Adenosine Hypothesis of Antipsychotic Drugs Revisited: Pharmacogenomics Variation in Nonacute Schizophrenia

Abstract

The existing antipsychotic therapy is based on dopamine hyperfunction and glutamate hypofunction hypotheses of schizophrenia. Adenosine receptors (ADORA) have a neuromodulatory role and can control dopaminergic and glutamatergic systems. To elucidate the effect of ADORA polymorphisms on psychopathological symptoms and adverse effects in patients with schizophrenia on long-term antipsychotic treatment, we examined 127 nonacute schizophrenia outpatients in a cross-sectional study using the Positive and Negative Symptoms Scale, Simpson-Angus Scale, Barnes Akathisia Rating Scale, and Abnormal Involuntary Movement Scale. All patients were genotyped for 18 polymorphisms in ADORA1, ADORA2A, and ADORA3. We found an association between ADORA1 rs3766566 and psychopathological symptoms (p = 0.006), in particular, with positive psychopathological symptoms (p = 0.010) and general psychopathological symptoms (p = 0.023), between ADORA2A rs2298383 and general psychopathological symptoms (p = 0.046), and between ADORA2A rs5751876 and akathisia (p = 0.015). Haplotype analysis showed an association between ADORA1 CTCAACG haplotype and overall psychopathological symptoms (p = 0.019), positive psychopathological symptoms (p = 0.021), and akathisia (p = 0.028). ADORA2A TCCTC haplotype was associated with parkinsonism (p = 0.014). ADORA3 CACTAC was associated with akathisia (p = 0.042), whereas CACTAT was associated with akathisia (p = 0.045) and tardive dyskinesia (p = 0.023). The results of this first comprehensive study on ADORA polymorphisms in patients with nonacute schizophrenia receiving long-term antipsychotic therapy suggest an important neuromodulatory role of ADORA receptors in both psychopathological symptoms and adverse effects of antipsychotics.

Introduction

Schizophrenia spectrum disorders are chronic psychotic disorders characterized by periods of remissions and relapses, and they usually require prolonged treatment. The current pharmacotherapy is mostly shaped by the two leading pathophysiological hypotheses of schizophrenia, that is, the dopamine hyperfunction and glutamate hypofunction hypotheses (Boison et al., 2012). These neurotransmitter pathways have been extensively studied, both biochemically and genetically, and were linked to morphological loci in the brain tissue, such as striatum in basal ganglia. The dorsal striatum receives motor information (dopaminergic afferent pathways from substantia nigra and glutamatergic afferent pathways from neocortex), whereas ventral striatum receives limbic information (dopaminergic afferent pathways from substantia nigra and glutamatergic afferent pathways from hippocampal formation, primary olfactory cortex, medial prefrontal cortex, and basolateral amygdaloid complex) (Ferre, 1997).

Positive symptoms of schizophrenia, such as delusions and hallucinations, are considered to be mainly related to dopamine system dysfunction, involving five known dopamine receptors (DRD). Among those, DRD2 has been demonstrated to have the biggest role in the occurrence of psychopathological symptoms and it is the main target of typical antipsychotic drugs, although while controlling those symptoms, they can also cause a substantial number of adverse effects (e.g., extrapyramidal symptoms, akathisia, and tardive dyskinesia) (Malhotra et al., 1993).

It has been observed that DRD1 and DRD2 form heterodimers with adenosine receptors (ADORA) ADORA1 and ADORA2A, respectively (Ferre et al., 2006). The ligand for ADORA is purine nucleoside adenosine that acts as a homeostatic modulator and is involved in an array of intracellular processes and intracellular signaling pathways (Cunha, 2001; Gomes et al., 2011). Adenosine also has a role as a neuromodulator in the central neural system, where it conducts the signal through neuronal circuits through neurotransmitter release control and plays a part in the process of neuronal excitability at both pre- and postsynaptic levels (Cunha, 2001; Fredholm et al., 2005; Gomes et al., 2011).

In neurons and neuroglia of the brain and spinal cord, four subtypes of ADORA (ADORA1, ADORA2A, ADORA2B, ADORA3) are expressed ( Burnstock et al., 2011; Burnstock and Verkhratsky, 2012; Fredholm et al., 2005), with low levels of both ADORA2B and ADORA3, as demonstrated through biochemical studies (Linden et al., 1993; Salvatore et al., 1993; Stehle et al., 1992). The dominating adenosine receptor is ADORA1, with its major action as a presynaptic inhibitor of the release of neurotransmitters (Burnstock and Verkhratsky, 2012).

Data on the adenosine systems in patients with schizophrenia are limited (Zhang et al., 2012). ADORA can modulate both dopaminergic and glutamatergic systems, and their neuromodulatory role supports the novel adenosinergic hypothesis of schizophrenia, which provides a link between the two aforementioned hypotheses (Gomes et al., 2011). Several indirect findings suggest that the adenosinergic activity might be deficient in schizophrenia (Gomes et al., 2011). It is noteworthy that activation of ADORA2A receptor reduces the effectiveness of DRD2 signaling, being the probable mechanism underlying the antipsychotic-like effects of adenosine agonists (Ferre, 1997; Ferré et al., 1994; Gomes et al., 2011).

