Downregulation of Circadian Genes, PER1, PER2, and PER3, in Chronic Myeloid Leukemia.

Circadian rhythm is present in all eukaryotic and some prokaryotic life forms. This time-keeping system is organized in a hierarchical fashion and composed of a self-sustained central pacemaker in the suprachiasmatic nuclei of the anterior hypothalamus and peripheral oscillators in most body cells. A recent study demonstrated that mice deficient in the clock gene mPer2 are cancer prone and display a deregulated temporal expression genes involved in cell cycle regulation, such as c-Myc, Cyclin D1, Cyclin A, and Mdm-2. These mice display salivary gland hyperplasia, and a large portion of them develops lymphomas. The results revealed that Per2 is an essential circadian gene and it is associated with proliferation control in mammals. In a more recent study, it was demonstrated that an age-dependent decay of the circadian clock both at the behavioral and the molecular levels was observed in mPer1^−/−mCry2^−/− double-mutant mice. To gain further insights into the roles of circadian genes in chronic myeloid leukemia (CML), we analyzed peripheral blood from 21 healthy individuals and 35 CML patients (18 in blastic crisis and 17 in chronic phase) by real-time quantitative RT-PCR for the expression of circadian genes PER1, PER2, and PER3. In blastic crisis cases, the expression levels of all three PER genes were significantly impaired than in healthy individuals (PER1, 1:5.97, p < 0.005; PER2, 1:13.51, p < 0.0001; PER3, 1:2181.33; p < 0.0001); whereas, in chronic phase, only the expression levels of PER2 and PER3 were significantly impaired (PER1, 1:1.23, p > 0.5; PER2, 1:2.47, p < 0.05; PER3, 1:14.22; p < 0.0001). Mutational analysis of the whole gene and methylation analysis of CpGs sites at the promoter area were further performed to investigate the possible mechanisms. No mutation was found within the coding regions of the three PER genes in all CML cases. Methylation analysis using methylation-specific PCR and direct sequencing showed no methylation in the promoter areas of both PER1 and PER2 genes. In contrast, most of the CpG sites were methylated in the promoter area of PER3 gene in CML cases and none of these CpG sites were methylated in healthy individual cases. In addition, the methylated CpG frequencies of PER3 gene differed in patients at blastic crisis and at chronic phase (CpG, 8.24 ± 0.73 vs. 4.48 ± 0.48, p < 0.0001; T/CpG, 10.47 ± 0.67 vs. 14.67 ± 0.46, p < 0.0001; TG, 1.24 ± 0.32 vs. 0.81 ± 0.16, p > 0.05). Demethylation by treatment of hypermethylated K562 cells with 5′-aza-2′-deoxycytidine resulted in partial reactivation of PER3 expression. Our results suggest that the downregulated PER3 expression in CML was due to the inactivation of PER3 gene by methylation and the differential expression was correlated to the ratio of methylated CpG sites at the promoter region. We hope to explore the role of circadian genes in the tumorigenesis of leukemia and to establish circadian genes as novel and useful markers for diagnosis and references for therapies in leukemia.