Short-term (1–3 days) constant light exposure (brief LL) potentiates nonphotic phase shifting induced by sleep deprivation and serotonin (5-HT) agonist stimulation. The present assessments reveal that exposure to brief LL markedly alters the magnitude and shape of the 5-HT1A,7 receptor agonist, 8-(+)2-dipropyl-amino-8-hydroxyl-1,2,3,4-tetrahyronapthalene (8-OH-DPAT) phase-response curve, facilitating (∼12 h) phase-advance shifts during the early morning when serotonergics have no phase-shifting effect. Brief LL also reduces the threshold for 8-OH-DPAT shifting at midday, evidenced by 5- to 6-h phase-advance shifts elicited by dosages that have no effect without the LL treatment. The brief LL-potentiated phase advances to intraperitoneal 8-OH-DPAT at zeitgeber time 0 (ZT 0) were blocked by the 5-HT1A antagonists, pindolol and WAY 100635, indicating that this shifting is mediated by 5-HT1A receptors. Antagonists with action at 5-HT7 receptors, including ritanserin and metergoline, were without effect. Although autoradiographic analyses of [3H]8-OH-DPAT binding indicate that brief LL does not upregulate suprachiasmatic nucleus (SCN) 5-HT1A receptor binding, intra-SCN microinjection of 8-OH-DPAT at ZT 0 in brief LL-exposed hamsters induced shifts similar to those produced by intraperitoneal injection, suggesting that SCN 5-HT1A receptors mediate potentiated 8-OH-DPAT-induced shifts during the early morning. Lack of shifting by intra-SCN 8-OH-DPAT at ZT 6 or 18 (when intraperitoneal 8-OH-DPAT induces large shifts), further indicates that brief LL-potentiated shifts at these time points are mediated by 5-HT target(s) outside the SCN. Significantly, sleep deprivation-induced phase-advance shifts potentiated by brief LL (∼9 h) at ZT 0 were blocked by pindolol, suggesting that these behavioral shifts could be mediated by the same SCN 5-HT1A receptor phase-resetting pathway as that activated by 8-OH-DPAT treatment.
- suprachiasmatic nucleus
- sleep deprivation
the suprachiasmatic nucleus (SCN) of the anterior hypothalamus is the site of the master circadian clock in mammals (21, 23, 40, 47). The SCN clock is entrained by photic input received principally from the retina via the retinohypothalamic tract and nonphotic signaling received from the intergeniculate leaflet via the geniculohypothalamic tract, and possibly from the midbrain raphe nuclei (28). Inputs from these nonphotic sources to the SCN are mediated by neuropeptide Y and serotonin (5-HT), respectively (34, 41).
In Syrian hamsters, nonphotic stimuli, in the form of behavioral manipulations (e.g., social interaction, cage changing, novel wheel exposure, or sleep deprivation) or pharmacological intervention (e.g., 5-HT agonists or benzodiazepines) can induce phase shifts similar in magnitude to photic shifts (∼1–2 h) (15, 31, 36). Recently, we demonstrated that nonphotic shifting responses of hamsters receiving sleep deprivation treatment or systemic application of the 5-HT1A,7 receptor agonist 8-OH-DPAT are greatly potentiated (∼10–12 h) by prior treatment with short-term (1–2 days) constant light exposure (brief LL) (24). The mechanism underlying this potentiating effect of brief LL is unknown, but could involve an upregulation of 5-HT receptor response and/or downstream processes of this or other nonphotic clock-resetting pathway(s).
Although the circadian phase-resetting actions of various serotonergic agents (8-OH-DPAT, in particular) are well documented (reviewed in Refs. 31 and 33; see also Refs. 14 and 44), the role of 5-HT in mediating nonphotic phase shifts remains speculative. This uncertainty stems from large inconsistencies between studies on the putative site(s) and receptor subtypes for the phase-resetting action of 5-HT, differences between in vivo and in vitro phase-resetting effects of 5-HT agonists, and marked species differences in phase-resetting response to such drugs. Also, reports that raphe lesions (29), 5-HT antagonist treatments (2), and SCN 5-HT depletion (6) do not block activity-induced phase shifting argue against a role of 5-HT in nonphotic behavioral circadian phase regulation.
On the other hand, there is a large body of evidence supporting a role of 5-HT in circadian clock resetting. Notably, robust phase-resetting effects of 5-HT agonists are evident under conditions in which serotonergic postsynaptic response is enhanced by suppressing 5-HT release. For example, inhibiting 5-HT synthesis with parachlorophenylalanine (p-CPA) significantly potentiates in vivo phase advances induced by intra-SCN perfusion of 8-OH-DPAT (14). Possibly related to this effect are the large phase-advancing responses of the deafferented SCN brain slice to 5-HT agonists (38, 42, 45, 46), which also have been attributed to depletion-induced hypersensitivity (37). Along similar lines, it has been proposed that the significant suppressive effect of brief LL exposure on peak nighttime 5-HT output in the SCN contributes to the dramatic hypersensitizing effects of this intervention on serotonergic phase resetting (24).
