Evidence suggests that glycogen synthase kinase-3β (GSK3B) activity is increased significantly in the brain of patients with major depressive disorders (MDD). Inhibition of GSK3B is thought to be a key feature in the therapeutic mechanism of antidepressants. To investigate whether common genetic variants in the GSK3B gene are associated with MDD and the therapeutic response to antidepressants, four polymorphisms (rs334558 (−50 T>C), rs13321783 (IVS7+9227 A>G), rs2319398 (IVS7+11660 G>T) and rs6808874 (IVS11+4251 T>A)) of the GSK3B gene were genotyped in 230 Chinese MDD patients and 415 controls. Among the MDD patients, 168 accepted selective serotonin reuptake inhibitor (SSRI) (fluoxetine or citalopram) antidepressant treatment and therapeutic evaluation for 4 weeks and 117 for 8 weeks. Significant association with MDD was not shown in the alleles and genotypes of single loci or four-locus haplotypes. However, three of the four polymorphisms investigated were significantly associated with 4-week antidepressant therapeutic effect (P=0.002–0.011). Of the four-locus haplotype analysis, the GSK3B TAGT carriers showed a poorer response to antidepressants in 4-week (P<0.0001) and 8-week (P=0.015) evaluation compared with other haplotype groups and would quite likely be the non-remitter to 8-week antidepressant treatment (P=0.006). Our findings show, for the first time, that GSK3B genetic variants play a role in the SSRI antidepressant therapeutic response and support the hypothesis that drugs regulating GSK3B activity may represent a novel treatment strategy for MDD.
Glycogen synthase kinase-3 (GSK3) is a ubiquitous cellular serine/threonine protein kinase. In mammals, two closely related isoforms, GSK3α (GSK3A) and GSK3β (GSK3B), are present. Each isoform is encoded by different genes but both share nearly identical sequences in their kinase domains. GSK3 was originally identified as a negative regulator of glycogen synthesis. Recently, it has become recognized as a broadly influential enzyme modulating many aspects of cellular functions, including gene expression, cellular architecture and apoptosis.1 GSK3 activity is tightly regulated by several mechanisms. Among them, the phosphorylation state of GSK3 is the most well-characterized mechanism that regulates its activity. Both GSK3A and GSK3B are significantly inhibited by phosphorylation on serine-21 and serine-9, respectively, and the downregulation of their constitutive activity may in turn activate the transcription of their target genes.2 Impairment of this inhibitory control of GSK3 can result in abnormally high GSK3 activity, which may have detrimental effects on neural plasticity and survival.1
Among the GSK3 isoforms, GSK3B is highly expressed in neural tissue where its expression is regulated during development. Several lines of evidence suggest that GSK3B may be implicated in the pathogenesis of major depressive disorders (MDD) and that inhibition of GSK3B activity may play a role in the therapeutic effects of antidepressants. Firstly, treatment with the antidepressants fluoxetine and imipramine was found to inhibit GSK3B activity by phosphorylation on serine-9 in the mouse’s brain.3 In another study, pretreatment with imipramine attenuated hypoxia-activating GSK3B in the mouse’s brain.4 Furthermore, lithium, a mood stabilizer that was shown to enhance the action of antidepressants, was found to inhibit GSK3B in vitro and in mouse brain5, 6 and in human peripheral blood mononuclear cells.7 Secondly, certain proteins that modulate the antidepressant response interact with GSK3B. CREB (the transcription factor cyclic AMP response element binding protein) and BDNF (the neurotrophin brain-derived neurotrophic factor) are key mediators of the therapeutic response to various classes of antidepressants, and downregulation of BDNF may contribute to the pathophysiology of MDD.8 It has been demonstrated that BDNF can inhibit GSK3B activity and that overexpression of GSK3B can abolish CREB phosphorylation induced by BDNF.9 Thirdly, two studies in 2004 demonstrated that rats treated with GSK3 inhibitors showed reduced duration of immobility when exposed to the forced swim test, a well-established model for antidepressant efficacy, suggesting the potential of GSK3 inhibitors as antidepressants.10, 11 Finally, a recent post-mortem study demonstrated that there was no change in GSK3B protein levels in the ventral prefrontal cortex of MDD patients but that GSK3B activity increased significantly compared with control subjects.12
The human GSK3B, located on chromosome 3q13.33, is about 268 kb in length and includes 12 exons. To investigate the pathological role of the GSK3B gene in MDD, the association between the disorder and genetic variants in the GSK3B gene were examined in a case–control study. Linkage disequilibrium (LD) measurement among the GSK3B polymorphisms and haplotype analysis between patient and control groups were conducted to assess the association between markers within the GSK3B gene and MDD. The association between the GSK3B polymorphisms and antidepressant therapeutic response in MDD patients was also tested.
