 472–473. [13] J.M. Kim, J.Y. Lee, H.J. Kim, et al., The LRRK2 G2385R variant is a risk factor for sporadic Parkinson’s disease in the Korean population, Parkinsonism Relat. Disord. 16 (2010) 85–88. [14] C.L. Ma, L. Su, J.J. Xie, J.X. Long, P. Wu, L. Gu, The prevalence and incidence of Parkinson’s disease in China: a systematic review and meta-analysis, J. Neural Transm. 121 (2014) 123–134. [15] J.L. Peters, A.J. Sutton, D.R. Jones, K.R. Abrams, L. Rushton, Comparison of two methods to detect publication bias in meta-analysis, JAMA 295 (2006) 676–680. [16] L. Yu, F. Hu, X. Zou, et al., LRRK2 R1628P contributes to Parkinson’s disease susceptibility in Chinese Han populations from mainland China, Brain Res. 1296 (2009) 113–116. [17] Y.S. Zhou, X.G. Luo, F.R. Li, et al., Association of Parkinson’s disease with six single nucleotide polymorphisms located in four PARK genes in the northern Han Chinese population, J. Clin. Neurosci. 19 (2012) 1011–1015. [18] J.P. Cai, Y. Lin, W.J. Chen, et al., Association between G2385R and R1628 P polymorphism of LRRK2 gene and sporadic Parkinson’s disease in a Han-Chinese population in south-eastern China, Neurol. Sci. 34 (2013) 2001–2006. [19] C.S. Lu, Y.H. Wu-Chou, M. van Doeselaar, et al., The LRRK2 Arg1628Pro variant is a risk factor for Parkinson’s disease in the Chinese population, Neurogenetics 9 (2008) 271–276. [20] Y.R. Wu, K.H. Chang, W.T. Chang, et al., Genetic variants of LRRK2 in taiwanese parkinson’s disease, PLoS One 8 (2013). [21] Y.H. Wu-Chou, Y.T. Chen, T.H. Yeh, et al., Genetic variants of SNCA and LRRK2 genes are associated with sporadic PD susceptibility: a replication study in a Taiwanese cohort, Parkinsonism Relat. Disord. 19 (2013) 251–255. [22] T. Pulkes, C. Papsing, S. Mahasirimongkol, M. Busabaratana, K. Kulkantrakorn, S. Tiamkao, Frequencies of LRRK2 variants in Thai patients with Parkinson’s disease: evidence for an R1628P founder, J. Neurol. Neurosur. Psychiatry 82 (2011) 1179–1180. [23] B.S. Connolly, A.E. Lang, Pharmacological treatment of Parkinson disease: a review, JAMA 11 (2014) 1670–1683. [24] C.L. Xie, J.L. Pan, W.W. Wang, et al., The association between the LRRK2 G2385R variant and the risk of Parkinson’s disease: a meta-analysis based on 23 case-control studies, Neurol. Sci. 35 (2014) 1495–1504. [25] C. Schulte, T. Gasser, Genetic basis of Parkinson’s disease: inheritance, penetrance, and expression, Appl. Clin. Genet. 4 (2011) 67–80. [26] I.F. Mata, J.M. Kachergus, J.P. Taylor, et al., Lrrk2 pathogenic substitutions in Parkinson’s disease, Neurogenetics 6 (2005) 171–177. [27] X. Wu, K.F. Tang, Y. Li, et al., Quantitative assessment of the effect of LRRK2 exonic variants on the risk of Parkinson’s disease: a meta-analysis, Parkinsonism Relat. Disord. 18 (July (6)) (2012) 722–730.