The potentiating effect of brief LL exposure on nonphotic clock resetting offers a unique approach for probing serotonergic mechanisms related to circadian phase regulation. Here, a series of experiments were undertaken to explore central elements of hypersensitized serotonergic and behavioral phase-resetting responses relating to changes in the 8-OH-DPAT dose- and phase-response curves, the identity of the 5-HT receptor involved in mediating large-magnitude (∼10–12 h) 8-OH-DPAT- and sleep deprivation-induced phase shifts, and the site of the 5-HT receptor mediating these potentiated responses.
MATERIALS AND METHODS
Adult male Syrian hamsters obtained from Harlan (Indianapolis, IN) were housed in light- and temperature-controlled (20–22°C) environmental chambers. Animals were individually housed in polystyrene cages and kept under a 14:10-h light-dark photocycle (LD) with light intensity of ∼250 lux. Food (Prolab 3000; PMI Feeds, St. Louis, MO) and water were provided ad libitum. The experiments were conducted using protocols that were in accordance with the National Institutes of Health Guidelines for the Care and Use of Laboratory Animals, and approved by the Institutional Animal Care and Use Committees at Kent State University and the University of Kentucky.
Circadian Activity Measurements
For all experiments, locomotor activity was measured for a minimum of 7 days under LD before experimentation. The circadian rhythm of general locomotor activity was recorded using overhead infrared sensors for the 8-OH-DPAT and sleep-deprivation experiments. Sensor output was interfaced with a computerized data acquisition system (ClockLab; Coulbourn Instruments, Allentown, PA).
Constant Light Procedure and Phase-Shift Analyses
The method for administering the brief LL regimen is the same as that described by Knoch et al. (24). This treatment begins at zeitgeber time 12 (ZT 12), and is maintained (by keeping the ∼250 lux room lighting turned on) over most of two consecutive light and dark phases using the same protocol as previously described (24). This constituted 50 and 56 h of continuous light extending from the last dark phase for treatments delivered at ZTs 0 and 6, respectively. Immediately following drug treatment, animals were released to constant darkness (DD) for 7–10 days to assess phase-resetting response using a modified Aschoff type II procedure (4). By definition, the 24-h time designation for animals held under LD is referred to as ZT (with ZT 12 designated as the time of lights off). For clarity of data presentation, however, ZT rather than the circadian time (CT) convention for free-running conditions with no ZT, was used to designate the circadian phase of treatments delivered during the brief LL exposures. There were no perceptible changes in phase or period of the free-running circadian activity rhythm during these brief exposures, so under these conditions, ZT is considered equivalent to CT as a phase marker.
Phase shifts were calculated as follows. A regression line based on the mean activity onset for a 5-day period under LD preceding treatment was extrapolated to the day of brief LL onset. Then a regression line based on days 3–10 posttreatment was back extrapolated to the same day of treatment. The difference between these two extrapolated lines on the day of treatment was considered the phase shift. Activity onset was defined as the first bout of activity sustained for at least 30 min.
Effects of brief LL on dose-response curves for 8-OH-DPAT.
To help determine how brief LL exposure affects sensitivity to 8-OH-DPAT, phase-advancing responses to multiple doses of this drug were assessed at two time points, ZTs 0 and 6, at which strong and less robust phase-advance shifts occur, respectively, under brief LL (24). Hamsters initially housed under LD with general locomotor activity sensors were placed in one of the two time point groups. Group 1 was exposed to the 2-day brief LL protocol and then treated with intraperitoneal injection at ZT 0 with either 0, 0.1, 0.5, 1.0, 2.5, or 5.0 mg/kg 8-OH-DPAT dissolved in DMSO. LD controls with no brief LL exposure received the same doses of 8-OH-DPAT at ZT 0. Group 2 was similarly exposed to the 2-day brief LL protocol but received intraperitoneal injection at ZT 6 with 0, 0.1, 0.5, and 5.0 mg/kg 8-OH-DPAT. LD controls with no brief LL received the same doses of 8-OH-DPAT at ZT 6. For each 8-OH-DPAT dosage group (n = 4–7), vehicle controls run with each drug group were pooled (n = 22–31).
Intra-SCN 8-OH-DPAT microinjection.