Genetic studies support the adenosine hypofunction hypothesis of schizophrenia. Genetic variability of the ADORA2A gene was found to be associated with schizophrenia susceptibility (Childs et al., 2008; Deckert et al., 1996, 1997; Gomes et al., 2011; Hong et al., 2005). In addition, ADORA1 polymorphisms may also represent a good candidate marker for schizophrenia since some associations with the disorder were previously reported in Japanese populations (Gotoh et al., 2009). Adenosine-mediated regulation of synaptic transmission in the central nervous system is most likely a result of the balance between the activation of ADORA1 and ADORA2A receptors, as the latter is also involved in synaptic control (Fredholm et al., 2005). Although ADORA3 is not biochemically colocalized with any DRD, we nevertheless decided to study its polymorphisms as its structure–activity relationship is very similar to ADORA1 (Salvatore et al., 1993).

Taking into account these evidences, the aim of our study was to elucidate if single-nucleotide polymorphisms (SNPs) in ADORA genes influence schizophrenia psychopathology in nonacute patients on long-term antipsychotic treatment. As a secondary aim, we investigated if the same variants were associated with the antipsychotic-induced extrapyramidal symptoms known to be related to the effects on the dopamine system.

Materials and Methods

Study design

The design of the study was cross sectional. Patients were recruited in a naturalistic setting from outpatient clinic population during their regular monthly visits, as previously described elsewhere (Dolzan et al., 2007; Plesnicar et al., 2006). They were all of Central European Caucasian (Slovenian) origin, between 14 and 78 years of age, and treated with antipsychotics for schizophrenia, schizophreniform disorder, or schizoaffective disorder as defined by DSM IV (American Psychiatric Association, 1994). The patients with current and past substance abuse or dependence, and/or with hepatic diseases were not included in the study. The Slovenian Ethics Committee for Research in Medicine approved this study and it was subsequently carried out according to the Declaration of Helsinki. Only the patients who agreed to participate in the study and provided written informed consent were included.

Patients were examined by an experienced psychiatrist who was blinded to the results of genotyping. Psychopathological symptoms were assessed with the Positive and Negative Symptoms Scale (PANSS) (Kay et al., 1987, 1989) in the last 2 months, and in the same period, the drug dosage was not to be altered. The following three PANSS subscales were evaluated: PANNSP to rate positive psychopathological symptoms such as delusions, conceptual disorganization, hallucinatory behavior, excitement, grandiosity, suspiciousness/persecution, and hostility, PANSS Negative to measure negative psychopathological symptoms (blunted affect, emotional withdrawal, poor rapport, passive/apathetic social withdrawal, difficulty in abstract thinking, lack of spontaneity and flow conversation, and stereotyped thinking), and Positive and Negative Symptoms Scale General (PANSSG) to rate general psychopathological symptoms, including somatic concern, anxiety, guilt feelings, tension, mannerisms and posturing, depression, motor retardation, uncooperativeness, unusual thought content, disorientation, poor attention, lack of judgment and insight, disturbance of volition, poor impulse control, preoccupation, and active social avoidance.

Parkinsonism was assessed with the Simpson-Angus Scale (Simpson and Angus, 1970) and diagnosed if a patient had a total score of >0.2. Akathisia was assessed with the Barnes Akathisia Rating Scale (BARS) (Barnes, 1989) and diagnosis was made if a patient scored at least “mild” on BARS. Tardive dyskinesia symptoms were assessed with the Abnormal Involuntary Movement Scale (Guy, 1976). The latter diagnosis was made if a patient scored at least “moderate” abnormal involuntary movements in one or more of the seven body areas, or at least “mild” in two or more areas. All patients were genotyped for a panel of SNPs in ADORA1, ADORA2A, and ADORA3.

SNP selection

The selection of SNPs was based on putative functionality of the polymorphisms from the National Center for Biotechnology Information SNP database and/or the literature, and block-tagging ability (Xu and Taylor, 2009). A total of 18 SNPs were selected on the basis of the information provided by the International HapMap Project Utah residents with ancestry from northern and western Europe (CEU; CEPH) data (Phase II, March 2008) with the following criteria: r² > 0.8 and the frequency of a minor allele greater than 5%. SNP selection was carried out with Tagger pair-wise mode using HaploView software version 4.1. (Barret et al., 2005)

The selected and analyzed SNPs were as follows:

• seven SNPs in ADORA1 (rs1874142, rs10920568, rs3766566, rs3766560, rs3753472, rs3766553, rs12744240),

• five SNPs in ADORA2A (rs2298383, rs2236624, rs5751876, rs35320474, rs17004921), and

• six SNPs in ADORA3 (rs3394, rs3393, rs2229155, rs35511654, rs1544223, rs2298191).

