Meta-Analysis Study on the Role of Bone-Derived Neurotrophic Factor Val66Met Polymorphism in Parkinson's Disease

Abstract

To evaluate a possible involvement of the brain-derived neurotrophic factor (BDNF) Val66Met polymorphism in susceptibility to Parkinson's disease (PD), we performed a meta-analysis of all studies on the topic published from 2002 to 2014. This article reviews and compares the data from two previous meta-analyses, including two studies not previously considered. We selected studies referring to a genetic comparison between PD patients and healthy controls, so 15 studies involving 3754 cases and 4026 controls were included in our meta-analysis. We found no association between the Val66Met polymorphism and the risk of developing PD in our overall analysis. The ethnicity-specific meta-analysis produced no significant association either. Our data do not support a major role for the BDNF Val66Met polymorphism in the pathogenesis of PD.

Introduction

Parkinson's disease (PD) is the most common neurodegenerative movement disorder, clinically characterized by resting tremor, rigidity, postural imbalance, and bradykinesia. The underlying pathological event in PD is the progressive loss of dopaminergic neurons in basal ganglia and their projections to the striatum and is of unknown etiology.¹

It has been proposed that the neurodegenerative process may result from an insufficient supply of neurotrophic factors, a family of proteins that aid in the differentiation, survival, and maintenance of specific neuronal populations, including the brain-derived neurotrophic factor (BDNF), which promotes dopamine uptake of midbrain dopaminergic neurons in vitro.² Decreased BDNF mRNA expression and protein content have been observed in the substantia nigra of PD patients,^3,4 making BDNF a logical candidate gene for susceptibility to PD. Over the past decade, the potential role of the BDNF gene in the pathogenesis of PD has been the object of much investigation, leading to inconsistent results. The most explored polymorphism in BDNF is a common non-conservative single-nucleotide polymorphism (SNP), Val66Met, located at nucleotide 196 (G/A) in the pro-protein of BDNF. This functional polymorphism produces an amino acid change at codon 66, resulting in abnormal intracellular distribution and low activity-dependent secretion of BDNF.^5,6 Numerous research groups investigated a probable association between this SNP and the risk of PD with conflicting results, and this discrepancy might be attributable to different factors, such as sample size, statistical power, clinical heterogeneity, or combination of them. To overcome the limitations of individual studies and resolve inconsistencies, this meta-analysis represents an important research strategy to combine different results and determine whether variations in effect among them are due to different analytical approaches, to variations in the samples used, or to random statistical fluctuation. Very recently, two meta-analyses have tested the pooled data on this issue and have come to different conclusions.^7,8 In an attempt to resolve these contradictory results, we performed the current meta-analysis including two new studies and comparing the data from the two previous meta-analyses.

Methods

Selection of studies

To identify appropriate studies for meta-analysis, we searched PubMed, Scopus, and the ISI Web of Knowledge using the keywords “Parkinson's disease,” “BDNF,” “G196A,” “Val66Met,” and their combinations in PubMed. We selected papers from 2002 to 2014 and without any language limitation. After reviewing the contents of this first selection, we also reviewed their reference lists to search for additional studies via Google Scholar. We followed the procedural steps indicated by Cochrane (www.cochrane-handbook.org).

In Fig. 1, we reported the flowchart that represents the inclusion and exclusion criteria of studies. We included only studies: (1) Showing comparative analyses of PD and healthy populations; (2) analyzing the association between G196A polymorphism of BDNF gene and PD; (3) in which raw data of allele or genotype frequencies or counts were available. The allele was G/A, and the genotype covered GG, AG, and AA. For studies published by the same group on the same gene and markers, only the first report was included.

FIG. 1.

The flowchart of the studies included in the meta-analysis. HWE, Hardy–Weinberg equilibrium.

Data extraction

Two reviewers independently extracted the following data from each study: Name of the first author, year of publication, ethnicity, and both in cases and controls sample size, mean age and percentage of female, and allele and genotype distribution.