It has been demonstrated using intra-SCN reverse microdialysis perfusion that potentiated 8-OH-DPAT phase-shifting effects in hamsters treated with p-CPA are registered directly by the SCN (14). To test whether brief LL exposure may also sensitize the SCN to serotonergic stimulation, phase-shifting responses were measured in brief LL-exposed animals treated with intra-SCN microinjection of 8-OH-DPAT. Animals were implanted with a microinjection reentry cannula (Plastics One, Roanoke, VA) stereotaxically aimed at the SCN (anteroposterior, +0.3 mm from bregma; lateral, +0.3 mm from the midline; horizontal, −7.0 mm from dura). After a minimum 5-day recovery period, the animals were subjected to the 2-day brief LL protocol and were given a 0.5-μl injection of 8-OH-DPAT in artificial cerebrospinal fluid (5.0 mg/ml) or vehicle at ZT 0, 6, or 18. LD controls received the same injections but were not exposed to brief LL. Phase shifts of the general locomotor activity rhythm were analyzed as described in Constant Light Procedure and Phase-Shift Analyses. For each treatment and vehicle control group under brief LL, n = 4–6; for each treatment and control group maintained under LD, n = 5–8. The location of the microinjection site was determined by histological evaluation of the probe tract from 10-μm-thick cryostat sections stained with cresyl violet.
Antagonist assessment of 5-HT receptor for 8-OH-DPAT phase resetting.
A variety of 5-HT receptor subtypes are implicated in mediating circadian serotonergic phase-resetting responses, including 5-HT1A, 5-HT7, and 5-HT5 subtypes (see Ref. 44). Notably, in an earlier study, Tominaga et al. (49) showed that intraperitoneal 8-OH-DPAT shifts were blocked using (-)-pindolol, implicating 5-HT1A receptor in mediating this agonist's effect. Here, various 5-HT antagonists were used in conjunction with 8-OH-DPAT to help determine which subtype mediates brief LL-potentiated serotonergic phase shifting. Hamsters treated with the 2-day brief LL protocol received an intraperitoneal injection of 8-OH-DPAT (0.5 mg/kg; 5-HT1A,7 agonist; n = 17), ritanserin (5-HT2,7 antagonist; 5 mg/kg; n = 5) vehicle (DMSO; n = 6) or a cocktail containing ritanserin (5 mg/kg), and 8-OH-DPAT (0.5 mg/kg) at ZT 0 (n = 5). Phase shifting was assessed as described above. Additional treatments included intraperitoneal injection of a cocktail of ritanserin and the 5-HT1,2,7 antagonist, metergoline [for a more comprehensive assessment of possible 5-HT7 receptor involvement (both 5 mg/kg; n = 9)], the 5-HT1A,B antagonist, (-)-pindolol (5 mg/kg; n = 6), or the mixed 5-HT1A antagonist, WAY 100635 (5 mg/kg; n = 4), 15 min before intraperitoneal injection of 8-OH-DPAT (0.5 mg/kg) at ZT 0 (to assess 5-HT1A receptor involvement). Controls received intraperitoneal injection of (-)-pindolol (n = 6) or WAY 100635 (n = 4) 15 min before intraperitoneal injection of DMSO vehicle.
Antagonist determination of 5-HT receptor for sleep deprivation-induced phase resetting.
It is possible that the receptor mediating brief LL-enhanced behavioral phase resetting is the same as that involved in brief LL-potentiated 8-OH-DPAT shifts. This trial was undertaken using an established sleep deprivation protocol described previously (18, 24). Hamsters were treated with the standard 2-day brief LL protocol and received intraperitoneal injection of (-)-pindolol (5 mg/kg; n = 8) or saline vehicle (n = 7) 5 min before placing them in a novel cage with clean bedding for 1 h at ZT 0 with the lights turned off. The animals were maintained in a waking state continuously over the 1-h treatment period by air puffs and gentle handling by personnel equipped with night vision goggles. Immediately following this exposure, the hamsters were placed back in their home cages and maintained under DD for 10 days to determine phase-resetting responses.
Serotonin Receptor Autoradiography
Autoradiography for the 5-HT1A receptor was conducted using the radioligand [3H]8-OH-DPAT ([Free] = 2.4 nM; PerkinElmer, Boston, MA), as described in detail previously (11, 13). Tissue samples were taken after the brief LL exposure at ZTs 0 and 6. Specific binding to the 5-HT1A receptors was defined as the difference between binding of [3H]8-OH-DPAT in the absence and presence of pindolol (500 nM; Sigma-RBI, St. Louis MO). Autoradiograms were generated by exposing X-ray films (Biomax MR; Kodak, Rochester, NY) to the tissue sections and radioactive standards (3H-labeled microscales; Amersham, Piscataway, NJ) for 8–10 wk. Exposure time ranged from 2 to 4 mo. The autoradiograms were quantified by computer-assisted microdensitometry (M4 System; Imaging, Ontario, Canada), as described previously (12). Data are reported as femtomole per milligram tissue equivalent.