ADORA genotyping

Per patient, a total of 10 mL of peripheral blood was collected into tubes with sodium citrate, and then stored short term at −20°C until DNA isolation. Genomic DNA was isolated from peripheral blood leukocytes using a standard salting-out procedure (Miller et al., 1988). Genotyping was performed using a fluorescence-based competitive allele-specific assay (KASPar; Kbiosciences). PCRs were performed using ABI 7900HT (Applied Biosystems) in a 4 μL reaction mix containing 0.055 μL of KASPar allele-specific PCR assay, 2 μL of KASPar PCR Master Mix (Kbiosciences), 2.2 mM MgCl₂, and 10 ng of DNA. In all reactions, nontemplate controls were included.

Statistical analyses

Statistical analyses were performed with the statistical software STATISTICA 7 for Windows (StatSoft, 1995). Basic descriptive statistics were used to show demographic and clinical features of the sample, including age (years), number of hospitalizations, duration of illness (years), duration of antipsychotic treatment (years), duration of current antipsychotic treatment (months), antipsychotic dose (converted to chlorpromazine equivalents) (Kroken et al., 2009; WHO Collaborating Centre For Drug Statistics Methodology, 2005), and clinical outcomes.

Categorical variables were analyzed with the χ² test. Associations between categorical and continuous variables were tested with the one-way analysis of variance (ANOVA). If the level of significance was <0.05, multivariate analysis of covariance (MANCOVA) was used to analyze covariance and exclude possible confounding factors. Deviation from the Hardy–Weinberg equilibrium (HWE) was calculated for each SNP using the standard χ² test. The Benjamini–Hochberg false discovery rate was used to account for multiple comparisons (Benjamini and Hochberg, 1995) and p-values less than 0.05 were considered significant after correction.

Haplotype analysis

ADORA1, ADORA2A, and ADORA3 haplotype blocks and linkage disequilibrium between the selected SNPs were illustrated using HaploView software version 4.1 (Barret et al., 2005). Frequencies of alleles and the HWE were established. Furthermore, to study the effect of genotypes and alleles on clinical outcomes, one-way ANOVA was performed within a linear regression model using R software (cran.r-project.org). Multivariable analysis included relevant clinical covariates.

Results

Patient characteristics

We recruited 53 male and 74 female outpatients (n = 127), diagnosed with schizophrenia (n = 101, 79.5%), schizoaffective disorder (n = 11, 8.6%), or other psychotic disorders (n = 15, 11.9%), and treated with maintenance antipsychotic therapy: oral risperidone (n = 21, 16.5%), oral haloperidol (n = 2, 1.6%), haloperidol decanoate (n = 6, 4.7%), oral fluphenazine (n = 5, 3.9%), fluphenazine decanoate (n = 21, 16.5%), oral zuclopenthixol (n = 1, 0.8%), zuclopenthixol decanoate (n = 69, 54.3%), or oral thioridazine (n = 2, 1.6%). Demographic and clinical data are shown in Table 1.

Table 1.

Demographic and Clinical Features of the Patients (n = 127) in the Study

	Minimum	Maximum	Mean	SD	95% CI
Age (years)	14	78	43.85	13.17	41.54–46.16
Hospitalizations (n)	1	24	5.28	4.97	4.40–6.15
Duration of illness (years)	1	38	10.83	7.11	9.59–12.08
Duration of therapy (years)	0.25	14	3.68	3.10	3.14–4.23
Current treatment duration (months)	3	168	44.20	37.10	37.75–50.65
Treatment dose (chlorpromazine equivalents mg/day)	25	996	210.54	161.20	182.23–238.84
PANSS (total)	30	91	46.29	11.41	44.29–48.30
PANSSP	7	26	9.36	3.22	8.80–9.93
PANSSN	0	30	13.85	4.35	13.09–14.61
PANSSG	16	42	23.04	5.36	22.10–23.98
Parkinsonism (SAS-total)	0	14	1.33	3.07	0.79–1.87
Akathisia (BARS-total)	0	9	0.27	1.31	0.04–0.50
Tardive dyskinesia (AIMS-total)	0	19	1.68	3.88	0.99–2.36

AIMS, Abnormal Involuntary Movement Scale; BARS, Barnes Akathisia Rating Scale; PANSS, Positive and Negative Symptoms Scale; PANSSG, Positive and Negative Symptoms Scale General; PANSSN, Positive and Negative Symptoms Scale Negative; PANSSP, Positive and Negative Symptoms Scale Positive; SAS, Simpson-Angus Scale.

All patients were successfully genotyped for all the investigated polymorphisms and their allele and genotype frequencies are shown in Supplementary Table S1 (all frequencies were in HWE).

The influence of ADORA polymorphisms on psychopathology

The associations between the investigated SNPs and psychopathology features are shown in Table 2.

Table 2.

