Abstract
Objective: To evaluate the relationship between patients’ clinical parameters, especially clinical specifiers, and the intron 4 VNTR variant of the endothelial nitric oxide synthase (NOS3) gene in bipolar disorder (BD) patients.
Methods: A sample of 95 patients with BD and 95 healthy volunteers were included in the case-control study. The patients consecutively admitted to the outpatient psychiatry clinic for 6 months and were evaluated with some scales for clinical parameters. In addition, PCR was used to determine the NOS3 intron 4 VNTR variant.
Results: The NOS3 genotype and allele frequency distributions of rapid cycling BD patients were significantly different from non-rapid cycling BD patients and the control groups. Furthermore, NOS3 genotype and allele frequency distributions of treatment-resistant BD patients were significantly different from treatment-responsive BD patients and the control groups. While BD patients carrying the b/b genotype and b allele had a lower risk of rapid cycling and treatment resistance, having the b/a genotype in BD patients was at higher risk in terms of rapid cycling and treatment resistance. In addition, the number of hospitalizations and the Clinical Global Impression-Improvement Scale scores of the BD group with the b/b genotype were statistically lower than the BD group with b/a and a/a genotypes.
Conclusions: We propose that the intron 4 VNTR variant of the NOS3 gene may be associated with rapid cycling and treatment resistance in Turkish patients diagnosed with BD.
Bipolar disorder (BD) is a severe and chronic psychiatric disorder that affects approximately 1.3% of the population and is characterized by varying clinical presentations of depression and mania in addition to manic-depressive episodes with rapid cycling and mixed features.1,2 While the involvement of inflammatory processes, oxidative stress, genetic factors, neurotransmitters, and psychosocial factors associated with BD have been investigated, the etiology of BD has not yet been identified. 3-5
Numerous studies have identified inflammation-related genes that may play roles in triggering BD and documented increased levels of circulating proinflammatory cytokines in different phases of BD.6-8 T lymphocytes, macrophages, and endothelial cells secrete polypeptide molecules that support critical functions, including the proliferation of B cells, synthesis of acute-phase reactants, activation of neutrophils, and increased vascular permeability.3 Many studies that showed an association between vascular pathologies and BD had demonstrated inflammation-related endothelial damage.9 Goldstein et al10 reported that BD patients are more likely to have hypertension and cardiovascular diseases and developed these disorders a decade earlier than non-BD participants.
Nitric oxide (NO) plays an essential role in various functions, including antibacterial and antitumoral functions, neurotransmission, vasodilation, and neurotoxicity related to learning and memory impairment.11,12 In addition, a large body of evidence suggests that NO is a critical factor in many psychiatric disorders and can play both a protective and destructive role depending upon NO’s production and interaction with other factors in the cell.13,14 Nitric oxide synthase 1 (NOS1), NOS2, and NOS3 genes encode 3 NOS enzymes that synthesize NO from L-arginine. NOS3 gene produces endothelial NOS (eNOS), mainly involves the endothelium and maintains the basal vascular tone and cerebral blood flow; neuronal NOS (nNOS) is produced by the gene NOS1, and inducible NOS (iNOS) is produced by the gene NOS2.12 Although different NOS3 variants have been determined, the variable number of tandem repeats (VNTR) in intron 4, the Glu298Asp variant in exon 7, and the T786C variant in the promoter region are the most critical polymorphisms of the NOS3 gene.15 The VNTR in intron 4 of the NOS3 gene is responsible for producing more than 25% of basal plasma NO.16 Several neuropsychiatric disorders and vascular disorders such as diabetes mellitus, hypertension, QTc prolongation, and spontaneous abortion were found to be related to the polymorphisms of these NOS genes.11,17
The NOS3 gene has been found on 7q36 and consists of 26 exons spanning approximately 21 kilobases.18 Genome-wide scan of BD and investigation of population stratification effects on linkage provided evidence supporting susceptibility loci at 4q21, 7q36, 9p21, 12q24, 14q24, and 16p13.19 Therefore, we hypothesized that the intron 4 VNTR variant of the NOS3 gene might be associated with more severe and treatment-resistant forms of BD in Turkish patients. We aimed to examine the association between BD clinical information and the intron 4 VNTR variant in the NOS3 gene by comparing genotype distributions of patients with healthy controls.