Serotonergic drugs used in this study included the following: ±8-OH-DPAT (Sigma), metergoline (Sigma), (-)-pindolol (Tocris), ritanserin (Sigma), and WAY 100635 (Wyeth Ayerst).
Data from the 8-OH-DPAT dose-response trials were analyzed using a two-way ANOVA followed by the Student-Newman-Keuls post hoc test to compare time of injection to drug treatment. Data from the intra-SCN 8-OH-DPAT microinjection trials were analyzed using a two-way ANOVA followed by the Student-Newman-Keuls post hoc test to compare time of injection and drug treatment. Statistical significance of the autoradiograms was assessed by a Student's t-test. For all experiments, P values <0.05 were considered significant.
Constant Light Potentiates 8-OH-DPAT Phase Advances
At ZT 0, intraperitoneal administration of 8-OH-DPAT to animals maintained under LD produced no phase shifting of the general locomotor activity rhythm at any of the doses tested (Figs. 1 and 2). In marked contrast, 8-OH-DPAT administered at this time to animals exposed to the 2-day brief LL treatment produced phase advances averaging as much as 12.1 ± 0.5 h (P < 0.01 vs. LD), with a peak shifting plateau occurring around 0.5 mg/kg. At ZT 6, 8-OH-DPAT administration to animals maintained under LD produced maximal phase-advance shifts averaging 2.13 ± 0.3 h (Figs. 3 and 4). This phase-shifting effect was evident only at the highest dose tested (5 mg/kg; P < 0.05 vs. vehicle). Administration of 8-OH-DPAT to animals exposed to the brief LL treatment produced significantly greater shifts, with maximal phase advances averaging 5.6 ± 0.4 h, with a peak shifting plateau of 0.1 mg/kg. From these results, it is evident that the brief LL treatment markedly enhances both the amplitude of 8-OH-DPAT-induced phase-advancing response and sensitivity to the drug (dose-response curve shifted to the left).
Constant Light Potentiates Intra-SCN 8-OH-DPAT Phase Advances at ZT 0
At ZT 0, intra-SCN microinjection of 8-OH-DPAT (2.5 μg/0.5 μl) into hamsters receiving the 2-day brief LL treatment induced large magnitude phase-advance shifts (12.7 ± 0.8 h vs. 3.9 ± 1.1 h for vehicle injection controls; P < 0.05; Fig. 5). However, similar intra-SCN 8-OH-DPAT injections undertaken at ZTs 6 and 18 did not elicit shifts that were different from vehicle injection controls. For animals maintained under LD, with no brief LL exposure, intra-SCN 8-OH-DPAT or vehicle injections did not have any significant phase-shifting effect at any of the ZTs tested.
Constant Light-Potentiated 8-OH-DPAT Phase Advances at ZT 0 Are Blocked by 5-HT1A Receptor Antagonists
Pharmacological evidence suggests that serotonergic phase resetting is mediated by 5-HT7 receptors (10, 14, 27, 45). Evident from the results summarized in Fig. 6 is that the large- magnitude phase-advancing effect of intraperitoneal injected 8-OH-DPAT (0.5 mg/kg) at ZT 0 in animals treated with brief LL (11.3 ± 0.3 h) is probably not mediated by 5-HT7 receptors, because this shifting was not inhibited by coinjection of a 10-fold higher administered concentration of ritanserin ([5.0 mg/kg] 10.9 ± 0.6 h, P < 0.5 vs. 8-OH-DPAT). Pretreatment with a cocktail containing 10-fold higher administered concentrations of ritanserin and metergoline (both 5.0 mg/kg) also had no suppressive effect on 8-OH-DPAT shifting (11.4 ± 0.8 h, P < 0.5 vs. 8-OH-DPAT). In marked contrast, pretreatment with either (-)-pindolol or WAY 100635 (both 5.0 mg/kg) completely abolished 8-OH-DPAT shifting response (1.7 ± 0.8 and 1.6 ± 0.3 h, respectively, both P < 0.01 vs. 8-OH-DPAT).
