Circadian neurogenetics and its implications in neurophysiology, behavior, and chronomedicine

Abstract

The circadian rhythm affects multiple physiological processes, and disruption of the circadian system can be involved in a range of disease-related pathways. The genetic underpinnings of the circadian rhythm have been well-studied in model organisms. Significant progress has been made in understanding how clock genes affect the physiological functions of the nervous system. In addition, circadian timing is becoming a key factor in improving drug efficacy and reducing drug toxicity. The circadian biology of the target cell determines how the organ responds to the drug at a specific time of day, thus regulating pharmacodynamics. The current review brings together recent advances that have begun to unravel the molecular mechanisms of how the circadian clock affects neurophysiological and behavioral processes associated with human brain diseases. We start with a brief description of how the ubiquitous circadian rhythms are regulated at the genetic, cellular, and neural circuit levels, based on knowledge derived from extensive research on model organisms. We then summarize the latest findings from genetic studies of human brain disorders, focusing on the role of human clock gene variants in these diseases. Lastly, we discuss the impact of common dietary factors and medications on human circadian rhythms and advocate for a broader application of the concept of chronomedicine.

1. Introduction

The circadian clocks are endogenous timekeeping mechanisms that allow organisms to synchronize their physiology with the ever-changing environment and perform biological processes at the most appropriate time of the day (Koronowski and Sassone-Corsi, 2021). Circadian rhythms control various biological processes in living organisms, spanning from bacteria to humans (Bell-Pedersen et al., 2005). Perhaps the most apparent manifestation of circadian rhythms is the daily sleep and wake cycle of animals; however, many other fundamental neurophysiological processes are also under regulation by the circadian clock, from body temperature control and hormone secretion to cognition and mood (Alvord et al., 2021). In mammals, the suprachiasmatic nucleus (SCN) in the hypothalamus acts as the “master clock” for the brain and body, regulating their physiological rhythms (Hastings et al., 2018). The rhythmic outputs from the SCN are responsible for orchestrating rhythms in various body systems. The complex interaction between the core clock gene and its protein byproducts is responsible for circadian rhythms at the cellular level.

Disruption of circadian rhythms by genetic or environmental insults can lead to disorders of various physiological processes (Hastings et al., 2003; Maywood et al., 2006). Genetic variations in circadian clock genes are increasingly associated with a variety of common human diseases. Therefore, the link between circadian physiology and other physiological processes has implications for human physiology and medicine. Notably, humans in modern society are increasingly disregarding natural environmental cues that regulate our body’s internal clock, resulting in aberrant body rhythms and an increased risk of developing certain diseases. It is therefore important to understand how these perturbations in circadian rhythms may affect our physiology and susceptibility to disease, to identify pathogenic mechanisms underlying circadian disruption so that novel therapeutics can be developed based on that knowledge.

The current review assembles recent findings that have begun to untangle the molecular mechanisms by which circadian biology influences neurophysiological and behavioral processes connected with human brain illnesses. We begin with a quick overview of how the ubiquitous circadian rhythms are regulated at the genetic, cellular, and brain circuit levels, based on substantial research on model species. We then discuss the most recent discoveries from genetic investigations of human brain illnesses, with a focus on the role of human clock gene variations in these diseases. Finally, we address the effects of common dietary variables and drugs on human circadian rhythms and argue for a broader use of the chronomedicine concept.

2. Circadian timekeeping mechanisms in mammalian cells

Circadian rhythms can be found within cells. Rhythmic cellular processes are driven by rhythmic gene expression, which involves several interlocking transcription-translation feedback loops (TTFLs) consisting of approximately a dozen so-called “clock genes” in mammals. Recent work has found that the intracellular signaling network is coupled to the core clock gene feedback loops, establishing an interphase between the clock machinery and a myriad of intracellular processes. Thus, these cellular processes are regulated by the circadian clock and, in turn, provide feedback to the circadian timing process, synchronizing the cellular clock with cell physiology and metabolism.

2.1. The TTFLs that drive rhythmic gene expression in mammalian cells

The circadian clock system encompasses a cell-autonomous transcript feedback loop comprising a basic set of evolutionarily conserved genes in animals (Takahashi, 2017). During the daytime, the basic helix-loop-helix PAS domain, involving the transcription element NPAS2 (or CLOCK), interacts with aryl hydrocarbon receptor nuclear translocator-like (ARNTL1/BMAL1) to form heterodimers (Vanselow and Kramer, 2010). The CLOCK/BMAL1 heterodimer binds to the E-box enhancers and activates the transcription of the Per1/2/3 and Cry1/2 genes. The Per and Cry1 mRNAs are translated into proteins, which form multiple protein complexes with CSNKIe/d in the cytosol. Once these protein complexes accumulate to a certain level, they translocate into the cell nucleus, interact with Clock/Bmal1 complexes, and inhibit gene transcription, thereby forming a negative feedback loop (Lowrey and Takahashi, 2000). At night, the PER and CRY proteins are phosphorylated and degraded through ubiquitination-mediated mechanisms, allowing BMAL1/CLOCK to initiate a new repeating cycle (Lee et al., 2001). In addition to this feedback loop, the CLOCK/BMAL1 heterodimer also activates transcription of nuclear receptor subfamily 1 group D members 1/2 (NR1D1/2 or REV-ERBα/β) and the retinoic acid receptor-associated orphan receptors (RORs) α/β/γ. REV-ERBs inhibit, and RORs activate Bmal1 transcription, respectively, creating additional feedback loops (Preitner et al., 2002). REV-ERBs and the retinoic acid RORs α/β/γ inhibit or activate Bmal1 transcription through the ROR response elements (RREs), respectively, creating additional feedback loops (Preitner et al., 2002). Apart from these core feedback loops, many other proteins modulate the clock proteins at post-transcriptional levels, including protein kinases, phosphatases, ubiquitin ligases, etc. Together, these mechanisms collaboratively fine-tune the speed and robustness of the circadian clock (Figure 1).

