Thank you for visiting nature.com. You are using a browser version with limited support for CSS. To obtain the best experience, we recommend you use a more up to date browser (or turn off compatibility mode in Internet Explorer). In the meantime, to ensure continued support, we are displaying the site without styles and JavaScript.

# Bright light improves sleep in patients with Parkinson’s disease: possible role of circadian restoration

## Abstract

Parkinson’s disease (PD) is one of the most common neurodegenerative disorders. Among the most common manifestations of PD are sleep problems, which are coupled with the adverse effects of dopaminergic therapies (DT). A non-pharmacological solution for these sleep problems has been sought to avoid additional pharmacological intervention. Here, we show that bright light therapy (BLT) is effective for improving sleep in Japanese PD patients receiving DT. Furthermore, experimental evaluation of peripheral clock gene expression rhythms revealed that most PD patients receiving DT who experienced improved sleep following BLT showed a circadian phase shift, indicating the existence of a correlation between circadian modulation and sleep improvement. Conversely, this result indicates that sleep problems in PD patients receiving DT may arise at least in part as a result of circadian dysfunction. Indeed, we found that chronic dopaminergic stimulation induced a rapid attenuation of autonomous oscillations of clock gene expression in ex vivo cultured mouse suprachiasmatic nucleus (SCN) at the single neuron level. In conclusion, BLT is a promising medical treatment for improving sleep in PD patients receiving DT. This BLT-induced improvement may be due to the restoration of circadian function.

## Introduction

Parkinson’s disease is one of the most common neurodegenerative disorders, affecting approximately 10 million people worldwide1. First-line treatment strategies involve implementation of dopaminergic therapies such as dopamine replacement therapy using levodopa which is converted to dopamine by enzymes in the brain2,3. Among the most common manifestations of PD are sleep problems, which reduce quality of life and impair daytime functioning. For example, in a large survey of non-motor symptoms, 64% of PD patients reported sleep problems, representing the second most common non-motor complaint4. Previous studies have suggested that PD symptoms and the adverse effects of medications may be some of the causes of sleep problems in PD patients5,6,7. Because of the risk of interference with treatment regimens, few treatment options for sleep problems in PD are available; in particular, chronic sleep problems are coupled with the adverse effects of the medications prescribed for PD. Therefore, preventing sleep problems in patients with PD requires the development of non-pharmacological approaches.

Although melatonin is a useful biological marker for the circadian clock in research settings21, reliable measurement requires that patients remain under dim light conditions during sample collection22. Importantly, because melatonin secretion is affected by aging and neurodegenerative disorders23,24, there is increasing concern about the limitations of inter-individual comparisons of circadian characteristics among elderly subjects with neurodegenerative disorders. Instead, clock gene transcripts can be used as alternative markers. Circadian rhythms are driven by negative feedback loops of transcription in clock gene expression25,26. Cell-autonomous transcriptional feedback loops are ubiquitous throughout the body, and the SCN functions as the central clock to orchestrate clock gene expression in peripheral tissues27. Peripheral clock gene transcripts in humans can be measured by collecting peripheral tissues such as white blood cells, oral mucosa and hair follicles28,29,30 and are useful optional markers in experimental evaluation of the human circadian clock. Previous studies using expression data of three clock genes demonstrated that sampling at 6‐hr intervals (four times a day) allowed the reliable mathematical estimation of circadian phase28,31. This low sampling frequency may be advantageous compared with melatonin measurement.

In the present study, we aimed to provide further evidence on whether BLT-mediated sleep improvement in DT-receiving PD patients involves functional modulation of the central circadian clock, and whether DT-induced suprachiasmatic dysfunction is a cause of sleep problems in these patients. To do this, given the circadian phase advance of melatonin secretion and increased frequency of early-morning awakening in DT-receiving PD patients, we investigated the effect of evening BLT on sleep problems. Simultaneously, we confirmed the effect of evening BLT on circadian phase by detecting peripheral clock gene expression rhythms using plucked hairs, and compared estimated circadian phases with sleep improvement. Furthermore, to obtain evidence on the negative effects of chronic dopamine exposure on the suprachiasmatic function, we performed ex vivo culture of mouse SCN and real-time monitoring of clock gene expression rhythms under chronic treatment with dopamine.

