Physiology & Behavior 139 (2015) 136–144

Contents lists available at ScienceDirect

Physiology & Behavior journal homepage: www.elsevier.com/locate/phb

Effects of lighting condition on circadian behavior in 5-HT1A receptor knockout mice Victoria M. Smith a,b, Ryan T. Jeffers a,b, Brendan B. McAllister a,b, Priyoneel Basu a,b, Richard H. Dyck a,b,c, Michael C. Antle a,b,d,⁎ a

Department of Psychology, University of Calgary, Calgary, Alberta, Canada Hotchkiss Brain Institute, University of Calgary, Calgary, Alberta, Canada Department of Cell Biology and Anatomy, University of Calgary, Calgary, Alberta, Canada d Department of Physiology and Pharmacology, University of Calgary, Calgary, Alberta, Canada b c

H I G H L I G H T S • • • • •

Serotonin regulates both photic and non-photic responses of the circadian clock 5-HT1A receptor deletion decreased levels of wheel running in LD and LL Re-entrainment to a shifted light dark cycle was delayed Re-entrainment to a shifted light dark cycle exhibited delayed transients Phase shifts to dark pulses were attenuated

a r t i c l e

i n f o

Article history: Received 17 July 2014 Received in revised form 31 October 2014 Accepted 3 November 2014 Available online 8 November 2014 Keywords: Mice Serotonin 5-HT1A Dark pulse Period Amplitude Circadian Behavior

a b s t r a c t Serotonin (5-HT) is an important regulator of the mammalian circadian system, and has been implicated in modulating entrained and free-running rhythms, as well as photic and non-photic phase shifting. In general, 5-HT appears to oppose the actions of light on the circadian system of nocturnal rodents. As well, 5-HT mediates, at least in part, some non-photic responses. The 5-HT1A, 1B and 7 receptors regulate these acute responses to zeitgebers. 5-HT also regulates some entrained and free-running properties of the circadian clock. The receptors that contribute to these phenomena have not been fully examined. Here, we use 5-HT1A receptor knockout (KO) mice to examine the response of the mouse circadian system to a variety of lighting conditions, including a normal light–dark cycle (LD), T-cycles, phase advanced LD cycles, constant darkness (DD), constant light (LL) and a 6 hour dark pulse starting at CT5. Relative to wildtype mice, the 5-HT1A receptor KO mice have lower levels of activity during the first 8 h of the night/subjective night in LD and LL, later activity onsets on transient days during re-entrainment, shorter free-running periods in LL when housed with wheels, and smaller phase shifts to dark pulses. No differences were noted in activity levels during DD, alpha under any light condition, freerunning period in DD, or phase angle of entrainment in LD. While the 5-HT1A receptor plays an important role in regulating photic and non-photic phase shifting, its contribution to entrained and free-running properties of the circadian clock is relatively minor. © 2014 Elsevier Inc. All rights reserved.

1. Introduction The master mammalian circadian clock in the suprachiasmatic nucleus (SCN) regulates daily rhythms in physiology and behavior [1]. Serotonin (5-HT) is an important regulator of the circadian system [2,3], ⁎ Corresponding author at: Department of Psychology, University of Calgary, 2500 University Drive NW, Calgary, Alberta T2N 1N4, Canada. Tel.: + 1 403 220 2574; fax: + 1 403 282 8249. E-mail address: [email protected] (M.C. Antle).

http://dx.doi.org/10.1016/j.physbeh.2014.11.005 0031-9384/© 2014 Elsevier Inc. All rights reserved.

modulating responses to both photic and non-photic zeitgebers. The SCN receives 5-HTergic input from the midbrain raphe nuclei [4–7], and contains a number of 5-HT receptor types, including the 1A, 1B, 2A, 2C, 5A, and 7 subtypes [8–14]. The intergeniculate leaflet (IGL), which innervates the SCN to provide both photic and non-photic signals, also receives 5-HTergic input from the midbrain raphe nuclei [15] and contains 5-HT1A and 5-HT7 receptors [11]. The number of receptors, as well as the number of regions in the circadian network that are affected by 5-HT, presents a challenge for understanding 5-HT's role in regulating circadian rhythmicity.

