Deletion of Clock Gene Period 2 (Per2) in Astrocytes Shortens Clock Period but Does Not Affect Light-Mediated Phase Shifts in Mice

Abstract

The circadian clock is a self-sustaining oscillator with a period of approximately 24 h, enabling organisms to anticipate daily recurring events, such as sunrise and sunset. Since the circadian period is not exactly 24 h and the environmental day length varies throughout the year, the clock must be periodically reset to align an organism’s physiology with the natural light/dark cycle. This synchronization, known as entrainment, is primarily regulated by nocturnal light, which can be replicated in laboratory settings using a 15 min light pulse (LP) and by assessing locomotor activity. An LP during the early part of the dark phase delays the onset of locomotor activity, resulting in a phase delay, whereas an LP in the late dark phase advances activity onset, causing a phase advance. The clock gene Period 2 (Per2) plays a key role in this process. To investigate its contributions, we examined the effects of Per2 deletion in neurons versus astrocytes using glia-specific GPer2 (Per2/GfapCre) knockout (KO) and neuronal-specific NPer2KO (Per2/NesCre) mice. All groups were subjected to Aschoff type II protocol, where an LP was applied at ZT14 or ZT22 and the animals were released into constant darkness. As control, no LP was applied. Phase shift, period, amplitude, total activity count, and rhythm instability were assessed. Our findings revealed that mice lacking Per2 in neurons (NPer2) exhibited smaller phase delays and larger phase advances compared to control animals. In contrast, mice with Per2 deletion specifically in glial cells including astrocytes (GPer2) displayed normal clock resetting. Interestingly, the absence of Per2 in either of the cell types resulted in a shorter circadian period compared to control animals. These results suggest that astrocytic Per2 is important for maintaining the circadian period but is not required for phase adaptation to light stimuli.

1. Introduction

The mammalian circadian system is an intricate internal timekeeping system that synchronizes physiological and behavioral functions with the 24 h environmental daily cycle. The master oscillator in mammals is the hypothalamic suprachiasmatic nuclei (SCN), which are responsible for the coordination of central and peripheral clocks. The SCN are aligned with environmental cycles such as the day/night cycle by cues known as “Zeitgebers” [1,2]. The most prominent Zeitgeber is light.

Photoreceptors in the retina detect light and relay the photic signal to the SCN, which transmit this signal to other central and peripheral oscillators. The molecular mechanism of the circadian oscillator is made up of a transcriptional/translational feedback loop. This process begins when the BMAL1 and CLOCK proteins (i.e., positive elements of the loop that activate the transcriptional/translational feedback loop) form a heterodimer (CLOCK:BMAL1) that drives the transcription of the negative elements of the loop, namely, Period (Per1, Per2, and Per3) and Cryptochrome genes (Cry1 and Cry2), leading to the expression of their respective proteins (PER, CRY). These negative elements, in turn, form their own heterodimer complex (PER:CRY) that suppresses their own transcription by inhibiting the activity of the CLOCK:BMAL1 complex [3,4].

In nocturnal rodents, nocturnal light signals can reset the circadian oscillator in the SCN, which leads to phase shifts of the circadian locomotor activity rhythm [5,6,7]. The responsiveness to light in nocturnal rodents is time-dependent. Wild-type (WT) animals exhibit phase delays when photic signals are applied early in the night and phase advances when signals are applied late at night, while no significant effect is observed when light is applied at midday or midnight [7].

The expression of the molecular clock genes Per1 and Per2 has been reported to be induced by light within the SCN and retina [8,9]. This modulation underscores their role in the light transduction pathway, light responsiveness, and subsequent re-entrainment of circadian rhythms.

Per2 is a critical component of the molecular clockwork in the SCN of mammals, playing a pivotal role in resetting the circadian clock in response to light. Studies have shown that mutant Per2 mice, as well as total and neuronal-specific Per2 knockout mice, are unable to delay the clock when exposed to a light pulse at early night (ZT14), unlike their wild-type (WT) counterparts [10,11]. Furthermore, mutant Per2 mice and total Per2 knockout mice display a shorter period compared to their controls with a transition to arrhythmic circadian activity under prolonged constant darkness [12,13]. Notably, the shorter period length is also observed in neuronal-specific Per2 knockout mice [11]. Interestingly, recent findings indicate that Per2 also participates in the CREB pathway, contributing to the activation of the Per1 gene in response to photic signals [14].

