1. ¹Department of Genetics, University of Pennsylvania, Philadelphia, PA, USA
2. ²Sleep and Performance Research Center, Washington State University, Spokane, WA, USA
3. ³Department of Psychiatry, University of Pennsylvania, Philadelphia, PA, USA
4. ⁴Penn Center for Bioinformatics, University of Pennsylvania, Philadelphia, PA, USA

Correspondence: Professor M Bucan, Department of Genetics, University of Pennsylvania School of Medicine, 528 CRB, Philadelphia, PA 19104-6141, USA. E-mail: bucan@pobox.upenn.edu

Received 24 June 2007; Revised 27 November 2007; Accepted 5 December 2007; Published online 26 February 2008.

Abstract

Previous studies have implicated the circadian system in the pathophysiology of bipolar disorder, but conclusive evidence for altered circadian clocks is lacking. Cultured fibroblasts harbor circadian clocks representative of those in the master clock resident in the suprachiasmatic nuclei, providing a new avenue to investigate the core clock machinery in patients with bipolar illness. We examined the rhythmic expression patterns of core clock genes (BMAL1, PER1, PER2, REV-ERB, DEC2, DBP) in fibroblasts from 12 bipolar patients and 12 healthy controls. Although we did not detect differences in the circadian period between bipolar patients and controls, the amplitude of rhythmic expression for BMAL1, REV-ERB and DBP, as well as the overall mRNA expression level for DEC2 and DBP was reduced in fibroblasts from bipolar patients. Bonferroni's correction for multiple comparisons still resulted in significantly reduced DBP expression level, and trends toward reduced overall expression level of DEC2 and circadian amplitude of BMAL1, in fibroblasts from bipolar patients. We next examined an expanded cohort of 18 bipolar patients and 35 healthy controls for mRNA expression levels of four kinases (CKI, CKI, GSK3 and GSK3) and the protein and phosphorylation levels of two of them (GSK3 and GSK3). We did not detect differences in steady-state mRNA levels or protein levels of these kinases between bipolar patients and controls, but the level of GSK3 phosphorylation was significantly reduced in bipolar patients within an Old Order Amish bipolar kindred. Our results suggest that the reduced amplitudes and overall expression levels of circadian genes, and the decreased phosphorylation level of GSK3 may lead to dysregulation of downstream genes, which could explain some pathological features of bipolar disorder.