Figure 1.

The circadian clock gene network in mammals is composed of overlapping transcriptional-translational feedback loops. A collection of fundamental clock genes form transcription-translation feedback loops in the circadian clock system. The circadian clock in mammals is made up of a major negative feedback loop that involves the genes Clock (and its paralog Npas2), Bmal1, Per1/2, and Cry1/2. CLOCK (or NPAS2) and BMAL1 are transcription factors with basic helix-loop-helix PAS-domains that stimulate transcription of the Per and Cry genes. The resultant PER and CRY proteins heterodimerize, translocate to the nucleus, and suppress transcription by interacting with the CLOCK-BMAL1 complex. After some time, the PER-CRY repressor complex degrades, allowing CLOCK-BMAL1 to initiate a new transcriptional cycle. Rev-Erb is a direct target of the CLOCK-BMAL1 transcription activator complex in the secondary autoregulatory feedback loop. REV-ERB binds RREs in the Bmal1 promoter and competes with an ROR to suppress Bmal1 transcription.

2.2. Intracellular signaling pathways that regulate clock gene expression

Cells respond to external stimuli and maintain intracellular homeostasis via various signal transduction pathways that couple extracellular and intracellular signals to gene expression. These signaling pathways regulate many aspects of cell biology and are essential for proper cell function. The cellular clock is constantly regulated by signals originating from outside the cell, which are transmitted through intracellular signaling pathways to regulate gene expression in the cell. Moreover, these signaling pathways can also interact with one another, forming a complex intracellular signaling network that forms an interface between the circadian timing process and cell physiology and metabolism. The known cell signaling pathways that regulate clock gene expression are summarized in Figure 2. Most of these discoveries were made in SCN neurons.

Figure 2.

Important signaling pathways and protein kinases have been demonstrated to function within the SCN. Several neurotransmitters and neuropeptides that affect SCN neurons can activate plasma membrane ion channels and receptors, resulting in intracellular signaling events. GPCR activation can signal through Gs, Gi, and Gq proteins to activate adenylyl cyclase (AC), inhibit AC, or activate PLC, respectively. AC increases cAMP synthesis, which activates PKA and EPAC (exchange proteins activated by cAMP). PKA and cAMP can both stimulate CREB-mediated transcription. The hydrolysis of PI (4,5) P2 into diacylglycerol (DAG) and IP3 is catalyzed by PLC.

DAG activates PKC at the plasma membrane, whereas IP3 diffuses into the cytosol and causes the ER to release intracellular Ca2+ reserves. Ionotropic receptors, G-protein-gated ion channels, voltage-gated ion channels, and RTKs can all cause an increase in cytosolic Ca2+ levels. The phosphorylation of the receptor by GRK2 results in the downregulation of GPCR signaling. The RAF, MEK1/2, and ERK1/2 pathways are activated at the plasma membrane via receptor-mediated Ras activation. PKC promotes Ras/MAPK signaling by activating Ras or deactivating RKIP-mediated Raf inhibition via phosphorylation of RKIP. MAPK/ERK stimulates RSK1 and MSK1, which induce CREB-mediated transcription. MAPK/ERK also acts as an upstream activator of mTOR, which increases translation by activating p70S6K and blocking 4E-BP1-mediated eIF4E repression. Ca2+-induced CaMKII activation can connect to cGMP synthesis via the NOS/GC pathway in terms of Ca2+ signaling. PKG is activated by cGMP, promoting CREB-mediated transcription. The NOS pathway also activates Dexras1, a small G protein that inhibits GPCR-mediated Gi activation and indirectly inhibits AC via ligand-independent Gi activation. Many upstream signaling events converge on the MAPK/ERK pathway, which plays a critical role in photic entrainment through its effects on CREB. SIK1 is a CRE-inducible gene that works as a feedback inhibitor of CREB signaling by suppressing CREB-dependent transcription coactivator 1 (CRTC1). Finally, various protein kinases, including CK1 and CK2, GSK3, SIK3, and GRK2, have been found to phosphorylate clock proteins PER and CRY.