Statistical analysis

To investigate the association of the G196A polymorphism with PD, the allele and genotype frequencies of the SNP between PD patients and healthy controls were compared and expressed in terms of odds ratio (OR) with the corresponding 95% confidence interval (95% CI). In each study, three ORs and their corresponding 95% CIs were calculated: (1) Dominant OR (GA+AA versus GG); (2) recessive OR (AA versus GA+GG); 3) OR A versus G; (4) OR GG versus AA. Between-study heterogeneity was assessed by the Q-statistic test and I² statistic. A p value <0.1 was considered statistically significant for the Q-statistic test. I² ranges between 0% and 100%; a value of 0% represents no heterogeneity, and larger values represent increasing heterogeneity. A fixed-effect model was performed to obtain the pooled OR, and if there was evidence of study heterogeneity, a random-effects model was applied.

Subgroup analysis was performed to evaluate the possible effect of ethnicity. We performed a separate meta-analysis combining the data of studies that categorized the PD sample based on cognitive impairment^9

–12 or specifically declared to exclude patients with neurological disease.^13,14 A possible effect on the OR of the difference in mean age and gender distribution between PD and control groups was investigated by performing a meta regression analysis. Sensitivity analysis was performed excluding studies that considered patients with PD family history. Publication bias and small-study bias were evaluated with a funnel plot of the data and with the Egger test. Temporal effect was also estimated with a cumulative meta-analysis. A fixed continuity correction was added when a study contained a zero cell. Statistical significance was defined as p<0.05. All statistical analyses were performed using STATA v. 10.1 (Stata Corp, College Station, TX).

Results

Description of studies

By the initial research, a total of 15 studies were eligible for inclusion in our analysis as described in a flowchart (Fig. 1). Characteristics of studies included in meta-analysis are reported in Table 1. Important articles have been excluded from our analysis for different reasons: Three studies were not included because the genotype distributions of controls was not in Hardy–Weinberg equilibrium (HWE).^15
–17 We left out the Foltynie et al. studies^18,19 because they did not report a healthy control group for comparison. Finally, the Karamohamed et al. study²⁰ was excluded because it did not report the genotype frequency of healthy controls.

Table 1.

Characteristics of the Studies Included in the Meta-Analysis

		PD patients				Controls
Reference	n	Gender (%F)	Mean age	Family history	n	Gender (%F)	Mean age	HWE p	Country
Hong et al. (2003)²¹	107	28	68.2±10.4	Negative	103	33	66.3±14.2	0.5	Asian
Hakanasson et al. (2003)²⁶	257	—	—	Negative	306	—	—	0.4	Caucasian
Masaki et al. (2003)²⁵	291	—	65.7±8.99	Negative	291	—	65.6±9.21	0.66	Asian
Nishimura et al. (2005)²²	327	60.5	66.4±9.9	Negative	275	61	68.4±10.5	0.1	Asian
Saarela et al. (2006)²⁷	52	52	—	Negative	101	55	—	0.09	Caucasian
Chen et al. (2007)²³	356	45.8	69.2±9.1	Negative	325	48	62.6±10.4	0.7	Asian
Guerini et al. (2009)¹¹	294	45.6	69.5 (33–86)	Positive 19%	233	65.7	61.5 (26–86)	0.9	Caucasian
Benitez et al. (2010)²⁹	100	49	60.1±12.6	Negative	136	53	62.4±9.5	0.4	American
Gao et al. (2010)¹⁰	193	42	66.1±10.4	Negative	300	37.6	61.7±12.2	0.9	Caucasian (except 3 patients)
Lin et al. (2011)¹⁴	442	56.9	57.9±11.7	Positive 5%	286	44.3	59.2±14.6	0.8	Asian
Karakasis et al. (2011)¹²	184	—	63.9±0.7	Negative	113	—	71.6±0.5	0.9	Caucasian
Chen et al. (2011)¹³	266	35	65.5±10	Negative	400	59	64.5±5.7	0.7	Asian
Liu et al. (2012)²⁴	464	40.1	66.7±10.3	Positive 2.4%	549	41.2	70±5.9	0.7	Asian
Svetel et al. (2013)²⁸	177	34.5	58.9±10.9	Positive 18.6%	366	—	—	0.4	Caucasian
Bialecka et al. (2013)⁹	244	50.8	64.23±9.42	Negative	242	53.2	64.95±9.41	0.04	Caucasian

HWE, Hardy–Weinberg equilibrium.