Circadian molecular clock genes are expressed not only in neurons of the SCN but also in various glial cells, including astrocytes [15,16]. Emerging evidence suggests that a functional and intact circadian molecular clockwork in astrocytes plays an essential role in regulating SCN outputs and behavioral rhythms—such as activity patterns and sleep homeostasis—in a manner that is autonomous and independent from neuronal clocks [17,18,19].

Astrocytic clock genes are reported to regulate the fluctuating expression of glutamate, ATP, and adenosine, which are crucial in controlling circadian system entrainment to photic cues and the rhythmic sleep/wake cycle [20,21]. Disruption of the astrocytic clock has been shown to alter various animal behaviors [22]. For instance, targeted deletion of Per2 in astrocytes decreases anxiety-like and depression-related behaviors [13]. Furthermore, specific deletion of Bmal1 in astrocytes leads to disruptions in cognitive functions and circadian locomotor activity rhythms in mice, resulting in longer period lengths compared to control groups [18,23].

Despite these findings, the role of the astrocytic Per2 gene in regulating the re-entrainment of the circadian clock in response to photic signals remains unexplored. Therefore, the current study aims to investigate the potential role of the astrocytic Per2 gene in the photic signal transduction pathway and circadian clock resetting in mice. This investigation will utilize astrocyte-specific Per2 knockout mice (GPer2) and neuron-specific Per2 knockout mice (NPer2) to compare the function of Per2 in astrocytes in circadian periodicity and phase resetting with its function in neurons.

2. Results

2.1. Effect of Constant Darkness (DD) Without LP on Circadian Locomotor Activity Rhythm

To determine the potential role of the Per2 gene in the light response process, we used glia-specific GPer2 (Per2/GfapCre) knockout (KO) [13] and neuronal-specific NPer2 (Per2/NesCre) KO [11] mice. Glial and neuronal Per2 deletion was verified by genotyping, immunohistochemistry, and Western blot techniques in brain tissues of mice [11,13]. These animals were compared with their corresponding cre-line GfapCre (Gcre) [24], NesCre (Ncre) [25], and littermate wild-type (WT) mice as control groups. The protocol is outlined in Figure 1 (see the Materials and Methods Section). We used an Aschoff type II protocol for the assessment of light-induced phase shifts [26,27].

To determine the possible effects of transition from a light/dark cycle (LD) to constant darkness (DD) in the different genotypes, the circadian locomotor activity parameters were analyzed. Transition from LD to DD did not elicit significant differences in activity onsets between the genotypes (p > 0.05) (Figure 2A–F). Interestingly, the periods of NPer2 and GPer2 animals were significantly shorter compared to WT, Ncre, and Gcre control mice (Figure 3A), which is in line with previous observations made for Per2 mutant and total Per2 KO animals [11,12,28]. Comparison of the amplitude between LD and DD conditions revealed that for all genotypes amplitude was significantly shortened in DD, except for WT (p = 0.051) (Figure 3B). In LD, both Ncre and NPer2 had a higher amplitude compared to WT; however, this was not significantly different (p = 0.07 and p = 0.08, respectively). No differences in the amplitude were observed between other genotypes (p > 0.05).

Next, we looked at the relative power of phase (FFT), a parameter used to indicate rhythm stability. Transition from LD to DD exhibited a significant decrease in FFT (indicating more rhythm instability) in all experimental groups, except WT (p= 0.06) (Figure 3C). However, the transition to DD did not affect the total locomotor activity counts in all investigated groups (p > 0.05, Figure 3D).

2.2. Effect of Neuronal and Astrocytic Per2 Deletion on Light Responsiveness and Circadian Clock Resetting

To investigate the possible effect of neuronal and astrocytic Per2 in light responsiveness and circadian clock resetting, all genotypes were subject to a light pulse (LP) either at ZT14 (early night) or at ZT22 (late night) (Figure 1). Phase shift and other locomotor activity parameters were analyzed.