## Materials and methods

### Patients

Japanese PD patients with a Unified Parkinson’s Disease Rating Scale (UPDRS) part III score of 9–45 who were aged 52–80 years and receiving DT were recruited in Osaka Toneyama Medical Center. No patient had severe cataracts, which could affect retinal light input. All subjects underwent a neuro-ophthalmic examination to confirm the absence of any abnormalities. In addition, all subjects underwent the Mini-Mental State Examination (MMSE) and Alzheimer’s Disease Assessment Scale (ADAS) to confirm that cognitive function was within the normal range. These patients spent most of the day indoors. Their daily schedule was controlled during hospitalization and re-hospitalization (wake-up time, 6:00; breakfast, 8:00; lunch, 12:00; dinner, 17:30; bedtime, 22:00). Sleep was evaluated using the Epworth Sleepiness Scale (ESS) and Parkinson Disease Sleep Scale 2 (PDSS-2). This study was conducted in accordance with the Declaration of Helsinki and was approved by the institutional review boards of Osaka Toneyama Medical Center and Senri Chuo Hospital, with all participants providing written informed consent.

### Experimental design

Patients were exposed to one hour of bright light once a day during the evening hours (19:00–21:00) using a table-top illuminator emitting 5,000-lux bright light at a distance of 50 cm with a color temperature of about 6,700 K. This bright light exposure was performed every day for a period of approximately three months in total: one week during hospitalization, more than 10 weeks at home after discharge and one week during re-hospitalization. To minimize the effects of environmental factors such as ambient light, all subjects stayed in a hospital room with blackout curtains during hospitalization and followed a regular daily schedule, as described above. Patient #4 received bright light exposure for a period of eight months due to difficulties with scheduling re-hospitalization. Before and after the period of bright light exposure, patients completed the UPDRS part III, ESS and PDSS-2, and hair follicles were collected from them at approximately 6‐hr intervals around the clock.

### non-PD subjects

Japanese non-PD subjects with an ESS score of 1–9 who were aged 59–82 years and had stayed for fracture rehabilitation in Senri Chuo Hospital were recruited. These subjects spent most of the day indoors before and during hair collection. Their daily schedule was controlled during hospitalization (wake-up time, 6:00; breakfast, 8:00; lunch, 12:00; dinner, 18:00; bedtime, 22:00).

### Statistical analysis

A Wilcoxon rank sum test was used for pairwise comparisons because it was not possible to assume that the data obtained were homoscedastic and normally distributed (Table 2). P values of less than 0.05 were considered to indicate statistical significance. All described P values are two-sided. Statistical analyses were performed using JMP 11 for Windows (SAS Institute Inc., U.S.A.).

### Peripheral clock gene expression

Hair follicles were collected by gripping and tugging the hair shaft with a pair of tweezers and were quickly soaked in Lysis Solution (RNAqueous-Micro Kit; Thermo Fisher Scientific, U.S.A.). Hair follicle cells attached to the hair shafts were stored at −20 °C until RNA purification. Approximately 2–10 scalp hair follicles were required to detect clock gene expression at each sampling timepoint. To minimize skin damage, samples were obtained from different regions of the scalp. The RNAqueous-Micro Kit was used with frozen cytolysis solution to purify total RNA. After checking the quality and concentration using a NanoDrop (LMS, Japan), total RNA was reverse-transcribed using a SuperScript VILO cDNA Synthesis Kit (Life Technologies, U.S.A.), and real-time PCR was performed using a TaqMan MGB probe (Applied Biosystems, U.S.A.) and a 1/20 volume of the reverse transcription product. Data were obtained using a PRISM7300 (Applied Biosystems, U.S.A.) and corrected by 18S ribosomal RNA (18S-rRNA), the expression of which is constant regardless of cell type and sampling time32. The sequences of the primers and probe for the Per3, Nr1d1 (Rev-erbα), Nr1d2 (Rev-erbβ) and 18S-rRNA transcripts are listed in our previous report28.