Materials and methods
The study population consisted of 230 psychiatric outpatients from southern Taiwan (male/female: 101/129; mean age: 40.11 years (s.d.: 15.71)) who met DSM-IV (Diagnostic and Statistical Manual of Mental Disorders-IV edition) criteria for MDD. A senior psychiatrist (YWY) who had training experience in using Structured Clinical Interview for DSM-IV (SCAN) made the diagnosis by interviewing patients, family members and obtaining records where possible. Other inclusion criteria were a minimum baseline score of 18 on the 21-item Hamilton Depression Rating Scale (HAM-D)13 and the presence of depressive symptoms for at least 2 weeks before entry into the study. Exclusion criteria were additional current DSM-IV Axis I diagnoses (including schizophrenia, bipolar disorder, substance abuse, generalized anxiety disorders, panic disorders or obsessive compulsive disorders), personality disorders, pregnancy, recent suicide attempt and major medical and/or neurological disorders. There were 194 drug-naive patients and 34 patients who had a history of at least one past treatment. For those patients with a past treatment history, a drug-free period of at least 2 weeks was required before entry into the study. Of the MDD patients, 121 had been enrolled in our previous study.14 An additional 415 normal subjects (male/female: 195/220; mean age: 41.47 years (s.d.: 16.14)) were enrolled as controls after undergoing a psychiatric interview to rule out psychotic or mood disorders. The sample consisted entirely of ethnic Chinese adults. The study approval was approved by the Institution Review Board in Taipei Veterans General Hospital and E-DA Hospital, and the study was carried out in accordance with the Declaration of Helsinki Principles. Informed consents were obtained from all subjects prior to commencement.
For the selective serotonin reuptake inhibitor (SSRI) pharmacogenetic study, daily doses of fluoxetine or citalopram were administered, starting at 20 mg day−1; on the basis of the clinical response, the investigator could increase the dosage to 40 mg day−1. The patients in the two treatment groups did not differ in their past treatment histories (drug-naive vs non-drug-naive, P=0.142), 4-week response (4-week responder vs non-responder, P=0.210) and 8-week response (8-week responder vs non-responder, P=0.278). No other psychotropic medications were permitted; however, anxiolytics were allowed for insomnia. Treatment efficacy was evaluated by one investigator (YWY), blind to patient genotype, who administered the HAM-D Scale before and after the 4-week and 8-week antidepressant treatment. Responders were defined as patients with at least a 50% decrease in HAM-D total score after antidepressant treatment.
Four single nucleotide polymorphisms (SNPs) in GSK3B were investigated in this study. For GSK3B, there were 186 genotyped SNPs in the Han Chinese population of Beijing (CHB) in the International HapMap Project. We first selected those SNPs with minor allele frequency greater than 10%, and there were 47 SNPs who fulfilled the selection criteria. Using the default algorithm from Gabriel et al.15 in Haploview, the 47 SNPs form a haplotype block and five of them (from 5′ to 3′: rs16830594 (intron 7), rs6805251 (intron 11), rs6808874 (intron 11), rs9826659 (intron 11) and rs2873950 (intron 11)) could be further selected as tagSNPs to capture a minimal haplotype variation greater than 5%. As rs6805251, rs6808874, rs9826659 and rs2873950 are all located in the intron 11, only rs6808874 was selected and rs6805251 was replaced by another two SNPs (rs13321783 and rs2319398) that are located in the middle region of the GSK3B gene and in strong LD with rs6805251 (absolute D′=1, r2=1). An additional SNP rs334558 (−50 T>C) was also included because it is located in the promoter region of the GSK3B gene, and the T allele has greater transcription activity compared with the A allele.16
For the genotyping of these four GSK3B polymorphisms, peripheral venous blood was withdrawn from the study subjects. Genomic DNA was isolated by using the PUREGENE DNA purification system (Gentra Systems Inc., Minneapolis, MN, USA). GSK3B polymorphism genotyping was performed using high-throughput matrix-assisted laser desorption ionization time of flight (MALDI-TOF) mass spectrometry. Briefly, primers and probes were designed with SpectroDESIGNER software (Sequenom, San Diego, CA, USA). A multiplex PCR was performed and unincorporated dNTPs (deoxyribonucleotide triphosphates) were dephosphorylated with shrimp alkaline phosphatase (Hoffman-LaRoche, Basel, Switzerland) followed by primer extension. The purified primer extension reaction was spotted onto a 384-element silicon chip (SpectroCHIP; Sequenom) and analyzed in the Bruker Biflex III MALDI-TOF SpectroREADER mass spectrometer (Sequenom). The resulting spectra were processed with SpectroTYPER (Sequenom).
