Impact of genetic variations in ADORA2A gene on depression and symptoms: a cross-sectional population-based study

Abstract

Genetic variants involved in adenosine metabolism and its receptors were associated with increased risk for psychiatric disorders, including anxiety, depression, and schizophrenia. Here, we examined an association between a single nucleotide polymorphism in A_2A receptor gene (ADORA2A, rs2298383 SNP) with current depressive episode and symptom profile. A total of 1253 individuals from a cross-sectional population-based study were analyzed by the Mini International Neuropsychiatric Interview 5.0. Our data showed that the TT genotype of ADORA2A rs2298383 SNP was associated with reduced risk for depression when compared to the CC/CT genotypes (p = 0.020). This association remained significant after adjusting for confounding variables such as smoking, gender, socioeconomic class, and ethnicity (OR = 0.631 (95% CI 0.425–0.937); p = 0.022). Regarding the symptoms associated with depression, we evaluated the impact of the ADORA2A SNP in the occurrence of sad/discouraged mood, anhedonia, appetite changes, sleep disturbances, motion changes, energy loss, feelings of worthless or guilty, difficulty in concentrating, and presence of bad thoughts. Notably, the TT genotype was independently associated with reduced sleep disturbances (OR = 0.438 (95% CI 0.258–0.743); p = 0.002) and less difficulty in concentrating (OR = 0.534 (95% CI 0.316–0.901; p = 0.019). The cross-sectional design cannot evaluate the cause-effect relationship and did not evaluate the functional consequences of this polymorphism. Our data support an important role for ADORA2A rs2298383 SNP in clinical heterogeneity associated with depression. The presence of the TT genotype was associated with decrease risk for current depression and disturbances in sleep and attention, two of the most common symptoms associated with this disorder.

Introduction

Depression is a leading cause of disability and is estimated to become the second most prevalent illness worldwide by 2020 [1, 2]. The etiology of this condition is complex and multifactorial, involving environmental, genetic, and psychosocial factors [3]. Genetic variability plays an important role in depression, and studies suggest a heritability of 25–29% for males and 42–49% for females. However, multiple genes with small effects seem to contribute to the clinical characteristics and progression of this disorder [4, 5].

According to the Diagnostic and Statistical Manual for Mental Disorders 5 (DSM-5), an episode of depression is characterized by the presence of depressed mood and/or anhedonia and other symptoms including changes in appetite or weight; sleep; fluctuations in general activity and energy; feelings of hopelessness, worthlessness, or guilt; diminished concentration; and suicide ideation [6]. Thus, depression is currently defined only on the basis of behavioral modifications found in patients, and the heterogeneity in clinical presentation might suggest the involvement of different neurochemical and genetic substrates [7]. Hence, there is a major socioeconomical pressure in order for the development of new and more effective strategies to identify, classify, and manage this condition.

In this context, the adenosinergic system appears as a neuromodulatory system able to control the release of neurotransmitters and hormones involved in the pathophysiology of depression, including the serotonergic and dopaminergic systems and the hypothalamic-pituitary-adrenal axis [8–10]. There is clinical evidence suggesting that the consumption of caffeine (a non-selective adenosine receptor antagonist) and caffeinated beverages improves the effect of classical antidepressant and is inversely correlated with the appearance of depressive symptoms [11–17]. In attempt to clarify the involvement of adenosinergic system in depression, preclinical studies have suggested that inhibition of adenosine A2A receptors using pharmacological or genetic approaches produced antidepressant-like effects in animal models of depression and improved the social behavior, a fundamental element affected in many psychiatric disorders [18–27]. Moreover, a recent finding also showed that increased levels and functionality of the A2A receptor induced by early-life hyperthermic seizures might induce depressive-like behavior in adult rats [28].

More recently, another line of evidence emerged, suggesting a possible influence of genetic variations in genes involved in adenosine metabolism and adenosine receptors in psychiatric conditions. In fact, a SNP in the nucleoside transporter gene, SLC29A3, was reported to compromise adenosine transport and predispose women to depression [29]. Besides that, polymorphisms in ADORA2A, the gene that encodes A2A receptors, have been associated with several psychiatric conditions, such as schizophrenia, psychosis, anxiety, and phobia [30–37]. Besides that, risk genotypes for ADORA2A, associated to anxiety and panic disorder, were also associated with increased binding and availability of adenosine A1 receptor in image studies in the human brain, which might suggest an imbalance between inhibitory and facilitatory regulations of adenosinergic system [38]. Thus, considering the clinical and preclinical findings that point to the involvement of the adenosinergic system in depression and related psychiatric disorders, it is important to evaluate the association of the ADORA2A rs2298383 SNP in depression and its impact on specific symptoms associated with this disorder. Our results might potentially contribute to the characterization of disease heterogeneity and identification of susceptible and vulnerable individuals.