V.M. Smith et al. / Physiology & Behavior 139 (2015) 136–144

Extracellular 5-HT levels in the rodent SCN are related to behavioral state, with levels peaking during the early night in nocturnal rodents [16]. Non-photic manipulations, such as wheel running or sleep deprivation during the daytime inactive period, increase 5HT output at both the SCN [16,17] and IGL [18]. 5-HT's function changes across the day, with daytime 5-HT contributing to nonphotic phase shifts [2], and nighttime 5-HT opposing photic input to the clock [3]. Animals in which the 5-HTergic system has been lesioned with neurotoxins or other pharmacological agents exhibit larger responses to phase shifting light [19–23] and weakened responses to some non-photic manipulations [24,25]. 5-HTergic modulation of the acute responses to light, as well as some non-photic effects, appears to be mediated, in part, by the 5-HT1A receptor [26]. 5-HT also modifies entrained and free-running properties of the circadian clock. Lesions of the median raphe advance activity onset, delay activity offset and expand the duration of the active phase [15,20,22,27,28]. More selective lesions of just the 5-HTergic input to the SCN only lead to advanced activity onsets [27]. Manipulations that alter 5-HT levels can also affect free-running period under certain circumstances. For instance, global 5-HT lesions lengthen the free-running period of hamsters housed in constant light (LL) [29]. Conversely, augmenting 5-HT output with chronic fluoxetine shortens the free-running period of mice housed in constant dark (DD) [30]. However, neither 5-HT lesions in hamsters housed under DD [29] nor tryptophan depletion in blind rats [31] alters free-running period. Overall, these results do indicate that entrained and free-running properties of the circadian clock are sensitive to 5-HT. However, there has been little investigation into which receptors might mediate these effects. The 5-HT1B receptors in the SCN are primarily located on retinal terminals [8]. Activation of these receptors is generally associated with diminished responses to retinal input [32]. 5-HT1B receptor KO animals exhibit a variety of other alterations to their circadian system, including longer freerunning periods in LL, narrower circadian limits of entrainment to T-cycles, and altered phase angle of entrainment [33,34]. Animals lacking the 5-HT 7 receptor re-entrain to an advanced LD cycle at the same rate as controls, have similar phase angles of entrainment, free-running periods, activity levels, and duration of activity [35]. The 5-HT2C receptor KO has not been systematically examined for circadian alterations, but they do exhibit higher activity levels in the dark [36,37]. We have previously reported that mice lacking the 5-HT1A receptor have larger phase advance responses to light [26], although phase delays to light are unaffected. This is consistent with the effects of drugs that bind to the 5-HT1A receptor, with agonists attenuating photic phase shifts [38] and antagonists enhancing photic advances [39–41]. These findings demonstrate that the 5-HT1A receptor is important for responses to brief light pulses, particularly in the phase advance region of the phase response curve (PRC). It is not clear how the 5-HT1A receptor might regulate responses to more long-term exposure to light such as might be experienced under a normal light–dark (LD) cycle. If the 5HT1A receptor regulates acute and long-term light exposure in a similar fashion, we would hypothesize that entrained responses to light could also be altered in 5-HT 1A knockout (KO) animals. If the 5HT 1A receptor plays a role in modulating light input to the clock, then we might also expect circadian responses to LL (lengthening of period and shortening of the duration of the active phase), but not to DD, to be altered in the 5-HT 1A KO mice. Finally, given the evidence that the 5-HT1A receptor plays a role in non-photic phase shifting as well, we hypothesize that phase shifts to dark pulses, which have both photic and non-photic components, will also be affected in 5-HT1A KO animals. To test these hypotheses, we examined the 5-HT1A KO mice under a variety of lighting conditions, including a normal 12:12 LD cycle, accelerated days (11.5:11.5 days and 11:11 days), a phase advance of their LD cycle, DD, LL, and a dark pulse when housed in LL.

137

2. Materials and methods 2.1. Animals and housing A total of 102 male mice were used for these experiments, consisting of 9 C57BL/6J mice obtained from the University of Calgary Biological Sciences breeding colony (Calgary, AB, Canada), as well as 49 5-HT1A receptor WT and 44 5-HT1A KO mice. The KO mice were generated from a breeding colony established in our laboratory from mice originally developed by Dr. Thomas Shenk (Princeton University, Princeton, NJ, USA), provided by Dr. Miklos Toth (Cornell University Medical College, New York, NY, USA) and backcrossed to the C57BL/6J line from the University of Calgary's local breeding program. For information regarding the generation of the 5-HT1A KO mice see Parks et al. [42]. Animals from our breeding colony were genotyped using a pair of KO primers (amplifying a 400 base pair product that included a portion of the neomycin cassette that replaced the start codon of the 5-HT1A receptor gene) as well as a pair of WT primers (amplifying a 238 base pair product that included the portion of the 5-HT1A receptor gene replaced by the neomycin cassette in the KO animals) to identify the presence or absence of each allele of the 5-HT1A receptor gene (KO up primer: CTT TAC GGT ATC GCC GCT CCC GAT TC; KO down primer: TGC AGG ATG GAC GAA GTG CAG CAC A; WT up primer: AGT GCA GGC AGG CAT GGA TAT GTT; WT down primer: CCG ATG AGA TAG TTG GCA ACA TTC TGA). Animals were individually housed in Nalgene Type L clear polycarbonate cages (30.3 cm long × 20.6 cm wide × 26 cm high; Nalg Nunc International, Rochester, NY, USA), equipped with a 24.2 cm diameter stainless steel running wheel. Mice were housed in a temperature — (21 ± 1 °C) and humidity-controlled room, with ad libitum access to food and water. Cages were changed approximately every 14 days (7– 10 days prior to and following a manipulation). All protocols were approved by the Life and Environmental Sciences Animal Care Committee at the University of Calgary, and adhered to the Canadian Council on Animal Care guidelines for the ethical use of animals. Every attempt was made to minimize both the number and the suffering of animals used in these experiments.

2.2. Activity rhythms Daily wheel-running activity was continuously monitored using magnetic switches mounted on running wheels, which were connected to a computer running the Clocklab data collection software package (Coulbourn Instruments, Allentown, PA, USA), with data being collected into 10-min bins. Actograms were generated and analyzed by Clocklab analysis software. Mice were allowed to entrain to the 1500 lx –0 lx LD cycle (WT n = 6, KO n = 12), or free-run in 0 lx DD (WT n = 25, KO n = 20), or 300 lx LL (WT n = 19, KO n = 23) for a minimum of 7 days prior to the start of manipulations. All of the animals from the LD condition and some from the LL condition (WT n = 1, KO n = 1) were also used for the DD condition. Some animals were also tested under an 11.5:11.5 LD cycle (T23, WT n = 6, KO n = 6) and an 11:11 LD cycle (T22, WT n = 6, KO n = 4, all from the T23 experiment). Onset of running wheel activity for manipulation day was predicted using a regression line fit to the activity onsets for the 7–10 days prior to testing day. General locomotor and body temperature rhythms were monitored in a subset of animals (c57BL/6J = 4, WT = 2, KO = 6) for a portion of the study using telemetry. During this portion of the study the animals were housed without access to a running wheel. Under surgical anesthesia (isoflurane, 2%) and using sterile techniques, they had a G2 EMitter (Bio-Lynx Scientific Equipment, Montreal, Quebec, Canada) surgically implanted into their peritoneal cavity. General cage activity and body temperature were recorded by a computer running Vital View Software (Bio-Lynx). Circadian parameters for c57BL/6J mice and