Photic information from the retina is transmitted to the SCN and triggers the release of neurotransmitters such as glutamate and pituitary AC cyclase-activating peptide (PACAP), which in turn activates a series of intracellular signaling cascades. Among these signaling pathways, the MAPK/ERK pathway is a key signaling pathway that couples photic stimulation to gene expression in the SCN (Obrietan et al., 1998). This MAPK cascade involves sequential phosphorylation-induced activation of Raf, MEK1/2, and ERK1/2. During the early or late subjective night, a brief burst of light triggers a significant increase in ERK1/2 phosphorylation in the SCN (Coogan and Piggins, 2003; Obrietan et al., 1998), which is essential for coupling light to SCN clock entrainment (Butcher et al., 2002). Phosphorylated ERK1/2, in turn, activates MSKs and ribosomal S6 kinases (RSKs) to regulate gene expression and clock entrainment (Butcher et al., 2005; Cao et al., 2013a). MSK1 regulates the photic entrainment of the circadian clock by facilitating light-induced histone phosphorylation, CREB phosphorylation, and Per1 expression (Cao et al., 2013a). Dexras1, an upstream modulator of ERK1/2 and a Ras-like protein, enhances photic responses and inhibits the non-photic response of the circadian clock (Cheng et al., 2004). The Raf-1 kinase inhibitor protein (RKIP) also plays a role in modifying MAPK/ERK signaling in the suprachiasmatic circadian clock (Antoun et al., 2012). Furthermore, light at night triggers phosphorylation of cAMP response element-binding protein (CREB) at Ser133 and Ser142, activating CRE-dependent gene expression in the SCN (Gau et al., 2002; Ginty et al., 1993). The JNK cascade represents another major MAPK signaling pathway that is activated by light at night (but not during the day), indicating that JNK1 may also be associated with photic entrainment (Pizzio et al., 2003). Indeed, the knockdown of JNK3 has been shown to alter behavioral rhythms, resulting in longer free-running periods and impaired phase shifts in response to light (Yoshitane et al., 2012).

Various studies have established a connection between CaMKII and the photic entrainment process. Light pulses delivered during the subjective night rapidly induced CaMKII phosphorylation in the SCN under entrainment and free-running conditions (Agostino et al., 2004). The activation of the CaMKII upstream modulator by light plays an important role in the phase-shifting and induction of Per1/2 genes in the hamster suprachiasmatic circadian clock (Yokota et al., 2001). Administration of a centralized CaMKII inhibitor significantly attenuated phase delays and advances caused by bright light pulses in hamsters (Golombek and Ralph, 1994). Consistent with these effects, CaMKII antagonists attenuated glutamate-induced delays in SCN firing rates, and systemic administration of calmodulin antagonists inhibited light-induced expression of c-Fos in the rat SCN (Fukushima et al., 1997). GSK3 is also associated with photic entrainment in the SCN. Late-night light pulses activate GSK3β activity across the SCN, and inhibiting GSK3 eliminates the continuous increase in the electrical activity of the SCN in response to late-night light pulses (Paul et al., 2017).

In addition to transcriptional regulation, light can regulate mRNA translation in the SCN through translational control pathways. The mechanistic/mammalian target of the rapamycin (mTORC1) pathway is activated by light at night as a downstream event of ERK MAPK activation (Cao et al., 2008; Cao et al., 2010). Importantly, the clock-resetting effects of light are sensitive to rapamycin regulation, probably through the regulation of PER1/2 protein levels in the SCN. The mTORC1 pathway also regulates the intracellular coupling of SCN neurons through translational control of vasoactive intestinal peptide (VIP) via its target 4E-BP1(Cao et al., 2013b). VIP is an important neuropeptide that couples SCN cells. In mice lacking 4E-BP1, SCN exhibits larger Per2::Luc rhythms, and mice demonstrate enhanced entrainment to a shifted light/dark cycle. Conversely, in Mtor^+/− mice, VIP expression is decreased, and SCN cell synchrony is impaired (Liu et al., 2018). The mTOR signaling cascade also regulates cellular clock properties, such as period length and amplitude, in various cells and tissues, indicating a significant role of translational control in regulating autonomous clock properties. Another important translational control pathway tat is regulated by light is the MNK/eIF4E phosphorylation pathway, which is downstream of the ERK MAPK pathway. Light at night triggers MNK activation and eIF4E phosphorylation at Ser 209 in the SCN. In mice, where eIF4E Ser209 is mutated to alanine and thus cannot be phosphorylated, light pulse-induced clock resetting and the capability of clock entrainment to T-cycles are impaired. eIF4E phosphorylation facilitates mRNA translation of Per1 and Per2, which may underlie the photic entertainment of the SCN clock (Cao et al., 2015).

