Peripheral Clock System Abnormalities in Patients With Parkinson's Disease

Abstract

Objective: To evaluate the altered expression of peripheral clock genes, circulating melatonin levels, and their correlations with sleep-wake phenotypes including probable rapid eye movement sleep behavior disorder (pRBD) symptoms in a relatively large population of Parkinson's disease (PD) patients. Methods: We determined the expression profiles of five principal clock genes, BMAL1, CLOCK, CRY1, PER1, and PER2, in the peripheral blood mononuclear cells (PBMCs) of PD patients (n = 326), and healthy controls (HC, n = 314) using quantitative real-time PCR. Melatonin concentration in the plasma of two groups was evaluated by enzyme-linked immunosorbent assay. Then we performed comprehensive association analyses on the PBMCs clock gene expression, plasma melatonin levels and sleep characteristics. Results: Our data showed that the expression levels of BMAL1, CLOCK, CRY1, PER1, and PER2 were significantly decreased in the PBMCs of PD as compared with that of HC (P < 0.05). PD patients had reduced plasma melatonin levels compared with HC (P < 0.0001). pRBD and excessive daytime sleepiness are common in these PD patients and are associated with the expression levels of all five clock genes (r = -0.344∼-0.789, P < 0.01) and melatonin concentration (r = -0.509∼-0.753, P < 0.01). Statistical analyses also revealed that a combination of five clock genes and melatonin could reach a high diagnostic performance (areas under the curves, 97%) for PD comorbid pRBD. Conclusion: This case-control study demonstrates that peripheral BMAL1, CLOCK, CRY1, PER1, PER2, and melatonin levels are altered in PD patients and may serve as endogenous markers for sleep and wakefulness disturbances of PD.

Keywords: Parkinson’s disease; circadian rhythm; clock gene; melatonin; sleep-wake disturbances.

FIGURE 1

Scatter plots of BMAL1, CLOCK, CRY1, PER1, and PER2 relative mRNA expression levels in the PBMCs of HC (n = 314) and PD (n = 326). (A–E) Significantly lower levels of BMAL1 (A, P < 0.0001), CLOCK (B, P < 0.01), CRY1 (C, P < 0.01), PER1 (D, P < 0.05) and PRY2 (E, P < 0.01) were seen in PD patients compared to those in HC. Horizontal bars represent mean and SE values. *P < 0.05, **P < 0.01, and ****P < 0.0001.

FIGURE 2

Scatter plots of BMAL1, CLOCK, CRY1, PER1, and PER2 relative mRNA expression levels in the PBMCs of HC and different subgroups of PD. (A–E) PD patients with pRBD (n = 88) had significantly lower expression levels of five clock genes compared with those in HC (n = 314, BMAL1, P < 0.0001; CLOCK, P < 0.01; CRY1, P < 0.05; PER1, P < 0.05; PER2, P < 0.01, respectively), and had decreased level of BMAL1 compared with that of PD patients without pRBD (n = 176, P < 0.05). (F–J) Expression levels of all five clock genes in PD patients with EDS (n = 160) were significantly lower than that of HC (n = 314, BMAL1, P < 0.0001; CLOCK, P < 0.001; CRY1, P < 0.01; PER1, P < 0.05; PER2, P < 0.01, respectively). Significantly differences of BMAL1 (P < 0.01) and CLOCK (P < 0.05) expression levels were seen between the subgroups of PD patients with EDS and without EDS. Horizontal bars represent mean and SE values. *P < 0.05, **P < 0.01, ***P < 0.001, and ****P < 0.0001.

FIGURE 3

Scatter plots of melatonin concentration in the plasma of HC (n = 156) and PD (n = 153). (A) Plasma melatonin concentration was significantly decreased in patients with PD than in HC (P < 0.0001). (B) PD patients with pRBD (n = 56) had significantly lower melatonin concentration compared with those in HC (n = 156, P < 0.0001) and PD patients those without pRBD (n = 97, P < 0.0001). A significant difference in melatonin concentration was found between HC and PD without pRBD (P < 0.0001). (C) Alterations of plasma melatonin concentration were found between PD with EDS (n = 92) and PD without EDS (n = 64, P < 0.0001), and both of two groups have significant lower melatonin concentration when comparing to HC (PD with EDS, P < 0.0001; PD without EDS, P < 0.01). ^∗∗P < 0.01 and ^****P < 0.0001.

FIGURE 4

Receiver operating characteristic (ROC) curves of BMAL1, CLOCK, CRY1, PER1, PER2 and melatonin levels for PD with pRBD vs. PD without pRBD, and those for PD with EDS vs. PD without EDS. (A) The AUCs values of BMAL1, CLOCK, CRY1, PER1, PER2, melatonin, and the combination of five clock genes and melatonin for pRBD were 0.94 (BMAL1, 95% CI, 0.9–0.97, P < 0.05), 0.8 (CLOCK, 95% CI, 0.73–0.88, P < 0.05), 0.81 (CRY1, 95% CI, 0.74–0.88, P < 0.05), 0.76 (PER1, 95% CI, 0.68–0.84, P < 0.05), 0.8 (PER2, 95% CI, 0.73–0.87, P < 0.05), 0.82 (melatonin 95% CI, 0.75–0.89, P < 0.05), and 0.97 (combined five clock genes and melatonin, 95% CI, 0.94–0.99, P < 0.05), respectively. (B) The AUCs values for EDS were 0.86 (BMAL1, 95% CI, 0.8–0.92, P < 0.05), 0.82 (CLOCK, 95% CI, 0.75–0.89, P < 0.05), 0.78 (CRY1, 95% CI, 0.71–0.86, P < 0.05), 0.73 (PER1, 95% CI, 0.65–0.81, P < 0.05), 0.7 (PER2, 95% CI, 0.61–0.78, P < 0.05), 0.83 (melatonin, 95% CI, 0.76–0.9, P < 0.05), and 0.91 (combined five clock genes and melatonin, 95% CI, 0.86–0.96, P < 0.05), respectively.

FIGURE 5

Expression levels of the five clock genes and melatonin in different sunlight intensity of HC and patients with PD. (A) A schematic diagram of the sunlight conditions when sampling in different groups of months. The time of blood sampling from the subjects was fixed at 6–7 a.m. Subjects were divided into 4 groups according to the local sunrise schedule. I: Sampling before the sunrise (January, February, and December, n (HC: PD) = 79: 80); II: Sampling around the sunrise (March and November, n (HC: PD) = 58: 64); III: Sampling 0.5 h after the sunrise (April, September, and October, n (HC: PD) = 57: 63); IV: Sampling 1 h after the sunrise (May, June, July, and August; n (HC: PD) = 120: 119). (B–F) PBMCs levels of BMAL1 (B), CLOCK (C), CRY1 (D), PER1 (E), and PER2 (F) expression in the group of PD and HC, with different intensities of sunlight. (G) Plasma melatonin concentration in PD and HC with different intensities of sunlight. The results are the mean ± SEM values. *P < 0.05, **P < 0.01, and ****P < 0.0001.