Statistical analysis was performed using SPSS 10.0. Differences in continuous variables were evaluated using Student's t-test. Categorical data were analyzed using the χ2-test or Fisher's exact test, if necessary. A logistic regression analysis was performed using antidepressant response as the dependent variable, and age, sex, antidepressant dose and haplotypes as the predictor variables. The probability of a type one error was set at a maximum level of 0.05. Data are presented as the mean (s.d.).
For haplotype analysis, the software SNP Alyze V3.2 (Dynacom Co. Ltd, Kanagawa, Japan) was used to evaluate the status of pairwise disequilibrium (LD) for the studied polymorphisms, to infer the haplotype frequency and to determine whether haplotype frequency varied between groups. The significance level of these analyses obtained from the SNP Alyze V3.2 was set as P<0.05 after 100 000 permutation tests.
The mean age (t=−1.026, df=636, P=0.305) and sex distributions (χ=0.536, df=1, P=0.453) of the MDD patients and the controls were similar. The genotypes and allele distribution of the four GSK3B SNPs for the MDD group and controls are shown in Table 1. The distributions of the genotypes for MDD cases and controls were in Hardy–Weinberg equilibrium (Table 1).
The analysis for single locus effects showed no significant association between the GSK3B SNPs and MDD (Table 1). Logistic regression analysis (with sex and age being entered as covariates) also failed to show significant association between each of the studied SNPs and MDD (all P>0.60). The association of the genetic sequences with MDD was further investigated by looking into the haplotypes. The four markers were found to be in strong LD to each other in both control (D′=0.78–1.0) and case (D′=0.83–1.0) groups (Figures 1b and c). The results of global case–control haplotype analysis and comparisons of individual haplotypes between groups are presented in Table 2. Global case–control haplotype analysis showed that there was a significant difference in haplotype distribution between the groups (P=0.026). However, individual haplotype analysis showed that the frequencies of all haplotypes were similar in the control and case groups.
A total of 168 (73.0%) and 117 (50.9%) of the MDD patients completed 4-week and 8-week fluoxetine/citalopram treatment, respectively, and underwent evaluation using the HAM-D. The mean fluoxetine/citalopram doses for these patients was 20.1 mg day−1 (s.d.: 2.7) at week 4, and 75 (44.6%) of these patients showed at least a 50% decrease in HAM-D total score after taking antidepressant medication for 4 weeks. There were no significant differences in gender (responder vs non-responder (M/F): 29/46 vs 47/46, χ=2.36, df=1, P=0.124) or baseline HAM-D score (responder vs non-responder: 28.2±5.3 vs 28.0±4.8, t=0.221, df=166, P=0.825) between the responder and the non-responder groups. However, the non-responder group had an older mean age compared with the responder group (responder vs non-responder: 39.2±14.2 vs 45.4±15.7 years, t=−2.674, df=166, P=0.008). Meanwhile, the baseline HAM-D scores and doses of antidepressant were comparable between the genotypes of each studied SNP (all P>0.1, Supplementary Table A). The GSK3B genotype distribution and allele frequencies in the SSRI responders and non-responders appear in Table 3. Response to SSRI was significantly associated with all the GSK3B SNPs except rs6808874. The associations between the first three SNPs (rs334558, rs13321783 and rs2319398) and the 4-week response to SSRI remained significant after Bonferroni correction for the four investigated markers. However, for the MDD patients (n=117) who completed 8-week SSRI treatment and evaluation, the analysis for single locus effects showed no significant association between the GSK3B SNPs and SSRI treatment response (data not shown).
