Circadian oscillators in the epithalamus

Abstract

The habenula complex is implicated in a range of cognitive, emotional and reproductive behaviors, and recently this epithalamic structure was suggested to be a component of the brain's circadian system. Circadian timekeeping is driven in cells by the cyclical activity of core clock genes and proteins such as per2/PER2. There are currently no reports of rhythmic clock gene/protein expression in the habenula and therefore the question of whether this structure has an intrinsic molecular clock remains unresolved. Here, using videomicroscopy imaging and photon-counting of a PER2::luciferase (LUC) fusion protein together with multiunit electrophysiological recordings, we tested the endogenous circadian properties of the mouse habenula in vitro. We show that a circadian oscillator is localized primarily to the medial portion of the lateral habenula. Rhythms in PER2:: LUC bioluminescence here are visualized in single cells and oscillations continue in the presence of the sodium channel blocker, tetrodotoxin, indicating that individual cells have intrinsic timekeeping properties. Ependymal cells lining the dorsal third ventricle also express circadian oscillations of PER2. These findings establish that neurons and non-neuronal cells in the epithalamus express rhythms in cellular and molecular activities, indicating a role for circadian oscillators in the temporal regulation of habenula controlled processes and behavior.

Suppl. Fig. S1

Suppl. Fig. S2

Suppl. Fig. S3

Fig. 1

Circadian rhythms in PER2::LUC expression in Hb (A, B) and SCN (C) slice cultures. (A) Detrended PMT recording of PER2::LUC emission (counts per minute) in a Hb slice culture. Exposure to forskolin (10 μM) restarted oscillations. (B) Relative PER2::LUC bioluminescence in a Hb slice in the presence of 0.5 μM TTX. (C) PER2::LUC emission from an SCN slice culture. (E, F) Circadian characteristics of Hb slice cultures in control (n=34) and TTX containing medium (n=8), recorded in PMTs. There are no significant differences in period (D), amplitude (E) or rate of damping (F) following culture with 0.5 μM TTX.

Fig. 2

Rayleigh vector plots showing the phase of peak PER2::LUC expression in vitro recorded in PMTs, calculated as the time of peak bioluminescence after culture preparation or geographical ZT, in Hb and SCN slice cultures prepared at different times throughout the LD cycle. The phase of peak PER2::LUC expression in the Hb is not correlated with ZT (A) or time after culture preparation (B), while in the SCN it is correlated with ZT (C) but not time after culture preparation (D). Filled circles indicate the phase of peak bioluminescence in individual slice cultures. Direction of arrow indicates the mean phase vector, its length indicates the significance of phase clustering, with the surrounding box indicating the variance of phase. The inner broken line indicates the significance threshold of P=0.05.

Fig. 3

PER2::LUC expression in the habenula and ependymal cell layer. (A) EM-CCD images overlaid on a transmitted light image from a Hb slice culture, illustrating PER2::LUC bioluminescence (green) in the LHb and ependymal cell layer. Inset highlights single cells. Calibration bar 250 μm. Plots of relative PER2::LUC expression delineated in the (B) ependymal cell layer, (C) LHb, (E) MHb and, (D) integrated across the whole Hb culture. (F) Bioluminescence emission from representative individual cells in the LHbM. (G) Rayleigh vector plots showing the phase of peak PER2::LUC expression in vitro in the LHb and ependymal cell layer (green) calculated as the time of peak bioluminescence after culture preparation or ZT. Filled circles indicate the peak phase of individual slice (n=10).

Fig. 4

Circadian expression of PER2::LUC in the Hb complex persists in the presence of TTX (0.5 μM). Plots of relative PER2::LUC expression in slices cultured with TTX, delineated in the (A) ependymal cell layer, (B) LHb, (C) integrated across the whole Hb complex and (D) MHb. (E) Bioluminescence emission from individual cells in the LHbM. (F) Period of circadian oscillations in the ependymal cell layer, LHb and in single cells in the LHb in slices cultured in control (n=11) or TTX (n=3) medium. * P<0.05, ** P<0.01 versus the ependymal cell layer.

Fig. 5

Temporal patterns of electrical activity in the LHb. Recordings from the LHb discriminated multiunit (MUA; A, C) and single unit electrical activity (SUA; B, D). MUA rhythms generally showed one peak (A) or peaks on consecutive days (C) with a circadian period. Single cells discriminated from the multiunit recordings in (A, C) are shown below them (B, D). Inset traces in (B, D) indicate the average spike waveforms for each cell; scale bars represent 15 μV (vertical) and 1 ms (horizontal). (E) Representative photograph of electrode positioning on the LHbM for electrophysiological recordings. (F) Rayleigh vector plot showing that the phases of peak electrical activity in the LHb are randomly distributed across the LD cycle. Filled circles indicate the phase of individual slices (n=18). For interpretation of the references to color in this figure legend, the reader is referred to the Web version of this article.