Time-courses of gene expression data were normalized to fit 24-h-period cosine curves with unity amplitude. The normalized gene expressions at time t were denoted as x(t), y(t), and z(t) for Per3, Nr1d1, and Nr1d2, respectively. Expression data for these three genes were then fitted to the following 24-h-period cosine curves under the constraint that phase difference between Per3 and Nr1d1 and between Per3 and Nr1d2 were Δθy and Δθz, respectively:

$$x(t)={A}_{x}\,\cos (\omega t+\theta )+{C}_{x},$$
(1)
$$y(t)={A}_{y}\,\cos (\omega t+\theta +\Delta {\theta }_{y})+{C}_{y},$$
(2)
$$z(t)={A}_{z}\,\cos (\omega t+\theta +\Delta {\theta }_{z})+{C}_{z}.$$
(3)

For four-point experimental measurements, the cost function was defined as follows:

$$\begin{array}{ccc}E(\theta ,A,C) & = & {{\Sigma }_{i=1}}^{4}[{\{x({t}_{i})-{A}_{x}\cos (\omega {t}_{i}+\theta )-{C}_{x}\}}^{2}+{\{y({t}_{i})-{A}_{y}\cos (\omega {t}_{i}+\theta +\Delta {\theta }_{y})-{C}_{y}\}}^{2}\\ & & +{\{z({t}_{i})-{A}_{z}\cos (\omega {t}_{i}+\theta +\Delta {\theta }_{z})-{C}_{z}\}}^{2}].\end{array}$$
(4)

To minimize the cost function, model parameters, {θ, Ax, Ay, Az, Cx, Cy, Cz}, were determined using the conjugate gradient method. Empirically, the differences in peak expression time between the clock genes are known to be around Δτy = 4 h and Δτz = 2 h, which translate to phase differences of Δθy = 2πΔτy/24 and Δθz = 2πΔτz/24, respectively. By varying the phase differences in the range of Δτy [3.5 h, 4.5 h] and Δτz [1.5 h, 2.5 h], the optimal set of phase differences giving the minimum cost function was selected. Finally, the peak expression time of Per3 was estimated as CTPer3 = 24(0.5π-θ)/(2π). To evaluate the cosine curve fitting, we computed the coefficient of determination r2 between the experimental measurements and cosine curves. We also calculated 95% confidence intervals for the estimated value from the inverse of the Hessian matrix of the cost function, based on the assumption that the estimation errors are uncorrelated and normally distributed. The time intervals of hair sampling were not identical among subjects but varied by a maximum of about ±1 h. However, we previously compared the difference in phase-estimation accuracy among various combinations of three data points28. Our mathematical analysis indicated high estimation accuracy for various sampling intervals (e.g., 8 h–8 h, 6 h–6 h and 3 h–3 h intervals). Small differences in sampling interval are therefore not expected to affect phase-estimation accuracy.

### Animals

Per2-luciferase knock-in mice (Per2Luc mice) were a kind gift from Dr. Joseph Takahashi33. Mice were maintained on a 12-h/12-h light dark cycle (light on at 9:00 a.m.) and allowed ad libitum access to food and water. All protocols for animal experiments were approved by the Animal Research Committee of Yamaguchi University. Animal studies were performed in compliance with the Yamaguchi University Animal Care and Use guidelines.