Constant Light-Potentiated Behavioral (Sleep Deprivation) Phase Advances at ZT 0 Are Blocked by 5-HT1A Receptor Antagonism
Consistent with our previous report (24), brief constant light treatment potentiates the phase-advancing effect of sleep deprivation at ZT 0. This behavioral shifting is thought to be mediated by the serotonergic system. The 5-HT receptor mediating such response has never been identified, but the present data suggest that it is the same as that mediating brief LL-potentiated 8-OH-DPAT shifts. Animals exposed to brief LL that received vehicle injection before sleep deprivation exhibited large-magnitude shifts averaging 8.7 ± 1.5 h (Fig. 7). In contrast, animals treated with brief LL that received (-)-pindolol injection (5.0 mg/kg) before sleep deprivation exhibited phase-advance shifts averaging only 2.3 ± 1.4 h (P < 0.01 vs. the vehicle + sleep deprivation group). Controls treated with brief LL and vehicle injection without sleep deprivation had advance shifts of only 1.3 ± 0.2 h and those treated with brief LL but not vehicle injection or sleep deprivation had advance shifts averaging 2.0 ± 0 h (both groups, P < 0.05 vs. vehicle + sleep deprivation controls).
Constant Light Does Not Affect SCN 5-HT1A Binding Sites
As shown by receptor autoradiography, the brief LL exposure did not alter the extent of this binding at ZT 6 or 0, relative to LD levels at the same time points (Fig. 8).
Exposure to brief LL markedly alters the magnitude and shape of the 8-OH-DPAT phase-response curve (PRC), facilitating large (Type 0) phase-advance shifts during the early morning when 8-OH-DPAT normally has no phase-shifting effect. This photic treatment also reduces the response threshold for 8-OH-DPAT shifting at midday, as evidenced by maximal 5–6 h phase-advance shifts elicited by small dosages having no effect under LD. The hypersensitized phase advances to intraperitoneal 8-OH-DPAT injection administered at ZT 0 in brief LL-exposed hamsters were blocked by the potent 5-HT1A antagonists, (-)-pindolol and WAY 100635, indicating that this shifting is mediated by 5-HT1A receptors. The total lack of effect of antagonists with activity at 5-HT7 receptors, including ritanserin and metergoline, on 8-OH-DPAT shifting argues against involvement of 5-HT7 receptors that have been strongly implicated in mediating other serotonergic clock-resetting responses (10, 14, 19, 27, 44, 45). The finding that intra-SCN microinjection of 8-OH-DPAT at ZT 0 in brief LL-exposed hamsters induces shifts of similar magnitude as those produced by intraperitoneal injection of the drug at this time implies further that 5-HT1A receptors residing in the SCN may mediate potentiated 8-OH-DPAT-induced shifts during the early morning. Notably, there were no shifting effects of intra-SCN 8-OH-DPAT at ZT 6 or 18, when systemic 8-OH-DPAT induces large advance and delay shifts, respectively (24), suggesting that shifts induced at these latter time points are mediated at target site(s) for 8-OH-DPAT outside the SCN. Lastly, the demonstration that brief LL-potentiated behavioral (sleep deprivation) advance shifts were also blocked by (-)-pindolol pretreatment indicates that these shifts were mediated by the endogenous activation of 5-HT neurons possibly acting via the same SCN 5-HT1A receptor pathway for nonphotic clock phase resetting as that stimulated in the 8-OH-DPAT trials.
Brief Constant Light Exposure Hypersensitizes Serotonergic Phase Resetting
The phase-advancing effect of systemic 8-OH-DPAT administration during the subjective day in Syrian hamsters is well documented (5, 24, 30, 49). At the highest dose most commonly used (5 mg/kg ip), 8-OH-DPAT injection at midday produces phase-advance shifts in the range of ∼1–2 h, which is similar in magnitude and circadian phase of response to nonphotic shifts caused by behavioral stimuli, including wheel-running, social interaction, and sleep deprivation (reviewed in Refs. 31 and 32). In the present study, brief LL-exposed animals exhibited dramatically enhanced phase-advance shifts averaging ∼12 h at ZT 0 and ∼6 h at ZT 6. At this latter time point, maximal shifting was elicited at a dose of only 0.1 mg/kg, which had no shifting effect in animals kept under LD. In the animals maintained under LD, only the highest dose of 8-OH-DPAT (5 mg/kg) produced shifts at ZT 6 (averaging ∼2 h), and no shifting was observed at any dose at ZT 0. From these data, it is apparent that the brief LL treatment substantially enhances the magnitude of 8-OH-DPAT-induced phase shifting, lowers the threshold for shifting response to the drug, and alters the shape of the PRC for 8-OH-DPAT phase resetting.