138

V.M. Smith et al. / Physiology & Behavior 139 (2015) 136–144

WT mice were compared and pooled if they were not significantly different. 2.3. Circadian behavior parameters Several circadian behavior parameters were examined under 3 different lighting conditions (LD, DD, LL). In LD, phase angle of entrainment was determined by calculating the average latency from lights-off to activity onset over a 7-day period, with positive values indicating an activity onset that preceded dark onset. Phase angle was also calculated in the T23 and T22 lighting conditions. If the animals entrained to these LD cycles, the last 7 days were averaged to calculate the phase angle. Patterns of activity were analyzed in all three lighting conditions. Average waveforms of activity profiles were created by first averaging the daily activity profile for each animal over approximately 10 days, followed by averaging these profiles over all animals of the same genotype. In LD, 10-min bins were aligned based on the bin that coincided to lights-off. For DD and LL, 10-min bins were aligned based on activity onset (designated as CT12 by convention), which was determined to be the first 10-min bin in which the activity count was greater than 20% of the animal's maximum activity in a 10-min bin and remained above this threshold for at least 3 of the following 6 10-min bins. For statistical analysis, activity was summed into 4-hour blocks. Average waveforms were also used to calculate duration of locomotor activity (alpha) under all 3 lighting conditions. This was done by subtracting the time of activity onset from the time of activity offset. Activity onsets and offsets under each lighting condition were determined as follows: LD) activity onset = first bin above 20% of maximum after lights off,

activity offset = last bin above 20% of maximum; DD) activity onset = same as used for aligning activity waveforms above, activity offset = last 10-min bin with activity above 20% of maximum after which activity drops below that threshold for at least 6 consecutive bins; and LL) activity onset = same as used for aligning activity waveforms above, activity offset = last 10-min bin with activity greater than 20% of maximum followed by at least 4 h without a bin reaching that threshold. Free-running period (tau) was calculated using the Clocklab analysis software under DD and LL, by applying the Lomb– Scargle periodogram function to 7–10 days of activity under each of those lighting conditions. 2.4. Re-entrainment to a shifted LD cycle Mice (n = 6 WT, 6 KO) were entrained to an LD cycle for 17 days, at which point their LD cycle was advanced by 6 h. They remained on the new LD cycle for 37 days. Animals were said to have entrained when their phase angle was within 20 min of their average phase angle from the 4 days prior to the LD advance. 2.5. Phase shifts to a 6-hour dark pulse A subset of the mice used to examine activity in LL (WT = 6, KO = 9) were used to examine responses to dark pulses. C57BL/6J mice (n = 5) were also used to increase the sample size of non-KO animals. Mice were allowed to free-run in (300 lx) for at least 2 weeks. They were then exposed to a 6-hour dark pulse from CT5 to CT11, by individually transferring them to a light-sealed box equipped with a recording

Fig. 1. Double-plotted actograms from representative (A) WT and (B) 5-HT1A KO mice housed in LD. Every horizontal line represents the data from 2 consecutive days, with subsequent days plotted both to the right and below the previous day. Shading represents times when the lights were off. Vertical deflections from the horizontal represent 10-minute bins during which wheel running occurred, with the height of the line being proportional to the amount of activity. (C) Waveforms of locomotor activity averaged over approximately 10 days and then averaged over all animals within that particular genotype (solid line = WT, dashed line = 5-HT1A KO mice). For statistical analysis, average activity (white bars for WT, gray bars for KO) was summed into 4 h bins (D). Error bars represent the standard error of the mean (SEM). * = p b 0.05.

V.M. Smith et al. / Physiology & Behavior 139 (2015) 136–144

cable so their running wheel activity could be monitored during the dark pulse. Wheel running activity was then monitored for at least 2 weeks following the dark pulse, and phase shifts were calculated using the actograms generated by Clocklab software. A regression line was fit to the activity onsets for the 10–14 days prior to the manipulation day, and another regression line was fit to the activity onsets for the 10–14 days following the treatment (excluding the 3 days immediately following the dark pulse to reduce the effect of transients). The resulting phase shift of the wheel running activity was calculated using the horizontal difference between the two generated lines on the day following the dark pulse. Total activity counts for each animal during the dark pulse were also calculated. Phase shifts for c57BL/6J mice and WT mice were compared and pooled if they were not significantly different. 2.6. Statistical analysis To determine if there was a significant difference in daily activity between the two genotypes, a 2 (WT, KO) by 6 (the six 4-hour time blocks) factorial analysis of variance (ANOVA) was conducted. A 2 (WT, KO) by 21 (day) factorial ANOVA was used to examine the daily activity onsets during the 6 h phase advance experiment. Student–Newman–Keuls post hoc analyses were used to examine pair-wise differences if significant interactions were detected by the ANOVAs. Independent sample t-tests were used to determine if there were significant differences between the two genotypes for each of phase angle of entrainment in LD, T23, T22, tau in DD, tau in LL, alpha in LD, alpha in DD,

Fig. 2. (A) Activity durations (alpha) under LD, DD, and LL from WT (white bars) and 5HT1A KO mice (gray bars). (B) Free-running period under DD and LL from WT (white bars) and 5-HT1A KO mice (gray bars). Error bars represent the SEM. * = p b 0.05.

