T102C polymorphism of serotonin-2A receptor gene in Turkish schizophrenia patients: Association with cognitive impairment and soft neurological signs

Abstract

Aim:

Previous studies have shown an association between the T102C polymorphism of the serotonin-2A receptor gene and schizophrenia. In addition, an association of this polymorphism with clinical phenotypes in schizophrenia such as treatment response and cognitive impairment has been observed.

Materials and Methods:

In this case-control study conducted in Turkish Caucasians, we compared T102C polymorphism genotype and allele frequencies in 76 schizophrenic patients and 165 healthy controls. We also investigated interaction of this polymorphism with clinical and cognitive variables in patients.

Results:

No significant difference was observed in the distribution of the three genotypes (T/T, T/C and C/C) and in the allele frequencies in controls and patients with schizophrenia. No evidence of association was detected at various clinical phenotypes including symptom severity, suicidality, treatment response, age of disease onset, number of hospitalizations and history of violence (in co-dominant, dominant, or recessive models). However, as compared to the C/C genotype, patients with 1 or 2 copies of the T allele were characterized by better stroop test performances and less “motor coordination” soft neurological signs.

Conclusion:

Further research is needed to elucidate the impact of T102C polymorphism on neurocognitive functions in both healthy and patient populations.

INTRODUCTION

Schizophrenia is a severe psychiatric disease with a significant genetic component.[1,2] Several lines of evidence suggest that abnormalities in the serotonin (5-HT) system may be related to the pathophysiology of schizophrenia. Among the various serotonin receptors, 5-HT2 receptors have received much attention because of their unique role in the therapeutic actions of clozapine and several other atypical antipsychotics.[3,4,5] Various studies have shown an association between the T102C polymorphism of the 5-HT_2A receptor gene and schizophrenia.[6] However, negative findings have also been reported.[7] The controversial results have been conceived as partially explained by ethnic factors. A recent meta-analysis demonstrated that the association between T102C polymorphism of the 5-HT_2A receptor gene and schizophrenia is stronger in European than in East Asian populations.[8] In addition, there have been numerous studies of the association of T102C polymorphism with various clinical aspects of schizophrenia, including disease outcome,[9] susceptibility to tardive dyskinesia,[10,11,12] and more recently neurocognitive functioning.[7,13,14,15]

There is emerging evidence for the role of genes in cognition in schizophrenia.[16] In this study, we compared the genotype distributions and allele frequencies of the T102C polymorphism of the 5-HT_2A receptor gene between patients with schizophrenia and unrelated healthy controls of Turkish descents. The patient sample was further characterized for certain clinical and neuropsychological features that might be potentially related to serotonin dysfunction.

MATERIALS AND METHODS

Subjects

All individuals taking part in the study were Turkish Caucasians. 76 biologically unrelated patients aged 18-60 meeting DSM-IV-TR diagnostic criteria for schizophrenia[17] were included in the study. Patients were recruited from the Department of Psychiatry, Cerrahpaşa Medical Faculty, Istanbul, Turkey. Consensus diagnosis by at least two psychiatrists was made for each patient, according to DSM-IV criteria. Patients with a history of alcohol and/or drug abuse, major medical illness that could affect brain functioning, clinically evident mental retardation, neurological disorder, a past head injury with loss of consciousness and a history of having received electroconvulsive therapy within the past year were excluded.

The control subjects were selected among healthy volunteers, and none of them had a history of past and/or present psychiatric diagnosis. A total of 165 biologically unrelated controls were recruited from sources including people from the hospital staff, and medical and nursing students. All subjects participated in the study gave informed consent. The local Ethic Committee of Cerrahpaşa Medical Faculty approved the study. Patients’ clinical and neurocognitive functions were assessed following the extensive method used in a previous study.[7]

Assessment of clinical profiles

The severity of clinical symptoms in patients with schizophrenia was assessed by the Positive and Negative Syndrome Scale (PANSS).[18] This scale consists of 30 items, with the total score consisting of the sum of the seven positive items, seven negative items, and 16 general psychopathology items. The Turkish version reliability and validity study was conducted by Kostakoğlu et al.[19]

The age of disease onset in individual patients was determined as the age when psychotic symptoms were first presented or of first contact with psychiatrist. Number of hospitalizations due to the psychiatric diagnosis was also determined.

Absence or presence of a history of suicidal behavior was recorded. History of violence was rated on a three-point scale: (no) indicated no physical violence and no significant verbal aggression (mild) referred to patients with significant and frequent verbal aggression or violence to object or mild physical aggression to person, and (severe) represented patients with physical violence to person resulting in medical consultation of the victims. Absence or presence of a family history of mental illness was determined according to information gathered from patients and their family members.

Response to treatment was evaluated retrospectively over the course of an individual patient's illness and was rated on a three-point scale. Good response referred to complete symptom remission and return to the level of premorbid functioning. Negligible response indicated persistent symptoms for more than 2 years despite optimal dosage of at least two different antipsychotic drugs for 6 weeks each. Patients with treatment outcome between good and poor response were considered as the partial response.

