Evaluation of the Relationship Between Dopamine Receptor D2 Gene TaqIA1 Polymorphism and Alcohol Dependence Risk

Abstract

Several studies are published, that investigated dopamine receptor 2 (DRD2) gene TaqIA polymorphism as a risk factor for alcohol dependence (AD) with positive and negative associations. To derive a more precise estimation of the relationship, a meta-analysis of case–control studies that examined the association between DRD2 gene Taq1A polymorphism and alcohol dependence was performed. Eligible articles were identified through a search of databases including PubMed, Science Direct, Springer link, and Google Scholar. The association between the DRD2 TaqIA polymorphism and AD susceptibility was conducted using odds ratios (ORs) and 95% confidence intervals (95% CIs) as association measures. A total of 69 studies with 9125 cases and 9123 healthy controls were included in the current meta-analysis. Results of the present analysis showed significant association between DRD2 TaqIA polymorphism and AD risk using five genetic modes (allele contrast model—OR 1.22, 95% CI 1.13–1.32, p < 0.0001; homozygote model—OR 1.35, 95%CI 1.18–1.55; p ≤ 0.0001; dominant model—OR 1.29; 95% CI 1.20–1.39; p < 0.0001; recessive model—OR 1.21; 95% CI 1.08–1.36; p = 0.0006). There was no significant association found in subgroup analysis, TaqIA polymorphism was not significantly associated with AD risk in the Asian population under all genetic models, but in the Caucasian population, TaqIA polymorphism was significantly associated with AD risk. Overall, results support the hypothesis that DRD2 Taq1A polymorphism plays a role in alcohol dependence.

Introduction

Alcoholism dependence (AD) is a complex behavioural and multifactorial disorder often associated with an increased risk of developing other behavioural and psychiatric disorders such as anxiety, depression, anti-social personality disorder, and bipolar disorder, by affecting various neural mechanisms and part of the brain with its effect according to the dose of consumption, genetic factors, etc. AD is a psychiatric disorder, with a life time population risk of approximately 5.4%. Family and twin studies support the role of a genetic component in AD. It is widely accepted that the dopaminergic system play a crucial role in the development of psychoactive substance dependence including opiates, cocaine, nicotine, and alcohol [1–3]. The dopaminergic system regulates the brain reward mechanisms [4, 5], so genes of the reward pathway especially dopamine receptors are considered a strong candidate for alcohol dependence. Alcohol stimulates dopamine receptors, which release dopamine in the ventral striatum leading to increased alcohol consumption through mechanisms involving incentive salience attributions and craving [6]. Aberrant dopamine signaling is implicated in several psychiatric and neurological brain disorders such as schizophrenia, depression, eating disorders, Parkinson’s disease, and addiction [7]. The dopamine D2 receptor (DRD2) gene is the most studied polymorphic site in the reward pathway in connection to AD. DRD2 receptor is present throughout the brain but the highest density was reported in the substantia nigra, ventral tegmental area, and nucleus accumbens [8–10]. DRD2 gene is present on chromosome 11q23.1 [11, 12], and after transcription, it produces two transcripts that encode two isoforms (D2L isoform and D2S isoform). Several polymorphisms are reported in the DRD2 gene, but the most studied polymorphism is TaqIA polymorphism (rs1800497 C32806T Glu713Lys) in the 3 flanking region of the gene [13]. The DRD2 gene spans approximately 270 kb with about a 250 kb intron separating the first and second exons [12]. The TaqIA polymorphism lies 10,541 bp downstream (C32806T) of the termination codon of the DRD2 gene and falls within ankyrin repeat and kinase domain containing 1 (ANKK1) gene [14, 15]. The two alleles are referred to as A2 (cytosine) and A1 (thymine) and the TaqIA genotypes are named A2/A2 (homozygous wid), A2/A1 (heterozygous), and A1/A1 (homozygous). DRD2 Taq1A polymorphism has been associated with reduced D2 receptor availability in the striatum [13, 16], lower mean relative glucose metabolic rate in dopaminergic regions and low receptor density [17].

Blum et al. [18] initially demonstrated an association of the minor Taql A allele (A1) of the DRD2 gene with AD. Since then, several studies have been published from different populations, with many affirming this finding [14, 19–24], while others have not [25–28]. Hence, the authors evaluated the association between DRD2 Taq1 polymorphism and alcohol dependence by performing of case–control.

Methods

Meta-analysis was carried out according to Meta-analysis of observational studies in epidemiology (MOOSE) guidelines [29].