139

alpha in LL, days to re-entrain to a shifted LD cycle, the magnitude of phase shift elicited by a 6-hour dark pulse and the total wheelrunning activity counts during the dark pulse. A Pearson's correlation was also calculated to determine if there was a significant correlation between the amount of activity during the dark pulse and the magnitude of the resulting phase shift. All means are reported ± SEM. 3. Results 3.1. Light–dark Both WT and KO animals entrained to the 12:12 LD cycle (Fig. 1A–B). KO mice had lower activity levels than WT mice in the first 8 h of darkness (genotype × time-block interaction, F(5,75) = 3.843, p = 0.004, Fig. 1C–D). There was no difference between WT and KO animals in terms of phase angle of entrainment in the 12:12 LD cycle (WT = 0.128 ± 0.05 h, KO = 0.203 ± 0.06 h, t(13) = 0.863, p = 0.404) or alpha (t(13) = 0.557, p = 0.587, Fig. 2A) under the 12:12 LD cycle. All mice of both genotypes entrained to the T23 LD cycle (Fig. 3). There

Fig. 3. Sample actograms from animals housed in both T23 and T22 LD cycles. Each actogram is plotted modulo the T-cycle period (i.e., 23 h on the left, 22 h on the right). Each horizontal line represents the data from a single T-cycle day, with subsequent days plotted below the previous days. Each row represents data from a single animal. The top three rows come from wildtype animals and depict the 3 cases observed: Entrained, bouncing and relative coordination. The bottom two rows are from 5-HT1A KO animals and represent the two patterns observed: entrained and relative coordination. Bouncing was never observed in the 5-HT1A KO animals.

140

V.M. Smith et al. / Physiology & Behavior 139 (2015) 136–144

was no significant difference in phase angle of entrainment between the genotypes (WT = −0.6 ± 0.15 h, KO = −1.7 ± 1.05 h, t(10) = 1.044, p = 0.321). One KO animal did entrain to the T23 cycle with an extreme phase angle of −6.91 h (Fig. 3). Two of four KO animals and three of six WT animals entrained to the T22 LD cycle (Fig. 3). Two WT animals exhibited “bouncing” [43] when housed under the T22 cycle. This was not observed with any of the KO animals. One WT animal had an extreme phase angle of entrainment (− 6.3 h) to the T22 cycle. Due to the small number of animals that entrained to the T22 cycle, statistical analyses were not performed. When the extreme phase angle animal was excluded, the average phase angles were virtually identical for the two genotypes (WT = −0.37 ± 0.3 h, KO = −0.37 ± 0.06 h). 3.2. Re-entrainment When subjected to a 6 h phase advance of the LD cycle, there was no significant difference in the number of days required to re-entrain (WT = 5.33 ± 0.21 days, KO = 6.0 ± 0.58 days, t(10) = 1.084, p = 0.304). However, the phase of activity onsets during the first 5 days was significantly delayed in the KO animals relative to the WT animals (genotype × day interaction, F(20,200) = 2.021, p = 0.008, Fig. 4).

either WT (r = −0.207, p = 0.541) or KO mice (r = 0.476, p = 0.23, Fig. 7E).

4. Discussion 5-HT is an important regulator of entrained and free-running rhythms [44], as well as photic and non-photic phase shifting of the circadian system [2,3]. The precise roles of the various 5-HT receptors in regulating entrained and free-running rhythms have yet to be determined. The 5-HT1A receptor acts as both a somatodendritic autoreceptor that regulates raphe output, as well as a postsynaptic receptor mediating 5-HT's effects on target cells. In the present study, we show that mice lacking the 5-HT1A receptor exhibit some alterations in circadian parameters under a variety of lighting conditions. Specifically, they have lower activity levels during the first two-thirds of the night/

3.3. Constant darkness Both WT and KO mice exhibited stable free-running rhythms under DD (Fig. 5A–B). Under DD, a wide range of free-running periods were observed in both WT (23.05 h to 23.85 h) and KO mice (23.2 h to 23.95 h). No significant difference in free-running period was detected between the genotypes, either when housed with wheels (t(38) = 1.136, p = 0.263, Fig. 2B) or without (t(9) = 1.16, p = 0.274, Fig. 2B). Furthermore, no significant main effect of genotype (F(1,180) = 0.928, p = 0.342) or significant interaction between genotype and timeblock (F(5,180) = 0.167, p = 0.974) was detected on locomotor activity levels (Fig. 5C–D). The duration of activity did not differ between WT and KO animals (t(39) = 0.883, p = 0.383, Fig. 2A). 3.4. Constant light Both WT and KO mice exhibited free-running rhythms under LL (Fig. 6A–B), although activity levels were severely diminished. KO mice had lower activity levels than WT mice in the first 8 h of the subjective night (genotype × time-block interaction, F(5,75) = 5.657, p b 0.001, Fig. 6C–D). Overall, the duration of the active phase under LL was shorter than under DD, but did not differ between the genotypes (t(38) = 0.13, p = 0.898, Fig. 2A). When housed with running wheels, the free-running period of KO mice was significantly shorter than that of WT mice (t(39) = 2.317, p = 0.025, Fig. 2B). This difference in freerunning periods in LL was no longer significant when animals were housed without running wheels (t(9) = 1.13, p = 0.285, Fig. 2B). 3.5. Dark pulses A 6-hour dark pulse starting 7 h before activity onset elicited running wheel activity, and resulted in large phase advances in both WT and KO mice (Fig. 7A–B). Three WT mice exhibited near inversion of their circadian rhythms, and these were classified as phase advances due to the presence of advancing transients in the 3 days following the dark pulse before steady state was achieved. While still quite large, the KO mice had significantly smaller phase shifts to the dark pulse than did the WT mice (t(18) = 2.135, p = 0.047, Fig. 7C). Due to equipment failure, activity during the dark pulse was not obtained for one KO animal. For the remainder, there was no significant difference between the genotypes on the amount of activity during the dark pulse (t(17) = 1.13, p = 0.274, Fig. 7D). There was also no significant correlation between activity during the dark pulse and the magnitude of the phase shift for

Fig. 4. Actograms from representative wildtype (A) and 5-HT1A KO (B) animals subjected to a 6 h advance of their LD cycle. The average time of activity onset in the 5-HT1A KO animals was significantly delayed (*p b 0.05) relative to the WT animals for the first 5 days of the shifted LD cycle (C). The number of days required to re-entrain (WT = 5.33 ± 0.21, KO = 6.0 ± 0.58) did not significantly differ between the genotypes.