Methods
Patient selection
The present study was designed as a case-control study and conducted at the Bakirkoy Mazhar Osman Mental Health and Neurology Training and Research Hospital outpatient clinic, Istanbul, Turkey. A sample of 95 patients diagnosed with BD was followed for 6 months in 2018; additionally, 95 age- and sex-matched controls were included in the study. The study was approved by the Clinical Research Ethics Committee of the Istanbul Faculty of Medicine under the ethical standard for human experimentation established by the Declaration of Helsinki (10/10.01.2018).20 Furthermore, all participants’ written consent was obtained to participate in the study.
Inclusion and exclusion criteria
Inclusion criteria for the study included the following: Aged 18-65 years, a DSM-IV diagnosis of BD type I or II according to the Structured Clinical Interview for DSM-IV Axis-I Disorders (SCID-I), no history of neurological/systemic illness, substance use, or event that might influence cognitive function. In addition, we excluded patients with a history of any comorbid Axis I disorders, neurodevelopmental disorders such as autism, mental retardation, cerebral tumor, and cerebrovascular disease.
Diagnostic tools and scales
First, the SCID-I was used to confirm the diagnosis according to the DSM-IV-TR criteria, and the presence of any psychiatric disorder in the healthy control group was the basis for exclusion from the study.21,22 Then, sociodemographic and clinical parameters data form was applied to BD patients. Subsequently, the Young Mania Rating Scale (YMRS) and Hamilton Depression Rating Scale (HAM-D) were administered to patients with BD to evaluate the severity of mania and depression, and the Clinical Global Impression Scale (CGI) was used to assess the severity of the disorder and response to the treatment.23-27
Criteria for treatment resistance and rapid cycling in BD
Sachs proposed the term “treatment-resistant bipolar disorder” to define patients who do not respond to at least 2 antimanic or antidepressant drugs with adequate dose and duration of treatment in a specific period, such as 6 weeks for mania or depression and 6 months or three-cycle lengths for maintenance.28 The DSM-5 defines rapid cycling BD as a pattern of presentation accompanied by four or more mood episodes in 12 months, with a typical course of mania, hypomania, or depression. The episodes must be demarcated by a complete or partial remission lasting at least two months or by switching to a mood state of the opposite polarity.29
DNA analyses
Blood samples for isolating DNA material were received from participants to analyze at the Istanbul Faculty of Medicine Laboratory of Medical Biology. The intron 4 VNTR NOS3 gene variants were genotyped by polymerase chain reaction (PCR) using the following primers: (forward) 5’-AGG CCC TAT GGT AGT GCC TTT-3’ and (reverse) 5’-TCT CTT AGT GCT GTG GTC AC-3’. Then, we separated the products on a 4% NuSieve GTG agarose gel and repeated the experimental process twice for each sample. We designed primers to amplify a 393 bp and/or 420 bp segment of the NOS3 intron 4 VNTR region, including the microsatellite repeat sequence. NOS3 intron 4 VNTR genotypic distributions were determined as 393 bp, 393, 420 bp and 420 bp for a/a, a/b, and b/b genotypes, respectively.30
Statistical analyses
We performed statistical analysis using IBM SPSS version 21.0 (IBM Corp. released 2012; Armonk, NY, USA). Descriptive statistics contained mean, standard deviation, frequency, and percentage. The comparison of genotype distributions of NOS3 intron 4 VNTR variants between groups was analyzed with the Pearson chi-square test. Again, the comparisons of genotype distributions of NOS3 intron 4 VNTR variants in BD patients in terms of rapid cycling or treatment resistance were analyzed with the Pearson chi-square test or Fisher’s exact test. We also calculated the odds ratio (OR) and the 95% confidence interval (CI). The Shapiro-Wilk test evaluated the suitability of continuous variables for a normal distribution. Since the variables did not have a normal distribution, the scale scores and clinical parameters were compared using Mann-Whitney U testing. We analyzed genotype distributions in participants due to the Hardy-Weinberg Equilibrium (HWE) and accepted p<0.05 as a statistical significance. The power analysis was performed with the “G*power” software (version 3.0.5, http://www.psycho.uni-duesseldorf.de/abteilungen/aap/gpower3/), post hoc goodness of fit χ2 test, with an “-error” probability of 0.05. The possible presence of population stratification bias has been calculated, according to Lee and Wang, regarding intron 4 VNTR NOS3 gene polymorphism frequencies documented for Turkish populations and incidence rate of BD in Turkey.31-33
Results
NOS3 intron 4 VNTR genotyping
Ninety-five patients with BD (56 female/39 male) were evaluated according to their clinical parameters, and the scale scores are presented in Table 1. According to the NOS3 intron 4 genotype distribution, 69.5% (n=66) of the patients diagnosed with BD had the b/b genotype, 26.3% (n=25) had b/a, and 4.2% (n=4) had a/a. Eighty-one point one percent (n=77) of the healthy controls had the b/b genotype, 17.9% (n=17) had b/a, and 1.1% (n=1) had a/a. Comparing the NOS3 genotype (b/b, b/a, a/a) and the allele frequency (b, a) distributions of patients with BD to the control group revealed a statistically significant difference between the allele frequency of the 2 groups. The BD patients had a higher frequency of the “a” allele than the control group (OR:1.892; 95% Cl: 1.033-3.463; p=.037) Table 2.