3. The master circadian pacemaker in mammals: the suprachiasmatic nucleus (SCN)

In mammals, the circadian system consists of a central pacemaker and peripheral oscillators found in almost all cell types in the body (Albrecht, 2012). The SCN’s primary input pathway is the retinohypothalamic tract (RHT), which originates from intrinsically photosensitive retinal ganglion cells. The SCN rhythmically regulates other brain regions as well as peripheral organs, through neuronal and humoral outputs. The pituitary gland is the only known portal system in the mammalian brain. In the mouse brain, a second portal pathway connects the capillary beds of the central brain clock in the SCN and the organum vasculosum of the lamina terminalis (OVLT) (Yao et al., 2021). Numerous physiological and behavioral processes exhibit daily rhythms, such as body temperature, blood pressure, hormone secretion, learning and memory, mood, etc.

3.1. The SCN neurons are heterogeneous and express different neuronal peptides

The SCN can be roughly divided into the ventrolateral “core” and the dorsomedial “shell” regions (Antle and Silver, 2005) (Figure 3). All SCN neurons are GABAergic; however, they express different neuropeptides. The ventrolateral SCN expresses VIP and gastrin-releasing peptide (GRP) and is directly responsive to photic input (Card et al., 1981; Mikkelsen et al., 1991). In contrast, the dorsolateral SCN expresses arginine vasopressin (AVP), responds secondarily to photic stimulation, and plays a crucial role in maintaining circadian rhythmicity (Antle and Silver, 2005). Extensive intracellular communication occurs within and between different SCN regions, allowing heterogeneous clock cells to synchronize with each other. VIP, expressed in the ventral SCN, plays an important role in the photic resetting of SCN oscillations (Aïoun et al., 1998). Application of VIP to ex-vivo SCN slices during the late subjective night advanced the phase of neuronal firing, stimulated the expression of Per1/2, and mimicked the effects of light at late night (Meyer-Spasche and Piggins, 2004; Nielsen et al., 2002). VIP also mediates circadian rhythmicity and synchrony in SCN neurons (Aton et al., 2005). The synchronization of SCN cells is disrupted in mice lacking Vip or Vpac2 (the gene encoding the receptor for VIP in the SCN), and animals show arrhythmic behavior in constant darkness. GRP transmits photic phase-shifting signals in the SCN by increasing the neurophysiological activity of SCN neurons (Kallingal and Mintz, 2014). Additionally, during the early and late nights, microinjection of GRP into the SCN induces Per1 expression throughout the dorsal and ventral SCN and behavioral phase shifts (Gamble et al., 2007). In contrast to VIP, AVP is thought to primarily function as an output signal of the SCN clock. Deletion of AVP receptors (V1a and V1b receptors) does not affect mouse free-running rhythms in constant darkness. However, these mice exhibit accelerated re-entrainment to shifted light/dark cycles, indicating the AVP signaling confers the SCN an intrinsic resistance to external perturbations (Yamaguchi, 2018). Prokineticin 2 (PK2) and its receptor (PKR2) are highly expressed in the SCN and are crucial for regulating circadian behavior (Li et al., 2018). However, overexpression of PK2 leads to a slight but significant reduction in transcript levels of Bmal1, Per1, Per2, and Cry2 in the SCN. Thus, PK2 signaling is necessary for the amplitude of rhythmic behavior, but it has minimal impact on the entrainment of the SCN and circadian pacemaking (Li et al., 2006; Li et al., 2018; Prosser et al., 2007).

Figure 3.

Neurochemical composition of the mammalian SCN. The SCN is divided into dorsal (shell) and ventral (core) regions. Most, if not all, of the ~20,000 SCN cells are GABAergic but differ in their neuropeptide content. AVP and VIP delineate the shell and core regions, respectively. GRP-expressing neurons are primarily localized in the medial core. PK2 is another SCN neuropeptide that is highly expressed by shell and core SCN neurons.

The regulation of SCN by glutamatergic signaling has been extensively studied. Both ionic and metabotropic glutamate receptors have been shown to modulate central pacemaker activity (Gannon and Rea, 1994). Glutamate application to SCN tissue explants has been shown to induce neuronal firing (Bos and Mirmiran, 1993), and optic nerve stimulation results in glutamate release (Liou et al., 1986). Similarly, the application of selective receptor antagonists can reversibly reduce SCN neuronal activity induced by RHT stimulation (Cahill and Menaker, 1989). A non-selective antagonist of excitatory amino acid receptors, cis-2,3-piperidine-dicarboxylic acid, can reversibly block the postsynaptic response evoked in the SCN by stimulation of the optic nerve. In addition, treatment with NMDA receptor antagonists perturbed light-induced nocturnal phase shifts in wheel-running activity as well as SCN neuronal activity (Colwell and Menaker, 1992). The administration of NMDA or AMPA receptor agonists during the subjective night recapped the phase-shifting effects of light in vitro (Shibata et al., 1994). Early-night glutamate treatment leading to phase delays involves the activation of cAMP/PKA signaling and increasing Per1 expression (Tischkau et al., 2000). In contrast, late-night glutamate stimulation-induced phase advances are mediated through cGMP/PKG signaling and are attenuated by cAMP/PKA activation (Tischkau et al., 2003). Brain-derived neurotrophic factor (BDNF) has also been shown to increase glutamate-induced phase shifts of neuronal activity in the SCN (Kim et al., 2006).