Associations Between Categorical (Single-Nucleotide Polymorphisms) and Continuous Variables (PANSS, PANSSP, PANSSN, PANSSG, SAS, BARS, AIMS)

p (ANOVA)	rs number	MAF	PANSS	PANSSP	PANSSN	PANSSG	SAS	BARS	AIMS
ADORA1	rs1874142	0.35	0.941	0.758	0.766	0.833	0.343	0.862	0.657
	rs10920568	0.37	0.719	0.232	0.891	0.757	0.62	0.991	0.484
	rs3766566	0.19	0.006 ^*	0.010 ^*	0.110	0.023 ^*	0.158	0.535	0.508
	rs3766560	0.12	0.819	0.438	0.717	0.720	0.662	0.613	0.800
	rs3753472	0.33	0.729	0.854	0.578	0.773	0.433	0.25	0.797
	rs3766553	0.42	0.772	0.921	0.765	0.687	0.732	0.886	0.837
	rs12744240	0.10	0.630	0.396	0.589	0.716	0.941	0.958	0.871
ADORA2A	rs2298383	0.31	0.181	0.448	0.376	0.046 ^*	0.212	0.501	0.151
	rs2236624	0.20	0.381	0.956	0.256	0.125	0.602	0.830	0.833
	rs5751876	0.29	0.712	0.700	0.657	0.479	0.763	0.015 ^*	0.272
	rs35320474	0.28	0.672	0.442	0.717	0.468	0.117	0.264	0.102
	rs17004921	0.12	0.535	0.830	0.704	0.427	0.357	0.932	0.533
ADORA3	rs3394	0.26	0.395	0.520	0.426	0.188	0.677	0.782	0.723
	rs3393	0.44	0.845	0.486	0.959	0.953	0.958	0.830	0.834
	rs2229155	0.25	0.635	0.604	0.258	0.824	0.653	0.733	0.750
	rs35511654	0.15	0.322	0.066	0.696	0.218	0.263	0.965	0.354
	rs1544223	0.41	0.233	0.089	0.757	0.199	0.725	0.856	0.591
	rs2298191	0.33	0.652	0.428	0.978	0.845	0.337	0.112	0.766

^*p-values before correction for multiple testing. Bold = statistically significant associations (p < 0.05).

ADOR, adenosine receptor; ANOVA, analysis of variance; MAF, minor allele frequency.

There was an association between ADORA1 rs3766566 and overall psychopathological symptoms as measured by the PANSS total score (p = 0.006). Exploratory analysis showed a main effect on positive symptoms (p = 0.010) and general symptoms (p = 0.023). Furthermore, there was an association between ADORA2A rs2298383 and general symptoms (p = 0.046).

All of the detected associations remained significant in the multivariate analysis after adjustment for age and gender (MANCOVA: p = 0.00187, p = 0.00008, p = 0.01522, and p = 0.01138, respectively), and remained significant after testing for Benjamini–Hochberg corrected level of significance (p < 0.05).

The influence of ADORA polymorphisms on extrapyramidal adverse effects

The associations between the investigated SNPs and extrapyramidal adverse effects are shown in Table 2.

ADORA2A rs5751876 was associated with akathisia (p = 0.015) and remained significant after testing for Benjamini–Hochberg corrected level of significance (p < 0.05).

The statistically significant associations between genotypes and scores are shown in detail in Table 3.

Table 3.

Mean (SD) Total PANSS, PANSSP, PANSSG, and BARS Scores in Genotypes Significantly Associated with the Outcomes

	rs number [location]	Genotype	PANSS, mean (SD)	PANSSP, mean (SD)	PANSSG, mean (SD)	BARS, mean (SD)
ADORA1	rs3766566	CC	47.51 (11.13)	9.58 (3.22)	23.64 (5.57)
	c.341 + 7045G>A	CT	41.95 (8.86)	8.36 (1.80)	21.13 (3.70)
	[intron]	TT	53.00 (7.04)	11.20 (0.75)	25.60 (3.07)
ADORA2A	rs2298383	TT			25.00 (8.06)
	c.-275 + 1797C>T	CT			22.62 (4.97)
	[intron]	CC			22.54 (3.97)
	rs5751876	TT				1.21 (2.98)
	c.1083T>C	CT				0.14 (0.74)
	[p.Tyr361 = ]	CC				0.16 (0.90)

Association of haplotypes of ADORA1, ADORA2A, and ADORA3 with clinical outcomes

Haplotype analysis was performed on 12 ADORA1, six ADORA2A, and eight ADORA3 haplotypes (26 in total). All the haplotype frequencies and their associations with clinical outcomes are shown in Supplementary Table S2.