### Explant cultures and bioluminescence measurement

Coronal brain slices including the SCN (300–400 μm thickness) were prepared from adult Per2Luc mice using a vibratome34,35. Paired SCNs were excised from coronal brain slices and placed on a culture membrane (Millicell-CM; Merck, Germany) in a covered and sealed culture dish filled with medium containing 100 μM luciferin. Bioluminescence was measured in realtime with a photomultiplier tube (LM2400; Hamamatsu, Japan). The data sets were detrended by subtracting the 24 h running average from the raw data. To perform single neuron imaging, a luminescence microscope optimized for live cell imaging (LV200; Olympus, Japan) was used.

## Results

Although BLT reportedly improves sleep in PD patients receiving DT, as described above, the number of participants and trials examined and research groups to have investigated the efficacy to date has been small. Further studies are therefore required to confirm the clinical usefulness of BLT. In the present study, we investigated the effect of BLT on sleep problems in East Asian (Japanese) PD patients. Given that PD patients receiving DT show early-morning awakening and circadian phase advance of melatonin secretion, we exposed patients to one hour of bright light once a day during the evening hours (Fig. 1). This bright light exposure was performed every day for a total period of approximately three months. PD patients participating in this study were aged 52–80 years, all of whom were receiving DT and had a UPDRS part III score of 9–45 (Table 1). Sleep was evaluated using the Epworth Sleepiness Scale (ESS) and Parkinson Disease Sleep Scale 2 (PDSS-2). Re-examination of these scales after the period of bright light exposure revealed no significant change in the UPDRS part III or ESS scores but showed a significant positive impact on sleep according to the PDSS-2 score (Table 2). Statistical comparison for each item of PDSS-2 between before and after BTL revealed a significant improvement in two items: nocturia and pain in arms or legs. To examine the effect of aging on BLT efficacy, we divided the PD patients into younger and older groups (n = 8, each) and performed the statistical analysis again. Interestingly, although the sample size is likely too small to make solid conclusions, the results suggest that BLT induced a statistically significant improvement in PDSS score only in the younger group. The disappearance of statistical significance for nocturia and pain in arms or legs is probably due to the reduced sample size after grouping. Together, these results show that BLT is a promising therapy for sleep problems in PD patients receiving DT.

Next, to evaluate the association between circadian modulation and sleep improvement in PD patients who received BLT, we examined the correlation between BLT-induced changes, namely, the change in circadian phase in clock gene expression versus the change in severity in sleep problems (Fig. 3). No significant improvement in sleep was detected using ESS in PD patients who received BLT (Fig. 3a), and a two-dimensional plot of the change in ESS score versus the change in circadian phase in clock gene expression showed no clear correlation between these parameters (Fig. 3b). In contrast, a significant improvement in sleep was detected using PDSS-2, with the exception of one patient (Fig. 3c). Further, a two-dimensional plot of the change in PDSS-2 score versus the change in circadian phase in clock gene expression showed a clear correlation between these parameters (Fig. 3d): 75% of PD patients with sleep improvement showed a circadian phase delay in clock gene expression. Although the causal link remains unclear, we speculate that functional modulation of the circadian clock is a potential mechanism underlying the BLT-mediated sleep improvement in PD patients receiving DT.

The hypothalamic SCN of PD patients receiving DT is chronically exposed to dopaminergic stimulation. To experimentally evaluate the possible effect of chronic dopaminergic stimulation on the function of the SCN, we examined the effect of chronic dopamine exposure to ex vivo cultured mouse SCN on circadian clock gene expression (Fig. 4). Autonomous circadian rhythms of the expression of the circadian protein Period2 (Per2) in ex vivo cultured SCN were detectable in real time using knock-in mice expressing Per2 fused to firefly luciferase (Per2Luc mice). Chronic exposure of the cultured SCN to dopamine induced a slight and strong attenuation of bioluminescence rhythms at concentrations of 15 and 30 μM, respectively (Fig. 4a). A decay constant was calculated to evaluate the statistical significance of the attenuation of bioluminescence rhythms, and a significant attenuation was confirmed at a concentration of 30 μM (Fig. 4b). To clarify whether this attenuation was attributable to desynchronization of circadian rhythms among SCN neurons or dampening of circadian oscillations in individual neurons, we examined the effect of chronic exposure to dopamine on the circadian rhythms of bioluminescence using highly sensitive imaging at the single neuron level (Fig. 4c). The data clearly demonstrated that chronic exposure of the cultured SCN to 30 μM dopamine caused a rapid attenuation of bioluminescence rhythms in individual neurons (Fig. 4d). Together, these results suggest that functional defects in the SCN caused by chronic exposure to dopamine may in part underlie the sleep problems observed in PD patients receiving DT.

