Abstract
Sleep depriving mice affects clock gene expression, suggesting that these genes partake in sleep homeostasis. The mechanisms linking wakefulness to clock gene expression are, however, not well understood. We propose CIRBP because its rhythmic expression is i) sleep-wake driven and ii) necessary for high-amplitude clock gene expression in vitro. We therefore expect Cirbp knock-out (KO) mice to exhibit attenuated sleep-deprivation (SD) induced changes in clock gene expression, and consequently to differ in their sleep homeostatic regulation. Lack of CIRBP indeed blunted the SD-incurred changes in cortical expression of the clock gene Rev-erbα whereas it amplified the changes in Per2 and Clock. Concerning sleep homeostasis, KO mice accrued only half the extra REM sleep wild-type (WT) littermates obtained during recovery. Unexpectedly, KO mice were more active during lights-off which was accompanied by an acceleration of theta oscillations. Thus, CIRBP adjusts cortical clock gene expression after SD and expedites REM sleep recovery.
Introduction
The sleep-wake distribution is coordinated by the interaction of a circadian and a sleep homeostatic process (Daan et al., 1984). The molecular basis of the circadian process consists of clock genes that interact through transcriptional/translational feedback loops. CLOCK/NPAS2:BMAL1 heterodimers drive the transcription of many target genes, among which Period (Per1-2), Cryptochome (Cry1, −2), and Rev-Erb (Nr1d1, −2). Subsequently, PER:CRY complexes inhibit CLOCK/NPAS2:BMAL1 transcriptional activity and thus prevent their own transcription. Through another loop, other clock components such as the transcriptional repressor REV-ERBα regulate the transcription of Bmal1 (Arntl), ensuring together with other transcriptional feedback loops a period of ca. 24 hours (Lowrey and Takahashi, 2011).
The sleep homeostatic process keeps track of time spent awake and time spent asleep, during which sleep pressure is increasing and decreasing, respectively. The mechanisms underlying this process are to date unknown. However, accumulating evidence implicates clock genes in sleep homeostasis [reviewed in (Franken, 2013)]. This is supported by studies in several species (i.e. mice, fruit flies and humans), showing that mutations in circadian clock genes are associated with an altered sleep homeostatic response to sleep deprivation (SD) [e.g. (Mang et al., 2016, Shaw et al., 2002, Viola et al., 2007, Wisor et al., 2002)]. Furthermore, SD affects the expression of clock genes such as Rev-erbα, Per1-3 and Dbp (Mongrain et al., 2010), but the mechanisms through which this occurs are unclear.
In this study, we examined one such mechanism and hypothesized that some of the SD-induced changes in clock gene expression occur through Cold-Inducible RNA Binding Protein (CIRBP). Decreasing temperature in vitro increases CIRBP levels (Nishiyama et al., 1997) and the daily changes in body temperature of the mouse are sufficient to drive robust cyclic levels of Cirbp and CIRBP (Morf et al., 2012) in anti-phase with temperature. Although the daily changes in cortical temperature (Tcx) appear circadian, more than 80% of its variance is explained by the sleep-wake distribution in the rat (Franken et al., 1992). Hence, the daily rhythms of cortical Cirbp become strongly attenuated when controlling for these sleep-wake driven changes in Tcx by SDs (see Figure 1, based on Gene Expression Omnibus number GSE9442 from Maret et al., 2007). Furthermore, Cirbp is the top down-regulated gene after SD (Mongrain et al., 2010, Wang et al., 2010) underscoring again its sleep-wake dependent expression. But how does CIRBP relate to clock gene expression?
Two independent studies showed that the temperature-driven changes in CIRBP are required for high amplitude clock gene expression in temperature synchronized cells (Morf et al., 2012, Liu et al., 2013). Therefore, we and others (Archer et al., 2014) hypothesized that changes in clock gene expression during SD are a consequence of the sleep-wake driven changes in CIRBP. We used mice lacking CIRBP (Cirbp KO) (Masuda et al., 2012) to test this hypothesis. We first assessed whether also in the mouse the daily changes in Tcx are driven by the sleep-wake distribution and what the contribution of locomotor activity (LMA) to these changes was. Next, we assessed SD-induced changes in clock gene expression in WT and KO mice. Because we expected that the response to SD in terms of clock gene expression differed in KO mice, and clock genes partake in sleep homeostasis (Franken, 2013), we also assessed the homeostatic regulation of sleep in Cirbp KO and WT mice.
Our experiments revealed that also in the mouse the sleep-wake distribution is the major determinant of changes in Tcx, with a significant albeit small contribution of LMA. In line with our predictions, we found that lack of CIRBP indeed attenuated the SD-induced changes in the cortical expression of Rev-erbα and the homeostatic response in REM sleep time. However, in contrast to our hypothesis, we observed that the changes in Per2 and Clock expression after SD were augmented in Cirbp KO mice. Unexpectedly, we discovered that Cirbp KO mice were substantially more active compared to their WT littermates without increasing their time spent awake. This dark-phase increase in LMA was accompanied by an acceleration of EEG theta oscillations during active waking. Altogether, our data show that Cirbp contributes to some of the SD-induced changes in clock gene expression, but also points to the existence of other sleep-wake driven pathways conveying sleep-wake state to clock gene expression.
Results
THE RELATION BETWEEN CORTICAL TEMPERATURE (TCX), SLEEP-WAKE DISTRIBUTION, AND LOCOMOTOR ACTIVITY (LMA)
The dependence of brain or cortical temperature on sleep-wake state has been demonstrated in a number of mammals (Alfoldi et al., 1990, Baker and Hayward, 1968, Deboer et al., 1994, Franken et al., 1992, Hayward and Baker, 1968) but has not been specifically addressed in the mouse. Moreover, no study so far specifically controlled for LMA when quantifying the contribution of sleep-wake state to brain temperature. We therefore measured cortical temperature (Tcx), LMA and sleep-wake state in WT and Cirbp KO mice during two baseline days, a 6hr SD and the following two recovery days. Because the relationship between Tcx, LMA, and waking in WT and KO mice was alike, we illustrated the results in WT mice only.
Fast changes in Tcx occur at sleep-wake state transitions
A representative example of a 96h recording of LMA, sleep-wake state and Tcx is depicted in Figure 2. Consistent with mice being nocturnal animals, the mouse shows more waking and LMA, and overall higher Tcx levels during the dark phase. SD on the third recording day (between Zeitgeber Time (ZT)0-6) led to an almost uninterrupted period of 6hr waking, during which LMA and Tcx reached values comparable to bouts of spontaneous wakefulness under undisturbed baseline conditions (i.e., ZT12-18). A closer inspection of the rapid changes in Tcx suggests that sleep-wake state transitions underlie these fluctuations. We further quantified these sleep-wake evoked changes in Tcx by selecting and aligning transitions between consolidated bouts of NREM and REM sleep and wakefulness during the two baseline days (Figure 2-B). When entering NREM sleep, Tcx consistently decreased, whereas at a transition into wake and REM sleep, Tcx increased. The latter transition was characterized by a fast and consistent change in Tcx; within 1.5 minutes, Tcx increased by 0.4°C. The subsequent transition from REM sleep into wake leads to an initial decrease in Tcx and contrasts with the waking-evoked increase in Tcx when transitioning from NREM sleep to wake. Altogether, these results provide evidence that sleep-wake state importantly contributes to changes in Tcx. The sleep-wake state evoked changes in Tcx did not differ between genotypes (2-way RM ANOVA, factors genotype (GT) and Time; GT: p>0.13, GTxTime: p>0.09).
Daily cycles in Tcx are determined by sleep-wake state
After having established that the rapid changes in Tcx are indeed evoked by changes in sleep-wake state, we next wondered if the large daily change in Tcx is also due to the daily rhythms in sleep-wake state and LMA, and therefore inspected these variables per hour. The LMA data were log2 transformed to allow for parametric assessment. Tcx, waking and LMA oscillated over the course of the 24h baseline (BL) in similar fashion in both genotypes (2-way RM ANOVA on BL1 and −2 averaged 1h intervals, Factor Time: F(23,207)=70.5; 27.2; 22.5; p<0.0001, respectively, Figure 3-A, Factor GT × Time (1,24); Tcx: F(23,207)=0.9, p=0.63; waking: 1.7, p=0.03, LMA: 1.21, p=0.24) and the amplitude of the BL change in Tcx did not differ between genotypes (WT: 2.34±0.1, KO: 2.33±0.1; t-test: t(9)=-0.02, p=0.98; average of the difference between the highest and lowest value hourly value in BL1 and −2). Importantly, the time course of waking and LMA both closely resembled that of Tcx. This observation was supported by the strong correlation between Tcx and waking (Figure 3-B left: WT: R2=0.76; KO: R2=0.81, p<0.0001) and between Tcx and LMA (Figure 3-B right: WT: R2=0.60; KO: R2=0.72, p<0.0001).