Four haplotypes were significantly correlated with clinical outcomes (Table 4). Particularly, the ADORA1 CTCAACG haplotype was associated with a higher PANSS total score (p = 0.019) and with positive psychopathological symptoms (p = 0.021). This haplotype was also associated with the presence of akathisia (p = 0.028). The ADORA2A TCCTC haplotype was associated with parkinsonism (p = 0.014). ADORA3 CACTAC and CACTAT haplotypes were associated with akathisia (p = 0.042, p = 0.045, respectively). The CACTAT haplotype was also associated with involuntary movement severity (p = 0.023). The influence remained significant after testing for Benjamini–Hochberg corrected level of significance (p < 0.05) for all associations, except for ADORA1 CTCAACG with the PANSS total score and ADORA2A TCCTC with parkinsonism.

Table 4.

Haplotype Frequencies of ADORA1, ADORA2A, and ADORA3, and Their Effect on Psychopathology Assessed with PANSS, and Extrapyramidal Side Effects Assessed with BARS, SAS, AIMS (p < 0.05)

Gene	Haplotype	Haplotype frequency		Haplotype score	p^*
ADORA1	CTCAACG	0.037	PANSS	2.35156	0.019
ADORA1	CTCAACG	0.037	PANSSP	2.31672	0.021
ADORA1	CTCAACG	0.037	BARS	2.20123	0.028
ADORA2A	TCCTC	0.517	SAS	−2.46754	0.014
ADORA3	CACTAC	0.267	BARS	−2.03258	0.042
ADORA3	CACTAT	0.181	BARS	2.00767	0.045
ADORA3	CACTAT	0.181	AIMS	2.27624	0.023

The single-nucleotide polymorphisms are ordered from 5′- to 3′- end (ADORA1 and ADORA2A) and from 3′- to 5′- end (ADORA3) as follows:

ADORA1: rs1874142 (chr1:203094406), rs10920568 (chr1:203098275), rs3766566 (chr1:203105355), rs3766560 (chr1:203118394), rs3753472 (chr1:203131533), rs3766553 (chr1:203133042), rs12744240 (chr1:203135674).

ADORA2A: rs2298383 (chr22:24825511), rs2236624 (chr22:24836024), rs5751876 (chr22:24837301), rs35320474 (chr22:24837909), rs17004921 (chr22:24840676).

ADORA3: rs3394 (chr1:112042131), rs3393 (chr1:112042149), rs2229155 (chr1:112042632), rs35511654 (chr1:112042787), rs1544223 (chr1:112046557), rs2298191 (chr1:112048264).

^*p-values before correction for multiple testing.

Discussion

We report for the first time the association between genetic polymorphisms of three ADORA (ADORA1, ADORA2A, and ADORA3) and both psychotic and extrapyramidal symptoms in schizophrenia patients receiving long-term maintenance antipsychotic treatment.

We found the correlation between ADORA1 rs3766566 and psychopathological symptoms as assessed by PANSS, particularly with positive and general psychopathological symptoms. This is a new finding, as there are no reports that this SNP could be associated with psychopathological symptoms and extrapyramidal side effects. We observed no association of ADORA1 SNPs with negative psychopathological symptoms or extrapyramidal side effects. However, ADORA1 haplotype CTCAACG was found to be associated with akathisia.

In our study, ADORA2A rs5751876 was associated with akathisia, whereas ADORA2A rs2298383 influenced general psychopathology. In silico analysis of rs2298383 has shown that this SNP is potentially functionally relevant since it may influence the regulation of gene expression (Goni et al., 2004). The highest expression of ADORA2A is in the striatum, where their highest levels are found in dendrites and dendritic spines that form asymmetric synapses that receive input from glutamatergic terminals and are excitatory in nature (Burnstock and Verkhratsky, 2012). It is suggested that ADORA2A combined with ADORA1 contributes to the hypoadenosinergistic state that may attribute to the pathophysiology of schizophrenia (Akhondzadeh et al., 2000; Lara, 2002; Lara et al., 2006; Salvatore et al., 1993). Furthermore, ADORA2A rs2298383 has been linked to caffeine-induced anxiety (Childs et al., 2008; Rogers et al., 2010) and was also associated with the pathogenesis of anxiety disorders and anxious personality (Hohoff et al., 2010).

General psychopathology in schizophrenia as measured with the PANSSG subscale includes assessment of anxiety, and our finding about ADORA2A rs2298383 may additionally confirm its potential role in evoking anxiety symptoms. Also, following treatment with antipsychotics, a significant upregulation of ADORA2A expression was found, and studies suggest that ADORA2A oversignaling results in typical hypokinetic symptoms of Parkinson's disease, because of the net result of dopamine depletion in the striatum (Gomes et al., 2011). This is consistent with our results that showed the effect of ADORA2A TCCTC haplotype on parkinsonism, although this finding is weak as it did not remain significant after adjustment for multiple comparisons. This may be due to the absence of severe symptoms of parkinsonism in most patients (Table 1).

Although ADORA3 in the brain is not extensively studied yet, there are some findings that show its presence in the striatum (Borea et al., 2009; Salvatore et al., 1993). In our study, none of the SNPs in ADORA3 was correlated to psychopathological symptoms or adverse effects of the antipsychotics. However, ADORA3 haplotype analysis revealed that three haplotypes did have an effect on extrapyramidal side effects and may present a higher risk for EPS. Particularly, haplotype CACTAC was associated with akathisia, while CACTAT was associated with akathisia and involuntary movements. This finding is valuable as it shows the importance of a combination of SNPs that could be used as markers, although more evidence is needed to confirm this assumption.