- Sociodemographic characteristics and the Scale Scores of BD patients.
- Comparison of genotype distributions of NOS3 intron 4 VNTR variants in BD patients with the control group.
Comparison of NOS3 genotype and allele frequency distributions of rapid cycling BD patients with non-rapid cycling BD patients and the control groups
Comparing the NOS3 intron 4 genotype (b/b, b/a, a/a) and allele frequency (b, a) distributions regarding the presence of clinical specifiers (atypical features, mixed features, seasonal pattern, rapid cycling, psychotic features, peripartum onset) in the BD patient group demonstrated that the NOS3 intron 4 genotype and allele frequency distributions were significantly different between the groups of patients with BD in terms of rapid cycling. BD patients without rapid cycling had a higher frequency of b/b genotypes and the b allele than BD patients with rapid cycling (OR: 0.258; 95% Cl: 0.092-0.722; p=.007; OR: 3.130; 95% Cl: 1.389-7.049; p=.004, respectively). Furthermore, the b/a genotype was found at a significantly higher frequency in BD patients with rapid cycling than in BD patients without rapid cycling (OR: 3.017; 95% Cl: 1.067–8.534; p=.033) Table 3.
- Comparison of NOS3 genotype and allele frequency distributions of rapid cycling BD patients with non-rapid cycling BD patients and the control groups.
Comparing the NOS3 intron 4 genotype and the allele frequencies of BD patients with rapid cycling to the control group showed that the NOS3 genotype and allele frequency distributions of BD patients with rapid cycling were significantly different from the control group. The control group had a higher frequency of the b/b genotype and b allele than the BD patients with rapid cycling (OR: 0.191; 95% Cl: 0.069-0.530; p=.001; OR: 0.231; 95% Cl: 0.102-0.521; p<.001, respectively). In addition, the b/a genotype was found to be significantly higher in the BD patients with rapid cycling compared to the control group (OR: 3.754; 95% Cl: 1.347–10.466; p=.016) Table 3.
Comparison of NOS3 genotype and allele frequency distributions of treatment-resistant BD patients with treatment-responsive BD patients and the control groups
When the NOS3 intron 4 genotype and allele frequency distributions of the treatment-resistant BD patients were compared with the treatment-responsive BD patients, there was a statistically significant difference between the NOS3 genotype and the allele frequency distributions of the two groups. The treatment-responsive BD patients had a higher frequency of the b/b genotype and the b allele than the treatment-resistant BD patients (OR: 3.656; 95% Cl: 1.396-9.575; p=.007; OR: 2.905; 95% Cl: 1.330-6.346; p=.006, respectively). In addition, the b/a genotype was found at a significantly higher frequency in the treatment-resistant BD patients compared to the treatment-responsive BD patients (OR: 0.318; 95% Cl: 0.119–0.850; p=.019) Table 4.
- Comparison of NOS3 genotype and allele frequency distributions of treatment resistant BD patients with treatment-responsive BD patients and the control groups.