In our meta-analysis, a pooled total of 3754 PD patients and 4026 healthy controls was analyzed. The mean age ranged from 57.9¹⁴ to 69.5 years¹¹ in PD patients, and from 59.2¹⁴ to 71.6 years¹² in controls. The percentage of women in PD groups varied from a minimum of 28%²¹ to a maximum of 60.5%²² and in control groups from 33%²¹ to 65.7%.¹¹ Seven studies^{13,14,21

–25} were conducted on Asian subjects, six studies^{9,11,12,26
–28} were conducted on Caucasian population, one study²⁹ was conducted on American population, and one had Caucasian background except three patients.¹⁰ Genotype and allele distribution of the G196A polymorphism of the BDNF gene from each study are reported in Table 2. All studies observed HWE. Four studies^11,14,24,28 also included in the PD group patients with a PD family history, so we repeated the analyses excluding them to verify the robustness of our results.

Table 2.

Distribution of BDNF Genotypes, and Their Allelic Frequency, for PD Patients and Controls

	PD				Controls				HWE	G/A	G/A
References	GG	GA	AA	N	GG	GA	AA	N	p	PD	Controls
Hong C et al. (2003)²¹	26	49	32	107	24	55	24	103	0.5	101/113	103/103
Hakanasson et al. (2003)²⁶	171	79	7	257	209	85	12	307	0.4	421/93	503/109
Masaki et al. (2003)²⁵	112	128	51	291	86	141	64	291	0.66	352/230	313/269
Nishimura et al. (2005)²²	92	171	64	327	88	140	47	275	0.1	355/299	316/294
Saarela et al. (2006)²⁷	42	10	0	52	81	17	3	101	0.09	94/20	179/23
Chen et al. (2007)²³	80	192	84	356	79	166	80	325	0.7	352/362	324/326
Guerini et al. (2009)¹¹	156	117	21	233	147	76	10	294	0.9	429/159	370/96
Gao et al. (2010)¹⁰	111	76	6	193	188	96	16	300	0.4	298/88	472/128
Benitez et al. (2010)²⁹	68	26	6	104	97	36	3	136	0.9	162/38	230/42
Lin et al. (2011)¹⁴	121	208	113	442	74	144	68	286	0.8	450/434	292/280
Karakasis et al. (2011)¹²	115	62	7	184	71	37	5	113	0.9	292/76	179/47
Liu et al. (2012)²⁴	110	245	109	464	151	278	120	549	0.7	465/463	580/518
Chen et al. (2011)¹³	63	133	70	266	107	197	96	400	0.7	259/273	411/389
Svetel et al. (2013)²⁸	121	52	4	177	252	106	8	366	0.4	294/60	610/122
Bialecka et al. (2013)⁹	176	62	6	244	168	65	9	242	0.04	420/74	401/83

BDNF, brain-derived neurotrophic factor; PD, Parkinson's disease; HWE, Hardy–Weinberg equilibrium.

Four studies stratified the PD sample based on cognitive impairment,^9

–12 two specifically declared to exclude patients with neurological diseases,^13,14 whereas the others omitted this clinical information.

Quantitative synthesis

Allele and genotype distributions of the G196A polymorphism from each study included in meta-analysis are shown in Table 2. Overall the results of our meta-analysis on the basis of different genetic models (dominant, recessive allelic, and homozygote) revealed no significant association between PD patients and the G196A polymorphism. In the fixed-effects model, the pooled OR for the dominant model (GA+AA vs. GG) was 1.07 (95% CI 0.97–1.19, p=0.157; I²=7%). A similar result was obtained when we considered the recessive model (AA vs. GA+GG) OR=1.04 (95% CI 0.91–1.19, p=0.552; I²=9%). No association between the BDNF (AA) genotype and PD was found, pooled OR was 1.06 (95% CI 0.91–1.24, p=0.456; I²=9%), and also between allele A and PD, pooled OR was 1.03 (95% CI 0.96–1.11, p=0.377; I²=24%). All ORs were pooled through a fixed-effects model (Fig. 2).