Besides glutamate, SCN neurons are regulated by PACAP and GABA. The application of PACAP to SCN tissue explants amplified the intensity of glutamate-induced phase delays of neuronal firing rhythms during the early night but inhibited glutamate-induced phase advances during the late night (Chen et al., 1999). While GABA is primarily recognized as an inhibitory neurotransmitter in the adult brain, it can also exert excitatory effects. GABA modulates photic and non-photic phase-shifting of the central clock by GABA_A and GABA_B receptors (Gillespie et al., 1997; Mintz et al., 2002). Almost all the SCN neurons are GABAergic (Moore and Speh, 1993). Studies have found that SCN neurons display excitatory GABAergic signaling to modulate light input and block GABAergic signaling to attenuate phase delays during the subjective early night (McNeill et al., 2018). In addition, GABAergic signaling is important for the synchrony of SCN neurons (Liu and Reppert, 2000). GABA can both phase shift and synchronize clock cells through GABA_A receptors. The GABAergic network comprises both excitatory and inhibitory connections (Freeman et al., 2013). Endogenous GABA_A signaling decreases the precision of circadian oscillations in SCN neuronal networks.

Besides receiving glutamatergic input from the RHT, the SCN, specifically its VIP-containing regions, is also innervated by serotonergic projections from the median raphe nuclei, via which serotonin facilitates the glutamate input during the daytime and inhibits it during the nighttime (Reghunandanan and Reghunandanan, 2006). Serotonin can modulate both photic and non-photic effects, including phase-shifting of SCN neuronal firing, locomotor activity rhythms, melatonin secretion, body temperature rhythms, and gene expression (Horikawa and Shibata, 2004; Horikawa et al., 2000; Paulus and Mintz, 2013; Smith et al., 2008). Acetylcholine (ACh) was one of the first neuromodulators presumed to play a key role in circadian rhythm regulation. Cholinergic signaling can modulate both photic and non-photic phase-shifting of the central clock (Basu et al., 2016; Yang et al., 2010). Glycine, the primary inhibitory neurotransmitter in the brain, can also exhibit excitatory transmission through allosteric activation of NMDA receptors (Johnson and Ascher, 1987). Glycine is capable of regulating the activity of clock neurons and resetting their rhythmic activity according to the stage of the daily cycle (Mordel et al., 2011). Glycine can also affect other aspects of SCN function by affecting NMDA receptors, particularly in the regulation of sleep and body temperature (Kawai et al., 2015).

3.2. An emerging role of glial cells in circadian timekeeping

Glial cells constitute a significant population of cells in the human nervous system. The main classes of glial cells include astrocytes, oligodendrocytes, and microglia. Among these, astrocytes are the most abundant and well-characterized type of glial cell in the mammalian brain, closely involved in the metabolic, circulatory, and neuromodulatory control of neurons (Ben Haim and Rowitch, 2017). Glial cells, in addition to neurons, also display circadian rhythms (Prolo et al., 2005). Among glial cells, astrocytes have been extensively studied in the context of circadian rhythms. Several studies have shown that the clocks within mammalian astrocytes can influence neuronal oscillators and subsequent behavioral rhythms, suggesting a harmonious interaction between SCN neurons and glial cells in shaping circadian rhythms. During the dark phase as opposed to the light phase, there is a reduced number of astrocytic processes connecting with synapses in the hippocampus, resulting in a slower removal of glutamate (McCauley et al., 2020). Cultured cortical astrocytes from mice exhibited Per1/2 expression, indicating that glial cells can show circadian rhythms in clock-gene expression (Prolo et al., 2005). A recent study revealed that SCN astrocytes constituted the other “half’ of the circadian clock by releasing glutamate and elevating extracellular glutamate levels during the subjective night (Brancaccio et al., 2017). Boosting extracellular glutamate levels activates pre-synaptic NR2C complexes on dorsomedial SCN neurons, increasing the inhibitory tone within the circuit and suppressing SCN neuronal activity (Brancaccio et al., 2017). During the subjective day, the decrease in glutamate release by astrocytes and the resulting decrease in extracellular glutamate levels attenuate the repression of the spontaneous activity of SCN neurons (Brancaccio et al., 2017). Co-culture of SCN neurons provides astrocytes with a slightly more stable rhythm, providing evidence for the mutually supportive effect between neurons and astrocytes on SCN rhythms (Brancaccio et al., 2017; Tso et al., 2017). Another study revealed that the knockdown of Bmal1 in astrocytes alters daily locomotor activity and cognitive function through GABA signaling (Barca-Mayo et al., 2017). One of the most impressive demonstrations of astrocyte potency is that it can autonomously encode circadian information and instruct its neuronal partners to initiate and indefinitely maintain circadian patterns of neuronal activity and behavior (Brancaccio et al., 2019).