Methodology

Study design and participants

This is a cross-sectional population-based study of people aged 18 to 35. The sample consisted of 1253 participants living in the urban Pelotas, Southern Brazil. Sample selection was performed by clusters (June 2011 to May 2013), considering a population of 97,000 people in that age range in the current census of 448 sectors in the city, provided by the Brazilian Institute of Geography and Statistics (IBGE). In order to ensure the necessary sample inclusion, 86 census-based sectors were systematically drawn. After identification, the subjects signed the informed consent and answered to a sociodemographic questionnaire. In order to ascertain current and lifetime psychiatric disorders, trained psychologists used a structured diagnostic interview, the Mini International Neuropsychiatric Interview according to DSM-IV criteria (M.I.N.I. 5.0, Brazilian version/DSM IV, Medical Outcome Systems Inc., Jacksonville, FL, USA). For this study, we considered individuals in current episode of depression. The study was approved by the Ethics Committee of our University (2010/15).

Genotyping assay and functional analysis

Blood samples were obtained by venipuncture after the clinical interview. Total DNA was extracted from peripheral blood leukocytes using a standardized salting-out procedure [39]. Subsequently, the total DNA has been quantified by spectrophotometry and stored at − 20 °C until molecular analysis. ADORA2A rs2298383 (C/T) SNP was genotyped using primers and probes contained in the 40x Human Custom TaqMan Genotyping Assay (Life Technologies, Foster City, CA, USA). The reactions were conducted in a 96-well plate, in a total 5-μl reaction volume using 2 ng of genomic DNA, TaqMan Genotyping Master Mix 1x (Applied Biosystems), and Custom TaqMan Genotyping Assay 1x, as determined by the manufacturer. Then, plates were positioned in a real-time PCR thermal cycler (7500 Fast Real PCR System; Applied Biosystems) and heated for 10 min at 95 °C, followed by 45 cycles of 95 °C for 15 s and 60 °C for 1 min. Fluorescence data files from each plate were analyzed using automated allele-calling software (SDS 2,0.1; Applied Biosystems). GTEx dataset (Release V7) was used to performed eQTL analysis to examine the effects of SNPs on mRNA expression (https://gtexportal.org).

Data analysis

The questionnaires were typed and coded directly into the EpiInfo6 program at the interview time, and statistical analysis was performed using the statistical package SPSS 23.0 for Windows. The genotypic and allelic frequencies were estimated by direct counting of the alleles, and the Hardy-Weinberg equilibrium was tested by the chi-square test. Comparisons between groups were performed by Student’s t test and chi-square test. A measure of magnitude of effect was calculated by logistic regression and presented as odds ratio (OR) and 95% confidence interval. Logistic regression analysis was adjusted for smoking, gender, socioeconomic class, and ethnicity. Values of p ≤ 0.05 are considered statistically significant.

Results

Our study encompassed 1253 participants, out of which 228 were in current episode of depression. The prevalence of depression in the sample was higher in women (70.6%; p < 0.001), non-Caucasian (30.7%; p = 0.01), smokers (38.9%; p < 0.001), and low-income individuals (33.5%; p < 0.001) (Table 1). When we analyzed the participants according to the genotype, we identified that the TT genotype of the rs2298383 SNP presents a reduced frequency in individuals with current depression episode (16.2%) compared with that in control (23.5%; p = 0.02) (Table 2).

Table 1.

Sociodemographic and clinical characteristics of the sample according to the current depression episode

Variables	Current depression episode
	No	Yes	p value
Female sex (%)	52.0	70.6	< 0.001*
Age (years)	25.62 ± 5.45	26.04 ± 5.45	0.286
Caucasian ethnicity (%)	77.5	69.3	0.0100*
Low socioeconomic class (%)	18.1	33.5	< 0.001*
Smokers (%)	18.4	38.9	< 0.001*

Data are presented as mean ± standard deviation or %

*p values were computed using chi-square test or Student’s t test, as appropriated

Table 2.

