Bipolar disorder is a highly heritable psychiatric disorder but the aetiology of the disease is complex, involving multiple genetic and environmental influences. Identifying genetic variants associated with bipolar disorder will increase the understanding of disease mechanisms and may lead to development of targeted therapeutics. Several genome-wide association studies have been conducted that together suggest that the genetic susceptibility of bipolar disorder is clearly polygenic in nature. Reference Craddock and Sklar1
One identified single nucleotide polymorphism (SNP) that has been linked to bipolar disorder is of particular interest: rs1006737, which is situated within intron 3 of the gene CACNA1C at chromosome 12, with the A allele associated with an increased risk. Reference Craddock and Sklar1 This gene codes the alpha 1C subunit (Cav1.2) of the L-type voltage-gated calcium channel. Reference Catterall2 Cav1.2 couples transient activation of inward calcium current to transcriptional regulation, which may play a role in dendritic development, neuronal survival, synaptic plasticity, memory formation, learning and behaviour. Reference Bhat, Dao, Terrillion, Arad, Smith and Soldatov3 The minor A allele has been associated with executive function deficits and has also been coupled to brain areas involved in affective regulation. Reference Bigos, Mattay, Callicott, Straub, Vakkalanka and Kolachana4 This allele has also been associated with schizophrenia and major depression, consistent with shared familial risks across psychiatric disorders. Reference Smoller, Ripke, Lee, Neale, Nurnberger and Santangelo5 However, the mechanisms underlying how genetic variants in CACNA1C may modify risk for psychiatric disorders and have an impact on cognition are unclear. One way to increase our understanding of the pathophysiology is to study the neurochemical correlates of CACNA1C gene variants. We have previously characterised the neurochemical profile of bipolar disorder by analyses of cerebrospinal fluid (CSF) markers of neuronal and glial function and degeneration in a large cohort of patients with bipolar disorder and healthy controls. Reference Jakobsson, Bjerke, Ekman, Sellgren, Johansson and Zetterberg6,Reference Jakobsson, Zetterberg, Blennow, Ekman, Johansson and Landen7 Here, we explored the association between the rs1006737 SNP and these CSF markers in patients with bipolar disorder and healthy controls.
Method
Our sample comprised 132 patients with bipolar disorder (type I: n = 66; type II: n = 44; other types n = 22) and 54 healthy controls. Patients were recruited from the St Göran Bipolar Project, enrolling patients from the bipolar unit at the Northern Stockholm Psychiatric Clinic, Stockholm, Sweden. All patients were assessed using a standardised interview protocol (the Affective Disorders Evaluation) Reference Sachs, Thase, Otto, Bauer, Miklowitz and Wisniewski8 previously used in the Systematic Treatment Enhancement Program of Bipolar Disorder (STEP-BD). The study was approved by the Regional Ethics Committee in Stockholm and conducted in accordance with the latest Helsinki Protocol. After a complete description of the study, all enrolled patients and controls consented orally and in writing to participate in the study. See the online supplement DS1 for further details.
The CSF sampling (lumbar puncture) was performed when the participants were in a stable euthymic mood. Participants fasted overnight before CSF collection, which took place between 09.00 and 10.00 h. CSF samples were divided into 1.0–1.6 ml aliquots that were stored at −80° C pending analysis. The CSF concentrations of neurofilament light chain (NF-L), S100B, myelin basic protein (MBP) and heart-type fatty acid binding protein (H-FABP), hyperphosphorylated tau (P-tau), total tau (T-tau), soluble amyloid precursor protein alpha (sAPP-α), soluble amyloid precursor protein beta (sAPP-β), Aβ1-42, AβX-38, AβX-40 and AβX-42, were analysed as described by their respective manufacturers. We also included several biomarker ratios in the analysis: AβX-42/AβX-40, AβX-42/AβX-38, and P-tau/T-tau. See online supplement DS1 for further details.
The CACNA1C rs1006737 SNP was genotyped with the KASPar PCR SNP genotyping system (KBioscience, Hoddesdon, UK; www.lgcgenomics.com). SPSS Statistics version 20 was used for all statistical analyses. Analysis of covariance (ANCOVA) with age and gender as covariates was used to analyse effects of rs1006737 on CSF marker concentrations. All P-values are presented as two-tailed. Bonferroni correction was used to correct for multiple comparisons (α = 0.05/15 = 0.00333).
