Responsiveness of behavioral circadian rhythms to light

Physiological and molecular experiments eventually intend to offer an explanation for the attributes of the system as they appear in the phenotype. For that reason, a description of the system's behavior can guide the design of experiments at other research levels. Complex attributes of the system that are not directly explainable in terms of their function are not less important in this respect. This section will discuss the several properties of the circadian system in its response to light.

Light-Induced Phase Shifts

Characteristically, the effects of light on the clock change as a function of time of day. During the night the circadian system responds to light with a phase adjustment, while during the day this response is absent. The changing responsivity of the clock to light is essential for photic entrainment and, in fact, is an analytical necessity for the pacemaker to become entrained. When mammals are kept in a constant environment, their activity pattern is still rhythmic but the period of the cycle deviates slightly from 24 hours. The characteristic light response of the circadian pacemaker is best illustrated when animals are kept in constant darkness and exposed to a short pulse of light (e.g., 5–15 minutes) at a particular phase of the circadian cycle. A light pulse that is presented shortly after a nocturnal animal's activity onset (which corresponds to the beginning of the animal's subjective night) will induce a phase delay of the circadian cycle. This means that on the next cycle the animal will start its activity later than was expected on the basis of its previous activity rhythm. A light pulse presented toward the end of the animal's activity (which corresponds to the end of the animal's subjective night) will induce a phase advance of the animal's rhythm (Fig. 38.1). During the animal's resting phase (which corresponds to the animal's subjective day), light does not affect the pacemaker's phase.

Figure 38.1..  

Phase-shifting effects of light on running wheel activity in the hamster. Consecutive days are plotted beneath one another. The presence of running wheel activity is recorded with a resolution of 1 minute. Following entrainment to a light-dark cycle (L:D 14:10), hamsters were released in constant darkness, resulting in a free-running rhythm with a period that is slightly deviating from 24 hours. A, Presentation of a 15 minute light pulse (100 lux) 2 hours after activity onset results in a phase delay of the free-running activity rhythm. B, An identical light pulse that is presented 7 hours after activity onset results in a phase advance of the rhythm. Parts A and B illustrate the phase dependence of the effects of light on the pacemaker. C, Ineffectiveness of light presentation 2 hours after activity onset to induce a phase delay in a blinded hamster. D, Ineffectiveness of light 7 hours after activity onset in producing a phase advance in a blinded hamster. Parts C and D illustrate that the effects of light on the pacemaker are dependent on the integrity of retinal input. E, Phase-response curve for the phase-shifting effects of light pulses on circadian running wheel activity rhythm in the hamster. Onset of behavioral activity is defined as circadian time 12 (CT 12). The phase-response curve shows that presentation of light in the first few hours after activity onset results in phase delays. Delays are maximal around CT 14. Toward the end of activity, light results in phase advances with maximal advances around CT 19. A dead zone exists in which the animals' circadian clock is not responding to light with a phase shift. (C and D based on Meijer et al., 1999; E based on Takahashi et al., 1984.)

The behavioral phase shifts that are induced by light pulses are permanent and an indication that the underlying pacemaker has shifted in phase (Fig. 38.1). The effects of light pulses can be summarized in a phase response curve by plotting the light-induced phase shift as a function of the phase of pulse application (Daan and Pittendrigh, 1976; Takahashi et al., 1984). The phase response curve illustrates how the pacemaker's responsiveness changes during the course of the cycle (Fig. 38.1).

For diurnal species, the phase response curve is nearly identical, and only the time of behavioral activity is shifted from night to day. Light pulses presented at the beginning of the animal's resting phase induce phase delays in diurnal animals, and light pulses at the end of the resting phase produce phase advances. The phase delaying and phase advancing responses at the beginning and end of the subjective night are common features of circadian pacemakers and ensure that animals entrain to a light-dark cycle. For example, when nocturnal animals leave their burrow too early in the evening, the evening light will induce a phase delay, causing them to leave their burrow later on the next cycle. In contrast, the circadian system is advanced when animals return to their burrow too late in the morning and see the morning light. In conclusion, photic entrainment relies on well-directed phase adjustments of the circadian pacemaker that follow from the phase-dependent responsiveness of the pacemaker to light.

Additional response properties of the circadian pacemaker to light further support the animal's ability to entrain. The magnitude of either a phase delay or a phase advance obtained at a particular phase of the cycle depends on (1) light intensity, (2) duration of the light pulse, and (3) wavelength of the light. For light pulses with durations of 5–15 minutes, light intensities up to a threshold level of about 0.1 lux or 10¹¹ photons Rev cm²/s for hamsters and 1 lux for rats do not phase shift the circadian clock (Meijer et al., 1986, 1992; Nelson and Takahashi, 1991). This threshold intensity is very high compared to the thresholds for scotopic vision and is similar to the thresholds for photopic vision. Above threshold intensity, the magnitude of the phase shifts increases in a nearly linear way with light intensity. At around 100 lux, saturation occurs and further increments in intensity do not result in a further increase in phase shift. The working range of about 2 log units is very small compared to the range of light intensities that occur in the environment in the course of a day. For example, on a sunny day, intensity levels are about 10⁵ lux. If we transform the environmental variations in light intensity using the intensity dependence of the circadian system, the daily cycle of light in the world becomes a rectangular waveform with transitions at dawn and dusk, allowing the system to discriminate between day and night effectively. In view of the function of light in entraining the system to the day-night cycle, this light responsiveness of the pacemaker makes sense.