The circadian clock regulates the activities of other glial cells, such as microglia and oligodendrocytes. Microglia, specialized immune cells in the brain, monitor the central nervous system through their intricate branching and dynamic projections (Wake et al., 2009). Deletion of Bmal1, a transcriptional activator of the circadian clock, in both neuronal and glial cells, induces severe age-dependent astrogliosis in the cortex and hippocampus, while preserving intact circadian and sleep-wake rhythms (Musiek et al., 2013). Deletion of Rev-erbα, a nuclear receptor and circadian clock component, leads to spontaneous activation of microglia in the hippocampus, increased expression of pro-inflammatory transcripts, and secondary astrogliosis (Griffin et al., 2019). Mice exposed to dim light at night exacerbate the inflammatory response of microglia with lipopolysaccharides (LPS) (Fonken et al., 2013). Moreover, SD rats injected with LPS in the light phase show robust hippocampal microglia production compared to rats injected in the dark phase (Fonken et al., 2015), indicating circadian regulation of microglia activation. Oligodendrocytes are crucial contributors to neuronal function and provide metabolic support to neurons (Lee et al., 2012). Bellesi and colleagues found that about 2% of oligodendrocyte genes fluctuate as a function of circadian time (Bellesi et al., 2013). Preclinical and clinical studies of oligodendrocytes are warranted to understand how circadian rhythms affect the maturation and function of oligodendrocytes (Colwell and Ghiani, 2020) and to aid in the development of new therapeutic strategies and standards of patient care.

3.3. The free-running property of the circadian clock

As aforementioned, the rhythmic expression of clock genes in cells is driven by self-sustained TTFLs (Rosbash, 2021). Thus, an important property of the circadian clock is that its daily oscillations are autonomous and do not rely on the input of environmental cues. This was originally demonstrated in the studies by French astronomer Jean-Jacques d’Ortous de Mairan, who found that the leaves of the Mimosa plant moved with a periodicity of 24-h, even when the plant was moved to a room of constant darkness (Roenneberg and Merrow, 2005). To demonstrate the existence of a functional circadian clock, constant environment is always used as an experimental condition. The term “zeitgeber” (time giver or time cue) refers to environmental variables that can act as circadian time cues. Zeitgeber time (ZT) means the time based on environmental cues, such as the 12h/12h light/dark cycles. Circadian time (CT), in contrast, means the time based on the endogenous biological clock when organisms are kept in free-running, constant conditions. The onset of activity of day-active organisms is defined as circadian time zero (CT0), and the onset of activity of night-active organisms is CT12 (Eelderink-Chen et al., 2015). For mouse experiments, constant darkness is usually required to reveal differences caused by the circadian clock. However, housing animals in constant darkness can cause difficulties in animal husbandry, and thus, ZTs are used in a lot of animal studies (Bittman et al., 2013).

4. Circadian rhythms in extra-SCN brain regions

The rhythmic expression of clock genes throughout the brain plays a critical role in regulating everyday brain functions (Adamantidis et al., 2007; Mohawk et al., 2012). The master pacemaker located in the SCN synchronizes circadian oscillators in extra-SCN brain regions, allowing neurophysiological processes to be orchestrated with external environmental cues, such as the light-dark cycle. The environmental signal is relayed to other brain regions through the SCN. Some of these receive direct innervation from the SCN pacemaker, while others get indirect input from the SCN through poorly understood neural or humoral mechanisms. The extra-SCN regions can modulate the activity of the SCN, contributing to the adaptability and flexibility of the circadian system (Begemann et al., 2020). To date, most studies on non-SCN clocks have focused on peripheral tissues (liver, muscle, tissue culture, etc.), and circadian oscillators in other brain regions are inadequately studied. Several extra-SCN brain oscillators have been characterized in terms of circadian regulation and output. However, their specific physiological functions and their roles in circadian clock regulation remain elusive.

4.1. Rhythmic gene expression in various brain regions

Although the SCN serves as the master pacemaker, rhythmic gene expression is also observed in many brain regions. The neuronal activity rhythms within the SCN modulate the rhythms in different brain areas via direct and indirect projections and the modulation of hormones, such as cortisol and melatonin (Logan et al., 2022a). However, not all brain regions share the same rhythmic gene expression pattern. The phase, robustness, and perhaps the circadian period differs among different brain regions, reflecting their distinct functions and anatomical connections.

The olfactory bulb (OB) is responsible for processing odor-related information. The SCN entrains but does not sustain the circadian rhythm of the OB (Granados-Fuentes et al., 2004). The mammalian brain includes a diverse set of Per1 circadian oscillators that are either SCN-dependent or SCN-independent and present in specific brain regions (Abraham et al., 2005). The OB contains a master circadian pacemaker that drives rhythms in the piriform cortex and interacts with SCNs to coordinate other daily behaviors in mice (Granados-Fuentes et al., 2006). VIP signaling regulates the output of OB to maintain circadian rhythms in the olfactory system of mammals (Miller et al., 2014).

The striatum is a part of the brain that is involved in dopamine (DA) signaling, movement, motivation, and reward. In psychiatric disorders, disruptions in the rhythmic and different expression of genes can lead to striatal dysfunction and psychosis(Ketchesin et al., 2023). These findings are consistent with the evidence of the regulation of core circadian clock genes in the striatum by DA (Hood et al., 2010; Imbesi et al., 2009). Variable restriction feeding disrupts the daily oscillations of Per2 expression in the limbic forebrain and striatum in rats (Verwey and Amir, 2012).