To assess the influence of waking on Tcx at a time of day when Tcx is normally low and mice spend most of their time asleep, mice were sleep deprived between ZT0 and ZT6. Mice were 98% of the 6hr SD awake, and thus more awake and active compared to the same time under BL conditions (paired t-test: waking [hours] BL: 2.2±0.1, SD: 5.9±0.04; t(10)=-38.1, p<0.0001; log2 [movements], BL: 13.1±2.4, SD: 39.4±2.1; t(10)=-15.2, p<0.0001). These changes led to sustained elevated Tcx (average ZT0-ZT6 [°C]: BL: 34.7±0.07, SD: 36.6±0.06, t(10)=-44.3, p<0.0001), suggesting that wakefulness and/or LMA drives changes in Tcx. Genotype did not contribute to or interact with these changes (2-way ANOVA, GT*SD/BL: p>0.39)
However, factors accompanying the SD other than extended waking, such as stress, could have contributed to the SD-induced changes in Tcx. To address this issue, we selected within each mouse the longest uninterrupted spontaneous waking bout occurring during BL (average bout length: 100±19 minutes). We then compared Tcx during the last 10 minutes of this spontaneous waking bout (to reduce any effects of differences in Tcx at bout-onset) with Tcx reached in the last 10 minutes of an equivalent time spent awake from the start of the SD on. Tcx reached during the SD and spontaneous wakefulness did not differ (Figure 3-C), also not in KO mice (t(5)=0.84, p=0.44), indicating that factors other than extended wakefulness (e.g. light exposure, circadian time, SD-associated stress) do not importantly contribute to the changes in Tcx during the SD.
Considering the strong correlation between LMA and Tcx (WT: R2=0.72; KO: R2=0.78; p<0.0001), it could be hypothesized that LMA explains partly the sleep-wake associated changes in Tcx. To investigate this further, the respective contribution of waking and LMA to changes in Tcx was quantified by a partial correlation analysis. Although LMA did significantly contribute, substantially more of the variance in Tcx was explained by waking in both genotypes (paired t-test on Fisher Z-transformed R2-values from each individual mouse’s partial correlation on hourly waking and Tcx, and on hourly LMA and Tcx: WT: t(5)=5.1, p=0.004; KO: t(5)=10.7, p=0.0001; see also Figure 3-D for R2-partial correlation coefficients which are based on hourly data from all WT mice combined). We then determined the variance that could not be explained by the correlation between waking to Tcx (i.e. the residuals) by calculating the difference between the observed Tcx in a given hour and the predicted Tcx based on the time-spent-awake in that hour. The linear regression overestimated and underestimated Tcx during the light phase and dark phase, respectively (Figure 3-E,F; BL1 and BL2), leading to negative residuals during the light phase and positive residuals during the dark phase. Fitting a sinewave through the residuals of the two baseline days revealed a ‘circadian’ distribution, with a mean amplitude of 0.29°C, which is almost twice the amplitude as was previously reported on in the rat [0.15°C] (Franken et al., 1992). Interestingly, when considering the time course of the residuals throughout the experiment, including the SD and recovery, a consistent parallel with the distribution of LMA expressed per unit of waking became evident (Figure 3-G).
Therefore, to determine if including LMA in addition to waking, could predict a larger portion of the variance in Tcx, we applied three Mixed Linear Models, where LMA was considered by expressing LMA per unit of waking (LMA/Waking). Model1 explained the variance in Tcx based on waking alone, Model2 also incorporated LMA/Waking, and Model3 considered additionally the interaction between Waking and LMA/Waking. Indeed, Model3 predicted best the variance in Tcx although in terms of explaining the variance in Tcx, the improvement is marginal over the two other models (Model1: R2c =0.84; Model2: R2c =0.85; Model3: R2c =0.86; chi-squared test: Model1 vs Model2: X2(5)=16.2; p<0.0001; Model2 vs Model3: X2(6)=25.0; p<0.0001). Thus, the sleep-wake distribution is the most important determinant of Tcx but LMA during waking is modestly contributing as well. Nevertheless, also the residuals of this model, depicted in Figure 3, supplement 1, still showed a similar pattern like the residuals in Figure 3-F, pointing towards the contribution of other (circadian) variables and/or a non-linearity of the association between the contribution of LMA and sleep-wake states to changes in Tcx.
THE INFLUENCE OF SD AND CIRBP ON TRANSCRIPTS IN CORTEX AND LIVER
After establishing that also in the mouse the sleep-wake distribution is the major determinant of Tcx, we assessed whether the SD-incurred changes in CIRBP participate in linking the effect of SD to clock gene expression. To this end, we quantified 11 transcripts from liver and 15 from cortex before and after SD by RT-qPCR. Genes of interest included transcripts affected by SD (Maret et al., 2007, Mongrain et al., 2010) and/or by the presence of CIRBP (Liu et al., 2013, Morf et al., 2012), with an emphasis on clock genes. Mice were sacrificed before SD at ZT0, or 6 hours later after SD (ZT6-SD) together with non-sleep deprived control mice that could sleep ad lib (ZT6-NSD). Statistics on ZT0 (t-test) and ZT6 (2-way ANOVA) can be found in Table 1.
From ZT0 to ZT6 under undisturbed conditions, Tcx decreased because mice spend more time asleep compared to the previous hours in the dark phase (see also Figure 3-A). This decrease in Tcx was accompanied by the expected increase of the expression of the cold-induced transcript Cirbp in WT mice (cortex: t(8)=3.2, p=0.01; liver: t(8)=2.7, p=0.03; Figure 4-A and Figure 4, supplement 1, compare also with the time course of Cirbp Figure 1). In contrast, SD during the same time span incurred a decrease in cortical and hepatic Cirbp relative to non-sleep deprived controls (cortex: Figure 4-A; liver: Figure 4, supplement 1), consistent with the wake-induced increase in Tcx during SD. No Cirbp mRNA was detected in KO mice.
RBM3 is another cold-inducible RNA Binding Protein and, like CIRBP, conveys temperature cycles into high-amplitude clock gene expression in vitro (Liu et al., 2013). A long and a short isoform of Rbm3 (Rbm3-long and –short, resp.) that differ in their 3’UTR length were discovered in the mouse cortex. Although both isoforms are referred to as ‘cold-induced’, they exhibit opposite responses to SD (Wang et al., 2010), with a decrease in the short isoform and an increase in the long isoform. We found that overall, the short isoform was more common than the long isoform in the cortex (PCR cycle detection number for all samples pooled: cortex: Rbm3-short: 25.6±0.2, Rbm3-long: 29.7±0.1, amplification efficiency Rbm3-short: 2.11 and Rbm3-long 2.07). In the liver, only the short isoform was detected (liver: Rbm3-short: 28.2±0.2, Rbm3-long: >32; i.e., beyond reliable detection limit). We confirmed that after SD, Rbm3-short was decreased in the cortex (Figure 4-A) and liver (Figure 4, supplement 1), whereas Rbm3-long was increased in cortex. The latter observation reached significance only in the KO mice (Figure 4-A).
As anticipated, cortical expression of the activity (and waking)-induced transcripts Homer1a, Dusp4, Hspa5/BiP, Hsp90b, and Hsf1 was increased by SD (Figure 4, supplement 2). Post-hoc tests revealed that the latter two were significantly increased only in Cirbp KO mice. Furthermore, the effect of SD on the transcripts Hsp90b and Hspa5 was significantly amplified in Cirbp KO mice compared to WT mice. Unexpectedly, no changes in the expression of heat shock transcripts incurred by SD or genotype were detected in the liver (Figure 4, supplement 1).