Of the 76 patients interviewed, the PANSS were completed for all patients. The age of disease onset of three patients, family history of mental illness of 14 patients, history of suicidal behavior of 12 patients, number of hospitalizations of 13 patients and violence of 13 patients could not be determined because the information was not documented in medical records, and/or the patients or their family members could not provide accurate information. The treatment response for 18 patients could not be assigned on the three-point scale because their duration of illness was less than 2 years or the available information was not sufficient.

Neurocognitive assessment

Patients’ degree of formal education was determined by information from patients and their family members. A semantic verbal fluency test was conducted for the “animal” category. Each patient was asked to generate as many exemplars in the animal category as possible in 1 min. The word output was then analyzed. The motor coordination soft neurological signs examination extracted from the Cambridge Neurological Inventory[20] was administrated. It consisted of finger thumb opposition, finger thumb tapping, dysdiadochokinesia, fist-edge-palm test, and Ozeretski test. The stroop test was conducted according to standard procedures.[21] Each patient was asked to perform two tasks. The first “color task” required the patient to read the words of color names, which was printed in colors different from the meaning of the words. The second “color-word task” required the patient to read the printed color of the words. Two minutes were allowed for each task. The numbers of correct and incorrect responses were recorded. The color score and color-word score was calculated from the number of correct responses minus incorrect responses in the color task and the color-word task, respectively. 12 patients refused to be examined for verbal fluency and soft signs, and 17 patients refused to be examined for stroop tasks.

DNA analysis

Genomic DNA was extracted from venous blood samples anticoagulated with EDTA. The genomic DNA extraction was performed by standard procedures. The polymerase chain reaction (PCR) amplification was carried out using oligonucleotide primers 5´-TCTGCTACAAGTTCTGGCTT-3 ´ and 5´-CTGCAGCTTTTTCTCTAG GG-3´.[7] The PCR was performed with 0.1 μg genomic DNA, 1.5 mg MgCl, 0.1 μm of each primer in a total volume of 50 μl. Genotyping was done by means of restriction fragment length polymorphism assay with the restriction enzyme (MSPI). Genotyping was performed blindly for cases and controls.

Statistical analysis

The genotype distributions of the T102C polymorphism of the 5-HT_2A receptor gene in patients and controls were evaluated for deviation from the Hardy–Weinberg equilibrium using the χ² test. Comparison between the patients and controls for the difference of genotype distributions, allele frequencies was analyzed by the χ² test. Further comparison of clinical profiles among patients with schizophrenia with different genotypes was carried out by one-way ANOVA for continuous variables or the χ² test for categorical variables. Similarly, the neurocognitive function among patients with various genotypes was compared by using ANCOVA controlled for age or degree of education as required. All the statistical tests were two-tailed with the significance level set at P < 0.05.

RESULTS

General characteristics

Age and gender distributions (male/female) of patients were 37.8 ± 10.2 and 43/33. Age and gender distributions (male/female) of controls were 42 ± 13 and 78/87. Patients and controls were not significantly different for gender distribution. However, mean age in the control group was significantly higher compared with the patient group (P = 0.009). Genotype distribution and allele frequencies of the T102C polymorphism of the HT_2A receptor gene among patients and controls are shown in Table 1. The genotype frequencies did not significantly deviate from the Hardy–Weinberg equilibrium in patients (χ² = 1.9, df = 1, P = 0.16) and controls (χ² = 0.05, df = 1, P = 0.8). There was no significant difference in genotype distributions or allele frequencies between patients and controls.

Table 1.

Genotype distributions and allele frequencies of the T102C polymorphism of the 5HT_2A receptor gene among schizophrenic patients and healthy controls

Clinical profiles

The associations between the clinical profiles of schizophrenic patients and their genotype distributions are shown in Table 2. No significant association was found between the clinical profiles and the genotype distribution in co-dominant, dominant or recessive models of the gene (statistics not presented).

Table 2.

The genotype distributions in association with clinical profiles among schizophrenic patients (n=76)

Neurocognitive dysfunction

Table 3 shows the comparison of neurocognitive performance among patients with various genotypes. As stroop test and verbal fluency test scores (i.e. mean color score and mean color word score) were found to be related with education, ANCOVA's controlled for education were used in comparisons of these neurocognitive variables. Soft neurological signs scores were related with subjects’ age, subsequently this variable was tested using ANCOVA's controlled for age. No statistically significant differences were observed between the three genotypes with respect to the scores for neurocognitive dysfunction. However, when the contrast of C/C versus C/T + T/T was examined in a recessive model of the C allele, patients with C/C genotype had significantly worse stroop test performance and more severe motor coordination soft neurological signs.

Table 3.

