Abstract
Background:
Recent studies explored polymorphisms of multiple genes as contributing to genetic susceptibility to psychosis in Parkinson’s disease (PDP).
Objective:
We aimed to examine the association of seven selected polymorphisms of genes related to dopamine pathways with PDP development. At the same time, demographic and clinical correlates of PDP were assessed.
Methods:
PD patients (n = 234), treated with levodopa for at least two years, were genotyped for the rs4680 in COMT, rs6277, rs1076560, and rs2283265 in DRD2, and rs1800497 and rs2734849 polymorphisms in ANKK1 genes. Also, variable number of tandem repeats polymorphism in the DAT gene was examined. Each patient underwent comprehensive neurological examination, assessment of psychosis, as defined by the NINDS/NIMH criteria, as well as screening of depression, anxiety, and cognitive status.
Results:
Diagnostic criteria for PDP were met by 101 (43.2%) patients. They had longer disease duration, were taking higher doses of dopaminergic agents, and had higher scores of the motor and non-motor scales than the non-PDP group. Multivariate regression analysis revealed LEDD≥900 mg, Unified Parkinson’s Disease Rating Scale III part score, the Hamilton Depression Rating Scale score≥7, the Hamilton Anxiety Rating Scale score > 14,and GG homozygotes of rs2734849 ANKK1 as independent predictors of the onset of PDP.
Conclusion:
Besides previous exposure to dopaminergic drugs, impairment of motor status, depression and anxiety, as well-established clinical risk factors for the development of PDP, GG rs2734849 ANKK1 could also be a contributing factor, which requires addressing by future longitudinal studies.
INTRODUCTION
Psychosis in Parkinson’s disease (PDP) can occur in up to 75%of patients during the disease course, usually after 10 or more years of diagnosis and significantly contribute to poor outcome with increased mortality, nursing home placement, and caregiver stress [1, 2]. The spectrum of Parkinson’s disease (PD) psychotic symptoms includes minor hallucinations (illusions, presence hallucinations, passage hallucinations), formed visual hallucinations (VH), nonvisual hallucinations, and delusions [3, 4]. Once psychotic features occur, they tend to be persistent and progressive [5]. The pathophysiology of PDP is complex and still insufficiently known. It was considered that long-term dopaminergic therapy was a primary risk factor for PDP, but recent studies have shown than antiparkinsonian medications alone were neither necessary nor sufficient to cause PDP in all patients [6, 7]. Other identified risk factors for the development of PDP were increasing age, longer duration and higher severity of PD, cognitive impairment, visual-processing abnormalities, depression, and sleep disturbances [6–9]. Also, polymorphisms of several genes such as dopamine receptor genes, dopamine transporter genes, apolipoprotein E, cholecystokinin system, HOMER gene, gene for angiotensin converting enzyme, and COMT gene were investigated as a potential genetic substrate of psychotic symptoms in PD [10], but the evidence is still inconclusive.
We investigated the associations of variations in four dopamine pathway-related genes with psychotic symptoms in a sample of 234 Serbian PD patients treated with levodopa for at least two years. The genes selected for this study are encoding dopamine receptors D2(DRD2), dopamine transporter (DAT1, SLC6A3 –solute carrier family 6 member 3), COMT enzyme, and regulators of dopamine signaling - ANKK1 (ankyrin repeat and kinase domain containing 1). Dopamine pathway-related genes are important determinants of dopamine function, and polymorphisms in these genes could be implicated in the genetic susceptibility to PDP.
METHODS
This cross-sectional study included 234 patients with PD at the Clinic of Neurology, Clinical Centre of Serbia (Belgrade), from 2016 to 2019. The study was approved by the local Ethical Committee. The diagnosis of PD was made according to the United Kingdom Brain Bank criteria [11]. Patients were recruited according to the following inclusion criteria: age at onset > 40 years, and on levodopa therapy for at least two years. The main exclusion criteria were presence of delirium before the study initiation as well as history of previous psychiatric psychotic illnesses (schizophrenia, schizo-affective disorders, mood disorders with psychotic characteristics) and presence of symptoms and signs suggestive of atypical and secondary parkinsonism.
Following the written informed consent, all patients completed a specially constructed questionnaire containing the essential demographic characteristics (sex, age); disease-related data (age of onset, duration, and a form of the disease); therapy-related data (current therapy, medications used and doses); and complication-related data (presence of dyskinesias, motor fluctuations, psychotic symptoms). For all patients, the levodopa equivalent daily dose (LEDD) was calculated according to the standard formula [12].