It must be noted that other types of large serotonergic circadian phase shifts have been reported. For example, systemic administration of the selective 5-HT1A receptor agonist MKC-242 significantly accelerates reentrainment to delay and advance shifts of the LD cycle and potentiates light-induced phase advances (35). Also, blocking raphe input to the SCN using the mixed 5-HT1A agonists/antagonists BMY 7378, S 15535 and MDL 73005 EF (16) markedly enhances photic phase-resetting responses. Such serotonergic potentiations of photic shifting are mechanistically distinct from the brief LL-potentiated 8-OH-DPAT shifts we report here, however, because the present effects involve an enhancement of nonphotic, rather than the potentiation of photic, phase-resetting responses.
The mechanism underlying the potentiating effect of brief LL exposure is not clear, but the various circadian-related effects of this treatment suggest various possibilities. The first of these relates to the highly suppressive effect of brief LL exposure on nocturnal in vivo 5-HT release measured in the SCN (24). Such a reduction in nighttime serotonergic tonus could theoretically produce a depletion-induced hypersensitization of postsynaptic serotonergic response the following day, thus enhancing phase-shifting responses to 8-OH-DPAT. This is supported by results from our previous study showing that increasing nighttime levels of SCN extracellular 5-HT under brief LL using reverse microdialysis dampens the potentiating effect of brief LL exposure on 8-OH-DPAT phase advances, presumably by decreasing hypersensitization of postsynaptic serotonergic response (24). Whereas the present data support a role for 5-HT1A receptors in mediating brief LL-potentiated 8-OH-DPAT and sleep deprivation phase shifts (discussed below), our receptor autoradiography data do not support the hypothesis that the hypersensitizing effect of brief LL exposure is due to an upregulation of these receptors. Therefore, the increased sensitivity to 8-OH-DPAT after brief LL exposure may involve a change in 5-HT1A receptor-mediated action at some signal transduction step downstream from ligand binding. It is also notable that 5-HT7 and 5-HT1B receptor binding in the SCN is not affected by brief LL as shown in our previous study (11). It is also possible that the reduced 5-HT release during the brief LL exposure provides a lower noise background relative to the exogenous administration of exogenous agonist, and this improved signal-to-noise is apparent in the larger phase shifts.
Another possibility is that the generalized suppressive effects of brief LL exposure on circadian-related functions including locomotor behavior, SCN 5-HT release (24), SCN Per mRNA (11, 48), and SCN electrical activity (43, 51) could reflect an attenuation of circadian pacemaker amplitude. Such an effect could enhance the effective strength of phase-shifting stimuli to levels sufficient for large-magnitude (Type 0) phase shifting. For oscillator systems, such as the circadian pacemaker, amplitude is considered to be an important determinant of phase-resetting response (22, 25, 50). If the circadian amplitude is made sufficiently small, a stimulus that would normally produce smaller (Type 1) shifts could produce Type 0 shifts like those seen here. This model is empirically supported by demonstrations that treatments suppressing circadian amplitude can produce Type 0 shifting of circadian rhythms (see Ref. 22). Theoretically, such a model would predict that the brief LL treatment would potentiate shifting responses to all types of clock-resetting stimuli. This seems to be not the case, however, as although shifts to 5-HT agonists and neuropeptide Y are potentiated by brief LL, a recent report has shown that the phase-resetting responses to intra-SCN injection of NMDA (a glutamate agonist mimicking photic stimulation) (26) or melatonin (Glass JD, unpublished observation) are not potentiated by brief LL treatment. These results are significant because it appears that brief LL exposure differentially sensitizes the SCN clock to the phase-resetting actions of various nonphotic entraining inputs by a mechanism likely unrelated to pacemaker attenuation.
Another notable aspect of the brief LL potentiation, apart from its enhancing effects on drug or behavioral shifting, is the phase-advancing effect of the LL-DD transition. This effect is evident in the vehicle control groups where phase advances of ∼2–4 h occurred after release to DD [part of the Aschoff type II procedure (4)]. This phenomenon is similar to that reported previously, where hamsters kept under LL for a longer duration (10 days) and then released to DD exhibited phase advances of similar magnitude (1). Also, hamsters housed under LL for periods long enough to promote free-running activity exhibited ∼2- to 3-h phase-advance shifts in response to a variety of nonphotic stimuli, including control injections (8). We have reported that the large potentiating effects of brief LL exposure on 8-OH-DPAT phase-advance shifts at ZT 6 is lost after 3 days of LL exposure, owing largely to the increased shifting responses of controls masking the drug effect (24). Interestingly, at ZT 18, control animals exhibit advance shifts, whereas 8-OH-DPAT-treated animals exhibit large delay shifts. Although the physiological basis for the shifting associated with the LL-DD transition is unknown, it likely involves a 5-HT1A receptor-mediated pathway, as this shifting is significantly attenuated by intraperitoneal (-)-pindolol injection (Glass JD, unpublished observation).