The prefrontal cortex (PFC), responsible for cognitive and executive functions, and the hippocampus, involved in learning and memory, show pronounced rhythmic patterns. The molecular rhythms of PFC undergo significant changes with age, which can impact cognition, sleep, and mood in later stages of life (Chen et al., 2016). The dorsolateral PFC plays a critical role in the core symptoms of psychiatric disorders and exhibits diurnal rhythms in gene expression, while most of the identified transcripts in the dorsolateral PFC are not rhythmic in schizophrenic patients (Seney et al., 2019). Thus, rhythmic genes might indirectly modulate PFC-dependent functions in schizophrenia. Robust sex differences in PFC shed light on the mechanisms underlying how neurotransmission and synaptic function are regulated in a sex-specific, circadian rhythm-dependent manner (Logan et al., 2022b).

Although the hippocampus and piriform cortex exhibit the highest levels of Per1 gene expression, they do not display intrinsic rhythmicity. In contrast, tissues with comparable (e.g., SCN and OB) or lower (e.g., arcuate nucleus) expression can oscillate in vitro with a cyclic pattern (Abe et al., 2002). The limbic forebrain, involved in the regulation of motivation and emotional states, shows rhythmic patterns of Per2 protein expression over time (Amir and Stewart, 2009). The Per2 gene in the hippocampus regulates long-term potential (LTP) and the recall of certain forms of learned behavior (Wang et al., 2009b) and might be modulated by the DA D1 receptor-PKA-CREB signaling pathway (Kim et al., 2022). Stress induces changes in Per1 protein expression in the limbic forebrain and hypothalamus of rats (Al-Safadi et al., 2014). The cAMP/MAPK/CREB signaling pathway in the hippocampus undergoes circadian oscillations, which correlate with the ability of mice to consolidate and maintain contextual memory (Eckel-Mahan et al., 2008b). The hippocampal-dependent long-term memory (LTM) depends on the SCN-regulated circadian oscillations of the MAPK/AC pathway in the hippocampus (Phan et al., 2011). The intrinsic excitability of the dentate gyrus (DG) and synaptic recruitment mediated by G-protein signaling are subject to circadian regulation (Gonzalez et al., 2023). Overall, rhythmic gene expression in various brain regions plays an important role in regulating the timing of physiological and behavioral processes, and its dysregulation is associated with various neurological and psychiatric disorders.

The circadian rhythm of Per2 protein expression reveals a novel SCN-controlled oscillator in the oval nucleus of the bed nucleus of the stria terminalis (BNST-OV) (Amir et al., 2004). The central (CeA) and basolateral nuclei of the amygdala (BLA) exhibit contrasting circadian rhythms of Per2 protein expression (Lamont et al., 2005). Moreover, thyroidectomy alters the daily pattern of Per2 protein expression in the BNST-OV and CeA in rats (Amir and Robinson, 2006). Circadian neurons in the paraventricular nucleus of the hypothalamus (PVN) synchronize and sustain the daily glucocorticoid rhythm (Jones et al., 2021). The differential transcriptome rhythms observed in the human striatum help to understand the normal functioning of circadian rhythms and the pathological consequences of their disruption (Ketchesin et al., 2021). Overall, the field of rhythmic gene expression in the brain is rapidly evolving, and further research is needed to fully understand the underlying mechanisms and the functional significance of these patterns in different brain regions.

4.2. Diurnal oscillation of neurophysiological processes

Rhythmic expression drives rhythmic neuronal activity. The activity of neuronal and glial populations changes throughout the day/night cycle, regulating sleep/wake rhythms, cognition, hunger, reward, and other functions (Frank, 2019; Prevot, 2022). Overall, the circadian system controls a diurnal rhythm that governs various neurophysiological processes in the body. Many neurophysiological processes exhibit robust diurnal oscillations, which are thought to be important for the normal functionality of the brain and mental well-being. It is assumed that circadian rhythms affect long-term memory formation. Memory is strongly influenced by the time of day of formation and retrieval (Chaudhury and Colwell, 2002; Eckel-Mahan et al., 2008a), and operations that disrupt the circadian system can significantly damage memory processes (Smarr et al., 2014). Memory oscillates in a diurnal cycle (Gerstner and Yin, 2010; Rawashdeh et al., 2014; Urban et al., 2021), usually peaking during the active phase in rodents, although this varies by task and laboratory (Liu et al., 2022; Tsao et al., 2022; Winocur and Hasher, 2004). Disruptions to the circadian rhythm, such as those caused by shift work or jet lag, can impair memory formation and impact cognitive performance. The time of day represents a crucial biological variable in both rodent and human studies (Nelson et al., 2021).