FIG. 2.

Forest plot of the bone-derived neurotrophic factor (BDNF) G196A polymorphism and Parkinson's disease (PD) risk. OR, odds ratio; CI, confidence interval.

The meta-regression analyses revealed no effect of the difference in the mean age of PD and control groups (dominant model, coefficient=1.0195% CI −0.2, 0.06, p=0.511; recessive model, coefficient=0.99, 95% CI 0.95, 1.04, p=0.659; AA vs GG: coefficient=1.00, 95% CI 0.93, 1.06, p=0.911; A vs. G, coefficient=1.01, 95% CI 0.98, 1.04, p=0.490), as well as for difference in sex distribution between the two groups (dominant model, coefficient=0.99, 95% CI 0.98, 1.0, p=0.192; recessive model, coefficient=1.00, 95% CI 0.98, 1.01, p=0.554; AA vs. GG, coefficient=0.99, 95% CI 0.97, 1.01, p=0.290; A vs. G, coefficient=0.99, 95% CI 0.99, 1.00, p=0.165). Sub-groups analysis showed no effect of the ethnicity in each genetic model. Overall, no evidence of publication bias was present; funnel plots seemed to be symmetrical, and the Egger regression model confirmed this assessment (dominant model, bias=0.61, 95% CI −3.21, 1.99, p=0.620; recessive model, bias=−0.26, 95% CI −1.33;0.80, p=0.602; AA vs. GG, bias=−0.34, 95% CI −1.69, 1.01, p=0.594; A vs. G, bias=0.97, 95% CI −1.14, 3.07, p=0.340). A temporal effect was excluded by cumulative analysis.

Association between the G196A polymorphism and PD with dementia was also calculated by performing a meta-analysis that considered only the four studies that stratified the PD patients on the basis of cognitive impairment^9

–12 and two studies that excluded patients with neurological diseases.^13,14 Information about allele G and A frequencies was absent in one study.¹⁰ For each genetic model, the random-effects model was applied, and the pooled ORs showed no significant association (Table 3).

Table 3.

Distribution of BDNF Genotypes and Alleles in PD Patients in Relation to Cognitive Impairment

	PD with dementia					PD without dementia
	No. of studies	Pooled ORs (D-L, random-effects model, 95% CI)	p value	I²	Test of heterogeneity (Cochran Q statistic)	No. of studies	Pooled ORs (D-L, random-effects model, 95% CI)	p value	I²	Test of heterogeneity (Cochran Q statistic)
AA+AG versus GG (dominant)	4	1.40 (0.69, 2.82)	0.348	60%	Chi-squared(3)=7.56, p=0.056	6	1.34 (0.95, 1.89)	0.093	79%	Chi-squared(5)=24.0, p<0.001
AA versus GG+AG (recessive)	3	3.71 (0.71, 19.53)	0.121	75%	Chi-squared(2)=8.11, p=0.017	5	1.55 (0.81, 2.97)	0.184	83%	Chi-squared(4)=23.7, p<0.001
AA versus GG	3	4.72 (0.63, 35.31)	0.131	81%	Chi-squared (2)=10.3, p=0.006	5	1.69 (0.77, 3.71)	0.19	86%	Chi-squared(4)=24.0, p<0.001

PD, Parkinson's disease; OR, odds ratio; D-L, DerSimonian-Laird; CI, confidence interval.

Sensitivity analysis

Sensitivity analysis was conducted to verify the robustness of our results. For each genetic model, the exclusion of studies where allele frequencies exhibited a significant deviation from the HWE as well as the exclusion of the studies that considered patients with PD family history showed no important different results.