When an LP was applied at ZT14, the three control groups (i.e., WT (blue), Ncre (rose), and Gcre (light green), Figure 4F) delayed activity approximately by 1 h (−57.6 ± 6.6 min, −55.1 ± 11.3 min, and −54.6 ± 8.1 min, respectively) with no significant differences observed between the three groups (p > 0.05). On the other hand, NPer2 displayed a significantly shorter phase delay (−15.5 ± 9.8) compared to the controls, WT and Ncre (p < 0.05). In contrast, GPer2 showed a normal phase delay (−72.6 ± 9.2) with a tendency to be longer but not significantly different compared to its controls, WT and Gcre (p > 0.05). In addition, both KO groups showed a highly significant different phase shift (p < 0.01, Figure 4A–F).

After an LP at ZT14 followed by a release into constant darkness, control groups showed shorter period length than 24 h (WT: 23.8 h ± 0.05, Ncre: 23.6 h ± 0.05, and Gcre: 23.8 h ± 0.04) and they were not significantly different (p > 0.05) (Figure 5A). Compared to their controls, NPer2 and GPer2 exhibited significantly shorter period length (NPer2: 23.3 h ± 0.09, p < 0.0001 vs. WT and p < 0.05 vs. Ncre; GPer2: 23.4 h ± 0.1, p < 0.01 vs. WT and p < 0.001 vs. Gcre). Both groups showed non-significant period lengths among them (p > 0.05, Figure 5A). In addition, the LP at ZT14 did not elicit any significant changes in amplitude, FFT, and total activity counts between all investigated genotypes (p > 0.05, Figure 5B–D).

When LP was applied at ZT22 followed by DD (Figure 6), the WT and Ncre groups exhibited approximately 30 min phase advance (31.4 ± 9.1 min and 23.63 ± 9.1 min, respectively) and were not significantly different (p > 0.05), whereas NPer2 revealed a significantly longer phase advance (72.4 ± 14.6 min) compared to its control, Ncre (p < 0.05). On the other hand, mice from Gcre and GPer2 could not display a distinct phase advance (9.91 ± 10 min and 16.3 ± 14.1 min, respectively). No significant differences were observed between both groups as well as with the WT group (p > 0.05); however, GPer2 showed a significant difference compared to NPer2 (p < 0.05, Figure 6A–F).

Additionally, after LP at ZT22, the three control groups (i.e., WT, Ncre, and Gcre) showed a non-significant shorter period length (23.6 h ± 0.04, 23.4 h ± 0.11, and 23.6 h ± 0.09, respectively; p > 0.05) (Figure 7A). Whereas KO groups (i.e., NPer2 and GPer2) had significantly shorter period lengths (22.9 h ± 0.08 and 22.96 h ± 0.12) than their controls (p < 0.0001 and p < 0.05 NPer2 with WT and Ncre, respectively, and p < 0.0001 GPer2 with WT and Gcre). No difference was observed between KO groups (p > 0.05; Figure 7A). Moreover, NPer2 exhibited a significantly higher amplitude than GPer2 (p < 0.05, Figure 7B). However, LP was applied at ZT22 and did not affect the other locomotor activity parameters (i.e., FFT and total activity count) between different genotypes (p > 0.05; Figure 7C,D).

3. Discussion

In this study, we investigated the impact of Per2 expression deficiency in astrocytes and neurons in mice on period and phase using wheel-running activity. Our findings revealed that the circadian period is shorter in mice lacking Per2 in either astrocytes or neurons. Additionally, while light-induced phase shifts in mice deficient in astrocytic Per2 were comparable to control animals, the absence of Per2 in neurons significantly affected phase shifts.