Genotype distribution of ADORA2A rs2298383 SNP according to current depression episode

	Current depression
	No	Yes	p value
Genotype (rs2298383 SNP)
CC	308 (30.0%)	70 (30.7%)	0.040*
CT	477 (46.5%)	121 (53.1%)
TT	241 (23.5%)	37 (16.2%)
Genotypes (recessive model)
CC/CT	785 (76.5%)	191 (83.8%)	0.020*
TT	241 (23.5%)	37 (16.2%)
Genotypes (dominant model)
CC	308 (30.0%)	70 (30.7%)	0.902
CT/TT	718 (70.0%)	158 (69.3%)

Data are presented as n (%)

*p values were computed using the chi-square test

In Table 3, sociodemographic and clinical characteristics according to ADORA2A rs2298383 SNP genotypes demonstrated no significant differences were found for gender (p = 0.190), age (0.104), socioeconomic class (p = 0.219), and smoking (p = 0.459). However, there was a significant difference in ethnicity, with a higher prevalence of the T allele in Caucasian individuals compared to CC genotype (p < 0.0001) (Table 3). The genotype distribution of rs2298383 SNP was in agreement with the predicted by the Hardy-Weinberg equilibrium (x² = 0.16; p = 0.689). To evaluate the functionality of the SNP rs2298383, we queried this SNP for alteration in gene expression in the GTEx dataset and found that rs2298383-TT was associated with increased mRNA expression of ADORA2A (p = 0.00013; Fig. 1a) in whole blood tissue. The gene eQTL Visualizer plot is shown in Fig. 1b. However, this SNP was not identified as an eQTL for ADORA2A in other tissues.

Table 3.

Sociodemographic and clinical characteristics of sample according to ADORA2A rs2298383 SNP genotype

	Genotype (rs2298383 SNP)
	CC	CT	TT	p value
Female sex (%)	51.9	55.9	58.8	0.190
Age (years)	25.1 ± 5.9	25.9 ± 5.1	25.8 ± 5.3	0.104
Low socioeconomic class (%)	24.3	20.4	17.3	0.219
Caucasian ethnicity (%)	69.0	77.5a	82.3a	< 0.001*
Smokers (%)	19.9	23.2	22.7	0.459

Data are presented as mean ± standard deviation or %

*p values were computed using the chi-square test or one-way ANOVA test followed by Bonferroni’s post hoc test, as appropriated

^aDifferences between groups

Fig. 1.

eQTL analysis of rs2298383. a Boxplots from GTEx data showing the effect of rs2298383 genotypes on ADORA2A expression in whole blood tissue. Numbers below each boxplot indicate the sample size of each genotype. b Gene eQTL Visualizer plot showing significant eQTL for ADORA2A in whole blood tissue. An eQTL appears as a circle and rectangle box on the heat map. The color and size of the circle represent the effect size and p value of the eQTL results, respectively. Additionally, the closer a SNP to the transcription end site (TES), the darker the box is. Numbers in black ovals indicate sample sizes. The TSS is drawn as a flag and the direction of the flag indicates the transcription direction

Table 4 depicted the specific symptom profile of the population according to genotype distribution for the rs2298383 SNP. Our results showed that individuals carrying TT genotype presented reduced sleep disturbances (48.6%) and less difficulty in concentration and decision-making (39.2%), when compared to the C allele carriers for both symptoms (68.2% and 55.1%, respectively). Logistic regression analysis adjusted for smoking, gender, socioeconomic class, and ethnicity showed that the polymorphism in the ADORA2A gene remains associated with a reduced risk for depression, revealing an independent effect of the polymorphism with the disorder (OR = 0.631 (95% CI 0.425–0.937); p = 0.022). The C allele of rs2298383 remained independently associated with sleep disturbances (OR = 0.438 (95% CI 0.258–0.743); p = 0.002) and difficulty in concentration and decision-making (OR = 0.534 (95% CI 0.316–0.901; p = 0.019) after adjustment for confounding variables such as smoking, gender, socioeconomic class, and ethnicity (Table 5).

Table 4.