V.M. Smith et al. / Physiology & Behavior 139 (2015) 136–144

141

Fig. 5. Double-plotted actograms from representative (A) WT and (B) 5-HT1A KO mice housed in DD, using the convention described in Fig. 1. (C) Waveforms of locomotor activity averaged over approximately 10 days and then averaged over all animals within that particular genotype (solid line = WT, dashed line = 5-HT1A KO mice). For statistical analysis, average activity (white bars for WT, gray bars for KO) was summed into 4 h bins (D). Error bars represent the SEM.

subjective night under LD and LL, but not DD. Their transient onsets for the first five days following a 6-hour advance of their LD cycle are delayed relative to those of wildtypes. They exhibit a shorter freerunning period under LL, but only when housed with wheels. While they still exhibit large phase shifts to dark pulses, these shifts are significantly smaller than those observed in WT mice. Neither alpha nor phase angle of entrainment were altered in the 5-HT1A KO mice. These data add to our previous findings that 5-HT1A KO mice have larger phase shifts to advancing but not delaying light pulses [26]. A major role of 5-HT in the circadian system is to oppose the action of light on the SCN. 5-HT and its agonists decrease the release of neurotransmitters by retinal terminals in the SCN [32,45] and attenuate the responsiveness of SCN cells to retinal input [46,47]. Activity attenuates photic phase shifts through 5-HT [48] and 5-HT agonists applied directly to the SCN diminish behavioral phase shifts to light [49]. These findings are consistent with what we have observed previously with 5-HT1A KO mice [26], in that phase advances to light pulses are larger. Overall, these findings all support the model that during the night, 5-HT opposes the effects of light on the circadian clock, and that this inhibition is diminished in mice lacking the 5-HT1A receptor. Entrained and free-running properties of the circadian clock are modified by manipulations that decrease or enhance 5-HTergic tone [5,20,27,30,50]. We observed decreased activity in 5-HT1A KO mice during the first 8 h of both the night in LD and the subjective night in LL. Activity levels in DD did not differ from those observed in WT mice. The difference in LL but not in DD might reflect relative levels of masking by light. If 5-HT acts through the 5-HT1A receptor to oppose the effects of light, then light should be having a greater effect in the

5-HT1A KO mice where photic input is less inhibited. Activity levels in LL do not differ in 5-HT1B receptor KO mice [33], suggesting that 5HT's modulation of the masking effects of light may be mediated postsynaptically in the SCN. The reason for the difference in activity levels in LD is less clear. Free-running period in LL was shorter in the KO mice, but only when housed with wheels. The difference was about the same when housed without wheels, but was no longer significant due to larger variability in periods when housed without wheels. When housed with wheels, the change in period could be due to a number of factors. We have previously demonstrated that 5-HT1A KO mice have larger photic phase advances but not delays. Full illumination of the PRC should lead to more advanced phase, so the shorter tau might reflect the simple contribution from a PRC with a larger advance zone than in wild-types. This explanation would be consistent with the lack of difference in DD. Alternatively, the difference in free-running period might be related to the concomitant difference in activity levels. The relationship between free-running period and locomotor activity levels is well established [51–56]. Support for this hypothesis is that the difference in period was no longer significant when the animals were housed without wheels. However, this hypothesis is unlikely as the 5-HT1A KO mice had significantly lower activity levels, which are consistently associated with longer, not shorter, free-running periods [52]. Large changes in phase angle of entrainment of both activity onset and offset have been observed in animals with 5-HT neurotoxic lesions directed at either the SCN [15,27] or the raphe nuclei [20], or pharmacological depletion of 5-HT [22]. In contrast, phase angle of entrainment to a 12:12 LD cycle is not altered in our 5-HT1A KO mice. Given that mice

142

V.M. Smith et al. / Physiology & Behavior 139 (2015) 136–144

Fig. 6. Double-plotted actograms from representative (A) WT and (B) 5-HT1A KO mice housed in LL, using the convention described in Fig. 1. (C) Waveforms of locomotor activity averaged over approximately 10 days and then averaged over all animals within that particular genotype (solid line = WT, dashed line = 5-HT1A KO mice). For statistical analysis, average activity (white bars for WT, gray bars for KO) was summed into 4 h bins (D). Error bars represent the SEM. * = p b 0.05.

normally require a daily phase delay to entrain to a 12:12 LD cycle, and that 5-HT1A KO mice have normal delays, but larger advances to light [26,41], the phase angle finding is not surprising. We tested these mice under T23 and T22 cycles. There was no difference in phase angle in T23, although one 5-HT1A KO mouse did entrain with a severely delayed phase angle, something not observed with the WT mice. Under a T22 cycle, half of each genotype entrained. Two of six WT mice exhibited bouncing [43], something not observed in the 5-HT1A KO mice. It is likely that the T23 was not challenging enough for either genotype, and that the T22 was too challenging. Given the small sample for this experiment, strong conclusions cannot be drawn. A larger sample examining an intermediate T-cycle (e.g., T22.5) might be better able to resolve entrainment differences between these genotypes. While 5-HT appears to be important for regulating phase angle of entrainment based on studies with more global depletion of 5-HT in rats and hamsters [15,20,22,27], our present findings suggest that this effect, if present in mice, is not mediated through the 5-HT1A receptor. When subjected to a 6-hour phase advance of their LD cycle, the 5HT1A KO mice did not significantly differ in the number of days required to re-entrain. They did differ in the time of activity onset on the first five days of the shifted cycle, with later onsets than the wildtype animals. This is interesting, as with larger advances to light [26,41], their onset might be expected to be earlier than wildtypes, rather than later. This may reflect differences in how these mice respond to brief light (15 min in our previous studies) rather than long term light (~ 6 h illuminating the advance portion of the PRC in the 6-hour phase advance experiment). Dark pulses present an interesting case, consisting of both photic and non-photic components [57]. Some non-photic shifts are enhanced by prior LL exposure [58], a phenomenon that appears to be mediated by