## Discussion

At present, there is no evidence-based international consensus or guideline for medical treatment of sleep problems in PD patients. These sleep problems, coupled with the adverse effects of medications prescribed for PD, limit the usefulness of available treatment strategies. Non-pharmacological treatments for sleep problems are highly desirable because additional pharmacological interventions may cause unwanted side effects via unpredictable interactions with dopaminergic drugs. In the present study, we demonstrated that BLT, a non-pharmacological and chronobiological treatment, was highly effective for improving sleep in Japanese PD patients receiving DT. Although objective scales such as polysomnography are more informative for sleep evaluation, PD patients often feel uncomfortable during sleep. We therefore used subjective scales such as ESS and PDSS-2 in this study, which have been proven to be excellent for assessing sleep in PD patients41,42. BLT is relatively easy to prescribe and incorporate into clinical practice, with patients simply required to expose themselves to bright light for one hour once a day. Patients can therefore simply and safely continue the treatment at home after discharge from hospital. Indeed, in the present study, patients performed BLT by themselves at home during the major part of the experimental period. Therefore, BLT is a promising treatment for sleep problems in PD patients receiving DT.

To establish and generalize BLT-based therapies in clinical settings, it is indispensable to elucidate and understand the mechanism by which BLT improves sleep in PD patients receiving DT. However, no reports have suggested potential mechanisms based on experimental evidence. According to the two-process model, sleep is regulated by both homeostatic function and the circadian clock: while homeostatic sleep pressure (process S) increases with time awake and declines during sleep, alertness levels are dependent on the internal circadian clock (process C)36,37,38. The interaction between these two processes generates the diurnal rhythms of the sleep-wake cycle. Because the circadian clock is functionally modulated by retinal light input, there has been speculation, although still lacking sufficient experimental evidence, that the circadian clock is part of the mechanism by which BLT improves sleep in PD patients16. In the present study, we successfully provided supportive evidence for this hypothesis. Briefly, experimental evaluation of peripheral clock gene expression rhythms revealed that most of the PD patients receiving DT who experienced improved sleep following BLT indeed showed a circadian phase shift, indicating the existence of a correlation between circadian modulation and sleep improvement. While these results do not demonstrate a causal link between circadian modulation and sleep improvement, they provide experimental evidence to support the hypothesis that BLT improves sleep via circadian function.