Article Search

Published articles examining the effect of the DRD2 gene Taq1A polymorphisms on the risk of alcohol dependence were identified through electronic database searches in Pubmed, Science Direct, Google scholar, and springer link. Databases searches were done from January 1990 (the year that DRD2 Taq1A polymorphisms were first reported as a risk factor for AD. Electronic database searches were supplemented by manual searches of references of reviews and published research articles. The following terms were used: ‘‘Dopamine receptor 2’’, ‘‘DRD2,’’ ‘‘Taq1A’’, ‘‘Alcoholism’’, ‘‘Alcohol dependence’’, ‘‘polymorphism’’. The literature search included all languages. If any study reported results on different subpopulations according to ethnicity, each subpopulation was included as a separate study in the meta-analysis.

Inclusion Criteria

Studies included in our meta-analyses were based on the following criteria: (1) the article should be published, (2) the article should reported sample size, the ancestry of samples, etc.; and (3) the sample not be duplicative of other reports. Studies were excluded if: (1) incomplete raw data/information and not providing complete information for the number of genotypes and/or allele number calculation, (3) studies based on pedigree, and (4) review, letter to editors, and book chapters.

Data Extraction

From each of the included articles, the following information was extracted: first author’s family name, year of publication, country, ethnicity, study design, number of controls and cases, and evidence of HWE in controls.

Statistical Analyses

Crude odds ratios (ORs) with 95% confidence intervals (CIs) were calculated under five genetic models: the allele model (mutant [M] allele vs. wild [W] allele), the dominant model (MM + WM vs. WW), the recessive model (MM vs. WM + WW), the homozygous model (MM vs. WW), and the heterozygous model (WM vs. WW). The OR was estimated by using fixed effects [30] and random effects [31] models depending upon heterogeneity. If higher heterogeneity between studies then the pooled OR is estimated using the RE model [32, 33]. The heterogeneity was tested using the Q-statistic and was quantified using the I² statistic [34]. Subgroup analysis based on the ethnicity of subjects was performed to evaluate the stability of the results and sensitivity analysis was done by removing the studies, not in Hardy–Weinberg equilibrium (HWE), and studies with small sample sizes. Control population of each study was tested for Hardy–Weinberg Equilibrium (HWE) using the chi-square test. The quality score of studies was assessed according to 10-point scoring method of Clark and Baudouin [35]. Two authors (PK and VR) calculated scores. Studies with higher score (> 6) were defined as high-quality studies.

Funnel plot and the Egger regression asymmetry test [36] were used for the assessment of publication bias. P values were two-tailed with a significance level of 0.05. All analyses were done by the program Meta-Analyst [37] and Mix version 1.7 [38].

Results

Literature Search

The flow chart of the study selection process is shown in Fig. 1. Initial search of Pubmed, Science Direct, Google Scholar, and Springer Link databases, total 511 articles were retrieved, but 301 articles did not meet the inclusion criteria after reviewing the abstract. Out of the remaining 210 articles, 142 articles were also excluded as not following inclusion criteria. After applying inclusion and exclusion criteria, total 68 studies were found suitable for the present meta-analysis [14, 18–28, 39–92]. Panduro et al. [92] analyzed samples from two populations, so in the present meta-analysis data from both the population were included as separate studies, hence the total number of studies included in the current meta-analysis was sixty nine (Table 1).

Fig. 1.

Flow diagram of Study Search and Selection Process

Table 1.