In vitro studies have shown that the presence of CIRBP is associated with longer 3’UTRs of its target genes, such as the transcript splice-factor proline Q (Sfpq), resulting in a higher prevalence of long isoforms (extended or ext) over all isoforms (common or com), and thus an increased ext/com ratio [see FigS4-S5 in (Liu et al., 2013)]. We therefore expected a lower ext/com ratio in mice lacking CIRBP. However, under baseline conditions [ZT0 and ZT6-NSD], Cirbp KO mice did not differ in their ext/com ratio from WT littermates (ZT0: liver: (t(8)=1.55, p=0.16; cortex: t(7)=2.0, p=0.09; ZT6-NSD: liver: t(8)=0.19, p=0.85, cortex: t(8)=1.4, p=0.20). Because also RBM3 partakes in determining the ext/com ratio (Liu et al., 2013), the lack of an effect of CIRBP on the ext/com ratio could be due to compensation by RBM3. We could test this by assessing the effect of SD on the ext/com ratio, because SD acutely suppresses both RBM3 and CIRBP. Indeed, SD significantly decreased the ext/com ratio in the liver in both genotypes (Figure 4-B; 2-way ANOVA, factor SD: F(1,16)=20.4, p=0.003). In the cortex, however, we observed an unexpected non-significant increase in WT mice and a significant decrease in KO mice, leading to a significant GT x SD interaction (cortex: F(1,16)=5.25, p=0.036). Therefore, our data are inconclusive in confirming a role for CIRBP, and possibly RBM3, in the in vivo determination of Sfpq’s ext/com ratio.
Our main question concerned the contribution of CIRBP to sleep-wake induced changes in clock gene expression. Previous studies evaluating the effects of SD on cortical clock transcripts showed a consistent increase in Per2 and a decrease in Dbp and Rev-erbα whereas the response of Clock and Npas2 varied among studies, but if any, tended to increase after SD (reviewed in (Mang and Franken, 2015)). Indeed, in the cortex of WT mice, SD increased cortical Per2, decreased Dbp and Rev-erbα and did not significantly affect Clock and Npas2 (Figure 4-C). In accordance with our hypothesis, CIRBP attenuated the SD induced changes of cortical Rev-erbα, a transcriptional repressor recently implicated in the sleep homeostat (Mang et al., 2016). This observation contrasts with the genotype-dependent changes in Per2, because when considering the lower levels of cortical Per2 in Cirbp KO mice at ZT0, the effect of SD was amplified (Figure 4-C, 2-way ANOVA, ZT0-ZT6[SD], interaction effect GT × SD: F(1,16)=12.4, p=0.003). Also, the expression of Clock in the cortex was significantly increased by SD in Cirbp KO mice and not in WT littermates.
Compared to the cortex, the clock gene expression in the liver appeared more resilient to the effects of SD, as only Dbp and Rev-erbα were significantly affected and not Per2 (Figure 4, supplement 1). The lack of CIRBP did not interfere with this response, nor did it contribute to genotype dependent changes of other (clock) gene transcripts in the liver.
Taken together, the absence of CIRBP modulated the SD induced changes in the cortical expression of the clock genes Rev-erbα, Clock and Per2. Furthermore, the expression of transcripts in the heat shock pathway were also affected in a genotypic manner by SD.
CIRBP CONTRIBUTES TO SLEEP HOMEOSTASIS
Because Cirbp KO mice showed a modulated response to SD in three out of the five cortical clock gene transcripts we quantified, and clock genes importantly partake in the sleep homeostatic process (reviewed in (Franken, 2013)), we hypothesized that Cirbp KO mice have a blunted sleep-homeostatic process. We quantified EEG power in the delta band [0.75 – 4.0 Hz] during NREM sleep, which is a proxy of NREM sleep pressure and reflects a homeostatically regulated sleep process, Process S (Daan et al., 1984). As additional sleep homeostatic measures, we calculated the amount of NREM and REM sleep recovered after SD relative to baseline sleep.
Baseline characteristics of sleep-wake behavior do not differ between Cirbp KO and WT mice
During the two baseline days, no significant differences in waking, NREM or REM sleep were observed. This was neither the case for time spent in these three behavioral states per light and dark phase (Table 2), nor for the distribution of sleep and waking across the day (see Figure 5-A and Figure 6-A). Noteworthy is that under constant darkness we did not detect a change in circadian period length (period (hours): WT (n=5): 23.8±0.03 and KO (n=7): 23.8±0.01).
Sleep homeostatic processes under baseline and recovery
The time course of delta power in the two genotypes was overall similar. In the dark phase, when mice spent most of their time awake and thus sleep pressure accumulates, delta power during NREM sleep was highest. This contrasts with the end of the light phase [ZT8-12], where NREM sleep delta power reached its lowest levels of the day due to the high and sustained prevalence of NREM sleep in the preceding hours. Despite the overall similarities in daily changes of NREM delta power, subtle differences were observed: delta power levels were higher during the dark phase in Cirbp KO compared to WT mice, and these differences reached significance during the dark periods of recovery (Figure 5-A, 2nd graph from top).
Differences in delta power can be attributed to changes in the dynamics of the underlying homeostatic process, Process S, and/or to changes in the sleep-wake distribution. Evidence supporting the latter possibility was observed because Cirbp KO mice tended to spend less time in NREM sleep (and more time awake) during the early dark phase compared to WT mice, reaching significance during the recovery (Figure 5-A; 3rd graph from top). To test if these changes in the sleep-wake distribution were indeed sufficient to raise NREM delta power above WT levels, we estimated the increase (τi) and decrease (τd) rate of delta power by a simulation of Process S based on the sleep-wake distribution. We assumed Process S to increase exponentially during waking and REM sleep by time constant τi and to decrease during NREM sleep by time constant τd (see Materials and Methods, and (Franken et al., 2001) for more details). This simulation not only captured well the overall dynamics (mean square of the measured-predicted differences, mean±SEM: WT: 10.1±0.3, KO: 10.4±0.4) but also the genotype differences in delta power (Figure 5-A; top graph). No differences in the time constants of Process S were detected (see Table 3). Hence, the reduction in NREM sleep in Cirbp KO mice in the beginning of the dark period caused the higher NREM EEG delta power values in subsequent hours, underscoring the notion that small differences in NREM sleep time can have large repercussions on delta power when waking prevails and thus Process S increases (Franken et al., 2001).
A different aspect of NREM sleep homeostasis concerns the regulation of time spent in this state. This can be quantified by accumulating relative differences in time spent in NREM sleep from corresponding baseline hours over the recovery period. At the end of the first recovery day, both KO and WT mice had gained ca. 40 minutes of NREM sleep relative to baseline (Figure 5-B, upper panel).
The amount of REM sleep is also homeostatically defended (Franken, 2002). At the end of REC1, both WT and KO mice spent more time in REM sleep compared to corresponding baseline hours. However, this increase in REM sleep was significantly attenuated by 46% in Cirbp KO mice (Figure 5-B, lower panel). Because no significant differences were detected during baseline in time spent in REM sleep (see also Table 1), this attenuated rebound in REM sleep resulted from less REM sleep during recovery, specifically in the first hours of the dark phase when the genotypic differences were most prominent (Figure 5-A, lowest graph).
Thus, although CIRBP did not affect the processes underlying NREM sleep intensity and NREM sleep time, it did contribute to REM sleep homeostasis by increasing the amount of REM sleep after SD.
AN UNANTICIPATED WAKING PHENOTYPE IN CIRBP KO MICE
While quantifying sleep-wake states, we observed that Cirbp KO mice were more active than their WT littermates during the dark phase (t(31)=-2.56, p=0.015, see also Table 2). More specifically, Cirbp KO mice were almost twice as active in the first 6hrs of the dark phase (movements: WT: 463.8±60.7, KO: 801.8±118.4, t(35)=-2.7, p=0.012; Figure 6-A). Interestingly, this pronounced increase was not associated with a significant increase in time spent awake during BL (per 12 hrs: t(35)=1.2, p=0.24, and see Table 2), and indeed Cirbp KO mice were more active per unit of waking (average in the dark phase, LMA [movements/waking(min)], WT: 1.3±0.13, KO: 2.1±0.28; t(35)=-2.7, p=0.01). Note that also Tcx was not significantly increased in Cirbp KO mice during the dark phase (Tcx: WT: 35.9±0.1, KO: 36.1±0.1, t-test, t(10)=1.3, p=0.24) despite the increased LMA at this time of the day, again underscoring the minimal contribution of LMA to Tcx.