Genetic sequence association with SSRI treatment response was further investigated by looking into the haplotypes. There was a significant difference in haplotype distribution between the responders and non-responders after the 4-week SSRI treatment (P=0.028) (Table 4A) and TAGT was the only haplotype significantly associated with treatment response (P<0.0001). The difference in TAGT haplotype distribution between responders and non-responders remained significant after the 8-week SSRI treatment (P=0.015) (Table 4B). There was no significant difference in the mean baseline HAM-D scores and antidepressant doses between the TAGT haplotype carriers and non-carriers (all P>0.1, Supplementary Table B). Thus, patients with the TAGT haplotype had a significantly poorer response to both the 4-week and the 8-week SSRI treatment than patients with other haplotypes. The prediction levels were calculated by logistic regression analysis using SSRI antidepressant response as the dependent variable, and age, gender, antidepressant dose and GSK3B haplotype (TAGT vs non-TAGT) as predictor variables. For the 4-week treatment, increased patient age (Wald=6.129, df=1, P=0.13) and carrying the GSK3B haplotype TAGT (P=0.001; odds ratio (OR)=3.08, 95% confidence interval (CI)=1.60–6.00) were significant predictors of a poor response to SSRI treatment. For the 8-week treatment, carrying the haplotype TAGT was still a predictor of a poor response to SSRT treatment (Wald=7.33, df=1, P=0.007, OR=3.10, 95% CI=1.34–7.02).
The relationship between GSK3B polymorphisms and antidepressant response was also evaluated with respect to 8-week treatment remitter, defined as total HAM-D scores less than 8 at the end of the 8-week antidepressant treatment. In single-marker-based analysis, none of the studied SNP showed significant association with the 8-week remitter of antidepressant treatment (data are not shown, all P>0.1). However, in the haplotype-based analysis, we found that the distribution of GSK3B haplotypes was significantly different between the 8-week remitters and non-remitters (Table 5, global P=0.034). The second major haplotype TAGT was associated with the non-remitters (Table 5, P=0.006) and the other two haplotypes, TGTA and CGGA, were significantly associated with the 8-week remitters (Table 5, P=0.025 and 0.004, respectively). Moreover, after controlling for the effect of age, gender and antidepressant doses, results from the logistic regression also showed that the TAGT carriers had increased risk to be 8-week non-remitters (Wald=3.938, df=1, P=0.047, OR=2.50, 95% CI=1.01–6.16).
Several lines of evidence indicate that GSK3B is a good candidate molecule for MDD susceptibility and antidepressant therapeutic effect. In this study, we investigated whether common SNPs in the GSK3B gene were associated with MDD and with the SSRI antidepressant response. Significant association with MDD was not shown in the alleles and genotypes of single loci. Subsequent analysis of four-locus GSK3B haplotypes was performed, but we still could not find the association between GSK3B haplotype and MDD susceptibility, suggesting that the GSK3B SNPs investigated did not play a major role in MDD pathogenesis.
We also investigated whether GSK3B haplotypes were associated with SSRI response. SSRIs are currently the first-line medication for treatment of MDD. Like other antidepressant therapies, there is great variability among patients in terms of response to SSRI treatment. Of the possible factors causing interindividual variability in SSRI response, genetic factors may play an important role and the identification of genetic factors underlying response to SSRI therapy may help to predict therapeutic response and facilitate determination of optimal drug selection.17 The major finding of this study is that three of the four GSK3B SNPs investigated are strongly associated with 4-week SSRI antidepressant therapeutic response. This association between the GSK3B genetic variants and therapeutic response to SSRIs is supported by data from animal studies showing that GSK3 inhibitors had antidepressant-like activity10, 11 and that fluoxetine inhibited GSK3B activity.3 Furthermore, a recent study of Italian bipolar depression patients found that the C/C homozygote for the GSK3B rs334558 polymorphism (−50C/T) SNP showed acute effects of total sleep deprivation treatment on perceived mood.18 This is similar to our findings that MDD patients with rs334558 C/C homozygote have a better response to 4-week SSRI treatment than carriers of the T allele (Table 3). It is unknown how these GSK3B polymorphisms affect the SSRI antidepressant effect. These SNPs may have a direct effect on the therapeutic mechanisms of SSRIs. Among the SNPs associated with 4-week SSRI response, the rs334558 polymorphism (−50C/T) identified by Russ et al.19 by sequencing over 3000 bp of the GSK3B putative promoter is in the region upstream of the start codon, which falls into the effective promoter region (nt −171 to +29) of the GSK3B gene. A recent study showed that this polymorphism is associated with transcriptional strength in vitro in which the T allele has greater activity.16 It is also possible that the association of GSK3B polymorphisms and SSRI antidepressant response is due to LD between these polymorphisms and nearby functional polymorphisms. This is supported by our findings that GSK3B haplotypes, but not single loci, are associated with 8-week SSRI treatment as well as with 8-week remission. For complex phenotypes such as treatment response, single-marker association studies are known to be prone to generate weak association and/or conflicting results.20 Because haplotype analysis combines information concerning two or more known polymorphisms in the same candidate gene, haplotypes are more specific risk markers than single alleles and are currently the focus of intensive research effort. As shown in the present study, although the rs334558 in the promoter has not been associated with 8-week SSRI treatment as a single locus risk factor, its interaction effect comes up in combination with the other three GSK3B SNPs.