changes in 5-HT sensitivity because of low 5-HT levels that results from LL exposure [59]. While most work examining dark pulses has been done in hamsters [57,60–62], dark pulses are also effective in mice [63]. The present results suggest that the 5-HT1A receptor is involved in phase shifting to dark pulses, as phase shifts were significantly smaller in the 5-HT1A KO mice. While the 5-HT1A receptor is involved, it is not strictly necessary since dark pulses were still able to elicit very large phase shifts in the 5-HT1A KO animals. One study examined the role of 5-HT in phase shifting to dark pulses [61]. While they used shorter dark pulses than we and others have used [60,62], yielding smaller phase shifts, their findings are still of interest. They observed that dark pulses activated 5-HTergic cells in the raphe, consistent with 5-HT's involvement. They then tested the role of the 5-HT1A and 5-HT7 receptors in phase shifts to dark pulses by administering selective antagonists. While the 5-HT1A antagonist diminished the size of the phase shift, it did not do so significantly. In contrast, pretreatment with a 5-HT7 receptor antagonist completely blocked the phase shift. It is important to note that phase shifts were not significantly different between the 5-HT1A and 5-HT7 antagonist-treated animals. Together with our findings, this suggests that the 5-HT1A receptor is necessary for a full-magnitude phase shift to dark pulses, although large phase shifts to dark pulses are still possible when 5-HT1A receptors are absent or blocked. It is important to be cautious when interpreting findings from animals genetically modified to lack a particular gene, as there can be changes to a variety of other systems that result. Comparing findings from KO animals with those from lesions or pharmacological manipulations is essential when developing an understanding of the role for a particular protein, such as the 5-HT1A receptor. The effect of the loss of the 5-HT1A receptor on levels or sensitivity of other 5-HT receptors is

V.M. Smith et al. / Physiology & Behavior 139 (2015) 136–144

143

Fig. 7. Representative actograms from WT (A) and 5-HT1A KO mice (B) housed in LL and then subjected to a 6 h dark pulse starting at CT5 (gray shading, plotting convention the same as in Fig. 1). (C) Average phase shifts to the dark pulse procedure for WT (white bars) and 5-HT1A KO mice (gray bars). (D) Average wheel revolutions during the dark pulse procedure (white bars for WT, gray bars for KO). (E) Scatter plot depicting the relationship between activity during the dark pulse manipulation and size of the resulting phase shift for WT (white circles) and 5-HT1A KO mice (gray circles). * = p b 0.05.

unclear. Some studies have reported that the 5-HT1B receptor is supersensitive in 5-HT1A KO mice [64], while other studies found that the 5HT1B receptor is unaffected [65,66]. While the firing rate of raphe cells is increased in the 5-HT1A KO mouse [66], this does not translate to higher 5-HT output [65,66].

5. Conclusions In conclusion, the present results demonstrate that 5-HT signaling through the 5-HT1A receptor regulates some circadian properties. The present results are consistent with the widely acknowledged role of 5HT in modulating circadian responses to light. Furthermore, the present findings support a role in regulating phase shifting to dark pulses. Differences between the 5-HT1A KO and wildtype mice are more subtle in entrained and free-running conditions (present study) than they are to discrete exposure to photic or non-photic zeitgebers [26,41]. However, circadian amplitude in LD and LL, as well as rate of re-entrainment to a shifted LD cycle and free-running period in LL are all significantly different in the 5-HT1A KO mice, highlighting a role for the 5-HTergic system and this receptor in these responses under entrained and freerunning situations.

Acknowledgments This work was supported by an NSERC Discovery Grant to MCA (311874-2010), an NSERC-CGS-D, a Killam Fellowship and an AI-TF Fellowship to VMS, and an Eyes High Postdoctoral Fellowship to PB.

References [1] Antle MC, Silver R. Orchestrating time: arrangements of the brain circadian clock. Trends Neurosci 2005;28:145–51. [2] Mistlberger RE, Antle MC, Glass JD, Miller JD. Behavioral and serotonergic regulation of circadian rhythms. Biol Rhythm Res 2000;31:240–83. [3] Rea MA, Pickard GE. Serotonergic modulation of photic entrainment in the Syrian hamster. Biol Rhythm Res 2000;31:284–314. [4] Hay-Schmidt A, Vrang N, Larsen PJ, Mikkelsen JD. Projections from the raphe nuclei to the suprachiasmatic nucleus of the rat. J Chem Neuroanat 2003;25:293–310. [5] Morin LP, Meyer-Bernstein EL. The ascending serotonergic system in the hamster: comparison with projections of the dorsal and median raphe nuclei. Neuroscience 1999;91:81–105. [6] Pickard GE. The afferent connections of the suprachiasmatic nucleus of the golden hamster with emphasis on the retinohypothalamic projection. J Comp Neurol 1982;211:65–83. [7] Yamakawa GR, Antle MC. Phenotype and function of raphe projections to the suprachiasmatic nucleus. Eur J Neurosci 2010;31:1974–83. [8] Belenky MA, Pickard GE. Subcellular distribution of 5-HT(1B) and 5-HT(7) receptors in the mouse suprachiasmatic nucleus. J Comp Neurol 2001;432:371–88. [9] Duncan MJ, Franklin KM. Expression of 5-HT7 receptor mRNA in the hamster brain: effect of aging and association with calbindin-D28K expression. Brain Res 2007; 1143:70–7. [10] Duncan MJ, Jennes L, Jefferson JB, Brownfield MS. Localization of serotonin(5A) receptors in discrete regions of the circadian timing system in the Syrian hamster. Brain Res 2000;869:178–85. [11] Duncan MJ, Short J, Wheeler DL. Comparison of the effects of aging on 5-HT7 and 5HT1A receptors in discrete regions of the circadian timing system in hamsters. Brain Res 1999;829:39–45. [12] Moyer RW, Kennaway DJ. Immunohistochemical localization of serotonin receptors in the rat suprachiasmatic nucleus. Neurosci Lett 1999;271:147–50. [13] Prosser RA, Dean RR, Edgar DM, Heller HC, Miller JD. Serotonin and the mammalian circadian system: I. In vitro phase shifts by serotonergic agonists and antagonists. J Biol Rhythms 1993;8:1–16. [14] Sprouse J, Li X, Stock J, McNeish J, Reynolds L. Circadian rhythm phenotype of 5-HT7 receptor knockout mice: 5-HT and 8-OH-DPAT-induced phase advances of SCN neuronal firing. J Biol Rhythms 2005;20:122–31.