The role of ADORA3 gene was hypothesized to be mainly attributed to ischemic and neuroinflammatory disorders in the brain (Borea et al., 2009). Our finding on ADORA3 haplotypes CACTAC and CACTAT may further support the role of neuroinflammation in the pathogenesis of schizophrenia. Indeed, van der Putten et al. (2009) reported that ADORA3 may act as a suppressor of ADORA2A-mediated inhibition within microglia through toll-like receptor-induced responses. Toll-like receptors in the brain and the innate immune responses of microglia are mediated with extracellular heat shock 70-kDa proteins (HSP70s) that have been associated with the pathophysiology of schizophrenia (Reus et al., 2015). Furthermore, other recent studies showed that cytokines, cellular immune components, and glial cells are in causal relationship with the changes in neurobiology and behavior (Bergink et al., 2014; Singhal et al., 2014). A study using N-ethylcarboxamidoadenosine (i.e., NECA, a nonselective ADORA agonist, more potent than adenosine) showed that ADORA2A-mediated signaling generates opposite effects from ADORA3-mediated signaling and, in addition, found that even in the absence of ADORA3-mediated signaling, activated microglia were more sensitive to ADORA2A-mediated inhibition (van der Putten et al., 2009).

Studies in patient cohorts and in various animal models on microglia, monocytes, and their products strongly suggest a key role for these cells of the mononuclear phagocyte system in the pathogenesis of major psychiatric disorders, including schizophrenia (Beumer et al., 2012a). Patients with chronic schizophrenia and not acute schizophrenia were reported to clearly show an activated monocyte/macrophage system as evidenced by raised serum levels of the chemokines (CCL2, CCL4) and the cytokines (IL-1β, TNF-α, IL-6, PTX3), where the most dominant linkage was found with the disease state of schizophrenia itself (Beumer et al., 2012b). Reports on the effects of antipsychotics on cytokine levels have been conflicting and no antipsychotic has been shown to have consistent anti-inflammatory or proinflammatory action (Na et al., 2014).

Further studies are needed to explore the role of genetic variability in ADORA2A and ADORA3. Nevertheless, we can speculate that the association between side effects of antipsychotics and ADORA3 polymorphisms could be explained in two ways: (1) either this is not a direct correlation, but occurs through the activity of ADORA2A signaling, (2) or it is a result of neurodegeneration as a consequence of long-term neuroinflammation.

There are some limitations of our study. We did not control the results for false-positive findings using a more conservative Bonferroni correction because of a relatively small sample, however, it was genetically homogenous and the patients were clinically well defined. Baseline data were not available due to the cross-sectional design of the study, and therefore, it was difficult to distinguish if the observed polymorphisms are related with the psychopathological symptoms or with the response to the treatment. To differentiate this relationship, it would be necessary to measure PANSS before the beginning of the treatment, however, the inclusion criteria demanded stable psychopathological symptoms and no therapy alterations at least 2 months prior the enrolment in the study, so that was not possible. Also, a possible modification of antipsychotics in the past, when the patients could have had more severe extrapyramidal side effects with other antipsychotics, was not among exclusion criteria and could influence the results (especially on parkinsonism), but nevertheless, there were also patients with extrapyramidal symptoms and those were clinically well assessed. The antipsychotics used in treatment were different, but the majority of patients were receiving intramuscular depot therapy, and therefore, the compliance issues are not considered to be a limitation.

Conclusions

The results of this first study on ADORA polymorphisms in nonacute patients with schizophrenia suggest an important neuromodulatory role of ADORA receptors in both psychopathological symptoms and adverse effects of antipsychotics. ADORA1 rs3766566 and ADORA2A rs2298383 and rs5751876, as well as four haplotypes (ADORA1 CTCAACG, ADORA2A TCCTC, and ADORA3 CACTAC and CACTAT), may be linked to both schizophrenia psychopathology (mainly with positive and general psychopathological symptoms) and to antipsychotic-induced extrapyramidal symptoms. An interesting effect of ADORA3 CACTAC and CACTAT haplotypes on akathisia and involuntary movements requires replication in further studies.

Footnotes

Acknowledgments

The authors thank Jure Koprivsek at University Clinical Centre Maribor, Maribor, Slovenia, for helping in recruitment of patients, and Matej Kastelic at University of Ljubljana, Ljubljana, Slovenia, for his help on genetic analysis. Also, the authors thank Nenad Bartonicek at Garavan Institute of Medical Research, Sydney Area, Australia, for advice on statistical analysis.

Author Disclosure Statement

No competing financial interests exist.