The observed shorter period in NPer2 and GPer2 knockout (KO) mice suggests that both astrocytes and neurons contribute to circadian clock period regulation. This aligns with earlier studies demonstrating that circadian behavior is regulated not only by neurons [29] but also by astrocytes [18,19,22]. Prior research further highlighted that astrocytes sustain circadian oscillations and influence the circadian period in the SCN, albeit with lower efficacy than neurons [30]. Interestingly, while deletion of Per2 in astrocytes shortens the behavioral period (Figure 3A), ablation of Bmal1 in astrocytes lengthens it [18,23]. This reinforces the importance of astrocytic clock components in behavioral period regulation.

To examine phase shifts, we employed an Aschoff type II protocol and first assessed whether transitions from light/dark (LD) to constant darkness (DD) without a light pulse were consistent across genotypes. No significant differences were observed in activity onset post LD to DD transition among all groups of animals (Figure 2). This suggests that the shorter period observed in NPer2 and GPer2 mice (Figure 3A) did not significantly affect the phase shifts observed (Figs. 4 and 6). Moreover, total activity levels across genotypes showed no significant differences that could potentially impact activity onset following a light pulse (Figure 3D). However, all genotypes exhibited reduced amplitude and stability of circadian locomotor activity rhythm in DD (Figure 3B,C).

When light was applied at ZT14, all genotypes except NPer2 mice displayed phase delays, with significantly reduced phase delays observed in NPer2 animals (Figure 4). This phenotype mirrors that of mice with global Per2 deficiency or Per2 mutation leading to unstable protein [10,31]. Notably, mice lacking Per2 in astrocytes (GPer2 mice) exhibited normal phase delays, suggesting that astrocytic Per2 does not influence phase delays, unlike neuronal Per2 (Figure 4). This finding corroborates previous studies indicating that astrocytes do not regulate circadian phase in the SCN [30]. After a ZT14 light pulse, NPer2 and GPer2 mice showed shorter periods compared to controls (Figure 5), consistent with observations in the absence of light pulses (Figure 3). Similarly, other parameters, such as amplitude, relative power of phase, and total activity, showed no significant differences across genotypes (Figure 5), indicating these factors did not account for phenotypic variations in light-induced phase delays between NPer2 and GPer2 mice.

At ZT22, light induced behavioral phase advances in all genotypes (Figure 6). Interestingly, NPer2 mice displayed significantly larger phase advances compared to all other genotypes, including GPer2 mice. This phenotype parallels previous findings in mice with global Per2 mutations, which showed greater phase advances relative to wild-type animals [10]. The enhanced phase advance in NPer2 mice is unlikely to result from their shorter circadian period (Figure 7A), as GPer2 mice also exhibit a shorter period but do not show increased phase advances (Figure 6F). This suggests that neuronal Per2, but not astrocytic Per2, influences phase advances—a conclusion consistent with observations of light-induced phase delays at ZT14 (Figure 4). These results further support the notion that astrocytes contribute to circadian period regulation but not phase modulation within the SCN [30].

When considering the results from the light pulses at ZT14 and ZT22 together, it is striking that mice lacking neuronal Per2—despite their impaired ability to delay the circadian clock—exhibit a disproportionately strong phase-advancing response to light during the late night. This suggests that neuronal Per2 plays a dual role—facilitating phase delays while concurrently restricting excessive phase advances.

One limitation of this study is the exclusive use of the Aschoff type II protocol, which does not exclude for potential influences of the preceding light/dark cycle prior to light pulse application and subsequent release into constant darkness. Future studies employing an Aschoff type I protocol in constant darkness could address this limitation.

In summary, our experiments shed light on the distinct roles of neuronal and astrocytic Per2 in regulating rapid behavioral phase shifts caused by brief light pulses. We conclude that while astrocytic Per2 contributes to circadian period regulation, neuronal Per2 is involved in both period and phase modulation.

4. Materials and Methods

4.1. Housing of Mice and Mouse Strains

First, 2–4-month-old male mice were placed in completely light-isolated cabinets, previously described in [32], with a 12:12 h light/dark cycle (LD 7:19 h). Temperature (22 ± 2 °C monitored by temperature sensor, Technoline WS-9410, Berlin, Germany), humidity (40–50% monitored by humidity sensor, Technoline WS-9410, Berlin, Germany), and illumination (1000 Lux monitored by Luxmeter, Testo, GmbH & Co, Titisee-Neustadt, Germany) were stable in all cabinets. Mice were housed individually in cages (L: 280 mm × W: 105 mm × H: 125 mm) containing a running wheel (made of steel and 115 mm in diameter, Trixie GmbH, Tarp, Germany) and were provided with enrichments including small amounts of woodchip bedding, nestlet (5 × 5 cm), piece of carton, and open-sided tube as a type of refinement. Food and water were provided ad libitum.