Discussion

Recently two important meta-analysis have investigated the role of the BDNF 196 G/A polymorphism as potential risk factor for PD reporting conflicting results.^7,8 This topic was already addressed in a previous meta-analysis, including six studies of 1419 PD patients and 1406 healthy subjects, suggesting no association between this polymorphism and susceptibility to PD.³⁰ This result should be taken with caution because either the sample size was relatively small or the genotype distribution in two included studies^15,16 was not in HWE. Therefore, we have focused only on the two most recent meta-analyses, aiming at resolving the contradictory results. More precisely, Lee and Song⁸ identified an association between the BDNF 196 AA+AG genotype and PD in Europeans, but not in Asians, conducting their analysis on 12 separate comparisons corresponding to 2891 patients and 3132 controls. Dai et al.,⁷ encompassing one additional study¹⁴ and covering 3333 PD patients and 3418 controls, found no association, even after stratifying the analysis on the basis of the ethnicity. This disagreement could be explained by the fact that Dai et al.⁷ have considered the PD cases from Gao et al.¹⁰ as a mixed population for the presence of two patients from North Africa and one with a gypsy background, whereas Lee and Song⁸ considered the overall sample as European.

Reviewing all of these data, we confirmed that the difference observed between the two meta-analyses was merely due to the diverse interpretation of the data reported by Gao et al.¹⁰ Concerning the studies on Asian samples, the two meta-analyses did not differ for the results, even though Lee and Song⁸ had omitted the Lin et al. study¹⁴ in their analysis.

In our meta-analysis, we included two recent case–control studies corresponding to 425 new PD patients and 608 controls and the overall results did not show significant association between 196 G/A polymorphism and PD risk, in line with data reported by both previous meta-analyses. In the sub-group analysis on the basis of the ethnicity, we still confirmed no association as described by Dai et al.⁷ and contrary to what was found by Lee and Song.⁸ We have repeated the analysis also excluding the Gao et al. study¹⁰ by Caucasian sample and we again obtain negative results. This evidence underlines the focal influence of ethnicity on the frequencies of 196 G/A polymorphism. In fact, the inclusion or exclusion of a sample from a specific ethnic group did provide different outcomes of the analysis. There are numerous findings in the literature regarding the ethnic differences in allelic frequencies of the 196 G/A polymorphism between Caucasian and Asian populations.³¹

Positive or negative associations between this polymorphism and some diseases, such as schizophrenia, depression, eating disorders, alcoholism, substance abuse, and anxiety, have been reported in Caucasian and Asian participants and might be explained by ethnic differences.^32

–36 BDNF plays a role in eating behavior, food intake, and energy expenditure.³⁷ Therefore, ethnic differences in the BDNF 196 G/A polymorphism might be associated with different dietary habits or be affected by exposure to various environmental factors.³⁸

The PD patients included in the meta-analysis were well defined, although they were not identical in terms of demographics and clinical symptoms, and this variability might be a possible confounder. To avoid this bias, we performed multivariate analyses to reveal possible effects due to difference in the age, ethnicity, sex distribution, or family history of PD patients and healthy subjects. None of these potential confounders seem to influence the results of the analyses. The effects of BDNF on cognitive processes are well known³⁹; therefore, we performed a meta-regression to examine the association between the BDNF G196A polymorphism and PD in patients with and without impaired cognitive abilities. Summarizing the clinical data available in six studies, we found no correlation between the BDNF G196A genotype and cognitive impairment. Heterogeneity is a potential problem that might have affected the interpretation of the results, even though it was not significant. In addition, funnel plots and Egger tests did not find potential publication bias. Results from sensitivity analysis and cumulative meta-analysis suggested high stability and reliability of our results.

In conclusion, this meta-analysis revises the previous incomplete data and supports the lack of association between the BDNF G196A polymorphism and the risk of PD.

Footnotes

Acknowledgments

R.S. thanks the National Research Council Aging Program 2012–2014; MIUR Cod. 1182/Ric/V o prot. 2010SH7H3F Functional connectivity and neuroplasticity in physiological and pathological aging [ConnAge]; FISM–Fondazione Italiana Sclerosi Multipla Cod.2011/R/32 Fatigue Relief in Multiple Sclerosis by transcranial Direct Current Stimulation (tDCS): Can we Differentiate stimulation Targets within the primary sensorimotor cortices? [FaReMuS DiCDiT]; and Ministry of Health Cod. GR-2008-1138642 Promoting recovery from Stroke: Individually enriched therapeutic intervention in Acute phase [ProSIA]; Ricerca Corrente, Italian Ministry of Health.