The primary limitation of this study is that the average age in our control group was relatively young, and thus parts of the participants in the control group might experience bipolar disorder in their later life. Recruiting relatively young age controls might be a possibility accounting for our negative findings for the association of GSK3B genetic polymorphisms with MDD, as this study is a cross-sectional one in nature. The second limitation of this study is that some factors possibly influencing treatment effect, such as a history of suicide, age of onset, drug side effects from treatment, prior treatment and dropout during treatment, were not excluded. The third limitation is the lack of a placebo control. It is well known that placebo response plays an important role as a notable component of therapeutic response to antidepressant agents.21 Thus, we cannot exclude the possibility that some patients in the responder group had responded to the placebo effect. The lack of a placebo control would limit whether and to what extent the findings in the present study can be attributed to a drug-related genetic or biological factor. The fourth limitation is that anxiolytics were allowed for insomnia in this study, which may confound the SSRI therapeutic evaluation. Finally, plasma levels of fluoxetine/citalopram were not analyzed in these outpatients. Thus, we do not know the patients’ drug compliance and the condition of drug metabolism. However, the later effect is minimal, as a previous study has demonstrated that there were no significant relationships between fluoxetine blood levels and clinical response in MDD patients.22 Despite these limitations, this study demonstrates that GSK3B genetic variants are associated with SSRI antidepressant therapeutic response. Confirmation of these preliminary findings in other populations might lead to GSK3B being considered as a useful candidate gene for studying antidepressant pharmacogenetics and the development of novel antidepressants.
In summary, inhibition of GSK3B is thought to play an important role in the therapeutic mechanism of antidepressants. The study, for the first time, demonstrated that GSK3B genetic variants are related to the SSRI antidepressant therapeutic response. Our findings support the hypothesis that drugs regulating GSK3B activity may represent a novel treatment strategy for MDD. In addition, the finding that the GSK3B genetic variants represent potential biological predictors of antidepressant response may help to establish tailored antidepressant treatment.
Jope RS, Johnson GV . The glamour and gloom of glycogen synthase kinase-3. Trends Biochem Sci 2004; 29: 95–102.
Frame S, Cohen P, Biondi RM . A common phosphate binding site explains the unique substrate specificity of GSK3 and its inactivation by phosphorylation. Mol Cell 2001; 7: 1321–1327.
Li X, Zhu W, Roh MS, Friedman AB, Rosborough K, Jope RS . In vivo regulation of glycogen synthase kinase-3beta (GSK3beta) by serotonergic activity in mouse brain. Neuropsychopharmacology 2004; 29: 1426–1431.
Roh MS, Eom TY, Zmijewska AA, De Sarno P, Roth KA, Jope RS . Hypoxia activates glycogen synthase kinase-3 in mouse brain in vivo: protection by mood stabilizers and imipramine. Biol Psychiatry 2005; 57: 278–286.
Klein PS, Melton DA . A molecular mechanism for the effect of lithium on development. Proc Natl Acad Sci USA 1996; 93: 8455–8459.
Stambolic V, Ruel L, Woodgett JR . Lithium inhibits glycogen synthase kinase-3 activity and mimics wingless signalling in intact cells. Curr Biol 1996; 6: 1664–1668.
Li X, Friedman AB, Zhu W, Wang L, Boswell S, May RS et al. Lithium regulates glycogen synthase kinase-3beta in human peripheral blood mononuclear cells: implication in the treatment of bipolar disorder. Biol Psychiatry 2007; 61: 216–222.
D’Sa C, Duman RS . Antidepressants and neuroplasticity. Bipolar Disord 2002; 4: 183–194.
Mai L, Jope RS, Li X . BDNF-mediated signal transduction is modulated by GSK3beta and mood stabilizing agents. J Neurochem 2002; 82: 75–83.