The following mouse strains were used: glia-specific GPer2 (Per2/GfapCre) knockout (KO) [13] and neuronal-specific NPer2 (Per2/NesCre) KO [11] mice. Glial and neuronal Per2 deletion was verified by genotyping, immunohistochemistry, and Western blot techniques in brain tissues of mice [11,13]. These animals were compared with their corresponding cre-line GfapCre (Gcre) [24], NesCre (Ncre) [25], and littermate wild-type (WT) mice as a control group. All experiments and procedures were performed according to the Schweizer Tierschutzgesetz guidelines and approved by the Canton of Fribourg and the cantonal commission for animal experiments (2022-32-FR, 35432).

4.2. Monitoring Circadian Locomotor Activity Rhythm

Circadian locomotor activity rhythm was evaluated using the running wheel. The running wheel revolutions were recorded by a magnetic circuit, previously described in [32,33], which is fixed vertically on the axis of the running wheel outside the cage. The circuit switches on and off according to the wheel’s revolution. These revolutions (indicating the locomotor activity) are captured in 1 min intervals by ClockLab 3 data acquisition system software (Acquisition Version 3.208, Analysis Version 6.0.36).

4.3. Experimental Design and Application of Light Pulse (Aschoff Type II Protocol)

The experimental design (Figure 1) was performed according to the Aschoff type II protocol [26,27]. For the 1st experimental condition, mice were kept in LD (12:12 h) for 2 weeks and locomotor activity was recorded until they had stable rhythms, then the mice were released into constant darkness (DD) without light pulse (LP) after the end of the last dark phase of the LD cycle by switching off the light in the cabinets using an automatic digital timer. Locomotor activity was recorded for at least 12 days. For the 2nd experimental condition, the mice readapted to the LD cycle for 2 weeks and were then released into DD with LP (using 2 visible light spectrum neon light sources: Lumilux cool daylight, 1000 Lux, 18 W, OSRAM, Munich, Germany) at ZT14 (early night) for 15 min on the last LD night, and again locomotor activity was recorded for at least 12 days. For the 3rd experimental condition, the previous step was repeated but with an LP at ZT22 (late night) instead of ZT14. During the constant darkness (DD) periods, mice were checked under dim-red light to control the water and food levels as well as their health status.

4.4. Analysis of Circadian Locomotor Activity Rhythm Parameters

Phase shift (i.e., phase resetting) in DD with/without LP as well as other circadian locomotor activity rhythm parameters (e.g., actograms, period length, amplitude, total locomotor activity count, and relative power of phase (FFT)) were evaluated using ClockLab analysis (Actimetrics) software version 6.0.36.

To analyze the phase shift in each condition, a fitting line was set through 7 consecutive activity LD onsets before the transition into DD with/without LP. Another fitting line was set through at least 10 consecutive activity DD onsets after the transition into DD with/without LP (the first two days after the transition phase were excluded from the calculations). The difference between the two fitting lines on the base of the day that follows the transition to the DD with/without LP is considered the phase shift value [33].

For other circadian locomotor activity rhythm parameters, 7 consecutive activity onsets before and after the transition into DD with/without LP were analyzed (the first two days after the transition phase were excluded from the calculations).

4.5. Statistical Analysis

GraphPad Prism software (version 10.3.1) was used for statistical analysis. After performing normality tests, parametric paired t-test was used to determine the significant difference before and after the transition phase in each group. In addition, the significant difference between three groups or more was performed using parametric one-way ANOVA followed by Dunnett’s test for multiple comparisons. Data are presented as mean ± SEM and are considered significant when the p value < 0.05.

The raw data appear in Appendix A.