Author Disclosure statement

No competing financial interests exists.

References

Guttman

, Kish

, Furukawa

. Current concepts in the diagnosis and management of Parkinson's disease. CMAJ, 2003; 168:293–301.

Hyman

, Hofer

, Barde

, Juhasz

, Yancopoulos

, Squinto

, Lindsay

. BDNF is a neurotrophic factor for dopaminergic neurons of the substantia nigra. Nature, 1991; 350:230–232.

Mogi

, Togari

, Kondo

, Mizuno

, Komure

, Kuno

, Ichinose

, Nagatsu

. Brain-derived growth factor and nerve growth factor concentrations are decreased in the substantia nigra in Parkinson's disease. Neurosci Lett, 1999; 270:45–48.

Parain

, Murer

, Yan

, Faucheux

, Agid

, Hirsch

, Raisman-Vozari

. Reduced expression of brain-derived neurotrophic factor protein in Parkinson's disease substantia nigra. Neuroreport, 1999; 10:557–561.

Chen

, Patel

, Sant

, Meng

, Teng

, Hempstead

, Lee

. Variant brain-derived neurotrophic factor (BDNF) (Met66) alters the intracellular trafficking and activity-dependent secretion of wild-type BDNF in neurosecretory cells and cortical neurons. J Neurosci, 2004; 24:4401–4411.

Egan

, Kojima

, Callicott

, Goldberg

, Kolachana

, Bertolino

, Zaitsev

, Gold

, Goldman

, Dean

, Lu

, Weinberger

. The BDNF val66met polymorphism affects activity-dependent secretion of BDNF and human memory and hippocampal function. Cell, 2003; 112:257–269.

Dai

, Wang

, Meng

, Zhang

, Fu

, Wu

, Bai

. Association between the BDNF G196A and C270T polymorphisms and Parkinson's disease: A meta-analysis. Int J Neurosci, 2013; 123:675–683.

Lee

, Song

. BDNF 196 G/A and 270 C/T polymorphisms and susceptibility to Parkinson's disease: A meta-analysis. J Mot Behav, 2014; 46:59–66.

Bialecka

, Kurzawski

, Roszmann

, Robowski

, Sitek

, Honczarenko

, Mak

, Deptula-Jarosz

, Golab-Janowska

, Drozdzik

, Slawek

. BDNF G196A (Val66Met) polymorphism associated with cognitive impairment in Parkinson's disease. Neurosci Lett, 2014; 561:86–90.

10.

Gao

, Diaz-Corrales

, Carrillo

, Diaz-Martin

, Caceres-Redondo

, Carballo

, Palomino

, Lopez-Barneo

, Mir

. Brain-derived neurotrophic factor G196A polymorphism and clinical features in Parkinson's disease. Acta Neurol Scand, 2010; 122:41–45.

11.

Guerini

, Beghi

, Riboldazzi

, Zangaglia

, Pianezzola

, Bono

, Casali

, Di Lorenzo

, Agliardi

, Nappi

, Clerici

, Martignoni

. BDNF Val66Met polymorphism is associated with cognitive impairment in Italian patients with Parkinson's disease. Eur J Neurol, 2009; 16:1240–1245.

12.

Karakasis

, Kalinderi

, Katsarou

, Fidani

, Bostantjopoulou

. Association of brain-derived neurotrophic factor (BDNF) Val66Met polymorphism with Parkinson's disease in a Greek population. J Clin Neurosci, 2011; 18:1744–1745.

13.

Chen

, Wang

, Xiao

, Wang

, Guo

, Zhao

, Hua

, Liu

, Zhang

. The 712A/G polymorphism of brain-derived neurotrophic factor is associated with Parkinson's disease but not major depressive disorder in a Chinese Han population. Biochem Biophys Res Commun, 2011; 408:318–321.