Characteristics of eligible studies considered in the present meta-analysis

Study	Country	Control number	Case number	p value of HWE	Quality Score
Blum et al., 1990	USA	35	35	0.65	6
Bolos et al., 1990	USA	127	40	0.02	4
Blum et al., 1991	USA	43	96	0.55	5
Comings et al., 1991	USA	108	104	0.19	6
Gelernter et al., 1991		68	44	0.81	5
Parsian et al., 1991	USA	25	32	0.75	5
Schwab et al., 1991	Germany	69	45	0.0007	4
Cook et al., 1992	Iowa	20	20	0.42	5
Goldman et al., 1992	USA	36	46	0.65	5
Amadeo et al., 1993	France	43	49	0.56	5
Arinami et al., 1993	Japan	35	78	0.17	5
Geijer et al., 1994	Sweden	81	93	0.33	6
Noble et al., 1994	USA	80	73	0.24	6
Saurez et al., 1994	USA	89	88	0.95	6
Neiswanger et al., 1995	USA	30	52	0.69	6
Sanders et al.,1995	Germany	113	270	0.95	7
Chen et al., 1996	Taiwan	41	73	0.14	6
Heinz et al., 1996	Germany	113	97	0.95	6
Lu et al., 1996, Han	Taiwan	24	20	0.32	5
Lu et al., 1996, Atyal	Taiwan	21	21	0.15	5
Lu et al., 1996, Ami	Taiwan	20	20	0.06	4
Chen et al., 1997, Atyal	Taiwan	31	36	0.33	5
Chen et al., 1997, Ami	Taiwan	23	24	0.43	5
Chen et al., 1997, Bunun	Taiwan	58	58	0.47	6
Chen et al., 1997, Paiwan	Taiwan	35	35	0.54	5
Chen et al., 1997, Han	Taiwan	66	50	0.1	5
Goldman et al., 1997	USA	161	276	0.26	7
Hietala et al., 1997	Finland	50	70	0.38	5
Kono et al., 1997	Japan	93	100	0.21	6
Lawford et al., 1997	Australia	46	201	0.42	7
Lee et al., 1997	Korea	100	67	0.27	7
Ishiguro et al.,1998	Japan	152	209	0.36	8
Gelernter and kranzler, 1999	USA	136	160	0.46	8
Ovchinnikov et al., 1999	Russia	76	42	0.55	6
Sander et al., 1999	Germany	196	310	0.82	9
Amadeo et al., 2000, French	France	57	69	0.13	7
Amadeo et al., 2000, Polynesian	France	22	34	0.82	6
Amadeo et al., 2000, Caucasian	France	23	26	0.26	6
Bau et al., 2000	Brazil	114	115	0.59	8
Garwood et al., 2000	France	75	113	0.57	8
Samochoviece et al., 2000	Germany	192	292	0.93	7
Anghelescu et al., 2001	Germany	98	243	0.65	6
Lu et al., 2001	Taiwan	85	63	0.4	6
Pastorelli et al., 2001	Italy	64	60	0.34	6
Shaikh et al., 2001	India	53	50	0.15	6
Limosin et al., 2002	France	107	120	0.84	6
Foley et al., 2004	Australia	109	87	0.54	7
Konoshi et al., 2004	USA	251	130	0.75	6
Berggren et al., 2006	Sweden	842	357	0.8	8
Freire et al., 2006	Brazil	112	100	0.69	7
Sakai et al., 2007	USA	151	239	0.8	7
Wang et al., 2007	Taiwan	158	73	0.4	6
Samochowiec et al., 2008	Poland	150	122	0.98	8
Kraschewski et al., 2009	Germany	368	360	0.09	8
Bhaskar et al., 2010	India	115	81	0.25	6
Berggren et al., 2010	Sweden	578	366	0.97	7
Kovanen et al., 2010	Finland	511	512	0	5
Lu et al., 2010	Taiwan	244	133	0.23	6
Prasad et al., 2010	India	60	90	0.36	6
Kasiakaogia-Worlley et al., 2011	UK	956	987	0.34	8
Ladgren et al., 2011	Sweden	32	84	0.58	6
Schellekens et al., 2012	Netherland	99	110	0.27	7
Mignini et al., 2012	Italy	280	280	0.059	4
Singh et al., 2013	India	286	129	0.47	8
Jasiewicz et al., 2014	Poland	157	169	0.19	8
Vasconcelos et al., 2015	Brazil	114	113	0.78	8
Ragia et al., 2016	Greece	74	72	0.31	7
Panduro et al., 2017, Guadalajara	Mexico	69	227	0	5

p = measure of probability (significance level); HWE = Hardy Weinberg Equation

Characteristic of Eligible Studies

The first study was published in 1990 [18] and the last study was published in 2017 [92]. In one study, authors mentioned only allele numbers. A total of 2591 subjects were involved in this meta-analysis, including 9125 AD cases and 9123 healthy controls. In cases, CC, CT, and TT genotypes were 3903, 3063, and 1020 respectively. In control group allele frequency of C and T was 72.42% and 27.58% respectively (Fig. 2). Out of 69 studies, the control population of four studies [38, 42, 83, 92] is not in Hardy Weinberg equilibrium. Twenty out of 69 studies, were conducted in Asian population, and the other 49 studies in Caucasian populations.

Fig. 2.