Results
The CACNA1C rs1006737 SNP was genotyped in the 132 patients with bipolar disorder. The frequency for the A allele was 0.40 with the genotypes distributed according to Hardy–Weinberg equilibrium (χ2 = 0.593, P = 0.441). AA (n = 23) and AG (n = 59) genotype carriers were grouped together for analysis of the effects of the A and G alleles on CSF biomarker levels. Demographics of the two groups are displayed in online Table DS1. The groups did not differ significantly with regard to age, gender, smoking status, medications, previous episodes of psychosis, Global Assessment of Functioning Reference Jones, Thornicroft, Coffey and Dunn9 score, Clinical Global Investigation Reference Guy10 score, diagnosis, duration of illness or number of episodes. A low P-tau/T-tau ratio was significantly associated with the A allele group (F(1,128) = 13.484, P<0.001, α= 0.00333) (online Fig. DS1), whereas the rs1006737 genotype had no effect on any of the other biomarkers (online Table DS2). We also found a significant association between the P-tau/T-tau ratio and the A allele under an additive model (β = −0.260, P = 0.002, age and gender as covariates). We next analysed whether this association was specific to bipolar disorder by analysing healthy controls (n = 54). The frequency for the A allele of rs1006737 in the control group was 0.30 with the genotypes distributed according to Hardy–Weinberg equilibrium (χ2 = 0.675, P = 0.411). In the control group, the rs1006737 SNP was not associated with the P-tau/T-tau ratio (F(1,50) = 0.275, P = 0.602) or with any of the other CSF biomarkers (online Table DS3).
Discussion
Variations in CACNA1C has previously been linked to various brain functions but it is unclear how these variations affect the brain on a chemical level. Here, we found a significant association between the rs1006737 SNP and the CSF P-tau/T-tau ratio in patients with bipolar disorder. No association was found in healthy controls, implying that the association is not a general physiological phenomenon but occurs in patients with a psychiatric illness.
Cav1.2 is primarily regulated through an interaction with Ca2+-bound calmodulin (CaM), which also mediates the downstream effects of Cav1.2. Reference Halling, Aracena-Parks and Hamilton11 Downstream effectors of CaM include the CaM-dependent protein kinase (CaMK) cascade and the mitogen-activated protein kinase (MAPK) pathway. Reference Bhat, Dao, Terrillion, Arad, Smith and Soldatov3 Interestingly, phosphorylation of tau is regulated by a range of proline-directed and non-proline-directed kinases, including CaMK and MAPK, Reference Avila, Lucas, Perez and Hernandez12 linking calcium signalling with phosphorylation of tau. Phosphorylation of tau reduces its binding to microtubules leading to destabilisation of microtubules and promoting cytoskeletal flexibility, which have been suggested to be important for axonal and synaptic growth/development and thus neurodevelopment and synaptic plasticity. Reference Avila, Lucas, Perez and Hernandez12 In addition, tau phosphorylation is markedly increased in brain tissue in pathological conditions (i.e. tauopathies), and in CSF in Alzheimer's disease (for a review see Blennow et al Reference Blennow, Hampel, Weiner and Zetterberg13 ). There are, however, no differences between patients with bipolar disorder and controls in either P-tau or T-tau concentrations. Reference Jakobsson, Zetterberg, Blennow, Ekman, Johansson and Landen7 Thus, the difference in P-tau/T-tau between rs1006737 risk allele carriers and non-risk allele carriers probably reflects alterations in the regulation of tau phosphorylation.
Importantly, this study links variations in the CACNA1C gene to neuroaxonal plasticity at the neurochemical level in people with bipolar disorder. Further studies are, however, needed to sort out the biological and clinical significance of altered tau phosphorylation in relation to CACNA1C polymorphism in bipolar disorder and other psychiatric disorders.
Acknowledgements
We thank Åsa Källén, Monica Christiansson, Sara Hullberg, Lobna Almasalmeh and Dzemila Secic for excellent technical assistance, the staff at the St Görans Bipolar Affective Disorder Unit, including the coordinator, Martina Wennberg, the study nurse, Agneta Carlswärd-Kjellin, and the data managers, Haydeh Olofsson and Mathias Kardell, for the diagnostic assessments and enrolling patients in this study. Yngve Hallström is acknowledged for performing lumbar punctures on patients and controls.
eLetters
No eLetters have been published for this article.