Main

We have previously produced transgenic mice carrying a luciferase (luc) reporter gene under the control of the promoter of the oscillating clock gene mPer1 (refs 2, 3). Using a two-dimensional photon-counting camera, we were able to detect a day–night variation in luciferase-mediated bioluminescence in the suprachiasmatic nucleus (SCN)3, the mammalian brain's circadian centre4, in fresh brain slices from these mPer1–luc mice.

We inserted a polymer optical fibre (500 μm in diameter; 0.5 numerical aperture) just above the SCN of a mPer1luc transgenic mouse using a stereotaxic frame; the other end of the optical fibre was connected to a photomultiplier tube operating as part of a photon-counting apparatus (Fig. 1a). As luciferase requires a substrate for light emission, we continuously infused luciferin dissolved in artificial cerebrospinal fluid through a lateral ventricle. The substrate concentration in the cerebrospinal fluid, sampled from the contralateral ventricle, reached a plateau at about 1.3 mM within 3 hours and remained more or less constant.

Figure 1: In vivo monitoring of bioluminescence from the suprachiasmatic nucleus (SCN).
figure 1

a, Experimental set-up. Bioluminescence is detected under constant darkness conditions by means of an optical fibre inserted above the SCN and connected to a photon-counting apparatus with a photomultiplier tube. Luciferin solution (10 mM in artificial cerebrospinal fluid) is infused continuously at 15 μlh −1 by a syringe pump. b, Circadian fluctuation of luminescence in the SCN of a representative mPer1–luc transgenic mouse previously housed under a 12-h light/12-h dark cycle. Each dot represents the average of luminescence counts collected over 5 min. Hatched and black bars along the x-axis represent subjective day and subjective night, respectively. c, Comparison of the phase of luminescence recorded from left, LD-entrained, and right, DL-entrained mice (see text for details of terminology). Time indicated is time since the start of monitoring. Bars and dots as in b.

With this system, we continuously recorded light emission from the SCN of the transgenic mice in vivo under constant darkness. The luminescence showed a clear circadian fluctuation (Fig. 1b), with a 1.5- to 2.5-fold amplitude and peaks and troughs at circadian time (CT) 4–6 and CT15–20, respectively (where CT0 is subjective dawn and CT12 is subjective dusk).

This temporal profile of bioluminescence strikingly resembles the expression profile of native mPer1 messenger RNA. The phase lag between transcription of luciferase mRNA and the production of luminescence was only 0–2 h. The circadian period calculated from the intervals between bioluminescence peaks was about 24 h, which is close to the period of locomotor activity of this transgenic mouse line (24.0 ± 0.05 h; n = 7).

To confirm that this fluctuation in luminescence represents an endogenous oscillation, we recorded the luminescence from two groups of mice housed for at least two weeks under alternating 12-hour light–dark cycles, with lights on at either 7:00 h (LD-entrained mice) or 19:00 h (DL-entrained mice). Both groups were hooked up to the monitoring system at 14:00 h and the luminescence emitted was recorded under constant darkness. As expected, the luminescence recorded from LD-entrained mice oscillated in the opposite phase to that of DL-entrained animals (Fig. 1c).

Capturing the effect from a luciferase reporter transgene with optical-fibre- mediated photon-counting technology offers a way to measure real-time gene expression profiles in the intact brain under physiological conditions. To our knowledge, this is the first time that a ticking clock gene has been monitored in a living mammal. Real-time optical imaging of gene expression, in combination with real-time measurement of electrophysiological, metabolic and behavioural parameters, should advance the study of mammalian brain function.