144

V.M. Smith et al. / Physiology & Behavior 139 (2015) 136–144

[15] Meyer-Bernstein EL, Morin LP. Differential serotonergic innervation of the suprachiasmatic nucleus and the intergeniculate leaflet and its role in circadian rhythm modulation. J Neurosci 1996;16:2097–111. [16] Dudley TE, DiNardo LA, Glass JD. Endogenous regulation of serotonin release in the hamster suprachiasmatic nucleus. J Neurosci 1998;18:5045–52. [17] Grossman GH, Mistlberger RE, Antle MC, Ehlen JC, Glass JD. Sleep deprivation stimulates serotonin release in the suprachiasmatic nucleus. Neuroreport 2000;11: 1929–32. [18] Grossman GH, Farnbauch L, Glass JD. Regulation of serotonin release in the Syrian hamster intergeniculate leaflet region. Neuroreport 2004;15:103–6. [19] Bradbury MJ, Dement WC, Edgar DM. Serotonin-containing fibers in the suprachiasmatic hypothalamus attenuate light-induced phase delays in mice. Brain Res 1997; 768:125–34. [20] Morin LP, Blanchard J. Depletion of brain serotonin by 5,7-DHT modifies hamster circadian rhythm response to light. Brain Res 1991;566:173–85. [21] Muscat L, Tischler RC, Morin LP. Functional analysis of the role of the median raphe as a regulator of hamster circadian system sensitivity to light. Brain Res 2005;1044: 59–66. [22] Penev PD, Turek FW, Zee PC. Monoamine depletion alters the entrainment and the response to light of the circadian activity rhythm in hamsters. Brain Res 1993;612: 156–64. [23] Smart CM, Biello SM. WAY-100635, a specific 5-HT1A antagonist, can increase the responsiveness of the mammalian circadian pacemaker to photic stimuli. Neurosci Lett 2001;305:33–6. [24] Edgar DM, Reid MS, Dement WC. Serotonergic afferents mediate activity-dependent entrainment of the mouse circadian clock. Am J Physiol 1997;273:R265–9. [25] Marchant EG, Watson NV, Mistlberger RE. Both neuropeptide Y and serotonin are necessary for entrainment of circadian rhythms in mice by daily treadmill running schedules. J Neurosci 1997;17:7974–87. [26] Smith VM, Sterniczuk R, Phillips CI, Antle MC. Altered photic and non-photic phase shifts in 5-HT(1A) receptor knockout mice. Neuroscience 2008;157:513–23. [27] Meyer-Bernstein EL, Blanchard JH, Morin LP. The serotonergic projection from the median raphe nucleus to the suprachiasmatic nucleus modulates activity phase onset, but not other circadian rhythm parameters. Brain Res 1997;755: 112–20. [28] Smale L, Michels KM, Moore RY, Morin LP. Destruction of the hamster serotonergic system by 5,7-DHT: effects on circadian rhythm phase, entrainment and response to triazolam. Brain Res 1990;515:9–19. [29] Morin LP, Blanchard J. Serotonergic modulation of the hamster wheelrunning rhythm: response to lighting conditions and food deprivation. Brain Res 1991;566: 186–92. [30] Possidente B, Lumia AR, McEldowney S, Rapp M. Fluoxetine shortens circadian period for wheel running activity in mice. Brain Res Bull 1992;28:629–31. [31] Kawai K, Yokota N, Yamawaki S. Effect of chronic tryptophan depletion on the circadian rhythm of wheel-running activity in rats. Physiol Behav 1994;55:1005–13. [32] Pickard GE, Smith BN, Belenky M, Rea MA, Dudek FE, Sollars PJ. 5-HT1B receptormediated presynaptic inhibition of retinal input to the suprachiasmatic nucleus. J Neurosci 1999;19:4034–45. [33] Sollars PJ, Ogilvie MD, Rea MA, Pickard GE. 5-HT1B receptor knockout mice exhibit an enhanced response to constant light. J Biol Rhythms 2002;17:428–37. [34] Sollars PJ, Ogilvie MD, Simpson AM, Pickard GE. Photic entrainment is altered in the 5-HT1B receptor knockout mouse. J Biol Rhythms 2006;21:21–32. [35] Gardani M, Biello SM. The effects of photic and nonphotic stimuli in the 5-HT7 receptor knockout mouse. Neuroscience 2008;152:245–53. [36] Hsu JL, Yu L, Sullivan E, Bowman M, Mistlberger RE, Tecott LH. Enhanced food anticipatory activity associated with enhanced activation of extrahypothalamic neural pathways in serotonin2C receptor null mutant mice. PLoS One 2012;5:e11802. [37] Rocha BA, Goulding EH, O'Dell LE, Mead AN, Coufal NG, Parsons LH, et al. Enhanced locomotor, reinforcing, and neurochemical effects of cocaine in serotonin 5hydroxytryptamine 2C receptor mutant mice. J Neurosci 2002;22:10039–45. [38] Weber ET, Gannon RL, Rea MA. Local administration of serotonin agonists blocks light-induced phase advances of the circadian activity rhythm in the hamster. J Biol Rhythms 1998;13:209–18. [39] Gannon RL. Serotonergic serotonin (1A) mixed agonists/antagonists elicit largemagnitude phase shifts in hamster circadian wheel-running rhythms. Neuroscience 2003;119:567–76.