The severity of parkinsonian signs was evaluated using the Unified Parkinson’s Disease Rating Scale (UPDRS) [13], while the Hoehn and Yahr PD staging scale (HY) [14] was used to determine the stage of the disease. Daily activities were assessed by the Schwab and England Activities of Daily Living Scale (S&E) [15]. Depression and anxiety were assessed by the Hamilton Depression Rating Scale (HDRS) [16] and the Hamilton Anxiety Rating Scale (HARS) [17], respectively, and for the cognitive evaluation, the Mini-Mental State Examination (MMSE) was applied [18]. The Non-Motor Symptoms Questionnaire for PD (NMSQuest) was used to estimate presence of non-motor PD symptoms [19]. The assessment of the presence of PPD was done by the standard NINDS/NIMH (National Institute of Neurological Disorders and Stroke/National Institutes of Mental Health) diagnostic criteria [20]. The study considered all psychotic manifestations, including minor phenomena, hallucinations, and delusions, taking thus into account the possibility that PDP is a continuum in which subtle symptoms progressively deteriorate.
All retrospective data provided by patients and caregivers were checked in available medical documentation (case histories and outpatient reports). Testing was performed during the “on” phase (the phase in which the optimal drug therapy and motor improvement were achieved in the patient).
Genotyping
Genetic testing was performed in the Laboratory for Genetic and Molecular Diagnosis of Neurological Diseases at Clinic of Neurology, Clinical Center of Serbia, Belgrade. Peripheral blood samples were obtained from 234 patients and 234 healthy controls for the purpose of genetic analysis. The control group included subjects who did not have PD, as well as other neurological and psychiatric diseases. The structure of the control group according to gender and age corresponds to the group of patients. A total of 4 ml of peripheral blood was collected and genomic DNA was isolated from peripheral blood by using standard methods. Genotyping of three polymorphisms in DRD2 gene (rs2283265, rs1076560, rs6277), two polymorphisms in the ANKK1 gene (rs1800497 and rs2734849), and one polymorphism in the COMT gene (Val158Met, rs4680) was performed using TaqMan SNP genotyping assays (Thermo Fisher Scientific, Foster City, CA) on the ABI Prism 7500 Fast Real-Time PCR System (Applied Biosystems, USA). The obtained results were processed using the 7500 Software program (Applied Biosystems, USA). VNTR polymorphism in the 3’UTR region of the DAT gene had been determined on 1%agarose gel electrophoresis.
Statistical analyses
Descriptive statistics were used to describe demographic and disease characteristics of patients. Categorical data are presented as numbers with percentages. Differences between groups were analyzed using Mann-Whitney U test for continuous variables, and the Pearson Chi-squared test for categorical variables. The univariate and multivariate logistic regressions were used to determine independent predictors for the development of psychosis in PD. Results were expressed as B, Wald Chi-Square, odds ratios (OR), and their 95%confidence intervals (CI). All tests were two-tailed. p < 0.05 was considered statistically significant. All analyses were conducted using the Statistical Package for the Social Sciences (IBM SPSS, version 21).
RESULTS
The demographic and clinical characteristics of the patients included in the study are presented in Table 1. Out of 234 PD patients, 101 (43.2%) patients met NINDS/NIMH criteria for the diagnosis of PDP. According to the results of the semi-structured interview minor hallucinations (MH) reported 24.7%, VH 35.9%, nonvisual hallucinations (auditory, olfactory, tactile) 20.9%, and delusion 14.1%. Auditory hallucinations were the most frequent among nonvisual (14.9%) and in only 6%of cases they were not accompanied by VH. Of the total number of patients with hallucinations 62.2%had retained insight.
Demographic, clinical and treatment data of the study group of PD patients (N = 234), as well as correlates of PDP and non-PDP patient group
Values are number of patients (percentage) and mean±standard deviation; NS, not significant *p values from Mann-Whitney U tests for continuous variables and the Pearson Chi-squared test for categorical variables comparing the two groups (patients with and without PD psychosis) are presented. Statistically significant difference is p < 0.05. Patients’ characteristics that differed significantly between groups (p < 0.05) are shown in boldface. UPDRS, Unified Parkinson‘s Disease Rating Score; HY stage, Hoenh and Yahr stage; S&E, Schwab-England Activities of Daily Living Scale; MMSE, Mini-Mental State Examination; HDRS, Hamilton Depression Rating Scale; HARS, Hamilton Anxiety Rating Scale; NMSQuest, Non-motor symptom questionnaire; LEDD, levodopa equivalent daily dose.
Significant differences between the PDP and non-PDP patients were observed for the age at onset and duration of PD (Table 1). PDP patients had significantly higher HY stage, S&E scores, higher scores of the UPDRS, both total and subscores. The HDRS, HARS, and NMSQuest total scores were significantly higher, while the mean MMSE scores were significantly lower in PDP group than in non-PDP group (p < 0.001, p < 0.001, p < 0.001, and p = 0.033, respectively). Patients with PDP more frequently reported sleeping problems like insomnia, vivid dreams, acting-out during dreams, and restless legs compared to non-PDP patients. There were no differences in the therapeutic profile between PDP and non-PDP group. Motor fluctuations and dyskinesias were more prevalent in PDP compared to non-PDP group. Qualitatively similar outcomes were obtained in the analysis that excluded patients on antipsychotics (Supplementary Table 1), demonstrating that use of this class of psychotropic drugs did not behave as a confounding factor.