Pharmacologic Assessment of 5-HT1A Receptors Mediating Constant Light-Potentiated 8-OH-DPAT Shifts at ZT 0
The hypersensitizing effect of brief LL on the phase-advancing effect of systemic 8-OH-DPAT application at ZT 0 is unique in terms of both its impact on phase-shift magnitude and circadian phase of shifting responsiveness. Normally under standard experimental LD conditions, ZT 0 is an unresponsive “dead zone” of nonphotic (including serotonergic) PRCs. It is thus possible that the shifting pathway influenced by the brief LL treatment at this time of day may be distinct from the 5-HT receptor-mediated pathway(s) mediating phase shifting at other times of day, and/or under LD conditions. The 5-HT7 receptor is regarded as a good candidate for mediating serotonergic phase resetting, because its activation inhibits circadian clock gene expression (10, 20) and application of 8-OH-DPAT to the SCN or dorsal raphe nucleus induces phase advances that are blocked by 5-HT7 antagonists (10, 14, 27, 30). The present data, however, do not support a role of the 5-HT7 receptor in mediating brief LL-potentiated 8-OH-DPAT shifts, as indicated by the lack of inhibitory effect of 10-fold higher administered concentrations of 5-HT7 antagonists, ritanserin and metergoline (although this does not necessarily mean 10-fold higher concentrations of the antagonists surrounding SCN neurons). Rather, the blockade of potentiated 8-OH-DPAT shifting by the potent 5-HT1A receptor antagonists, (-)-pindolol and WAY 100635, support the involvement of 5-HT1A receptors. In this regard, other in vivo studies have shown that 8-OH-DPAT shifts in hamsters housed under LL for several days are blocked by (-)-pindolol (49), and 8-OH-DPAT shifts in wild-type and in 5-HT7 knockout mice are blocked by WAY 100635 (44), suggesting that 5-HT1A receptors are primarily responsible for mediating these shifts. It should be noted that metergoline also reportedly has antagonist action at the 5-HT1A receptor (44) and blocks 8-OH-DPAT shifts in the mouse slice (37). Thus it is puzzling why here metergoline (mixed with ritanserin) had no attenuating effect on brief LL-potentiated 8-OH-DPAT shifts. Possibly this could be due to a lack of complete antagonism of 8-OH-DPAT binding to 5-HT1A receptors by metergoline, with residual agonist binding sufficient for inducing large shifts under LL-sensitized conditions. The pharmacokinetics of such interactions are currently unknown, however.
SCN Is a Target for Brief LL-Potentiated 8-OH-DPAT Action
The observation that intra-SCN perfusion of 8-OH-DPAT at ZT 0 causes large shifts in brief LL-exposed hamsters is significant for two reasons: 1) this suggests that the 5-HT1A receptors implicated in mediating this shifting (as discussed above) reside in the SCN, and 2) this is the first in vivo demonstration of large-magnitude type 0 shifts induced by direct serotonergic stimulation of the SCN. Considering the first point, there is little published data supporting a nonphotic phase-resetting role of 5-HT1A receptors in the SCN. In the rat for example, 5-HT1A mRNA expression in the SCN is sparse (39), and administration of 5-HT1A antagonists, including pindolol, directly to cultured SCN brain slices do not block 8-OH-DPAT advance shifts (27). Also in hamsters and rats, pharmacological data indicate that 8-OH-DPAT exerts its phase-resetting action via 5-HT7 receptors (and possibly 5-HT5A receptors) (46), rather than by 5-HT1A receptors (10, 14, 27, 45). However, there is a recent report from work in the C57BL/6J mouse SCN brain slice where 8-OH-DPAT shifts were blocked with WAY 100635, implicating 5-HT1A receptors in mediating serotonergic phase resetting (44). Given that the phase-resetting effect of 8-OH-DPAT is less robust in vivo (3), compared with the SCN slice preparation (37, 44), it is possible that a similar type of neurochemical enhancement of serotonergic response could occur in the brief LL-exposed hamster SCN, resulting in enhanced sensitivity to 8-OH-DPAT.