Kaidanovich-Beilin O, Milman A, Weizman A, Pick CG, Eldar-Finkelman H . Rapid antidepressive-like activity of specific glycogen synthase kinase-3 inhibitor and its effect on beta-catenin in mouse hippocampus. Biol Psychiatry 2004; 55: 781–784.
Gould TD, Einat H, Bhat R, Manji HK . AR-A014418, a selective GSK-3 inhibitor, produces antidepressant-like effects in the forced swim test. Int J Neuropsychopharmacol 2004; 7: 387–390.
Karege F, Perroud N, Burkhardt S, Schwald M, Ballmann E, La Harpe R et al. Alteration in kinase activity but not in protein levels of protein kinase B and glycogen synthase kinase-3beta in ventral prefrontal cortex of depressed suicide victims. Biol Psychiatry 2007; 61: 240–245.
Hamilton M . Development of a rating scale for primary depressive illness. Br J Soc Clin Psychol 1967; 6: 278–296.
Yu YW, Tsai SJ, Chen TJ, Lin CH, Hong CJ . Association study of the serotonin transporter promoter polymorphism and symptomatology and antidepressant response in major depressive disorders. Mol Psychiatry 2002; 7: 1115–1119.
Gabriel SB, Schaffner SF, Nguyen H, Moore JM, Roy J, Blumenstiel B et al. The structure of haplotype blocks in the human genome. Science 2002; 296: 2225–2229.
Kwok JB, Hallupp M, Loy CT, Chan DK, Woo J, Mellick GD et al. GSK3B polymorphisms alter transcription and splicing in Parkinson's disease. Ann Neurol 2005; 58: 829–839.
Lerer B, Macciardi F . Pharmacogenetics of antidepressant and mood-stabilizing drugs: a review of candidate-gene studies and future research directions. Int J Neuropsychopharmacol 2002; 5: 255–275.
Benedetti F, Serretti A, Colombo C, Lorenzi C, Tubazio V, Smeraldi E . A glycogen synthase kinase 3-beta promoter gene single nucleotide polymorphism is associated with age at onset and response to total sleep deprivation in bipolar depression. Neurosci Lett 2004; 368: 123–126.
Russ C, Lovestone S, Powell JF . Identification of sequence variants and analysis of the role of the glycogen synthase kinase 3 beta gene and promoter in late onset Alzheimer's disease. Mol Psychiatry 2001; 6: 320–324.
Cardon LR, Bell JI . Association study designs for complex diseases. Nat Rev Genet 2001; 2: 91–99.
Rausch JL, Johnson ME, Fei YJ, Li JQ, Shendarkar N, Hobby HM et al. Initial conditions of serotonin transporter kinetics and genotype: influence on SSRI treatment trial outcome. Biol Psychiatry 2002; 51: 723–732.
Burke WJ, Hendricks SE, McArthur-Campbell D, Jacques D, Stull T . Fluoxetine and norfluoxetine serum concentrations and clinical response in weekly versus daily dosing. Psychopharmacol Bull 1996; 32: 27–32.
This work was supported by Grant NSC 92-2314-B-075-087 from the National Science Council, Taiwan and Grant VGH-92-161 from the Taipei Veterans General Hospital. We thank Jer-Yuan Wu, the National Genotyping Center and the National Clinical Core at Academia Sinica, Taiwan, for genotyping support.
About this article
Cite this article
Tsai, SJ., Liou, YJ., Hong, CJ. et al. Glycogen synthase kinase-3β gene is associated with antidepressant treatment response in Chinese major depressive disorder. Pharmacogenomics J 8, 384–390 (2008). https://doi.org/10.1038/sj.tpj.6500486
- glycogen synthase kinase-3β
- major depressive disorders
Interactions Between Glycogen Synthase Kinase-3β Gene Polymorphisms, Negative Life Events, and Susceptibility to Major Depressive Disorder in a Chinese Population
Frontiers in Psychiatry (2021)
Current challenges and possible future developments in personalized psychiatry with an emphasis on psychotic disorders
NRG1, PIP4K2A, and HTR2C as Potential Candidate Biomarker Genes for Several Clinical Subphenotypes of Depression and Bipolar Disorder
Frontiers in Genetics (2020)
Systems Approach to Identify Common Genes and Pathways Associated with Response to Selective Serotonin Reuptake Inhibitors and Major Depression Risk
International Journal of Molecular Sciences (2019)