Bar diagram showing A2 and A1 alleles percentage in control group of total studies, Asian studies and Caucasian studies

Meta-Analysis

Table 2 summarizes the ORs with corresponding 95% CIs for the association between DRD2 TaqIA polymorphism and risk of alcohol dependence in allele contrast ((T vs. C/A1 vs. A2)), homozygote (TT vs. CC/A1A1 vs. A2A2), dominant (TT + CT vs. CC/A1A1 + A1A2 vs. A2A2), recessive (TT vs. CT + CC/ A1A1 vs. A1A2 + A2A2) and co-dominant (CT vs. CC/A1A2 vs. A2A2) models.

Table 2.

Summary estimates for the odds ratio (OR) of Taq1A polymorphism in various allele/genotype contrasts, the significance level (p value) of heterogeneity test (Q test), and the I² metric and publication bias p value (Egger Test)

Genetic models	Fixed effect OR (95% CI), p	Random effect OR (95% CI), p	Heterogeneity p value	I2 (%)	p-value publication bias
Allele contrast (T vs. C/A1 vs. A2)	1.18 (1.12–1.24), < 0.0001	1.22(1.13–1.31), < 0.0001	0.0003	41.02	0.003
Co-dominant (CT vs. CC/A1A2 vs. A2A2)	1.26(1.17–1.36), < 0.0001	1.27(1.15–1.40), < 0.0001	0.005	33.2	0.19
Homozygote (TT vs. CC/A1A1 vs. A2A2)	1.35(1.18–1.55), < 0.0001	1.39(1.19–1.61), < 0.0001	0.27	8.95	0.42
Dominant (TT + CT vs. CC/A1A1 + A1A2 vs. A2A2)	1.29(1.20–1.39), < 0.0001	1.29(1.18–1.43), < 0.0001	0.002	36.57	0.18
Recessive (TT vs. CT + CC/ A1A1 vs. A1A2 + A2A2)	1.21(1.08–1.36),0.0006	1.26(1.11–1.43), 0.0004	0.32	6.45	0.06

OR odds ratio, CI confidence Interval, p = measure of probability (significance level); I² = measure of heterogeneity

Allele contrast meta-analysis revealed a modest significant association between AD and DRD2 gene T allele (T vs. C) with both fixed effects (OR 1.18, 95% CI 1.12–1.24, p < 0.0001) and random effects model (OR 1.22, 95% CI 1.13–1.32, p < 0.0001) (Table 2). Heterogeneity was less than 50% (I² = 41.02), hence fixed effect mode was adopted. Results showed an increased risk of AD among mutant homozygote variants (TT vs.CC; homozygote model), with both fixed (OR 1.35, 95% CI 1.18–1.55; p ≤ 0.0001) and random (OR 1.39; 95% CI 1.19–1.61; p < 0.0001) effect models (Table 2).

Mutant genotypes (TT + CT vs.CC; dominant model) showed a positive significant association with AD using both fixed (OR 1.29; 95% CI 1.20–1.39; p < 0.0001) and random (OR 1.29; 95% CI 1.18–1.453; p < 0.0001) effect models (Table 2, Fig. 3). Similarly, the recessive genotypes model (TT vs. CT + CC) also showed a significant association with AD with both fixed (OR 1.21; 95% CI 1.08–1.36; p = 0.0006) and random (OR 1.26; 95% CI 1.11–1.43; p = 0.0004) effect models (Table 2).

Fig. 3.

Random effect Forest plot of dominant model (TT + CT vs. CC) of total 69 studies of DRD2 TaqIA polymorphism

Sub-group Analysis

The authors also performed ethnicity- based sub-group analysis. Out of 69 studies, 20 studies were from the Asian and 49 studies were from the Caucasian population. In the Asian population (number of studies = 20; 1410/1700 cases/controls), meta-analysis using all five genetic modes did not show any significant association (allele contrast: OR 1.23, 95% CI 1.11–1.37, p = 0.87; dominant model: OR 1.31, 95% CI 1.13–153, p = 0.71) (Fig. 4). Results of the Caucasian studies (number of studies = 49; 7715/7423 cases/controls) meta-analysis using five genetic models indicated significant association with both fixed and random effects model, but fixed effect mode was adopted due to less heterogeneity (allele contrast: OR 1.23, 95% CI 1.23–1.35, p = 0.00; dominant model: OR 1.31, 95% CI 1.16–147, p = 0.00) (Fig. 5).

Fig. 4.

Random effect Forest plot of dominant model (TT + CT vs. CC) of total Asian studies of DRD2 TaqIA polymorphism

Fig. 5.