Abbreviations Used

References

Akhondzadeh

, Shasavand

, Jamilian

, Shabestari

, and Kamalipour

. (2000). Dipyridamole in the treatment of schizophrenia: Adenosine-dopamine interactions. J Clin Pharm Ther, 25, 131–137.

American Psychiatric Association. (1994). Diagnostic and statistical manual of mental health disorders, 4th ed., Washington, DC.

Barnes

. (1989). A rating scale for drug-induced akathisia. Br J Psychiatry, 154, 672–676.

Barret

, Fry

, Maller

, and Daly

. (2005). Haploview: Analysis and visualization of LD and haplotype maps. Bioinformatics, 21, 263–265.

Benjamini

, and Hochberg

. (1995). Controlling the false discovery rate: A practical and powerful approach to multiple testing. J R Stat Soc B, 57, 289–300.

Bergink

, Gibney

, and Drexhage

. (2014). Autoimmunity, inflammation, and psychosis: A search for peripheral markers. Biol Psychiatry, 75, 324–331.

Beumer

, Drexhage

, De Wit

, Versnel

, Drexhage

, and Cohen

. (2012a). Increased level of serum cytokines, chemokines and adipokines in patients with schizophrenia is associated with disease and metabolic syndrome. Psychoneuroendocrinology, 37, 1901–1911.

Beumer

, Gibney

, Drexhage

, et al. (2012b). The immune theory of psychiatric diseases: A key role for activated microglia and circulating monocytes. J Leukoc Biol, 92, 959–975.

Boison

, Singer

, Shen

, Feldon

, and Yee

. (2012). Adenosine hypothesis of schizophrenia–opportunities for pharmacotherapy. Neuropharmacology, 62, 1527–1543.

10.

Borea

, Gessi

, Bar-Yehuda

, and Fishman

. (2009). A3 adenosine receptor: Pharmacology and role in disease. In: Handbook of Experimental Pharmacology. Springer-Verlag, Berlin, Heidelberg. Vol. 193; 297–327. Eds: Wilson and Mustafa.

11.

Burnstock

, Krügel

, Abbracchio

, and Illes

. (2011). Purinergic signalling: From normal behaviour to pathological brain function. Prog Neurobiol, 95, 229–274.

12.

Burnstock

, and Verkhratsky

. (2012). Purinergic Signalling and the Nervous System. Springer-Verlag, Berlin, Heidelberg.

13.

Childs

, Hohoff

, Deckert

, Xu

, Badner

, and de Wit

. (2008). Association between ADORA2A and DRD2 polymorphisms and caffeine-induced anxiety. Neuropsychopharmacology, 33, 2791–2800.

14.

Cunha

. (2001). Adenosine as a neuromodulator and as a homeostatic regulator in the nervous system: Different roles, different sources and different receptors. Neurochem Int, 38, 107–125.

15.

Deckert

, Nöthen

, Bryant

, et al. (1997). Mapping of the human adenosine A2a receptor gene: Relationship to potential schizophrenia loci on chromosome 22q and exclusion from the CATCH 22 region. Hum Genet, 99, 326–328.

16.

Deckert

, Nöthen

, Rietschel

, et al. (1996). Human adenosine A2a receptor (A2aAR) gene: Systematic mutation screening in patients with schizophrenia. J Neural Transm, 103, 1447–1455.

17.

Dolzan

, Plesnicar

, Serretti

, et al. (2007). Polymorphisms in dopamine receptor DRD1 and DRD2 genes and psychopathological and extrapyramidal symptoms in patients on long-term antipsychotic treatment. Am J Med Genet B Neuropsychiatr Genet, 144B, 809–815.

18.

Ferre

. (1997). Adenosine-dopamine interactions in the ventral striatum. Psychopharmacology, 133, 107–120.

19.

Ferre

, Ciruela

, Canals

, et al. (2006). Adenosine A_2A–dopamine D₂ receptor-receptor heteromers. Targets for neuro-psychiatric disorders. Parkinsonism Relat Disord, 10, 265–271.

20.

Ferré

, O'Connor

, Snaprud

, Ungerstedt

, and Fuxe

. (1994). Antagonistic interaction between adenosine A2A receptors and dopamine D2 receptors in the ventral striopallidal system. Implications for the treatment of schizophrenia. Neuroscience, 63, 765–773.

21.

Fredholm

, Chen

, Cunha

, Svenningsson

, and Vaugeois

. (2005). Adenosine and brain function. Int Rev Neurobiol, 63, 191–270.

22.

Gomes

, Kaster

, Tome

, Agostinho

, and Cunha

. (2011). Adenosine receptors and brain diseases: Neuroprotection and neurodegeneration. Biochim Biophys Acta, 1808, 1380–1399.

23.

Goni

, de la Cruz

, and Orozco

. (2004). Triplex-forming oligonucleotide target sequences in the human genome. Nucleic Acids Res, 32, 354–360.

24.

Gotoh

, Mitsuyasu

, Kobayashi

, et al. (2009). Association analysis of adenosine A1 receptor gene (ADORA1) polymorphisms with schizophrenia in a Japanese population. Psychiatr Genet, 19, 328–335.