When the NOS3 intron 4 genotype and the allele frequency distributions of the treatment-resistant BD patients were compared with the control group, the NOS3 genotype and allele frequency distributions of the treatment-resistant BD patients were significantly different from the control group. The control group had a higher frequency of the b/b genotype and the b allele than the treatment-resistant BD patients (OR: 0.216; 95% Cl: 0.085-0.551; p=.001; OR: 0.259; 95% Cl: 0.120-0.559; p<.001, respectively). Additionally, the b/a genotype was significantly higher in the treatment-resistant BD patients compared to the control group (OR: 3.605; 95% Cl: 1.397–9.304; p=.006) Table 4.
Comparison of scale scores and clinical parameters regarding the NOS3 intron 4 genotype and the allele frequencies of BD patients
Comparing the scale scores (HAM-D, YMRS, CGI-S, CGI-I) and clinical parameters (age of onset, duration of disorder, number of hospitalizations, number of manic episodes, depressive episodes, and total episodes) regarding the NOS3 genotypes of BD patients, the number of hospitalizations and CGI-I scores were significantly different between the genotype groups (b/b vs. b/a and a/a). In addition, the number of hospitalizations and CGI-I scores of the b/b group were statistically lower than the b/a and a/a group (p=.040; p=.037, respectively) Table 5.
- Comparison of scale scores and clinical parameters regarding the NOS3 intron 4 genotype distributions of BD patients.
Discussion
In this study, we found that the BD patients carrying the NOS3 intron 4 b/b genotype and b allele had a significantly lower frequency of rapid cycling and treatment resistance. Similarly, the number of hospitalizations and the CGI-I scores of the BD group with the b/b genotype were statistically lower than the BD group with b/a and a/a genotypes. Currently, progress in biotechnology, neuroimaging, and molecular genetics have contributed to new perspectives on the etiopathogenesis of BD. Numerous studies have identified genes that may be involved in triggering BD. Alves et al34 conducted a literature review of 129 articles and reported that 79 genes are associated with BD. The 5 genes that are the most mentioned in the literature are calcium voltage-gated channel subunit alpha1 C (CACNA1C), ankyrin 3 (ANK3), disrupted in schizophrenia 1 (DISC1), D-amino acid oxidase (DAOA), and tryptophan hydroxylase 2 (TPH2).34 Several oxidative stress markers have also been hypothesized to be associated with the pathophysiology of BD. Many research studies have reported oxidative damage to deoxyribonucleic acid (DNA), ribonucleic acid (RNA), lipids, and proteins. A meta-analysis by Brown et al35 reported DNA or RNA damage as well as increased NO and lipid peroxidation levels in BD. The BD studies have demonstrated that NO levels are enhanced during different mood episodes, especially in depressive episodes.36 Therefore, it seems likely that significantly higher concentrations of plasma nitrite in BD patients might be related to acute changes in regional eNOS activity. Reif et al16 also reported that the NOS3 genotype might be a modest genetic risk factor for developing BD.
Rapid cycling, a well-recognized specifier, is reported to exist in 5–15% of patients with BD. Rapid cycling BD patients seem to have higher rates of medical comorbidity, suicidal risk, treatment-resistant, and family history for BD, as well as increased susceptibility to DNA damage or mRNA hypo-transcription compared to BD patients without rapid cycling.37 In the present study, BD patients carrying the NOS3 intron 4 b/b genotype and b allele had a lower risk of rapid cycling, while BD patients with the b/a genotype had a higher risk of rapid cycling. Kirov et al. found that the frequency of the Met allele in the catechol-O-methyltransferase (COMT) gene was significantly higher in a rapid cycling BD group than in a non-rapid cycling group.38 In contrast, their following study did not detect a statistically significant difference in the rate of low-activity monoamine oxidase-A (MAO-A) alleles between ultra-rapid cycling and non-rapid cycling BD patients.39
Several critical studies have investigated the relationship between brain-derived neurotrophic factor (BDNF) gene polymorphisms and rapid cycling BD. For example, Muller et al40 showed a higher frequency of the Val allele of BDNF in rapid cycling BD compared to non-rapid cycling BD. Liu et al41 demonstrated a relationship between the BDNF gene polymorphism (rs7127507) and rapid cycling BD. Similarly, Munkholm et al42 reported that rapid cycling BD patients had increased BDNF plasma levels compared to healthy controls. Recently, they also reported an association between increased DNA damage and impaired repair mechanisms in rapid cycling BD patients compared to healthy controls.43 Our study contributes to these studies by demonstrating a relationship between the NOS3 intron 4 gene and rapid cycling BD for the first time.