Because Cirbp KO mice are not more awake (Table 2 and Figure 6), we wondered if their increased LMA is associated to the prevalence of sub-states of waking. Theta-dominated waking (TDW) is a sub-state of waking that correlates with activity, prevails during the dark phase and SD, and is characterized by the presence of EEG theta-activity (Buzsáki, 2006, Vassalli and Franken, 2017). Despite their increased LMA, Cirbp KO mice did not spend more time in TDW during the dark phase of the BL (see Table 1, t(31)=-1.22, p=0.23). If not time spent in TDW, does the increased LMA in Cirbp KO mice relate to changes in brain activity during dark phase TDW?
Although in both genotypes the TDW EEG showed the characteristic theta activity [6.5-12.0 Hz], subtle differences between genotypes were detected in the spectral composition of the EEG signal. Slow [32-45Hz] and fast [55-80Hz] gamma power were both reduced during TDW in Cirbp KO mice (Figure 6-B), and this reduction was observed throughout the experiment (Figure 6, supplement 1; and see time course in Figure 6, supplement 2), indicating that these spectral genotype differences are robust across different light conditions, circadian times and throughout the SD.
In contrast, the spectral composition of the EEG during ‘quiet’ waking (i.e. all waking that is not TDW) was remarkable similar between the two genotypes (Figure 6, figure supplement 1), demonstrating that the changes in spectral composition of TDW EEG are not the result of a general effect of CIRBP on the waking EEG.
Moreover, we observed a decrease in slow and a non-significant increase in faster theta activity in the TDW EEG of Cirbp KO mice, together hinting at an acceleration of theta peak frequency (TPF; lower panel in Figure 6-B). TPF during TDW in BL was indeed increased in KO mice (+0.15Hz) although our significance threshold was not met (t(35)=2.0, p=0.05o6). No suggestions for accelerated TPF in REM sleep during BL, the other sleep-wake state characterized by distinct theta oscillations in the EEG, were detected (WT: 7.43±0.06; KO: 7.56±0.05, t(35)=1.7, p=0.10). During locomotion, increased LMA correlates well with increased TPF (Jeewajee et al., 2008). In accordance with this observation, mean log2-transformed LMA levels per mouse during the dark phase predicted well the mean TPF observed during TDW at the same time of day (WT and KO combined; R2=0.52, p<0.0001), although this relationship remained significant only in KO mice when assessing the two genotypes separately (Figure 6-C). This genotype-dependent association between TPF and LMA was not confirmed by a significant difference in slope between the genotypes (ANCOVA, F(1,29)=3.8, p=0.059).
Because the group correlation did not account for inter-individual differences in LMA levels, we also assessed the correlations between TPF and LMA within individual mice. Moreover, to test if this association depended on the lighting condition, we analyzed this during the dark and light phase separately (i.e. 24 values per mouse per lighting condition; see also Table 4). In the dark phases, this correlation was significant in all but one mouse (a KO), and both the slope and the predictive power of this correlation did not significantly differ between genotypes (slope: WT: 0.15±0.01, KO: 0.14±0.01, t(31)=0.37, p=0.72; R2: WT: 0.81±0.03; KO: 0.79±0.04, t-test on the Fisher Z-transformed R2-values: t(31)=-0.49, p=0.62). In the light phases, this association was weaker (dark vs. light: paired t-test: slope: t(32)=7.8, p<0.0001; Fisher Z-transformed R2-values: t(32)=5.9, p<0.0001), but again did not differ between genotypes (WT: 0.07±0.01, KO: 0.09±0.01, t(31)=1.2, p=0.23; R2: WT: 0.59±0.04; KO: 0.67±0.05, t-test on the Fisher Z-transformed R2-values: t(31)=1.2, p=0.23). However, during the light phases, when LMA and TDW are substantially reduced and estimates of TPF during TDW are less precise, we found more non-significant associations in both genotypes (KO: 3/ 16; WT: 3/ 17 mice). Altogether, these results provide further evidence that LMA contributes to TPF and suggests that CIRBP, through its effects on LMA, reduces TPF.
The SD altered the distribution of waking during the recovery relative to BL (3-way RM ANOVA, factor Time, GT and BL/REC, factor BL/REC: REC1: F(1,558)=42.7, p<0.0001; REC2: F(1,1514)=441.8, p<0.0001; see triangles in REC1 and REC2, Figure 6-A). Surprisingly, while time spent awake was overall decreased compared to baseline, we observed several intervals during the recovery in which TDW was increased (3-way RM ANOVA, factor Time, GT and BL/REC, factor BL/REC: REC1: F(1,558)=13.9, p=0.0002; REC2: F(1,1514)=233.8, p<0.0001; Figure 6-A, upwards pointing triangles). This was true for both genotypes. Moreover, genotype differences in the distribution of waking and TDW became significant during the dark phases of both recovery days, with Cirbp KO mice spending more time awake and in TDW than WT mice (Figure 6-A; see post-hoc tests indicated by red line), as if SD amplified the non-significant genotype differences during BL (3-way RM ANOVA on hourly values: factor GTxTimexSD: total waking: F(41,1271)=1.4, p=0.04; TDW: F(41,1271)=1.4, p=0.056) but not for LMA (F(41,1271)=1.0, p=0.48).
The EEG spectra during TDW in REC1 and REC2 showed similar profiles as during BL (see Figure 6, figure supplement 2), although there were some changes that in recovery reached significance such as the increase in the delta power band. Along those lines, the non-significant increase in TPF in Cirbp KO mice during the BL dark phases became significant in the REC dark phases (REC1: WT: 8.1±0.05, KO: 8.4±0.07, t(35)=2.7, p=0.01; REC2: WT: 8.2±0.05, KO: 8.5±0.08, t(35)=2.6, p=0.01). Also, the non-significant difference in slope at the group level between TPF and LMA during BL (Figure 6-C), became significant after SD (ANCOVA, F(1,29)=5.8, p=0.02), providing further evidence that the suggestive genotype differences under baseline conditions become more pronounced after a challenge of the sleep homeostat.
Taken together, Cirbp KO mice were more active during the dark phase, which partly explains the faster TPF. Moreover, KO mice had less EEG power in the gamma band of TDW, and the 6-hour SD strengthened genotype differences in the sleep-wake distribution and EEG activity.
Discussion
In this study, we showed that, like in other rodents, the sleep-wake distribution is the major determinant of Tcx in the mouse. Because of the well-established link between temperature and CIRBP levels, it is likely that the equally well-known sleep-wake driven changes in Cirbp expression in the brain are conveyed through the sleep-wake driven changes in brain temperature. As predicted, the SD-incurred changes in the expression of clock genes was modulated by the presence of CIRBP. However, only for Rev-Erbα did we observe the anticipated attenuated response to SD in Cirbp KO mice, whereas the changes in the expression of Per2 and Clock were amplified compared to WT mice. Moreover, we did discover evidence of altered dynamics of the process regulating time spent in REM sleep. Unexpectedly, Cirbp KO mice are more active during the dark phase, and have during TDW reduced power in the gamma band and increased TPF.
CHANGES IN CORTICAL TEMPERATURE ARE SLEEP-WAKE DRIVEN
When sleep and waking occur at their appropriate circadian times, the changes in both brain and body temperature have a clear 24-hour rhythm and therefore appear as being controlled directly by the circadian clock. However, sleep-wake cycles contribute significantly to both the daily changes in brain and body temperature. In humans, this involvement is powerfully illustrated by spontaneous desynchrony, where body temperature follows both a circadian and an activity-rest (and presumably, sleep-wake) dependent rhythm (Wever, 1979). The contribution of sleep-wake state to the daily dynamics in body temperature is further supported by forced desynchrony studies, such as (Dijk and Czeisler, 1995), estimating that ‘masking’ effects of rest-activity and sleep-wake cycles contributed between 30% and 50% to the amplitude of the circadian body temperature rhythm (Hiddinga et al., 1997, Dijk et al., 2000). Not only in humans but also in smaller animals like rats, a circadian and rest-activity component contribute to the circadian fluctuations in body temperature (Cambras et al., 2007). Thus, the circadian amplitude of body temperature is amplified when wake and sleep occur at the appropriate phase of the circadian rhythm.