Symptom profile in sample according to ADORA2A rs2298383 SNP genotype

Symptoms	Genotype (rs2298383 SNP)		p value
	CC/CT	TT
Sad and discouraged mood
Yes	257 (26.3%)	71 (25.5%)	0.858
Anhedonia
Yes	196 (20.1%)	44 (15.8%)	0.134
Appetite changes
Yes	174 (59.6%)	47 (63.5%)	0.629
Sleep disturbances
Yes	199 (68.2%)	36 (48.6%)	0.003*
Emotion changes
Yes	160 (54.8%)	41 (55.4%)	1.000
Tiredness and loss of energy
Yes	188 (64.4%)	42 (56.8%)	0.281
Feelings of worthless or guilty
Yes	127 (43.5%)	28 (37.8%)	0.455
Difficulty in concentrating
Yes	161 (55.1%)	29 (39.2%)	0.020*
Bad thoughts
Yes	111 (38.0%)	21 (28.4%)	0.160

Data are presented as n (%)

*p values were computed by the chi-square test comparing depressive symptoms with different genotypes

Table 5.

Logistic regression analysis of ADORA2A rs2298383 SNP and clinical covariates in current depression episode and specific associated symptoms

Polymorphism	Current depression		Sleep disturbances		Difficulty in concentrating
rs2298383	Unadjusted OR (95% CI)/p	Adjusted OR (95% CI)/p	Unadjusted OR (95% CI)/p	Adjusted OR (95% CI)/p	Unadjusted OR (95% CI)/p	Adjusted OR (95% CI)/p
Recessive model
CC/CT	1	1	1	1	1	1
TT	0.631 (0.431–.923)/0.018*	0.631 (0.425–0.937)/0.022*	0.443 (0.264–0.743)/0.002*	0.438 (0.258–0.745)/0.002*	0.524 (0.312–0.883)/0.015*	0.533 (0.316–0.901)/0.019*

Data are presented as odds ratios (OR and 95% CI) for unadjusted and adjusted analysis (gender, ethnicity, socioeconomic class and smoking)

*p values were computed by logistic regression

Discussion

In the present study, we investigated the impact of ADORA2A rs2298383 SNP in current depression and associated symptom profile. We found a reduced risk of the TT genetic variant for depression. In addition, this genotype was also associated with protection against sleep dysfunction and attentional impairment, two symptoms frequently observed in patients with depression.

The ADORA2A locus is located on human chromosome 22 and contains a set of SNPs. The most studied SNP in the ADORA2A gene, rs5751876, is the synonymous exonic exchange Tyr361Tyr (formerly designated as 1976C/T or 1083C/T). This SNP has been observed as risk factor for panic disorder, blood-injury phobia [34], autism spectrum disorders [33], and caffeine- or amphetamine-induced anxiety [32, 36, 40, 41]. However, a study including 192 individuals failed to found an association between ADORA2A rs5751876 SNP and major depression or bipolar disorder [42]. Since the rs5751876 SNP is a synonymous exchange with no obvious functional impact on the protein, other polymorphisms in strong linkage disequilibrium are probably responsible for the underlying functional variation. In fact, the rs5751862, rs2298383, and rs3761422 SNPs were associated with different anxiety-related personality scores [35]. In the present study, we evaluated the rs2298383, which is in high linkage disequilibrium with the rs5751876 SNP and presents functional potential [35]. The rs2298383 SNP is located in putative promoter region upstream of the newly identified exon 1 variants [43] and serves as tagging SNPs according to the HapMap database. In addition, rs2298383 has a position within a predicted regulatory element suggested to display high regulatory potential (UCSC database) computed from alignments of human, chimpanzee, macaque, mouse, rat, dog, and cow [44, 45].

However, the functional impact of this SNP on A2A receptors was not determined yet. A recent finding has pointed out that other ADORA2A SNPs in strong linkage disequilibrium with rs2298383 are associated with increased binding and availability of A1 receptors in encephalic regions related to emotion, which might contribute to their effects on anxiety and panic disorder [38]. Increased A1 receptor binding was also detected in the hippocampus of two different rat models of stress, using unpredictable chronic mild stress or chronic restraint stress, demonstrating that A1 receptor regulation is involved on depressive-like behavior [46]. Beyond that, in silico analyses (http://pupasuite.bioinfo.cipf.es/) indicated that the rs2298383 SNP occurs within a triplex-forming oligonucleotide (TFO) target sequence which is involved in gene expression regulation [47]. Therefore, TFO binding to the C allele (identified as the risk allele) potentially leads to transcription inhibition and thus drastically decreased amounts of A2A receptor molecules. On the other hand, in TT individuals, the TFO sequence is disrupted, leading to a normal expression of the A2A receptors. Moreover, functional analysis of rs2298383 using eQTL database showed that the rs2298383-TT genotype increases the ADORA2A expression in whole blood tissue. In this scenario, our results showed reduced risk of rs2298383-TT genotype for depression suggesting the relevant role of ADORA2A gene expression as well as an imbalance between inhibitory and facilitatory regulations of adenosinergic system in mood regulation.