[40] Rea MA, Barrera J, Glass JD, Gannon RL. Serotonergic potentiation of photic phase shifts of the circadian activity rhythm. Neuroreport 1995;6:1417–20. [41] Smith VM, Hagel K, Antle MC. Serotonergic potentiation of photic phase shifts: examination of receptor contributions and early biochemical/molecular events. Neuroscience 2010;165:16–27. [42] Parks CL, Robinson PS, Sibille E, Shenk T, Toth M. Increased anxiety of mice lacking the serotonin1A receptor. Proc Natl Acad Sci U S A 1998;95:10734–9. [43] Pittendrigh CS, Daan S. A functional analysis of circadian pacemakers in nocturnal rodents. IV. Entrainment: pacemaker as clock. J Comp Physiol 1976;106:291–331. [44] Morin LP. Serotonin and the regulation of mammalian circadian rhythmicity. Ann Med 1999;31:12–33. [45] Rea MA, Glass JD, Colwell CS. Serotonin modulates photic responses in the hamster suprachiasmatic nuclei. J Neurosci 1994;14:3635–42. [46] Ying SW, Rusak B. Effects of serotonergic agonists on firing rates of photically responsive cells in the hamster suprachiasmatic nucleus. Brain Res 1994;651:37–46. [47] Ying SW, Zhang DX, Rusak B. Effects of serotonin agonists and melatonin on photic responses of hamster intergeniculate leaflet neurons. Brain Res 1993;628:8–16. [48] Mistlberger RE, Antle MC. Behavioral inhibition of light-induced circadian phase resetting is phase and serotonin dependent. Brain Res 1998;786:31–8. [49] Antle MC, Ogilvie MD, Pickard GE, Mistlberger RE. Response of the mouse circadian system to serotonin 1A/2/7 agonists in vivo: surprisingly little. J Biol Rhythms 2003; 18:145–58. [50] Glass JD, Selim M, Srkalovic G, Rea MA. Tryptophan loading modulates light-induced responses in the mammalian circadian system. J Biol Rhythms 1995;10:80–90. [51] Deboer T, Tobler I. Running wheel size influences circadian rhythm period and its phase shift in mice. J Comp Physiol A 2000;186:969–73. [52] Koteja P, Swallow JG, Carter PA, Garland Jr T. Different effects of intensity and duration of locomotor activity on circadian period. J Biol Rhythms 2003;18:491–501. [53] Edgar DM, Martin CE, Dement WC. Activity feedback to the mammalian circadian pacemaker: influence on observed measures of rhythm period length. J Biol Rhythms 1991;6:185–99. [54] Mrosovsky N. Further experiments on the relationship between the period of circadian rhythms and locomotor activity levels in hamsters. Physiol Behav 1999;66: 797–801. [55] Yamada N, Shimoda K, Ohi K, Takahashi S, Takahashi K. Free-access to a running wheel shortens the period of free-running rhythm in blinded rats. Physiol Behav 1988;42:87–91. [56] Yamada N, Shimoda K, Takahashi K, Takahashi S. Relationship between free-running period and motor activity in blinded rats. Brain Res Bull 1990;25:115–9. [57] Rosenwasser AM, Dwyer SM. Circadian phase shifting: relationships between photic and nonphotic phase-response curves. Physiol Behav 2001;73:175–83. [58] Mistlberger RE, Belcourt J, Antle MC. Circadian clock resetting by sleep deprivation without exercise in Syrian hamsters: dark pulses revisited. J Biol Rhythms 2002; 17:227–37. [59] Knoch ME, Siegel D, Duncan MJ, Glass JD. Serotonergic mediation of constant lightpotentiated nonphotic phase shifting of the circadian locomotor activity rhythm in Syrian hamsters. Am J Physiol Regul Integr Comp Physiol 2006;291:R180–8. [60] Boulos Z, Rusak B. Circadian phase response curve for dark pulses in the hamster. J Comp Physiol 1982;146:411–7. [61] Mendoza J, Clesse D, Pevet P, Challet E. Serotonergic potentiation of dark pulseinduced phase-shifting effects at midday in hamsters. J Neurochem 2008;106: 1404–14. [62] Reebs SG, Lavery RJ, Mrosovsky N. Running activity mediates the phase-advancing effects of dark pulses on hamster circadian rhythms. J Comp Physiol A 1989;165: 811–8. [63] Marston OJ, Williams RH, Canal MM, Samuels RE, Upton N, Piggins HD. Circadian and dark-pulse activation of orexin/hypocretin neurons. Mol Brain 2008;1:19. [64] Boutrel B, Monaca C, Hen R, Hamon M, Adrien J. Involvement of 5-HT1A receptors in homeostatic and stress-induced adaptive regulations of paradoxical sleep: studies in 5-HT1A knock-out mice. J Neurosci 2002;22:4686–92. [65] Bortolozzi A, Amargos-Bosch M, Toth M, Artigas F, Adell A. In vivo efflux of serotonin in the dorsal raphe nucleus of 5-HT1A receptor knockout mice. J Neurochem 2004; 88:1373–9. [66] Richer M, Hen R, Blier P. Modification of serotonin neuron properties in mice lacking 5-HT1A receptors. Eur J Pharmacol 2002;435:195–203.