In contrast to body temperature, brain temperature in rodents is much more determined by sleep-wake state: 80% of its variance can be explained by the sleep-wake distribution ((Franken et al., 1992) and this study). Likewise, the sleep-wake driven changes in brain temperature are still present in arrhythmic animals (Edgar et al., 1993, Baker et al., 2005), pointing to a more important sleep-wake dependency of brain temperature compared to body temperature. In our study, we also estimated the contribution of LMA to changes in Tcx and found that waking with higher LMA is associated with higher Tcx. Although significant, the contribution of LMA to the daily changes in Tcx was modest and explained only 2% more of the variance compared to waking alone. Can we optimize the prediction of Tcx? A non-linear relationship between sleep-wake state and Tcx was assumed previously (Franken et al., 1992) and could have improved the prediction of our model further. This is supported by the residuals from the complete model (see Figure 3, supplement 2), that exhibit under baseline conditions a circadian distribution, whereas during the SD, they remain increased as during the dark phase. Thus, the model overestimates Tcx during periods with little waking (light phase) and underestimates Tcx during periods that are dominated by waking (dark phase and SD), suggesting a non-linear relationship between these two variables.
Important to consider is that the influence of LMA on Tcx is likely affected by the type of activity; for example, rats on a running wheel activity can increase their brain temperature by 2°C within 30 minutes (Fuller et al., 1998). Also exercise in men leads to an increase in (proxies) of brain temperature (Nybo et al., 2002). Thus, although in our study the effect of LMA to Tcx was very modest compared to the effect of waking on Tcx, these contributions likely differ with various types of physical activity.
LMA-DEPENDENT AND INDEPENDENT CHANGES IN WAKING CHARACTERISTICS
Little is known about the role of CIRBP in neuronal and behavioral functioning. It was therefore unanticipated that Cirbp KO mice were more active during the dark phase. Neither were the changes in neuronal oscillations during TDW: a reduction in low- and high gamma power and an increase in TPF. Because increased running speed correlates with increased hippocampal TPF (Jeewajee et al., 2008), and our measured TPF is mainly of hippocampal origin (Buzsáki, 2006), we indeed can relate the increased TPF to the increase in LMA in KO mice. In contrast to TPF, the literature has not consistently reported on a relation between the general decrease in gamma power during active waking and its relation to LMA. Some studies have found that increased speed of movement relates to increased power in the gamma band (Furth et al., 2017, Niell and Stryker, 2010, Vinck et al., 2015), whereas others found that this association is only present in higher gamma frequencies [>60Hz] (Zheng et al., 2015). Thus, it is unclear if LMA relates to changes in gamma power. However, there is a clear increase in power of the high gamma band specifically during the SD (see Figure 6, supplement 2), as noted previously (Vassalli and Franken, 2017). This increase was present in both genotypes suggesting that while KO mice seem to have a reduced capacity to produce fast gamma activity, SD is still able to activate their fast-gamma circuitry. These results, together with the observation that during the light phase the decreased power in the gamma bands was still present at a time of day when LMA did not significantly differ, argue against an association between the decreased power in the gamma bands of Cirbp KO mice and their increased LMA.
Interestingly, gamma oscillations are associated with a palette of cognitive processes [reviewed in (Bosman et al., 2014)]. This is further supported by associations between behavioral impairments and changes in gamma power. For example, mice with abnormal interneurons are impaired at the behavioral level (e.g. lack of cognitive flexibility) and have a reduction in task-evoked gamma power in their EEG. Pharmacological stimulation of inhibitory GABA-neurons augmented power in the gamma band and rescued the behavioral phenotype of the mutants (Cho et al., 2015).
In the hippocampus, gamma-theta coupling, i.e. the occurrence of gamma oscillations at a specific phase of the theta oscillation, has been suggested to aid processes underlying memory [for review see (Colgin, 2015)]. Because CIRBP slows down TPF and increases power in the gamma bands, further analyses and experiments can address if Cirbp KO mice have altered phase coherence between these two frequency bands. Together with the postulated function of gamma power in cognitive flexibility, it would be interesting to assess if the spectral phenotype in Cirbp KO mice is associated with behavioral abnormalities.
Several aspects of waking that appeared to differ between Cirbp KO and WT mice under baseline dark conditions but were non-significant, reached significance during the recovery dark phase. For example, during baseline Cirbp KO mice were 4% more awake and 13% more in TDW compared to their WT littermates, which was amplified to 8% and 20%, respectively, during recovery. Also, TPF and the genotype-dependent association between overall TPF and LMA reached significance during the recovery. This suggest that SD amplified the genotypic differences. Other sleep deprivation studies found evidence for similar phenomena, where sleep disturbance can amplify molecular and behavioral phenotypes of Alzheimers’ mouse models (for review, see (Musiek and Holtzman, 2016)) and sensitivity to pain (Sutton and Opp, 2014). Our data indicates that a similar phenomenon occurs in Cirbp KO mice, where a single 6-hr SD reveals the suggestive baseline genotypic differences. It would be interesting to understand the dynamics of this change; e.g. if they are reversible or if a second SD could augment genotypic differences further.
CIRBP ADJUSTS CLOCK GENE EXPRESSION AND REM SLEEP RECOVERY FOLLOWING SD
CIRBP modulated the cortical response to SD in the expression of three out of the five clock genes quantified. As anticipated, the SD incurred decrease in cortical Rev-erbα was attenuated in Cirbp KO mice. REV-ERBα acts as a transcriptional repressor of positive clock elements such as BMAL1 (Preitner et al., 2002). Mice lacking both Rev-erbα and its homolog Rev-erbβ have a shorter and unstable period under constant conditions and deregulated lipid metabolism (Cho et al., 2012). We recently established that Rev-erbα also partakes in several aspects of sleep homeostasis: Rev-erbα KO mice accumulate at a slower rate NREM sleep need and have reduced efficiency of REM sleep recovery in the first hours after SD (Mang et al., 2016).
The expression of the clock genes Per2 and Clock was also modulated in the absence of CIRBP, suggesting that parts of the core clock are sensitive to the presence of CIRBP in response to SD. Given the role of clock genes in sleep homeostasis (Franken, 2013), the modulated clock gene expression in KO mice could have contributed to the REM homeostatic sleep phenotype. This is supported by studies showing that mutations in clock genes incurred a loss in REM sleep recovery [i.e. CLOCK (Naylor et al., 2000)], or impacted the initial efficiency of REM sleep recovery [i.e. DBP (Franken et al., 2000), PER3 (Hasan et al., 2011), and REV-ERBα (Mang et al., 2016). Follow-up studies can address if indeed the changes in clock gene expression in Cirbp KO mice are functionally implicated in the REM sleep phenotype.
The other aspects of the homeostatic regulation of sleep that we inspected, NREM EEG delta power and time spent in NREM sleep after sleep deprivation, were unaffected in Cirbp KO mice. Thus, CIRBP participates specifically in REM sleep homeostasis, whereas we do not find evidence for its participation in NREM sleep homeostatic mechanisms.
OTHER MECHANISMS LINKING SLEEP-WAKE STATE TO CLOCK GENE EXPRESSION
Our results show that other pathways besides CIRBP must contribute to the sleep-wake driven changes in clock gene expression. Some suggestions for such pathways are shortly discussed below, as well considerations that could potentially account for the absence of a more widespread CIRBP dependent change in clock gene expression that we expected based on in vitro results.
Rbm3 (RNA Binding Motif Protein 3), another cold-inducible transcript which is closely related to CIRBP, also conveys temperature information into high amplitude clock gene expression in vitro (Liu et al., 2013). Like Cirbp, its expression is sleep-wake driven (Wang et al., 2010). Thus, RBM3 might be another mechanism through which changes in sleep-wake state are linked to changes in clock gene expression and could therefore also explain the absence of a widespread CIRBP dependent SD-incurred change in clock gene expression. A follow-up study could address this possibility by quantifying the SD-evoked changes in clock gene expression in double Cirbp-Rbm3 KO mice.
Heat shock factor 1 (Hsf1) is a member of the heat shock pathway and in vitro studies showed that it conveys temperature information to the circadian clock by initiating Per2 transcription through binding to Per2’s upstream heat shock elements (Tamaru et al., 2011). Under undisturbed conditions, both Hsf1 mRNA and protein levels are constitutively expressed, but the protein exhibits daily re-localization during the dark phase to the nucleus where it acts as a transcription factor (Reinke et al., 2008). Interestingly, CIRBP binds to the 3’UTR of Hsf1 transcript [(Morf et al., 2012), see supplementary data therein], although it is unclear if this affects the transcriptional activity of HSF1. We found that SD induced a significant increase in Hsf1 only in KO mice, which is congruent with the observation that the expression of two other transcripts downstream of HSF1, Hsp90b and Hspa5/BiP, was significantly amplified in KO mice after SD. Altogether, this suggests that the increased expression of Per2 in KO mice might be linked to increased Hsf1 expression, and underscores the presence of other temperature (and thus sleep-wake) driven pathways that can ultimately affect clock gene expression.