Interestingly, our results also demonstrated an association between the TT genotype of the rs2298383 SNP and protection against sleep dysfunction. Adenosine is well-known endogenous sleep-promoting substance, and extracellular concentrations of adenosine in most brain areas are decreased during sleep compared to those during wakefulness [48]. These high levels of adenosine interact with both A1 and A2A receptors, which play a crucial role in the modulation of wakefulness and sleep cycle, and pharmacological studies showed that A2A receptor agonists produced significant increases in the total amounts of non-REM and REM sleep [49].

A dysregulation in sleep patterns is highly associated with many psychiatric disorders, including major depression, and more than 70% of all depressed patients report difficulties in onset or in maintenance of sleep [50–52]. In a study including 1423 subjects, a variation in the nucleoside transporter gene (SLC29A3) was reported to compromise adenosine transport and predispose women with sleep disturbances to depression [29]. Depression may also be observed as an effect of sleep disorders, which highlights an important relationship of reciprocity between them. Thus, we can speculate that impairment in A2A receptor expression and function in individuals carrying the C allele of the rs2298383 SNP might be associated with predisposition to sleep dysfunction and, consequently, depressive episodes.

In addition, in the present study, we observed that TT genotype of the rs2298383 SNP was also associated with a reduced impairment in concentration, another important symptom frequently observed in patients with depression. Attention and concentration are behavioral substrates particularly sensitive to the modulation by adenosine. Pharmacological manipulation of adenosine receptors plays a crucial role for concentration and decision-making [53, 54]. In fact, blockade of A2A receptors by caffeine and other A2A receptor antagonists is effective in preventing attentional and memory deficits in an animal model of attention deficit/hyperactivity disorder [55–57]. Thus, despite the well-established beneficial role of A2A receptor antagonists in attention, in our study, individuals carrying the C allele, associated with reduced expression of the A2A receptors, were more prompt to deficits in attention and depression. A functional evaluation of the A2A receptor in these individuals is still required, but we can speculate the normal levels of A2A receptors are important to attentional regulation.

Since the ability of the adenosinergic signaling to regulate sleep function, attention, and mood, our results highlight a relevant aspect of genetic variants in the ADORA2A in depression. Our study has some limitation, especially the fact that the functional consequences of these polymorphisms to expression and function of A2A receptors were not determined yet. In addition, with our cross-sectional design, we cannot evaluate a causal effect of impairments in sleep and attention and susceptibility to depression or the consumption of caffeine in individuals with different genotypes and consequences to mood regulation. In this context, further clinical studies should be developed to better comprehend the role of ADORA2A genes in depression. Future studies will be needed to address some of the limitations inherent in our cross-sectional design and to fill the gaps about the functional impact of this SNP on depressive disorder.

Authors’ contributions

All authors mentioned in the paper have significantly contributed to the research. RAS, KJ, LDMS, and DRL conceived and supervised the clinical evaluation. MPK and GG and supervised the collection and processing of biological samples. CRB and SO performed the DNA extraction and genotyping. SO, APA, GG, and MPK performed the statistical analysis and wrote the article. All authors approved the final manuscript.

Funding information

This study was supported by Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq), Coordenação de Aperfeiçoamento de Pessoal de Ensino Superior (CAPES), and Programa de Apoio a Núcleos de Excelência-Fundação de Amparo a Pesquisa do Estado do Rio Grande do Sul (PRONEX-FAPERGS 08/2009 - Pronex 10/0055-0). RAS, KJ, LDMS, and MPK are CNPq Research Fellows. APA and CRB receive a fellowship from CAPES.

Conflicts of interest

Sílvia Oliveira declares no conflict of interest.

Ana Paula Ardais declares no conflict of interest.

Clarissa Ribeiro Bastos declares no conflict of interest.

Marta Gazal declares no conflict of interest.

Karen Jansen declares no conflict of interest.

Luciano de Mattos Souza declares no conflict of interest.

Ricardo Azevedo da Silva declares no conflict of interest.

Manuella Pinto Kaster declares no conflict of interest.

Diogo Rizzato Lara declares no conflict of interest.

Gabriele Ghisleni declares no conflict of interest.

Ethical approval

The study was approved by the Ethical Committee of the Catholic University of Pelotas, Brazil (protocol number 2010/15), and all participants signed the informed consent.