Random effect Forest plot of homozygote model (TT vs. CC) of total Caucasian studies of DRD2 TaqIA polymorphism

Sensitivity Analysis

In allele contrast meta-analysis, sensitivity analysis was performed by the exclusion of five studies in which the control population was not in HWE. The result of meta-analysis by exclusion of these five studies not in HWE [38, 42, 83, 91] showed that the odds ratio was not increased. In addition, the influence of each study on the pooled OR was examined by repeating the meta-analysis while omitting each study one at a time. The results suggested that no individual study significantly affected the pooled ORs.

Publication Bias

Begg’s funnel plot and Egger’s test were performed to assess the publication bias. The shape of funnel plots did not reveal any evidence of obvious asymmetry in all genetic models except allele contrast (Fig. 6). Egger’s test was used to provide statistical evidence and p values of Egger’s tests were more than 0.05 (p = 0.003 for T vs. C; p = 0.42 for TT vs. CC; p = 0.19 for CT vs. CC; p = 0.18 for TT + CT vs. CC; and p = 0.06 for TT vs. CT + CC). The results did not show any evidence of publication bias except in allele contrast meta-analysis of total studies. The detailed data were given in Table 2.

Fig. 6.

Funnel plot- Precision by log odds ratio for dominant model (TT + CT vs. CC) of total 69 studies of DRD2 TaqIA polymorphism

Discussion

It is well recognized that there is individual susceptibility to AD even within the same environmental exposure. Main factor, including polymorphisms of genes involved in reward cascade pathway, may have accounted for this difference.

Taq1A1 polymorphism which leads to an amino acid substitution (p. Glu713Lys) in the 11th ankyrin repeat of the protein is postulated to modify substrate-binding specificity of the protein [93]. The minor (A1,T) has been reported to associated with low expression of DRD2 gene and also with decreased number of dopamine D2 receptor density/availability in the brain [93]. In addition, positron emission tomography (PET) studies have shown that Taq1A1 allele is also associated with decreased mean relative glucose metabolic rate in dopaminergic human brain regions [93]. To compensate less number of receptors or to find pleasure, individuals starts heavy drinking of alcohol to be happy and satisfied [94].

Four meta-analyses are published and demonstrated a moderate effect of the Taq1A polymorphism on AD [95–98]. The present meta-analysis included the largest number of studies so far investigating the association between Taq1A polymorphism and AD. Sixty-nine eligible studies published up to 2017 were considered, including 9,125 cases and 9,123 healthy controls. Our results provide strong evidence of the association between the Taq1A polymorphism and AD, especially in the European population. Heterogeneity and publication bias are not present and sensitivity analysis showed that the results from both allelic and genotypic meta-analyses are stable and not influenced by any individual study.

Meta-analysis offers a powerful method to synthesize information from independent studies with similar targets [99]. Several meta-analysis are published, which evaluated the effects of polymorphism in susceptibility of diseases/disorders-down syndrome [100, 101], cleft lip and palate [102], Glucose-6-phosphate dehydrogenase deficiency [103], male infertility [104], schizophrenia [105], obsessive compulsive disorder [106], depression [107], epilepsy [108], Alzheimer’s disease [109], ovarian cancer [110], endometrial cancer [111], and uterine leiomyoma [112].

Along with strengths, there were also a few limitations in the current meta-analysis like—(1) crude OR was used in the meta-analysis, (2) other risk factors among the subjects in the available studies, such as smoking status, diet, etc. were not considered and (3) only single gene polymorphism was considered and gene–gene or gene-environment interactions were not considered.

In conclusion, pooled analysis of data from 69 separate populations indicates that the DRD2 Taq1A polymorphism is associated with a significant risk of AD. Results of subgroup analyses showed that TaqIA polymorphism is not a risk factor for AD in the Asian population, but this polymorphism is provided susceptibility for AD in the Caucasian population. Future large-scale, population-based association studies are now the need of the hour to investigate potential gene–gene and gene–environment interactions involving the DRD2 Taq1A polymorphism in determining the risk of AD.

Author Contribution

PK and AC had searched the electronic databases for the suitable articles and extracted data independently from the downloaded articles, and PK has written the rough draft of manuscript. VR had checked the data and written the final manuscript of the article.

Declarations

Conflict of interest

All the authors declare that there is no conflict of interest.

Ethical Standards

Human and Animals Participants Informed consent and Ethical Clearance is not required, as there was no human or animal involvement in the study.

Sources of Financial Support

Nil.

Ethical Approval

Present manuscript is a meta-analysis/review, hence, ethical clearance is not required for the present manuscript, no human blood/ tissue samples are used in the present manuscript.