The duration of light pulses is also an important determinant of light-induced phase shifts (Meijer et al., 1992; Nelson and Takahashi, 1991). In hamsters, durations ranging from milliseconds to hours can reset circadian rhythms, but the optimal duration is about 5 minutes (Nelson and Takahashi, 1991). Short pulses on the order of seconds are not effective stimuli and require very high intensities of light, whereas long pulses on the order of hours saturate the system and so are also less efficient on a quantum basis. Indeed, the intensity-response functions for durations between seconds and hours can all be explained by integration of the total quanta during the pulse. Under these conditions, reciprocity between intensity and duration holds over the range from 1 to 45 minutes, suggesting that the circadian photic entrainment system acts as a photon counter.

Wavelength sensitivity is determined largely by the retinal pigments that mediate photic entrainment (to be discussed later in this chapter). The circadian system shows sensitivity to green, blue, and red light (Provencio and Foster, 1995; Takahashi et al., 1984; Yoshimura and Ebihara, 1996). Sensitivity is greatest for blue/green light, with a maximum sensitivity of about 480–500 nm. Phase-shifting responses have also been reported to be induced by pulses of ultraviolet radiation (Amir and Robinson, 1995; von Schantz et al., 1997).

Complexity in Light Response

Several complexities exist in an animal's behavioral responses to light. A phase shift is often not completed on the first circadian cycle after a light pulse but can exhibit transients. Especially for phase advances, several transient cycles occur in which the phase shift gradually grows after a single stimulating light pulse (Fig. 38.1). A two-pulse paradigm has been used to investigate whether the transient cycles reflect the pacemaker's position or whether, alternatively, the pacemaker is fully shifted on the first day after the pulse. In the latter case, transients would be a reflection of secondary mechanisms, that resynchronize to the SCN only slowly (Best et al., 1999; Watanabe et al., 2001). The data indicate that the pacemaker is fully shifted on day 1, at least the light-sensitive part of the pacemaker, while interaction with secondary downstream oscillators either inside or outside the SCN determine the delayed response of the behavioral activity pattern.

The immediate phase-shifting responses of activity onset and activity offset are different. While the activity onset shows transient cycles following a phase-advancing light pulse, the activity offset shows phase shifts that are complete on the first circadian cycle. In other words, the immediate shift in activity offset is larger that the shift in activity onset (Elliot and Tamarkin, 1994; Honma et al., 1985; Meijer and De Vries, 1995). In the course of several cycles, the shift in activity onset grows and the shifts in onset and offset become similar. For phase delays, the shifts in activity onset tend to exceed the shifts in activity offset, and transients are visible in the offset only. Several lines of research indicate that the pacemaker of the SCN consists of different subsets of oscillators (Jagota et al., 2000; Pittendrigh and Daan, 1976). Differences in response properties of these oscillators to light could cause differential phase shifts in onset and offset.

Repeated or prolonged light exposure leads to saturation of the phase shift. The pacemaker shows a reduction in light response in terms of its phase shifting capacity that persists for at least 1 hour after a subsaturating light pulse (Nelson and Takahashi, 1999). Two hours after a light pulse, the pacemaker's responsiveness seems to recover (Best et al., 1999). The decreased responsiveness to light does not correspond with the pacemaker's ability to track day length and suggests that day length is coded for in a different way.

Light pulses cause period changes in addition to phase shifts (DeCoursey, 1989; Hut et al., 1999). The light-induced changes in period depend on the circadian phase of pulse application. The changes in period occur such that they may contribute to photic entrainment (i.e., period lengthening occurs after light pulses during early night, and period shortening occurs following pulses during the late night).

Light is the most important but not the only phase-resetting stimulus since behavioral activity of the animal can also cause phase shifts (Maywood, 1999; Mrosovsky et al., 1989; Reebs and Mrosovsky, 1989). Behavioral activity of the animal causes phase shifts when the activity is triggered by, for instance, cage cleaning or by presenting a dark pulse against an otherwise illuminated background. Forced activity of the animal is less effective in producing a phase shift. Behavioral activity has phase-shifting effects when it occurs during the day but not when it occurs during the night. Nevertheless, it has the capacity at night to inhibit light-induced phase shifts. Light-induced phase advances but not delays are strongly attenuated and in some cases completely blocked by simultaneous running wheel activity of the animal (Mistlberger and Antle, 1998; Ralph and Mrosovsky, 1992). The effect of behavioral activity on the light-induced phase shift is correlated with the amount of running wheel activity of the animal. The data indicate that physical activity affects the circadian system and its responsiveness to light.

The intensity-response curves for phase shifting and melatonin suppression are different. Melatonin production in the pineal gland is high during the night and low during the day, and still cycles when animals are kept in constant darkness (Illnerova, 1991; Illnerova and Sumova, 1997). The nighttime elevation of melatonin can be suppressed by light. The SCN mediates these light effects on melatonin production. Light suppression of melatonin appears substantially more sensitive to irradiance compared to phase shifting (Nelson and Takahashi, 1991). The difference is 1.4 log units when half-saturation values are compared. As light information reaches the pineal gland via the SCN, the increased sensitivity cannot readily be explained and indicates differences in the organization of input pathways.