Structural variation in the glycogen synthase kinase 3β and brain‐derived neurotrophic factor genes in Japanese patients with bipolar disorders

Abstract Background Lithium is the first‐line drug for the treatment of bipolar disorders (BDs); however, not all patients responded. Glycogen synthase kinase (GSK) 3β and brain‐derived neurotrophic factor (BDNF) play a role in the therapeutic action of lithium. Since structural variations were reported in these genes, it is possible that these genomic variations may be involved in the therapeutic responses to lithium. Method Fifty patients with BDs and 50 healthy subjects (mean age 55.0 ± 15.0 years; M/F 19/31) participated. We examined structural variation of the GSK3β and BDNF genes by real‐time PCR. We examined the influence of structural variation of these genes on the therapeutic responses to lithium and the occurrence of antidepressant‐emergent affective switch (AEAS). The efficacy of lithium was assessed using the Alda scale, and AEAS was evaluated using Young Mania Rating Scale. Results Although we examined structural variations within intron II and VII of the GSK3® gene and from the end of exon IV to intron IV and within exon IX of the BDNF gene, no structural variation was found in BDs. Whereas 5 of 50 patients exhibited three copies of the genomic region within exon IV of the BDNF gene, all healthy subjects had two copies. No difference in the therapeutic efficacy of lithium was found between patients with three and two copies. No difference in the occurrence of AEAS was found between the two groups. Conclusion The amplification of the BDNF gene influenced neither the therapeutic responses to lithium nor the occurrence of AEAS.


| INTRODUC TI ON
Although it is well known that mood stabilizers, such as lithium and valproic acid, are effective in the treatment of bipolar disorders (BDs), [1][2][3][4][5] a growing body of evidence shows that certain patients with BDs respond poorly to these 2 mood stabilizers. 6,7 Based on pharmacological studies of lithium and valproic acid, it has been suggested that a common therapeutic action of these mood stabilizers is the inhibition of glycogen synthase kinase (GSK) 3 ® activity through increased serine phosphorylation. 9,10 In addition, chronic treatment with lithium and valproic acid has been reported to increase the levels of brain-derived neurotrophic factor (BDNF) in rodent brain and human blood. [11][12][13][14][15] Together, the inhibitory action of GSK 3 ® and upregulation of BDNF are closely involved in the therapeutic mechanism of these mood stabilizers. In this context, it is hypothesized that structural genomic variations in the GSK 3 ® or BDNF genes may affect the therapeutic efficacy of lithium and valproic acid. Patients with BDs who have structural genomic variations in these genes exhibit poor responses to lithium and valproic acid. In this study, we first examined whether CNVs in the GSK 3 ® and BDNF genes based on the DGV could be found in patients with BDs as assessed by quantitative real-time PCR (qRT-PCR) with TaqMan Ⓡ Copy Number Assays (Applied Biosystems), and if found, we compared the prevalence of the CNVs between patients with BDs and healthy subjects. Second, we examined the influence of the CNVs found in this study on the age of onset in patients with BDs and subtypes of BDs. Third, we also examined the influence of CNVs on the therapeutic responses to lithium and antidepressant-emergent affective switch.

| Subjects
Fifty patients with BDs (age: mean ± SD = 55.0 ± 15.0 years, gender: M/F = 19/31) and 50 age-and sex-matched healthy subjects participated in this study. All subjects were Japanese. The characteristics of the patients are shown in Table 1. All patients were diagnosed by trained psychiatrists according to DSM-IV-TR criteria (American Psychiatric Association, 1994), on the basis of unstructured interviews and information from medical records. Forty-seven patients with BDs received lithium treatment. The therapeutic response to lithium was evaluated using the Alda scale. 22

| Selection of the genomic region in the GSK3β and BDNF gene
With regard to CNVs in the GSK3 ® gene, we focused on nsv829696 (Chr3: 119848821-119992558, GRCh38:hg38) in the DGV. The region of this CNV (from intron II to intron IX) is relatively identical to a structural variant (variation_HU1; from 5ʹ-noncoding region to exon IX) which was reported to be amplified in a patient with BD. 21 Also, Ronai et al 21 demonstrated amplification of the shorter region of the GSK3 ® gene (from exon V to exon IV) in 7 patients with BD.
We therefore examined structural variation within intron II and intron VII using real-time PCR with TaqMan Ⓡ Copy Number Assay exon VI that has been registered in the DGV, we used TaqMan Ⓡ Copy Number Assay (Hs0925412_cn, and Hs0925549_cn) to examine CNVs in this region of nsv95132 ( Figure 2). In addition, because it has been reported that exon IX is a transcribed exon, we also measured the CNVs in nsv832095 (chr11: 27538626-27715739) covering the entire of exon IX (DGV) using TaqMan Ⓡ Copy Number Assay (Hs01542529_cn).

| Statistical analyses
Significance in the occurrence of CNV between patients with BDs

| RE SULTS
According to the DGV, we measured CNVs within intron II and intron  Table 2). All patients exhibited two copies of the genomic region within exon IX.
We next assessed the influence of the amplified type within exon IV of the BDNF gene on the clinical features of BDs. We found no significant difference in the occurrence of the amplified type between patients with BP I and BP II (Table 3). We also found no significant difference in the age of onset between patients with the amplified and wild-type gene (Table 4).
We then assessed whether the amplified type affected the therapeutic responses to lithium and antidepressants in this patient population. First, we evaluated the therapeutic response to lithium using the Alda scale and compared the responses between the amplified and wild type. There were no significant differences in the Alda scale A, B, or A-B rates between these two groups (Table 5). Second, we compared the usage rates of combination therapy with lithium and antidepressants between these two groups and found no significant  (Table 6).

| D ISCUSS I ON
The results of the present study demonstrated that while no CNV was found in intron II or intron VII of the GSK3 ® gene in 50 patients In contrast with these previous studies, none of the participants in the present study were known to have a neurodevelopmental disorder. In this context, to our knowledge, this is the first study demonstrating the involvement of structural variation of the BDNF gene in the pathogenesis of BDs in the absence of a neurodevelopmental disorder. Five of 50 patients with BD exhibited amplification of the genomic region within exon IV of the BDNF gene. It is of interest that BDNF exon I-and IV-containing transcripts were reported to be the most upregulated BDNF mRNAs in response to neuronal depolarization. 25 Since it is well known that chronic administration of lithium increases the levels of BDNF in the brain, [33][34][35] we hypothesized that this amplification may affect the therapeutic efficacy of lithium, and we compared the therapeutic responses to lithium between BD patients with and without the amplification, using the Alda scale. 22 However, we failed to find any differences in the therapeutic efficacy of lithium between these two groups. In addition, it is also well known that regulation of BDNF expression is tightly involved in the therapeutic action of antidepressants. So, we next examined whether this amplification leads to the exaggerated response to antidepressants, in In summary, we found that 5 of 50 patients with BDs exhibited three copies of the genomic region within exon IV of the BDNF gene,

CO N FLI C T O F I NTE R E S T
The authors have no conflicts of interest to declare.

DATA R E P O S I TO RY
The raw data belonged to the present study cannot be made publicly available, because the disclosure of personal data was not included in the informed consent of the present study.

This study was approved by the ethics committees of Kochi Medical
School and University of Tokushima Graduate School.

I N FO R M E D CO N S E NT
All subjects received a description of the study and gave written informed consent.

R EG I S TRY A N D TH E R EG I S TR ATI O N N O. O F TH E S TU DY/ TR I A L
n/a.

A N I M A L S TU D I E S
n/a.