Conversely, if this hypothesis holds true, sleep problems in PD patients receiving DT should arise at least in part as a result of circadian dysfunction. By conducting animal experiments in the present study, we successfully demonstrated the possibility that PD patients receiving DT may indeed exhibit circadian dysfunction. Because almost all PD patients chronically receive dopaminergic drugs, the SCN of the hypothalamus is exposed to chronic dopaminergic stimulation. Given that dopamine is released into the synaptic cleft at several tens to hundreds of micromolar concentrations, we chronically administered 30 μM (or less) dopamine to ex vivo cultured SCN to experimentally reproduce this in vivo situation and examine the effect on SCN function. We found that chronic dopaminergic stimulation induced a rapid attenuation of autonomous oscillations of clock gene expression in the SCN at the single neuron level. Importantly, our findings are consistent with those of a recent study. Following expression of the D1 dopamine receptor (Drd1) in the SCN, Grippo et al. found that selective activation of dopaminergic input to the SCN resulted in acceleration of the circadian phase shift, and mice lacking the functional Drd1 gene showed defects in the light-induced circadian phase shift20. Light-induced dopamine release therefore plays a physiological role in modulating SCN function, while DT-induced artificial and chronic dopaminergic stimulation may cause functional defects in the SCN. Indeed, not only decreased amplitude of the circadian rhythms of melatonin, an endocrine marker of circadian rhythms in the SCN, but also altered clock gene expression in peripheral blood mononuclear cells suggests functional defects in the SCN in PD patients receiving DT8,9. Given that regular and timed light input resets and potentiates suprachiasmatic circadian rhythms, these findings suggest that BLT may restore SCN function disturbed by DT, which in turn improves sleep. Although we performed BLT during the late evening hours in the present study, BLT scheduled at other times of day may also be effective for restoring SCN function: a report demonstrated improved sleep in PD patients who received one hour of BLT twice a day (in the morning at 9–11 AM and in the afternoon at 5–7 PM)16.

In conclusion, we demonstrated that BLT is a promising medical treatment for improving sleep in PD patients receiving DT and that this BLT-induced sleep improvement may arise due to restoration of circadian function. Further studies such as those that aim to determine the optimal time of day and minimum duration of bright light exposure, and the optimal illumination intensity and wavelength of irradiation light are required to increase the clinical usefulness of BLT. In addition, development of inexpensive wearable devices for bright light exposure will improve the generalizability of BLT. For example, while we used a table-top illuminator in the present study, low-cost and glasses-type wearable devices will increase compliance to the therapy in a homecare setting due to reduced limitations to patients’ behavior.

## References

1. 1.

Ball, N., Teo, W. P., Chandra, S. & Chapman, J. Parkinson’s Disease and the Environment. Front Neurol 10, 218 (2019).

2. 2.

Curtis, L., Lees, A. J., Stern, G. M. & Marmot, M. G. Effect of L-dopa on course of Parkinson’s disease. Lancet 2, 211–212 (1984).

3. 3.

Lewitt, P. A. Levodopa for the treatment of Parkinson’s disease. N Engl J Med 359, 2468–2476 (2008).

4. 4.

Barone, P. et al. The PRIAMO study: A multicenter assessment of nonmotor symptoms and their impact on quality of life in Parkinson’s disease. Mov Disord 24, 1641–1649 (2009).

5. 5.

Paus, S. et al. Sleep attacks, daytime sleepiness, and dopamine agonists in Parkinson’s disease. Mov Disord 18, 659–667 (2003).

6. 6.

Cantor, C. R. & Stern, M. B. Dopamine agonists and sleep in Parkinson’s disease. Neurology 58, S71–78 (2002).

7. 7.

Hobson, D. E. et al. Excessive daytime sleepiness and sudden-onset sleep in Parkinson disease: a survey by the Canadian Movement Disorders Group. JAMA 287, 455–463 (2002).

8. 8.

Bordet, R. et al. Study of circadian melatonin secretion pattern at different stages of Parkinson’s disease. Clin Neuropharmacol 26, 65–72 (2003).

9. 9.

Breen, D. P. et al. Sleep and circadian rhythm regulation in early Parkinson disease. JAMA Neurol 71, 589–595 (2014).

10. 10.

Coomans, C. P., Ramkisoensing, A. & Meijer, J. H. The suprachiasmatic nuclei as a seasonal clock. Front Neuroendocrinol 37, 29–42 (2015).

11. 11.

Vriend, J. & Reiter, R. J. Melatonin feedback on clock genes: a theory involving the proteasome. J Pineal Res 58, 1–11 (2015).

12. 12.

Joyce, D. S., Feigl, B., Kerr, G., Roeder, L. & Zele, A. J. Melanopsin-mediated pupil function is impaired in Parkinson’s disease. Sci Rep 8, 7796 (2018).