Neurocognitive function of schizophrenic patients (n=76) in association with their genotypes

DISCUSSION

This study found an association of T102C polymorphism of 5-HT_2A receptor with executive cognitive and soft neurological motor functions in patients with schizophrenia. Patients which were homozygous for the C allele as opposed to the patients carrying the T allele (i.e. patients with CT and TT genotypes) had significantly worse stroop test performance and more severe motor deficits. Conversely, parameters of disease severity such as negative and positive symptom severity, age of disease onset, number of hospitalizations, history of suicide/violence, family history of psychosis, and degree of treatment response did not differ among T102C genotypes.

Relatively few studies have investigated the role of T102C polymorphism of serotonin receptor on cognitive functions in schizophrenia and normalcy. In one study from Turkey,[15] patients of schizophrenia with the heterogeneous genotype (TC), compared to those who were genotype CC or TT did fairly worse in the Wisconsin Card Sorting (WCST) and the continuous performance tests measuring sustained attention span and freedom from distractibility.[22] Researchers also reported that the T allele, as opposed to the C allele was associated with worse performance in several cognitive domains investigated in that study. However, in a recent study from Russia with large sample sizes of patients and healthy controls[14] it was reported in a combined cohort of patients with schizophrenia and healthy individuals that overall productivity and immediate reproduction processes of short-term memory were significantly worse in presence of the CC genotype. Thus, our finding of a worse stroop test performance in CC genotype is largely in parallel with the findings of this latter study. An earlier and smaller study of the same group also demonstrated that male patients with schizophrenia of the CC genotype had lower verbal fluency than their counterparts.[13] Another study that investigated the effect of 5-HT_2A T102C polymorphism on cognition was performed in Chinese healthy individuals.[23] This study reported that subjects with the TT genotype had the fewest perseverative errors, while subjects with TC had the poorest performances in WCST. Overall, there is possibility that the effect of T102C polymorphism on cognitive performance is also observable in some healthy individuals with significant susceptibility to schizophrenia. Early onset schizophrenia may be particularly informative as it is associated with higher genetic loading and also with generalized cognitive impairments across a broad array of neurocognitive functions.[24] However, in a recently published study, Vyas et al. investigated the association of the T102C polymorphism of the 5-HT_2A receptor gene on visual sustained attention in patients with early onset schizophrenia and their nonpsychotic siblings. They found an association of the TC/CC genotype of the 5-HT_2A T102C polymorphism with poor performance on sustained visual attention in probands with early onset psychosis, but not in healthy siblings.[25] Yet, it is a rare form schizophrenia which occurs in only %5 of all schizophrenia cases. Future studies with a larger sample in healthy people at risk for schizophrenia should investigate the association of C allelle with cognitive measures.

We also found in our study that patients with the CC genotype had significantly more severe soft neurological (motor) signs than patients with the T allele. There is considerable evidence in the literature for a strong association between soft neurological signs and poor (executive) cognitive performance in schizophrenia.[26] Hence, our findings regarding the soft neurological signs and stroop test performances in patients correlates well with each other. This is, to the best of our knowledge, the first evidence of interaction between T102C polymorphism of serotonin receptor and soft neurological signs in a Caucasian population. Another single study that examined this issue have found less severe motor coordination soft neurological signs in patients with genotype TC and the most severe of these signs in the TT genotype, however in a population of Chinese Han patients.[7] Significant differences in the genotype distribution of the T102C polymorphism between the investigated populations (Turkish vs. Chinese) might be responsible for the above discrepancy. As mentioned above, a recent meta-analysis revealed different genotype distributions of the T102C polymorphism gene in schizophrenia and noncombinability of data between European and East Asian populations.[8]

Our above findings are in line with the recent evidence which indicates that T102C polymorphism located near the promoter region of 5-HT_2A receptor gene might have regulatory effects on gene expression.[27] For instance, 5-HT_2A receptor expression was reported to be less in the presence of C allele as opposed to the T allele in normal individuals.[28] There is also converging evidence which indicates that serotonin transmission, by interacting with the dopaminergic system plays an important role in prefrontal cognitive functions.[29] In rats, 5-HT_2A receptors have been reported to regulate prefrontal execution of primed responses and in healthy volunteers, 5-HT_2A agonists impair the continuous performance task, which is mainly determined by prefrontal function.[30] Conversely, clozapine, a strong antagonist to the 5-HT_2A receptor was found to improve cognitive functions including attention and verbal fluency in patients with schizophrenia.[31]

CONCLUSION

Our study in addition to the few studies mentioned above provide clues for the involvement of T102C polymorphism of serotonin receptor gene on cognition in schizophrenia; however a straightforward comparison between findings of these studies was not possible due to the fact that investigated subjects were from different ethnic origins and/or various cognitive tests had been used to assess their cognition. Our study is limited by the relatively small sample size of the patients in which cognitive functions were assessed. Furthermore, limited domains of cognition were tested as opposed to a global assessment. Another limitation of our study is that the possible effects of antipsychotics and anticholinergic agents on the cognitive function of the investigated patients cannot be excluded. Despite the above limitations, it seems reasonable in the future to determine the possible effects of T102C polymorphism of serotonin receptor gene on cognition in various psychiatric populations and healthy controls.