Beyond temperature, many other physiological changes occur during wakefulness that can subsequently affect clock gene expression. For example, oxygen consumption changes with sleep-wake state (Jung et al., 2011) and oxygen levels can modulate the expression of clock genes through HIF1α (Adamovich et al., 2017). Moreover, during SD, corticosterone levels increase which subsequently amplifies the expression of some, but not all, clock genes (Mongrain et al., 2010).
Another factor to consider is that the SD-incurred changes in clock gene expression depend also on an intact clock gene circuitry. For example, Npas2 KO mice showed a reduced increase in Per2 expression in the forebrain after SD (Franken et al., 2006), while Cry1,2 double-KO mice display a larger increase in Per2 expression after SD (Wisor et al., 2008). Thus, differences in clock gene circuitry, as suggested by the in vitro data (Liu et al., 2013, Morf et al., 2012), could also have contributed to the observed changes in clock gene expression after SD in Cirbp KO.
We could not corroborate the hepatic increase in heat shock transcripts (Hsf1, Hsp90b and Hspa5) and in Per2 after SD as reported in other studies (Diessler et al., 2018, Maret et al., 2007), whereas we did confirm the SD-induced changes in Cirbp, Rbm3-short, Dbp and Rev-erbα. We cannot readily explain this lack of confirmation.
Finally, we would like to briefly address an obvious shortcoming. The hypothesis of this study is based on results obtained in a relatively simple biological model (i.e., immortalized fibroblasts) and applied to a much more complex model (i.e., cortices and livers of male mice). Unpublished observations on the circadian dynamics of the expression of CLOCK-BMAL1 target genes in the liver, such as Rev-erbα and Dbp, show an increased circadian amplitude in Cirbp KO mice; i.e. the opposite phenotype from that observed in vitro (Schibler et al., 2015). Along those lines, we could not consistently reproduce the importance of CIRBP in determining the ext/com ratio of Sfpq (see also Figure 4-B). Thus, in vitro findings will not always predict in vivo results, which could account for the lack of a widespread CIRBP-dependent change in clock gene expression after SD.
CONCLUSION
This hypothesis-driven study explored whether the SD-induced changes in clock gene expression could be mediated through the cold-induced transcript CIRBP. After SD, the cortical expression of Rev-erbα, which we recently identified as a player in the sleep homeostat (Mang et al., 2016), was attenuated in Cirbp KO mice, whereas the expression of two other clock genes, Per2 and Clock, was amplified. Thus, the SD induced changes in clock gene expression are, in part, modulated by CIRBP.
A large body of evidence has shown that clock genes are crucial for metabolism (reviewed in (Panda, 2016)). This is supported by the observation that disturbance of clock gene expression, through for example genetic manipulations in mice or shift work in humans, can lead to the development of metabolic disorders (Rudic et al., 2004) (Karlsson et al., 2001). Not only sleeping at the wrong time, but also sleeping too little or of poor quality can induce disturbed metabolic state both in rats (Barf et al., 2010) and humans (Copinschi et al., 2014). Because sleep loss affects clock gene expression (Franken, 2013), we propose that this could represent a common pathway through which both sleep and circadian disturbances lead to metabolic pathologies. It is thus of importance to determine the pathways through which a disturbed sleep-wake distribution affects clock gene expression. We show that temperature and CIRBP partake in this process, and we identified the expression of Rev-erbα as one of the genes affected by CIRBP. Genetic (Delezie et al., 2012) and pharmacological (Solt et al., 2012) studies have shown that this transcriptional repressor is important for healthy metabolic functioning. Further experiments could address the metabolic consequences of the attenuated response in Rev-erbα to sleep loss.
Material and Methods
MICE AND HOUSING CONDITIONS
Cirbp KO mice, kindly provided by Prof Jun Fujita (Kyoto University, Japan), were maintained on a C57BL6/J background. In these mice, Cirbp exons were replaced by a TK-neo gene through homologous recombination in D3 embryonic stem cells, resulting in the absence of the Cirbp transcript and protein (Masuda et al., 2012). Breeding couples or trios consisted of heterozygous male and female mice. WT littermates were used as controls. Throughout all the experiments, mice were individually housed in polycarbonate cages (31×18×18 cm) with food and water ad libitum and exposed to a 12 h light/12 h dark cycle (70–90 lux). All experiments were approved by the Ethical Committee of the State of Vaud Veterinary Office Switzerland under license VD2743 and 3201.
EEG/EMG AND THERMISTOR SURGERY
At the age of 9 to 13 weeks, 17 KO and 20 WT male mice were implanted with electroencephalogram (EEG) and electromyogram (EMG) electrodes (8 experimental cohorts). The surgery took place under deep xylazine/ ketamine anesthesia supplemented with isoflurane (1%) when necessary; for details see (Mang and Franken, 2012). Briefly, six gold-plated screws (diameter 1.1 mm) were screwed bilaterally into the skull over the frontal and parietal cortices. Two screws served as EEG electrodes and the remaining four anchored the electrode connector assembly to the skull. As EMG electrodes, two gold wires were inserted into the neck musculature. Of all EEG/EMG implanted mice, 8 KO and 9 WT mice were additionally implanted with a thermistor (serie P20AAA102M, General Electrics (currently Thermometrics), Northridge, California, USA) which was placed on top of the right cortex (2.5 mm lateral to the midline, 2.5 mm posterior to bregma). The EEG and EMG electrodes and thermistor were soldered to a connector and cemented to the skull. Mice recovered from surgery during 5–7 days before they were connected to the recording cables in their home cage for habituation, which was at least 6 days prior to the experiment. In total no less than 11 days were scheduled between surgery and start of experiment.
EXPERIMENTAL PROTOCOL AND DATA ACQUISITION
EEG and EMG signals, Tcx and LMA were recorded continuously for 96 h. The recording started at light onset; i.e., Zeitgeber Time (ZT)0. During the first 48 h (days BL1 and BL2), mice were left undisturbed to establish a baseline. Starting at ZT0 of day 3, mice were sleep deprived by gentle handling for 6 hours (ZT0–6), as described in (Mang and Franken, 2012). The remaining 18 h of day 3 and the entire day 4 were considered as recovery (days REC1 and REC2, respectively). The analog EEG and EMG signals were amplified (2,000×), digitized at 2 kHz and subsequently down sampled to 200 Hz and stored. The EEG was subjected to a discrete Fourier transformation yielding power spectra (range: 0–100 Hz; frequency resolution: 0.25 Hz; time resolution: consecutive 4-sec epochs; window function: Hamming). Thermistors were supplied with a constant measuring current (Iconst = 100 microA), and voltage (V) was measured at 10 Hz to calculate the median resistance (Rt) per 4-s epoch as in eq. (1).
Each thermistor has an individual material constant, β. The resistance was measured at 25°C (R25°C) and 37°C (R37°C) by the manufacturer, and used to determine β as in eq.(2), with T values in Kelvin (°C + 273.15).
Following on eq. (2), the temperature (t) in °C can be calculated as described in eq. (3).
The EEG, EMG, and voltage across the thermistor were recorded with Hardware (EMBLA) and software (Somnologica-3) purchased from Medcare Flaga (EMBLA, ResMed, USA). LMA was detected with passive infrared sensors (Visonic Ltd, Tel Aviv, Israel) and quantified with ClockLab software (ClockLab, ActiMetrics, Wilmette, Illinois, USA).
ANALYSIS OF LMA
To inspect the time course of LMA corrected for time-spent-awake, raw LMA was expressed per unit of waking in percentiles to which an equal amount of time-spent-awake contributed (as in Figure 3-G). The number of percentiles per recording period were chosen according to the prevalence of wakefulness, where 6 percentiles were used during the light phase and 12 during the dark phase, with the exception for 6 sections during the SD and 3 sections during the remaining 6hrs of the light phase of REC1. To assess genotype differences in LMA (Figure 6), the absolute number of movements were inspected. The LMA recordings of four mice (3 WT, 1 KO) were interrupted due to technical problems during the experiment, leaving data from 17 WT and 16 KO mice for analyses involving LMA.