13. 13.

La Morgia, C., Ross-Cisneros, F. N., Sadun, A. A. & Carelli, V. Retinal Ganglion Cells and Circadian Rhythms in Alzheimer’s Disease, Parkinson’s Disease, and Beyond. Front Neurol 8, 162 (2017).

14. 14.

Mitolo, M. et al. Effects of Light Treatment on Sleep, Cognition, Mood, and Behavior in Alzheimer’s Disease: A Systematic Review. Dement Geriatr Cogn Disord 46, 371–384 (2018).

15. 15.

Willis, G. L., Moore, C. & Armstrong, S. M. A historical justification for and retrospective analysis of the systematic application of light therapy in Parkinson’s disease. Rev Neurosci 23, 199–226 (2012).

16. 16.

Videnovic, A. et al. Timed Light Therapy for Sleep and Daytime Sleepiness Associated With Parkinson Disease: A Randomized Clinical Trial. JAMA Neurol 74, 411–418 (2017).

17. 17.

Martino, J. K., Freelance, C. B. & Willis, G. L. The effect of light exposure on insomnia and nocturnal movement in Parkinson’s disease: an open label, retrospective, longitudinal study. Sleep Med 44, 24–31 (2018).

18. 18.

Willis, G. L., Boda, J. & Freelance, C. B. Polychromatic Light Exposure as a Therapeutic in the Treatment and Management of Parkinson’s Disease: A Controlled Exploratory Trial. Front Neurol 9, 741 (2018).

19. 19.

Doi, M. et al. Impaired light masking in dopamine D2 receptor-null mice. Nat Neurosci 9, 732–734 (2006).

20. 20.

Grippo, R. M., Purohit, A. M., Zhang, Q., Zweifel, L. S. & Guler, A. D. Direct Midbrain Dopamine Input to the Suprachiasmatic Nucleus Accelerates Circadian Entrainment. Curr Biol 27, 2465–2475 e2463 (2017).

21. 21.

Pandi-Perumal, S. R. et al. Dim light melatonin onset (DLMO): a tool for the analysis of circadian phase in human sleep and chronobiological disorders. Prog Neuropsychopharmacol Biol Psychiatry 31, 1–11 (2007).

22. 22.

Matsumura, R., Node, K. & Akashi, M. Estimation methods for human circadian phase by use of peripheral tissues. Hypertens Res 39, 623–627 (2016).

23. 23.

Pandi-Perumal, S. R. et al. Physiological effects of melatonin: role of melatonin receptors and signal transduction pathways. Prog Neurobiol 85, 335–353 (2008).

24. 24.

Wu, Y. H. & Swaab, D. F. Disturbance and strategies for reactivation of the circadian rhythm system in aging and Alzheimer’s disease. Sleep Med 8, 623–636 (2007).

25. 25.

Schibler, U. & Sassone-Corsi, P. A web of circadian pacemakers. Cell 111, 919–922 (2002).

26. 26.

Reppert, S. M. & Weaver, D. R. Coordination of circadian timing in mammals. Nature 418, 935–941 (2002).

27. 27.

Mohawk, J. A., Green, C. B. & Takahashi, J. S. Central and peripheral circadian clocks in mammals. Annu Rev Neurosci 35, 445–462 (2012).

28. 28.

Akashi, M. et al. Noninvasive method for assessing the human circadian clock using hair follicle cells. Proc Natl Acad Sci USA 107, 15643–15648 (2010).

29. 29.

Boivin, D. B. et al. Circadian clock genes oscillate in human peripheral blood mononuclear cells. Blood 102, 4143–4145 (2003).

30. 30.

Cajochen, C. et al. Evening exposure to blue light stimulates the expression of the clock gene PER2 in humans. Eur J Neurosci 23, 1082–1086 (2006).

31. 31.

Yamaguchi, A. et al. Normal peripheral circadian phase in the old-old with abnormal circadian behavior. Genes Cells 23, 849–859 (2018).