When we defined treatment resistance in BD according to Sachs’s criteria mentioned in the method section, we found that the NOS3 intron 4 genotype or allele frequency distributions were significantly different between the BD groups in terms of treatment response. Comparing the clinical parameters and scale scores regarding the NOS3 intron 4 distributions in BD patients, the number of hospitalizations and the CGI-I scores of the BD group with the b/b genotype were significantly lower than the BD group with b/a and a/a genotypes. As previous evidence has reported, the number of hospitalizations is thought to be associated with the treatment response in BD.44 Additionally, the CGI-I is used to evaluate the improved or worsened condition of the patient by comparing a baseline state before the treatment.45 Therefore, we can speculate that the Turkish BD patients with the b/b genotype and b allele have a lower risk of developing treatment resistance. A literature review on BD and treatment resistance identified studies on patient responses to lithium treatment. One of these studies showed that the lithium response might be associated with the BDNF Val66Met polymorphism and serum BDNF levels.46 A short allele of the serotonin-transporter-linked polymorphic region (5-HTTLPR) has also been related to both unipolar and bipolar affective disorders and inadequate antidepressant response.47 In another study, Rybakowski et al48 found that the lithium response has been associated with the dopaminergic receptor D1 (DRD1) gene polymorphism. Lithium may also affect the cyclic adenosine monophosphate (cAMP) pathway, which is an intracellular signaling system; Mamdani et al. reported a relationship between the lithium response for BD and 2 polymorphisms of the cAMP-responsive element-binding protein 1 (CREB1) gene located at chromosome 2q32–34.49 Few studies in the literature have examined the response to non-lithium mood stabilizers in patients with BD. One study identified a relationship between -116C/G polymorphism of X-box binding protein 1 (XBP1) and treatment response to prophylactic valproate in BD.50 In our study, when all mood stabilizers were evaluated together, a significant difference was found between treatment-resistant and treatment-responsive groups regarding NOS3 intron 4 polymorphism. In addition, 70% of patients diagnosed with rapid cycling BD (n=14/20) were also resistant to treatment with any mood stabilizer. This result is also crucial for showing the association between treatment resistance and rapid cycling in BD.
Several limitations should be considered in the present study. First, the small sample size of BD patients was a limitation in the current study; this research question should be studied with a larger group of participants to verify the outcomes. Secondly, in our study, the VNTR polymorphic region of the NOS3 in intron 4 was examined, but it was not possible to know how 27-bp repeat polymorphism would link with other single nucleotide polymorphisms of NOS3 gene (-786T/C and 894G/T polymorphisms). Therefore, considering the multi-genetic nature of BD, our study’s third limitation is that it was about single nucleotide polymorphism. Lastly, we did not measure nitrite or nitrate levels that may have established a direct correlation between the investigated polymorphism.
In conclusion, we found that a VNTR variant in the NOS3 intron 4 gene was associated with rapid cycling and treatment resistance in Turkish patients diagnosed with BD. Confirming the current results with further coding region variants in other populations who live in more extensive geographical regions will contribute to a better understanding of the relationship between NOS3 intron 4 gene polymorphism, rapid cycling, and treatment resistance in BD. The genetics and pharmacogenetics of NOS and other genes involved in NO metabolism have not yet been sufficiently investigated in BD. Increasing sample sizes and accounting for ethnic origin are necessary for high-level evidence. We believe that augmenting understanding of eNOS signaling pathways and developing new technologies will help identify eNOS as a primary target for treating BD.
Acknowledgments
The authors gratefully acknowledge Cambridge Proofreading LLC for native English editing.
Footnotes
Disclosure. The authors declare no conflicting interests, support or funding from any drug company.
- Received April 4, 2022.
- Accepted July 3, 2022.
- Copyright: © Neurosciences
Neurosciences is an Open Access journal and articles published are distributed under the terms of the Creative Commons Attribution-NonCommercial License (CC BY-NC). Readers may copy, distribute, and display the work for non-commercial purposes with the proper citation of the original work.