Frequencies for the seven investigated genetic polymorphisms of COMT, DRD2, ANKK1, and DAT genes for the patients group as well as for the healthy controls are given in Table 2. The genotype distribution did not deviate significantly from Hardy–Weinberg equilibrium. There were no significant differences in genotype distributions of rs4680 COMT, rs1076560, rs2283265 DRD2, and rs1800497 ANKK1 genes when comparing PDP and non-PDP group of patients (p = 0.680, p = 0.850, p = 0.614, and p = 0.538, respectively). Also, there was no significant difference between carriers of 9 and 10 VNTR DAT and carriers of more than 10 VNTR regarding the presence of psychosis (p = 0.638). Also, we found no association between DAT VNTR 9 repeats and PDP (p = 0.287). Higher frequency of PDP was found among TT homozygotes of rs6277 DRD2 gene compared to CT and CC rs6277 carriers (p = 0.020). Also, we found higher frequency of GG genotype of rs2734849 ANKK1 gene among PDP patients (p = 0.009). The analysis that excluded patients on antipsychotics (Supplementary Table 2) demonstrated that this class of psychotropic drugs was not a confounding factor in the interpretation of genetic polymorphisms.
Genotype frequencies of COMT, DRD2, ANKK1, and DAT1 polymorphisms in PD patients with and without psychosis
Values are number of patients with frequencies shown in brackets. NS, not significant. SNP, single nucleotide polymorphism; PPD, Parkinson’s disease psychosis; COMT, catechol-O-methyltransferase, DRD2, dopamine receptor D2, ANKK1, Ankyrin repeat and kinase domain containing 1, DAT1, dopamine transporter 1, SLC6A3, Solute Carrier Family 6, member 3, VNTR, Variable Number of Tandem Repeats.
Univariate analysis revealed 20 statistically significant associations with PDP, as shown in Supplementary Table 3. Multivariate regression analysis was performed on 10 selected variables. The highly correlated variables were excluded, as inclusion of all such variables in a model would be redundant, not adding any extra information. Our multivariate analysis demonstrated that independent predictors of the onset of PDP were: LEDD≥900 mg (OR =2.041, 95%CI: 1.089–3.824, p = 0.026), the UPDRS part III score (OR = 1.025, 95%CI: 1.001–1.050, p = 0.044), the HDRS score≥7 (OR = 2.846, 95%CI: 1.289–6.286, p = 0.010), the HARS score > 14 (OR = 2.236, 95%CI: 1.061–4.710, p = 0.034), and GG homozygotes of rs2734849 ANKK1 gene (OR =2.588, 95%CI: 1.325–5.054, p = 0.005) (Table 3).
Logistic regression model examining predictors of psychosis
B, B coefficient; CI, confidence interval; LEDD, levodopa equivalent daily dose; UPDRS III, Unified Parkinson‘s Disease Rating Score, III part; MMSE, Mini-Mental State Examination; HDRS, Hamilton Depression Rating Scale; HARS, Hamilton Anxiety Rating Scale.
DISCUSSION
In our cohort, patients with PDP were younger at PD onset, had longer disease duration, were taking higher doses of levodopa, also higher LEDD, had more severe disease, lower MMSE scores and worse all other motor and non-motor scale scores (the UPDRS total and subscales, S&E, NMSQuest, HDRS, HARS) than those who were not psychotic. Among genetic factors, we found significant associations between rs6277 DRD2 gene and rs2734849 ANKK1 gene polymorphisms and PDP. The risk of developing psychosis was 2.3 times higher in TT rs6277 DRD2 carriers and 2.2 times higher in GG rs2734849 ANKK1 carriers. Multivariate analyses revealed that LEDD≥900 mg/day, motor status, depression, and anxiety, as well as GG genotype of rs2734849 ANKK1 were independent factors of PDP. Depression was the strongest independent risk factor that almost 3 times increased the risk of PDP.
