Excessive daytime sleepiness in a model of Parkinson's disease improved by low-frequency stimulation of the pedunculopontine nucleus

Abstract

Patients with Parkinson's disease often complain of excessive daytime sleepiness which negatively impacts their quality of life. The pedunculopontine nucleus, proposed as a target for deep brain stimulation to improve freezing of gait in Parkinson's disease, is also known to play a key role in the arousal system. Thus, the putative control of excessive daytime sleepiness by pedunculopontine nucleus area stimulation merits exploration for treating Parkinson's disease patients. To this end, two adult nonhuman primates (macaca fascicularis) received a deep brain stimulation electrode implanted into the pedunculopontine nucleus area along with a polysomnographic equipment. Stimulation at low frequencies and high frequencies was studied, in healthy and then MPTP-treated nonhuman primates. Here, we observed that MPTP-treated nonhuman primates suffered from excessive daytime sleepiness and that low-frequency stimulation of the pedunculopontine nucleus area was effective in reducing daytime sleepiness. Indeed, low-frequency stimulation of the pedunculopontine nucleus area induced a significant increase in sleep onset latency, longer continuous periods of wakefulness and thus, a partially restored daytime wake architecture. These findings may contribute to the development of new therapeutic strategies in patients suffering from excessive daytime sleepiness.

Fig. 1. Surgery report, PD scores and experimental design.

a X-ray of the final implantation of the electrode in the PPN area for M1, in both coronal and sagittal view with the internal landmarks AC-PC line determined by ventriculography (LV: lateral ventricle). b Micrographs of Cresyl-violet stained sections through the PPN of M1 and M2 showing postmortem reconstruction of the electrode track and position of the contacts, (-) is the active contact. CGMB = central gray substance of midbrain; PnO= oral pontine reticular nucleus; ml= medial lemniscus. c Graphs illustrating the longitudinal progression of parkinsonian syndrome for M1 and M2, induced by injection of chronic low doses of MPTP, based on weekly observations. The solid black line shows the PD score and the gray dotted line represents the cumulative dose of MPTP with each dot corresponding to an injection of 0.2–0.5 mg/kg. Note 3 key periods: the healthy period (black), the MPTP treatment period (gray) and the stable parkinsonian period (orange). During the healthy period the PD scores were 0/25. During the stable parkinsonian period the score was 13.8 ± 0.2 / 25 for M1 and 18.9 ± 0.1 / 25 for M2. d Micrographs showing tyrosine hydroxylase (TH) immunostaining, at the level of the striatum (framed image, scale bar= 2000µm) and the substantia nigra (scale bar= 2000µm) for M1, M2 and the control animal (C). The circled areas correspond to the striatal structures (putamen in blue and caudate nucleus in purple) and the substantia nigra in black. e Graphs showing the percent loss of TH expression in the putamen, the caudate nucleus and the substantia nigra of M1 and M2 compared to the control animal. Percent loss are calculated based on optical density method using the region of interest outlined in d. f Design of the modified multiple sleep latency test (mMSLT) with 20 min light-OFF at 10:00 h (1), 11:00 h (2) and 12:00 h (3), performed in behavioral cage and under different conditions: healthy or stable parkinsonian states with PPN-DBS OFF or ON (LFS vs. HFS). g Design of long-term recordings of 12 h nighttime (from 19:00 h to 7:00 h) and daytime (from 7:00 h to 19:00 h), performed in home cage and under different conditions: healthy or stable parkinsonian states with PPN-DBS OFF or ON (LFS only). h Polysomnographic recordings for wake/sleep stages analysis. Thirty seconds epochs showing active wake (AW), quiet wake (QW), non-REM sleep stage 1 (N1), stage 2 (N2), stage 3 (N3) and REM sleep (R).

Fig. 2. Panel of different sleep parameters obtained from mMSLTs.

Sleep parameters were obtained in light-OFF sessions for each animal M1 and M2, OFF-stimulation versus ON-stimulation conditions in healthy (white) and parkinsonian (orange) states with low-frequency stimulation (LFS, vertical lines) and high-frequency stimulation (HFS, dots). a Sleep latency, expressed in minutes (mean ± SEM), for each light-OFF session (1) from 10:00 h to 10:20 h, (2) from 11:00 h to 11:20 h, (3) from 12:00 to 12:20 h for healthy (black line) and parkinsonian (orange line) states with LFS (bold dotted line) and HFS (thin dotted line), b Mean time of sleep latency, expressed in minutes (mean ± SEM), for all light-OFF sessions pooled together by condition, p < 0.05 * different from healthy OFF-stimulation; # different from parkinsonian OFF-stimulation: Kruskal–Wallis test followed by Dunn’s multiple comparisons test. c Occurrence of at least 30 seconds of sleep during all light-OFF sessions, expressed in %, p < 0.05 ǂ different from OFF-stimulation: Fisher’s exact test. d Mean time of sleep duration, expressed in minutes (mean ± SEM), for all light-OFF sessions pooled together by condition. Note in M1 parkinsonian condition the appearance of small episodes of REM sleep materialized in black on the histogram. p < 0.05 * different from healthy OFF-stimulation; # different from parkinsonian OFF-stimulation: Kruskal–Wallis test followed by Dunn’s multiple comparisons test. e Typical 30-s example of EEG and EMG tracings showing cortical desynchronization associated with high muscle tone observed in PPN-LFS condition and, f cortical synchronization associated with low muscle tone observed in PPN-HFS condition.

Fig. 3. Naps architecture and wake/sleep stages transitions.

a, b Representative daytime hypnogramm for M1 and M2 respectively, during healthy, parkinsonian PPN-LFS OFF and parkinsonian PPN-LFS ON conditions. The line position indicates the sleep stage represented in y-axis, with X = artifacts; A = active wake; W = quiet wake; 1,2,3= non-REM sleep stage N1, N2 and N3; R = REM sleep. Note the naps disorganization in parkinsonian state with nearly no continuous active wake periods and the reorganizing effect of PPN-LFS with more active wake periods (circled periods).

Fig. 4. Evaluation of PPN-LFS on excessive daytime sleepiness.

These evaluations were performed during healthy (white), parkinsonian PPN-LFS OFF (orange) and parkinsonian with PPN-LFS ON (orange with vertical lines) conditions. a Latency to the first sleep episode in the morning for M1 and M2, expressed in minutes (mean ± SEM). b Total sleep time during the day for M1 and M2, expressed in minutes (mean ± SEM). c Duration of continuous active wake periods for M1 and M2, expressed in minutes (mean ± SEM). d Stage shift index for M1 and M2, expressed in number of transition from one stage to another per hour (mean ± SEM). * p < 0.05 Kruskal–Wallis test followed by Dunn’s multiple comparisons test. e Typical example of matrices of transitions probabilities at healthy, parkinsonian PPN-LFS OFF and parkinsonian PPN-LFS ON. W = wake; 1,2,3= non-REM sleep stage N1, N2 and N3; R = REM sleep. f Transitions between wake and sleep stages, source stage (vertical) to destination stage (horizontal) according to the parkinsonian PPN-LFS OFF state (left) and the parkinsonian PPN-LFS ON state (right). Numbers in the matrices are the significant p values obtained with Wilcoxon rank sum test.