14.

Lin

, Wu

, Tai

, Chen

, Hu

. Lrrk2 S1647T and BDNF V66M interact with environmental factors to increase risk of Parkinson's disease. Parkinsonism Relat Disord, 2011; 17:84–88.

15.

Momose

, Murata

, Kobayashi

, Tachikawa

, Nakabayashi

, Kanazawa

, Toda

. Association studies of multiple candidate genes for Parkinson's disease using single nucleotide polymorphisms. Ann Neurol, 2002; 51:133–136.

16.

Parsian

, Sinha

, Racette

, Zhao

, Perlmutter

. Association of a variation in the promoter region of the brain-derived neurotrophic factor gene with familial Parkinson's disease. Parkinsonism Relat Disord, 2004; 10:213–219.

17.

Toda

, Momose

, Murata

, Tamiya

, Yamamoto

, Hattori

, Inoko

. Toward identification of susceptibility genes for sporadic Parkinson's disease. J Neurol, 2003; 250(Suppl 3):III40–III43.

18.

Foltynie

, Cheeran

, Williams-Gray

, Edwards

, Schneider

, Weinberger

, Rothwell

, Barker

, Bhatia

. BDNF val66met influences time to onset of levodopa induced dyskinesia in Parkinson's disease. J Neurol Neurosurg Psychiatry, 2009; 80:141–144.

19.

Foltynie

, Lewis

, Goldberg

, Blackwell

, Kolachana

, Weinberger

, Robbins

, Barker

. The BDNF Val66Met polymorphism has a gender specific influence on planning ability in Parkinson's disease. J Neurol, 2005; 252:833–838.

20.

Karamohamed

, Latourelle

, Racette

, Perlmutter

, Wooten

, Lew

, Klein

, Shill

, Golbe

, Mark

, Guttman

, Nicholson

, Wilk

, Saint-Hilaire

, DeStefano

, Prakash

, Tobin

, Williamson

, Suchowersky

, Labell

, Growdon

, Singer

, Watts

, Goldwurm

, Pezzoli

, Baker

, Giroux

, Pramstaller

, Burn

, Chinnery

, Sherman

, Vieregge

, Litvan

, Gusella

, Myers

, Parsian

. BDNF genetic variants are associated with onset age of familial Parkinson disease: GenePD Study. Neurology, 2005; 65:1823–1825.

21.

Hong

, Liu

, Lin

, Cheng

, Tsai

. Brain-derived neurotrophic factor (BDNF) Val66Met polymorphisms in Parkinson's disease and age of onset. Neurosci Lett, 2003; 353:75–77.

22.

Nishimura

, Kuno

, Kaji

, Kawakami

. Brain-derived neurotrophic factor gene polymorphisms in Japanese patients with sporadic Alzheimer's disease, Parkinson's disease, and multiple system atrophy. Mov Disord, 2005; 20:1031–1033.

23.

Chen

, Chen

, Chang

, Chen

, Lyu

, Liu

, Hu

, Chao

, Lee-Chen

, Wu

. Nuclear receptor NR4A2 IVS6+18insG and brain derived neurotrophic factor (BDNF) V66M polymorphisms and risk of Taiwanese Parkinson's disease. Am J Med Genet B Neuropsychiatr Genet, 2007; 144B:458–462.

24.

Liu

, Zhou

, Wang

, Zheng

, Chan

. Brain-derived neurotrophic factor (BDNF) genetic polymorphism greatly increases risk of leucine-rich repeat kinase 2 (LRRK2) for Parkinson's disease. Parkinsonism Relat Disord, 2012; 18:140–143.

25.

Masaki

, Matsushita

, Arai

, Takeda

, Itoyama

, Mochizuki

, Kamakura

, Ohara

, Higuchi

. Association between a polymorphism of brain-derived neurotrophic factor gene and sporadic Parkinson's disease. Ann Neurol, 2003; 54:276–277.

26.