Using the NINDS/NIMH criteria for PDP [20], 43.2%of our patients had PDP, similar to the previous data from literature (16–75%) [7, 22]. Assessment of drug-related risk factors in this study revealed that PDP patients were treated with higher doses of levodopa, and mean LEDD were significantly higher in PDP group when compared to non-PDP group. In multivariate model values of LEDD≥900 mg emerged as an independent predictor which doubled the risk of PDP. Contrary to our findings, some cross-sectional studies did not find significant difference in levodopa daily doses between PD patients with and without hallucinations [7, 24]. The concept of drug-induced psychosis in PD has been supported by the clinical observations that psychotic symptoms can be resolved with reduction or cessation of dopaminergic medication, although the data in the studies were not consistent. Goetz et al. were the first to show that VH did not relate simply to high levels of levodopa or to sudden changes in plasma levels [25], leading to a new concept that PDP was not purely “drug-induced”. There were some other factors that might be associated and contribute to PDP. This recent concept could be corroborated with findings of one recent longitudinal study that psychotic symptoms reemerged within less than 1 year in half of patients with PDP after dose reduction or initiation of antipsychotic drugs [26]. Also, hallucinations which were mainly auditory occurred in only 1%of patients treated with DA for pituitary tumors [27]. Furthermore, PDP cases were also reported in drug naïve PD patients [28]. In summary, antiparkinsonian medications alone were neither necessary nor sufficient to cause PDP and dopaminergic medications probably “triggered” psychotic symptoms in the context of underlying disease [5].
Since it has been known that dopaminergic system was implicated in psychiatric disorders such as drug addiction and schizophrenia, we wanted to evaluate the genetic variants of dopamine pathway-related genes as markers of susceptibility for PDP. We found that TT genotype of rs6277 DRD2 and GG genotype of rs2734849 ANKK1 gene were more frequent in PDP patients when compared to non-PDP.
ANKK1 and DRD2 genes are located close to each other on chromosome 11, and there is high linkage disequilibrium (LD) between SNPs in both genes [29]. SNP of genes selected for this study were important for D2 receptor functioning. Thus, T alleles rs6277 DRD2 gene have been associated with decreased affinity of the striatal D2 receptors. Polymorphisms rs1076560 and rs2283265 shifted splicing from D2 short (D2S) to D2 long (D2L) isoform, reducing thereby the expression of the D2S receptor variant. D2S is mainly expressed presynaptically, has autoinhibitory function and prevents excessive extracellular dopamine levels [30]. Finally, rs2734849 and rs1800497 polymorphisms are in the coding region of ANKK1 gene and they can affect DRD2 receptor density by altering the level of NF-κB expression, which is a necessary factor for DRD2 transcription [30]. Hence, these SNPs alter the density and affinity of D2 receptors, as well as synaptic dopamine levels.
To our knowledge, there has been only one study which explored similar SNP of DRD2/ANKK1 genes in the context of developing hallucinations in PDP [31]. This study showed significant association of TaqIA polymorphism and late-onset hallucinations (hallucinations that developed after 5 years of initial PD symptoms), but not the early ones, suggesting that it was a risk factor whose clinical significance expressed only with longer disease duration [31]. Besides, VH in PD were associated with DRD3 Ser9Gly polymorphism in one study [32].
We found no association between rs4680 of COMT gene and PPD, in accordance with retrospective study of Camicioli et al. [33] with autopsy proven PD patients and another study of Creese et al. [34] which included demented PD patients. These data suggested that involvement of dopamine in the pathogenesis of PDP may be limited [34].
DAT1 mediates rapid reuptake of dopamine from the presynaptic cleft, mainly in the striatum. The most studied polymorphism in DAT gene is VNTR that has been identified in the non-coding 3’ untranslated region (3’ UTR) and consists of a sequence of 40 base pairs (bp) that is repeated from 3 to 11 times; alleles with 9 and 10 repeats being the most common. The number of repeats may affect gene expression and thus retrieval of dopamine from the synaptic cleft [10, 30]. We did not find association between VNTR and PDP, in line with inconclusive data from other studies [35]. A retrospective study on 183 PD patients found that the 9 repeats allele occurred much more frequent in PD patients with psychosis or dyskinesia [36].
Our aim was a comprehensive evaluation of multiple clinical, drug-related, and genetic risk factors that may play a role in the development of psychosis, and we believe that is the important strength of this study. Previous studies examined mainly VH, but not the other psychotic symptoms in PD. Our study, using valid criteria for the diagnosis of PDP, considered all types of psychotic symptoms including minor phenomena, hallucinations, and delusions. Although it is still hard to tell whether psychosis is a continuum in which minor phenomena progress to hallucinations with retained insight, and then to hallucinations without insight and delusions, there nonetheless are data that support this concept [37]. However, the cross-sectional design, as the main limitation of the current study, disables the authors from determining true risk factors for the development of psychosis.
CONCLUSION
We found rs6277 of DRD2 gene and rs2734849 of ANKK1 gene might contribute to the development of PDP. Multivariate analysis suggested the LEDD, motor status, presence of depression and anxiety, as well as GG genotype of rs2734849 ANKK1 gene to be possible independent factors associated with PDP. Longitudinal, hypothesis-driven studies, with an appropriate number of PD patients, are needed to provide fuller insight into the role of genetic polymorphisms in the emergence of PPD.
The datasets generated during and/or analyzed during the current study are available from the corresponding author on reasonable request.