We determined in 5 WT and 7 KO mice after the EEG-based sleep phenotyping their circadian period under at least two weeks of constant darkness. Period length was determined by Chi-squared test with ClockLab software (ClockLab, ActiMetrics, Wilmette, Illinois, USA).
DETERMINATION OF BEHAVIORAL STATES
Offline, the mouse’s behavior was visually classified as ‘Wakefulness’, ‘REM sleep’, or ‘NREM sleep’ for consecutive 4-sec epochs based on the EEG and EMG signals, as previously described (Mang and Franken, 2012). Wakefulness was characterized by EEG activity of mixed frequency and low amplitude and variable muscle tone. NREM sleep was defined by synchronous activity in the delta frequency range (1–4 Hz), and low and stable muscle tone. REM sleep was characterized by regular theta oscillations (6–9 Hz) with low EMG activity. Waking was further differentiated into ‘quiet waking’ and ‘theta-dominated waking’ (TDW). TDW was determined based on the relative importance of power in the 6.5 to 12.0 Hz range to the overall power in the EEG of an artefact-free epoch scored as wakefulness, as described in (Vassalli and Franken, 2017). We refer to waking that is not classified as TDW as ‘quiet’ waking. Epochs containing EEG artefacts were marked according to the state in which they occurred and excluded from EEG spectral analysis but included in the sleep-wake distribution analyses. During the four day recording, 7.0 ± 0.9%, 2.1 ± 0.3% and 2.5 ± 0.2% of the epochs were scored as an artefact in waking, NREM, and REM sleep, respectively, and this did not differ between genotypes (t-tests, t(35)=1.77, p=0.09; t(35)=0.64, p=0.53; t(35)= 0.99, p=0.33, respectively).
ANALYSIS OF CORTICAL TEMPERATURE
The raw Tcx data showed unexpected variation. Therefore, we inspected the inter-individual variation in daily amplitude and absolute Tcx levels. The latter was determined in two ways: i) by averaging Tcx during the last five hours of SD, thus minimizing the sleep-wake state incurred differences in Tcx, and ii) by averaging Tcx during the 12h baseline light phase. These measures were highly correlated (R2=0.99; p<0.0001). Variation in the daily amplitude was quantified by averaging the difference between the highest and lowest hourly mean of Tcx of each of the two baseline days. No effect of genotype on absolute average Tcx or amplitude was detected (t-test, t(12)=0.61, p=0.55; t(12)=-0.63, p=0.54, respectively). Two mice (one of each genotype) exhibited a ca. 2-fold reduction in amplitude together with 2°C higher values during the SD relative to the other mice (Figure 7, pink symbols). Therefore, we excluded these two mice from subsequent Tcx analysis. Three other mice (2 WT and 1 KO) showed normal amplitude but overall lower absolute values (Figure 7, blue symbols). We corrected for this difference by raising their Tcx values by the difference between the Tcx reached in each of these 3 mice during the SD to the average Tcx reached over the same recording period in the remaining 9 mice. Of note, most of our Tcx analysis focuses on its relative sleep-wake dependent changes, which are not affected by differences in absolute Tcx values. Finally, the baseline Tcx data that was used to construct Figure 2B was based on 7 WT and 7 KO mice. During the recording, one KO mouse and one WT mouse had random fluctuations of Tcx beyond physiological reach and were therefore excluded from analysis involving the daily dynamics of Tcx (Figure 3). In the recovery, a KO mouse was excluded due to aberrant high Tcx that could not be accounted for by the sleep-wake distribution, leaving 6 WT and 5 KO mice for analyses involving REC1 and REC2.
Because visualization of all 4-s epochs occurring in 24-hour day is not compatible with the resolution of Figure 2-A, sleep-wake states were averaged per minute and assigned to either wake, NREM or REM sleep. Because REM sleep occurs less and in shorter bouts than waking and NREM sleep, this stage is slightly underrepresented in the hypnogram of Figure 2-A. Tcx was averaged per minute.
We inspected Tcx 1.5 min before and after sleep-wake transitions (i.e., transitioning from wake to NREM sleep, NREM to REM sleep, REM sleep to wake and NREM sleep to wake). A sleep-wake transition was selected when the state before and after the transition lasted at least 8 epochs (i.e. >32 sec). With this criterion, an average of 38 wake to NREM sleep, 101 NREM sleep to REM sleep, 28 REM sleep to wake and 32 NREM sleep to wake transitions per mouse during the two baseline days was detected. The temperature profile of Tcx before and after the transition was constructed by depicting Tcx relative to the mean Tcx at a given sleep-wake transition (i.e., the average of Tcx in the epoch before and after the sleep-wake transition). We subsequently constructed an individuals’ average change in Tcx for each sleep-wake transition. For this average, at least 10 traces were contributing at a given point in time to prevent skewing of the average individual temperature profile by few Tcx traces. Thus, the further from the sleep-wake state transition, the less epochs contributed to the average individual Tcx profile. One WT mouse exhibited an extreme drop in Tcx (−0.2°C in a 4-second epoch) after the transition from NREM sleep to wake in its average Tcx trace, but not in other sleep-wake transitions. We attributed this observation to a technical artefact and therefore this mouse was excluded from the NREM sleep to wake transitions.
The residuals of the correlation between waking and Tcx exhibited a circadian pattern under baseline conditions. We visualized the properties of this pattern further by fitting a sinewave through the data (Prism, non-linear regression; sine-wave with non-zero baseline; least squares fit).
ANALYSIS OF EEG BASED ON BEHAVIORAL STATE
Unless otherwise stated, all mice (20 WT and 17 KO) were included in the analyses based on the EEG data. Spectral content of the EEG within sleep-wake states was calculated as follows. To account for inter-individual differences in overall EEG power, EEG spectra were expressed as a percentage of an individual reference value calculated as the total EEG power across 0.75-45 Hz and all sleep-wake states in the 48h baseline. This reference value was weighted so that for all mice the relative contribution of the three sleep-wake states (wake, NREM and REM sleep) to this reference value was equal.
Theta peak frequency (TPF) was calculated by determining the frequency at which power density peaks per 4-s epoch and subsequently averaged per individual. Power density peaks were quantified from 6.5 to 12.0 Hz band and from 5.5 to 12.0 Hz band for TDW and REM sleep, respectively.
Time course analysis of EEG delta power (i.e., the mean EEG power density in the 0.75–4.0 Hz range in NREM sleep) during baseline and after SD was performed as described previously (Franken et al., 1999), and similar to the analysis of LMA per unit of waking. The light periods of BL1, BL2, and REC2 were divided into 12 percentiles, the REC1 light period (ZT6–12) into 8 sections, and all dark periods into 6 sections. The timing of these percentiles was based on the prevalence of NREM sleep. EEG delta power values in NREM sleep were averaged within each percentile and then expressed relative to the mean value reached in the last 4hr of the two main rest periods in baseline between ZT8–12. This reference was selected because delta power reaches lowest values at this time of the day and is least influenced by differences in prior history of sleep and wakefulness (see also (Franken et al., 1999)). In the time course of NREM delta power, one mouse (KO) demonstrated a strong decrease over the course of the experiment which could not be attributed to changes in the sleep-wake distribution. 9 out of the 12 delta power values during the light phase of REC2 in this mouse were outliers (MAD outlier test, consult (Leys et al., 2013) for details). This mouse was excluded from the analyses involving sleep homeostasis (Figure 5).
The effect of 6hr SD on subsequent time spent in NREM and REM sleep was assessed by calculating the recovery-baseline difference in sleep time per 1hr intervals.
SIMULATING NREM EEG DELTA POWER [PROCESS S]
We applied a computational method to predict the change in delta power during NREM sleep based on the sleep-wake distribution as described before (Franken et al., 2001). Process S is exponentially increasing with time constant τi during waking and REM sleep, and exponentially decreasing by τd during NREM sleep (eq. (4) and (5), respectively).
In these simulations, UA represents the upper asymptote, LA the lower asymptote and dt the time step of the iteration (4 seconds). Both asymptotes were estimated for each individual mouse. The upper asymptote was based on the 99% level of the relative frequency distribution of delta power reached in all 4s epochs scored as NREM sleep in the 4-day recording. As an estimate of the lower asymptote, the intersection of the distribution of delta power values in NREM sleep with REM sleep was taken. At the start of the simulation, an iteration through the first 24-hr (BL1) was performed with S0=150 at t=0. The value reached after 24-hrs is independent of S0 at t=0 and, assuming a steady state during baseline, reflects Process S at the start of the baseline for a given combination of time constants.