Hakansson

, Melke

, Westberg

, Shahabi

, Buervenich

, Carmine

, Klingborg

, Grundell

, Schulhof

, Holmberg

, Ahlberg

, Eriksson

, Sydow

, Olson

, Johnels

, Nissbrandt

. Lack of association between the BDNF Val66Met polymorphism and Parkinson's disease in a Swedish population. Ann Neurol, 2003; 53:823.

27.

Saarela

, Lehtimaki

, Rinne

, Huhtala

, Rontu

, Hervonen

, Roytta

, Ahonen

, Mattila

. No association between the brain-derived neurotrophic factor 196 G>A or 270 C>T polymorphisms and Alzheimer's or Parkinson's disease. Folia Neuropathol, 2006; 44:12–16.

28.

Svetel

, Pekmezovic

, Markovic

, Novakovic

, Dobricic

, Djuric

, Stefanova

, and Kostic

. No association between brain-derived neurotrophic factor G196A polymorphism and clinical features of Parkinson's disease. Eur Neurol, 2013; 70:257–262.

29.

Benitez

, Forero

, Arboleda

, Granados

, Yunis

, Fernandez

, Arboleda

. Exploration of genetic susceptibility factors for Parkinson's disease in a South American sample. J Genet, 2010; 89:229–232.

30.

Zintzaras

, Hadjigeorgiou

. The role of G196A polymorphism in the brain-derived neurotrophic factor gene in the cause of Parkinson's disease: A meta-analysis. J Hum Genet, 2005; 50:560–566.

31.

Pivac

, Kim

, Nedic

, Joo

, Kozaric-Kovacic

, Hong

, Muck-Seler

. Ethnic differences in brain-derived neurotrophic factor Val66Met polymorphism in Croatian and Korean healthy participants. Croat Med J, 2009; 50:43–48.

32.

Lang

, Sander

, Lohoff

, Hellweg

, Bajbouj

, Winterer

, Gallinat

. Association of the met66 allele of brain-derived neurotrophic factor (BDNF) with smoking. Psychopharmacology (Berl), 2007; 190:433–439.

33.

Pezawas

, Verchinski

, Mattay

, Callicott

, Kolachana

, Straub

, Egan

, Meyer-Lindenberg

, Weinberger

. The brain-derived neurotrophic factor val66met polymorphism and variation in human cortical morphology. J Neurosci, 2004; 24:10099–10102.

34.

Sen

, Nesse

, Stoltenberg

, Li

, Gleiberman

, Chakravarti

, Weder

, Burmeister

. A BDNF coding variant is associated with the NEO personality inventory domain neuroticism, a risk factor for depression. Neuropsychopharmacology, 2003; 28:397–401.

35.

Shimizu

, Hashimoto

, Iyo

. Ethnic difference of the BDNF 196G/A (val66met) polymorphism frequencies: The possibility to explain ethnic mental traits. Am J Med Genet B Neuropsychiatr Genet, 2004; 126B:122–123.

36.

Sklar

, Gabriel

, McInnis

, Bennett

, Lim

, Tsan

, Schaffner

, Kirov

, Jones

, Owen

, Craddock

, DePaulo

, Lander

. Family-based association study of 76 candidate genes in bipolar disorder: BDNF is a potential risk locus. Brain-derived neutrophic factor. Mol Psychiatry, 2002; 7:579–593.

37.

Russo-Neustadt

. Brain-derived neurotrophic factor, behavior, and new directions for the treatment of mental disorders. Semin Clin Neuropsychiatry, 2003; 8:109–118.

38.

Gratacos

, Gonzalez

, Mercader

, de Cid

, Urretavizcaya

, Estivill

. Brain-derived neurotrophic factor Val66Met and psychiatric disorders: Meta-analysis of case-control studies confirm association to substance-related disorders, eating disorders, and schizophrenia. Biol Psychiatry, 2007; 61:911–922.

39.

Schindowski

, Belarbi

, Buee

. Neurotrophic factors in Alzheimer's disease: Role of axonal transport. Genes Brain Behav, 2008; 7(Suppl 1):43–56.