25.

Guy

. (1976). ECDEU Assessment Manual for Paychopharmacology: Revised. Rockville, MD. National Institute of Mental Health.

26.

Hohoff

, Mullings

, Heatherley

, et al. (2010). Adenosine A_2A receptor gene: Evidence for association of risk variants with panic disorder and anxious personality. J Psychiatr Res, 44, 930–937.

27.

Hong

, Liu

, Liao

, and Tsai

. (2005). Association studies of the adenosine A2a receptor (1976T>C) genetic polymorphism in Parkinson's disease and schizophrenia. J Neural Transm, 112, 1503–1510.

28.

Kay

, Fiszbein

, and Opler

. (1987). The positive and negative syndrome scale (PANSS) for schizophrenia. Schizophr Bull, 13, 261–276.

29.

Kay

, Opler

, and Lindenmayer

. (1989). The positive and negative syndrome scale (PANSS): Rationale and standardisation. Br J Psychiatry Suppl. 59–67.

30.

Kroken

, Johnsen

, Ruud

, Wentzel-Larsen

, and Jørgensen

. (2009). Treatment of schizophrenia with antipsychotics in Norwegian emergency wards, a cross-sectional national study. BMC Psychiatry, 9, 24.

31.

Lara

. (2002). Inhibitory deficit in schizophrenia is not necessarily a GABAergic deficit. Cell Mol Neurobiol, 22, 239–247.

32.

Lara

, Dall'Igna

, Ghisolfi

, and Brunstein

. (2006). Involvement of adenosine in the neurobiology of schizophrenia and its therapeutic implications. Prog Neuropsychopharmacol Biol Psychiatry, 30, 617–629.

33.

Linden

, Taylor

, Robeva

, et al. (1993). Molecular cloning and functional expression of a sheep A3 adenosine receptor with widespread tissue distribution. Mol Pharmacol, 44, 524–532.

34.

Malhotra

, Litman

, and Pickar

. (1993). Adverse effects of antipsychotic drugs. Drug Saf, 9, 429–436.

35.

Miller

, Dykes

, and Polesky

. (1988). A simple salting out procedure for extracting DNA from human nucleated cells. Nucleic Acids Res, 16, 1215.

36.

, Jung

, and Kim

. (2014). The role of pro-inflammatory cytokines in the neuroinflammation and neurogenesis of schizophrenia. Prog Neuropsychopharmacol Biol Psychiatry, 48, 277–286.

37.

Plesnicar

, Zalar

, Breskvar

, and Dolzan

. (2006). The influence of the CYP2D6 polymorphism on psychopathological and extrapyramidal symptoms in the patients on long-term antipsychotic treatment. J Psychopharmacol, 20, 829–833.

38.

Reus

, Fries

, Stertz

, et al. (2015). The role of inflammation and microglial activation in the pathophysiology of psychiatric disorders. Neuroscience, 300, 141–154.

39.

Rogers

, Hohoff

, Heatherley

, et al. (2010). Association of the anxiogenic and alerting effects of caffeine with ADORA2A and ADORA1 polymorphisms and habitual level of caffeine consumption. Neuropsychopharmacology, 35, 1973–1983.

40.

Salvatore

, Jacobson

, Taylor

, Linden

, and Johnson

. (1993). Molecular cloning and characterization of the human A3 adenosine receptor. Proc Natl Acad Sci USA, 90, 10365–10369.

41.

Simpson

, and Angus

. (1970). A rating scale for extrapyramidal side effects. Acta Psychiatr Scand Suppl, 212, 11–19.

42.

Singhal

, Jaehne

, Corrigan

, and Baune

. (2014). Cellular and molecular mechanisms of immunomodulation in the brain through environmental enrichment. Front Cell Neurosci, 8, 97.

43.

Stehle

, Rivkees

, Lee

, Weaver

, Deeds

, and Reppert

. (1992). Molecular cloning and expression of the cDNA for a novel A2-adenosine receptor subtype. Mol Endocrinol, 6, 384–393.

44.

van der Putten

, Zuiderwijk-Sick

, van Straalen

, et al. (2009). Differential expression of adenosine A3 receptors controls adenosine A2A receptor-mediated inhibition of TLR responses in microglia. J Immunol, 182, 7603–7612.

45.

WHO Collaborating Centre for Drug Statistics Methodology. (2005). ATC Index with DDD. Accessed March 2016 .

46.

, and Taylor

. (2009). SNPinfo: Integrating GWAS and candidate gene information into functional SNP selection for genetic association studies. Nucleic Acids Res, 37, W600–W605.

47.

Zhang

, Abdallah

, Wang

, et al. (2012). Upregulation of adenosine A2A receptors induced by atypical antipsychotics and its correlation with sensory gating in schizophrenia patients. Psychiatry Res, 200, 126–132.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.03 MB

0.02 MB