The fit was optimized by minimizing the mean squared difference of simulated and observed NREM delta power for a range of Ti: 1-25 h, step size 0.125h; Td: 0.1-5.0 h, step size 0.025h; i.e. the simulation was run for all 38’021 combinations of Ti and Td for each mouse. The combination of Ti and Td giving the best fit was used to assess differences in process S between genotypes.
We noted a subtle but consistent linear discrepancy in the alignment of the simulated Process S to the measured NREM delta power values at the end of the light phase on BL1, BL2 and REC2 (Pearson correlation, slope≠0: 1 sample t-test; t(35)=-4.38, p=0.0001). This change correlated well with the day-to-day changes in total spectral power in the EEG calculated across all sleep-wake states in BL1, BL2, and REC2 (Pearson correlation: R2=0.70, p<0.0001; n=36). There was no effect of genotype on slope (Δ delta power %/h; students’ t-test; t(34)=0.62; p=0.54; WT:-0.086±0.027; KO:-0.065±0.021) or intercept (t(34)=-0.88; p=0.38; WT: 101.5±0.62; KO: 100.7±0.56; WT: n=20, KO: n=17). We attributed these linear changes to be of non-biological origin and detrended the measured NREM delta power values before optimizing the fit between observed and simulated delta power.
GENE EXPRESSION IN LIVER AND BRAIN
Five mice of each genotype (n=15 per genotype in total) were sacrificed either prior to SD (ZT0), at ZT6 allowing them to sleep ad lib (i.e. without SD; ZT6-NSD), or at ZT6 after 6h SD (ZT6-SD) across four experimental cohorts. Mice were randomly assigned to one of the three experimental conditions. Genes of interest included transcripts affected by SD (Maret et al., 2007, Mongrain et al., 2010) and/or by the presence of CIRBP (Liu et al., 2013, Morf et al., 2012) with a special interest for clock genes. Specific forward and reverse primers and Taqman probes were designed (seeSupplementary File 1) to quantify mRNA.
Upon sacrifice, both the cerebral cortex and liver were extracted and immediately flash frozen in liquid nitrogen. Samples were stored at −80°C. RNA from cortex was extracted and purified using the RNeasy Lipid Tissue Mini Kit 50 (QIAGEN, Hombrechtikon, Switzerland); RNA from liver was extracted and purified using the RNeasy Plus Mini Kit 50 (QIAGEN, Hombrechtikon, Switzerland), according to manufacturer’s instructions. RNA quantity (NanoDrop ND-1000 spectrophotometer; Thermo Scientific, Wilmington, NC, USA) and integrity (Fragment Analyzer, Advanced Analytical, Ankeny, IA, USA) was measured and verified for each sample. 1000 ng of purified total RNA was reverse-transcribed in 20μL using a mix of First-strand buffer, DTT 0.1M, random primers 0.25μg/μl, dNTP 10mM, RNAzin Plus RNase Inhibitor and Superscript II reverse transcriptase (Invitrogen, Life Technologies, Zug, Switzerland) according to manufacturers’ procedures. The cDNA was diluted 10 times in Tris 10 mM pH 8.0, and 2μL of the diluted cDNA was amplified in a 10μL TaqMan reaction in technical triplicates on an ABI PRISM HT 7900 detection system (Applied Biosystems, Life Technologies, Zug, Switzerland). Cycler conditions were: 2 min at 50°C, 10 min at 95°C followed by 45 cycles at 95°C for 15 s and 60°C for 1 min. Standard curves were calculated to determine the amplification efficiency (E). A sample maximization strategy was used where all biological replicates of one tissue were amplified for two genes per plate. Gene expression levels were normalized to two reference genes (cortex: Eef1a and Gapdh: M=0.23, CV=0.09 and liver: Gadph and Tbp; M=0.32, CV=0.11) using QbasePLUS software (Biogazelle, Zwijnaarde, Belgium). Rbm3 isoforms were in a separate run quantified in liver and cortex, again with their housekeeping genes (same as previously; cortex: M=0.22, and CV=0.08; liver: M=0.13, CV=0.05). Transcripts with an average Ct-value>30 were omitted from analysis (in KO and WT livers: Rbm3, Dusp4, Homer1a and Npas2; in cortex and liver of KO mice: Cirbp). Results are expressed as normalized relative quantity (NRQ) which based on the overall mean expression per gene, which was set at 1.0 (Hellemans et al., 2007).
CIRBP affects the poly-adenylation sites of several transcripts (Liu et al., 2013). We explored if this newly discovered role of CIRBP could be corroborated in our study by focusing on the transcript Splicing factor, proline and glutamine rich (Sfpq) which exhibits CIRBP-dependent alternative poly-adenylation (APA) ((Liu et al., 2013), see their Supplemental Fig4-5). We calculated the ratio of the prevalence of the external 3’UTR region over the common region according to eq. (6), where E is the amplification efficiency and Ctext and Ctcomm the number of cycles for the detection of the extended and common isoform, respectively.
STATISTICS
Statistics were performed in R (version 3.3.2) and Prism (version 7.0). The threshold of significance was set at p=0.05, and all statistics were solely performed on biological replicates. To be more specific, our RT-qPCR data stems from 5 biological replicates, whereas amplification of cDNA from one biological replicate is performed in three technical replicates. Deviations from the mean are representing standard error of the mean. The distribution of the LMA data was normalized by a log2 transformation on the hourly values, allowing for subsequent parametric analyses on the relationship between Tcx and LMA as in Figure 3. Time course data were analyzed by 1- and 2-way repeated measures (RM) analysis of variance (ANOVA) with as factors ‘time’ and ‘genotype’ (GT). Upon significance, post-hoc Fisher LSD tests were computed. Differences between BL and REC values within genotype were computed by paired t-tests. EEG spectra were also analyzed by 1- and 2-way RM ANOVA with as factors ‘time’ or ‘frequency’ and ‘GT’. When GT or its interaction with time or frequency reached significance, post-hoc t-tests were computed. The above-mentioned analyses were all performed in Prism. Missing values in the sleep-wake transition data of Figure 2-B led to exclusion of all data further away from the transition to perform a RM ANOVA.
Correlation coefficients of linear regression were calculated in Prism over all hourly values of LMA, Tcx and waking per genotype (96 per mice). To compare slopes of regression lines between genotypes, an ANCOVA was applied based on (Zar, 1984) and run in Prism. To quantify the contribution of waking and LMA independent from each other to Tcx, a partial correlation was performed (R software; package ‘ppcor’, function pcor.test). Mixed model analysis was performed with factors LMA (log2 transformed), waking, and genotype (R packages ‘lme4’, ‘lmer’, ‘lmerTest’, and ‘MuMIn’). Model1 quantified the predictive power of waking, Model2 of waking and LMA per unit of waking (LMA/Waking) and Model3 of waking, LMA/Waking and its interaction, to predict Tcx. Predictive power of models was compared with Chi-squared tests by assessing the statistical significance in the reduction of residual sum of squares between two models ordered by complexity; i.e. Model1 was compared to Model2, and upon significance, Model2 was compared to Model3. Goodness-of-fit was assessed by the marginal R-squared (R2m) which explains the effect of the fixed factors only, and the conditional R-squared (R2c), which considers the individual variance as well and is therefore more biological relevant. Hence, in the results section only the R2c values are reported.
For the molecular data, the qPCR NRQ values were log2-transformed to normalize the distribution. Genotype differences at ZT0 were tested with a t-test. The effect of SD and genotype at ZT6 was assessed by 2-way ANOVA with post-hoc Fisher LSD tests upon significance. One outlier (WT, cortex) in the ext/com ratio analyses was detected by the Grubbs outliers test (α < 0.05) and excluded.
Supplementary File 1: sequences of the forward and reverse primer and probe used for the RT-qPCR
Acknowledgements
We would like to thank Maxime Jan for his help in constructing the linear mixed model, our colleagues for their assistance with the sleep deprivations, Hannes Richter from the Genomic Technology Facility, for his support when setting up the RT-qPCR, David Gatfield and Bulak Arpat for fueling insightful discussions, and Jun Fujita (Kyoto University, Japan) for sharing the Cirbp KO mice. This study was performed at the University of Lausanne, Switzerland, and supported by the Swiss National Science Foundation (SNF n°146694 to PF supporting MMBH) and the state of Vaud (supporting MMBH, YE and PF).