Related to the issue of hamster SCN sensitivity to 5-HT are observations that behavioral shifts to intra-SCN 8-OH-DPAT treatment in this species are absent or small (<40 min) (7, 14, 30). It has been proposed that the capacity of the SCN to respond robustly to serotonergic phase-advancing influences (i.e., as demonstrated in vitro) depends upon its degree of postsynaptic sensitivity to 5-HT, which could be modulated by endogenous (or experimental) changes in serotonergic tonus. For example, in the deafferented rat and mouse SCN brain slice, where 5-HT release is presumably reduced, direct application of 8-OH-DPAT to the SCN elicits phase advances (∼2–4 h) that are considerably larger than those in vivo (37, 38, 42, 44). Also, phase advances to intra-SCN 8-OH-DPAT perfusion are significantly larger (∼1.6 h) in p-CPA-treated hamsters with depleted 5-HT compared with non-p-CPA-treated controls. Nevertheless, these larger depletion-potentiated shifts are still modest compared alongside the ∼12-h shifts of brief LL-exposed animals. This suggests that depletion-induced sensitization in itself may contribute to, but not fully account for, the large-magnitude serotonergic circadian shifting responses produced by brief LL exposure. It is notable that there were no potentiated phase-shifting effects of the intra-SCN 8-OH-DPAT treatments at ZT 6 or 18. The lack of effect at ZT 18 is of particular interest, because systemic application of 8-OH-DPAT at this time point in brief LL-treated animals causes ∼12-h delay shifts (24). Notably, there was no tendency toward a phase-delaying response at this time point, with the animals exhibiting small (∼2 h) advance shifts similar to the SCN vehicle perfusion controls and intraperitoneal vehicle controls in the former study. Thus the SCN may be directly responsive to the brief LL-potentiated 8-OH-DPAT shifting effect only during early morning, with the phase-resetting actions of the drug mediated by loci outside the SCN at other times of the day. Under LD conditions, the intra-SCN 8-OH-DPAT perfusion did not produce phase shifts at any of the time points that were different from vehicle perfusions, which is in general agreement with other reports discussed above.
5-HT1A Antagonist Blockade of Constant Light Potentiated Behavior-Induced Phase-Shifting
As noted above, brief LL exposure markedly enhances the phase-resetting response to behavioral stimuli, including sleep deprivation and novel wheel exposure (24, 26). Interestingly, large phase-advance shifts have also been reported in Syrian hamsters group housed under LD with no running wheels and then transferred individually during the light phase to a cage with a wheel under DD (17). Because the large shifting responses in both studies are of similar magnitude and phase, they may share a common pathway. The nature of this pathway is unclear. However, the present demonstration that (-)-pindolol abolished the phase-resetting effect of sleep deprivation in brief LL-exposed animals strongly suggests that this shifting involves the endogenous activation of 5-HT. These data indicate further that the sleep-deprivation phase-shifting response could be mediated by the same SCN 5-HT1A receptor pathway for nonphotic clock phase resetting as that stimulated by 8-OH-DPAT in the foregoing experiments. We have shown previously that behavioral stimuli, including novel wheel-running (9) and sleep deprivation (18) significantly stimulate the release of 5-HT in the SCN at midday via activation of serotonergic neurons in the midbrain raphe complex. Possibly, such 5-HT release from SCN terminals of median raphe neurons is responsible for mediating the brief LL-potentiated sleep deprivation phase-resetting response. It is notable that although (-)-pindolol blocks 8-OH-DPAT shifts in hamsters (present results and Ref. 49), treatment of hamsters under LD with 5-HT1A antagonists (intraperitoneal injection of NAN-190 and WAY 100135) does not block novel wheel-induced phase advances at midday (2). Therefore, the neural substrate mediating the brief LL-potentiated nonphotic shifting at ZT 0 may be distinct from that which mediates nonphotic advance shifts during the day in animals under LD.
In summary, the results of the present study underscore the strong potentiating effect of brief LL exposure on serotonergic phase resetting in hamsters. This treatment greatly increases the amplitude of 8-OH-DPAT-induced advance shifts and lowers the threshold for drug-induced phase resetting. Moreover, it renders animals highly sensitive to stimulation at a portion of the nonphotic PRC (at ZT 0) that is normally insensitive to serotonergic treatment. The brief LL-potentiated shifting effect of 8-OH-DPAT at ZT 0 is unique in that it appears to be mediated by 5-HT1A receptors located in the SCN, which is not the case at other time points. Moreover, the present sleep-deprivation trials indicate that these receptors may also mediate the potentiated effects of behavioral activation at this time point. Although the mechanism underlying these hypersensitizing effects of brief LL exposure is not yet fully elucidated, further study of this phenomenon is useful, as it could help in developing strategies for enhanced circadian clock resetting in humans.
This research was supported by National Institute of Neurological Disorders and Stroke Grant NS-35229 (to J. D. Glass), and National Institute on Aging Grant AG-013418 (to M. J. Duncan).
We are grateful to Brandon J. Riley for assisting with the behavioral stimulation trials and Verda A. Davis for assisting with the receptor autoradiography.
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
- Copyright © 2006 the American Physiological Society