32. 32.

Matsumura, R., Okamoto, A., Node, K. & Akashi, M. Compensation for intracellular environment in expression levels of mammalian circadian clock genes. Sci Rep 4, 4032 (2014).

33. 33.

Yoo, S. H. et al. PERIOD2::LUCIFERASE real-time reporting of circadian dynamics reveals persistent circadian oscillations in mouse peripheral tissues. Proc Natl Acad Sci USA 101, 5339–5346 (2004).

34. 34.

Yamazaki, S. & Takahashi, J. S. Real-time luminescence reporting of circadian gene expression in mammals. Methods Enzymol 393, 288–301 (2005).

35. 35.

Akashi, M., Hayasaka, N., Yamazaki, S. & Node, K. Mitogen-activated protein kinase is a functional component of the autonomous circadian system in the suprachiasmatic nucleus. J Neurosci 28, 4619–4623 (2008).

36. 36.

Crowley, S. J., Acebo, C. & Carskadon, M. A. Sleep, circadian rhythms, and delayed phase in adolescence. Sleep Med 8, 602–612 (2007).

37. 37.

Beersma, D. G. Models of human sleep regulation. Sleep Med Rev 2, 31–43 (1998).

38. 38.

Borbely, A. A. & Achermann, P. Sleep homeostasis and models of sleep regulation. J Biol Rhythms 14, 557–568 (1999).

39. 39.

Skene, D. J. Optimization of light and melatonin to phase-shift human circadian rhythms. J Neuroendocrinol 15, 438–441 (2003).

40. 40.

Campbell, S. S. et al. Light treatment for sleep disorders: consensus report. I. Chronology of seminal studies in humans. J Biol Rhythms 10, 105–109 (1995).

41. 41.

Kumar, S., Bhatia, M. & Behari, M. Excessive daytime sleepiness in Parkinson’s disease as assessed by Epworth Sleepiness Scale (ESS). Sleep Med 4, 339–342 (2003).

42. 42.

Trenkwalder, C. et al. Parkinson’s disease sleep scale–validation of the revised version PDSS-2. Mov Disord 26, 644–652 (2011).

## Acknowledgements

Per2Luc knock-in mice were kind gifts from Joseph Takahashi. We thank Junko Sumino for technical assistance. We express our great appreciation to Ai Yamaguchi, Hiroaki Naritomi in Senri Chuo Hospital and the nursing staff in Osaka Toneyama Medical Center for general support. We also acknowledge the support of fellowships from the Japan Society for the Promotion of Science. This work was supported by JSPS KAKENHI Grant Number JP17K01387.

## Author information

Authors

### Contributions

T.E., S.S. and M.A. designed the research. T.E., R.M. and T.Y. performed experiments. R.M., I.T. and Y.S. analyzed data. K.N. contributed reagents and materials. M.A. wrote the paper. All authors reviewed the manuscript.

### Corresponding author

Correspondence to Makoto Akashi.

## Ethics declarations

### Competing interests

The authors declare no competing interests.

Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

## Rights and permissions

Reprints and Permissions

Endo, T., Matsumura, R., Tokuda, I.T. et al. Bright light improves sleep in patients with Parkinson’s disease: possible role of circadian restoration. Sci Rep 10, 7982 (2020). https://doi.org/10.1038/s41598-020-64645-6

• Accepted:

• Published:

• ### Variable abnormality of the melanopsin-derived portion of the pupillary light reflex (PLR) in patients with Parkinson’s disease (PD) and parkinsonism features

• Bruce I. Gaynes
• , Raman Yousefzai
• , Mario Chazaro-Cortes
• , Kalea Colletta
• , Sandra L. Kletzel
• , Mary Beth Jost
• , Youngsook Park
• , Jasvinder Chawla
• , Mark V. Albert
•  & Ting Xiao

Neurological Sciences (2021)