We provide a detailed list of findings related to the role of circadian cycles in rodents, with a particular focus on learning and memory (Table 1). In brief, the circadian rhythm plays a key role in processing learning and memory, but its interaction with different types of learning and memory can yield inconsistent results. For instance, classic studies using the passive avoidance task have demonstrated that nocturnal rodents performed better during daylight times compared to the nighttime (Davies et al., 1973, 1974; Sandman et al., 1971; Stephens et al., 1967). In contrast, nocturnal rodents showed peak performance in the Morris water maze (MWM) test of spatial memory when they were active in the dark phase (Valentinuzzi et al., 2004b). Similarly, diurnal grass rats demonstrated optimal performance in the MWM during the light period (active phase) (Martin-Fairey and Nunez, 2014b). These findings suggest that the circadian rhythm differentially affects key brain circuits, such as the hippocampus and amygdala, which play crucial roles in different tasks. McCauley and his colleagues have shown that the circadian rhythm affects the surface NMDA receptors and astrocyte-synapse interactions of hippocampal pyramidal neurons, which in turn affects learning that depends on the hippocampus (McCauley et al., 2020). The impact of the circadian rhythm on these circuits depends on the specific nature and demands of the tasks being performed. For instance, visually guided behavior in mice reaches its peak sensitivity and stability at night without affecting the retinal ganglion cells, demonstrating how the day/night cycle mediates search strategies for visual cues (Koskela et al., 2020). The expression of core clock genes in the SCN is similar between nocturnal and diurnal rodents but differs in downstream regions such as the hippocampus and amygdala (Martin-Fairey and Nunez, 2014b; Ramanathan et al., 2010; Wang et al., 2009a). The persistence of memory depends on the time of training; nocturnal rodents showed better retention when trained at night in spatial learning and operant learning tasks (Gritton et al., 2012a). This effect appears to be dependent on the daily corticosterone rhythm, as adrenalectomy eliminates circadian changes in conditioned fear extinction learning (Woodruff et al., 2015a). The circadian rhythm has a complex impact on the corticosterone level, which is an important factor in controlling synaptic strength within the hippocampus (McCauley et al., 2020). In contrast, mice trained during their inactive phase in cued fear conditioning acquired the conditioning faster and had longer recall compared to mice trained during the active phase (Price and Obrietan, 2018). However, extinction was more likely to occur during the dark phase (Chaudhury and Colwell, 2002). A 24-h rhythm is observed in passive avoidance memory, with performance being better at a 24-h interval compared to the intervals in between (Holloway and Wansley, 1973a; Holloway and Wansley, 1973b). The effects of motor activity and risk assessment behaviors on passive avoidance memory formation and applied behavioral strategies are temporal and sex-dependent (Meseguer Henarejos et al., 2020).

Table 1.

The circadian regulation of rodent behaviors.

Task	Animal	Treatment	Experimental Time	Intertrial interval	Results	References
Passive avoidance	Swiss Webster mouse	ZT time	12:12h LD, ZT9 or 17	24h	Light phase groups performed better than dark ones.	(Stephens et al., 1967)
Passive avoidance	Rat	ZT time; melanocyte - stimulating hormone	12:12h LD, ZT6 or 13	48h	Light phase groups performed better than dark ones. Melanocytes improved their performance in the dark.	(Sandman et al., 1971)
Passive avoidance	Sprague-Dawley rat	ZT time	12:12h LD, every 4 h	48h	Light phase groups performed better than dark ones.	(Davies et al., 1973)
Passive avoidance	Rat	ZT time; ITI	12:12h LD, ZT2, 8, 14, 20	15min, 6h, 12h, 18h, 24h, 30h, 36h, 48h, 72h	15min and all 24h groups performed better than others.	(Holloway and Wansley, 1973b)
Passive avoidance	Rat	ZT time; ITI	12:12h LD, ZT1-4.5 or 5.5-8	30sec-36h	30 sec, 15min, 12h, 24h, and 36h groups performed better than others.	(Wansley and Holloway, 1976)
Passive avoidance	Sprague-Dawley rat	ZT time; ITI	12:12h LD; ZT1-4 or 7-10	15min or every 3-h up to 24h	15min, 9h, 12h, and 24h groups performed better than others.	(Hunsicker and Mellgren, 1977)
Passive avoidance	Sprague-Dawley rat	SCN lesions; ITI	12:12h LD; ZT6-7	18h, 24h, or 30h	24h performance was better than others; SCN lesions did not follow this trend.	(Stephan and Kovacevic, 1978)
Passive avoidance	Rat	+12h, +6h, or −6h shifts	12:12h LD, ZT0.5-2.5	48-h or 7 days	Phase-shifted groups had impaired performance.	(Tapp and Holloway, 1981)
Passive avoidance	Wistar rat	+4h or −6h shifts	14:10h LD→DL or DD	24h, 48h, 96h, 144h	Phase-shifted groups had impaired performance.	(Fekete et al., 1985)
Passive avoidance	Wistar rat	−6h shifts; ACTH analog ORG-2766 or DGAVP	14:10h LD→DL	24h, 48h,168h	Phase-shifted groups had impaired performance; ACTH and DGAVP attenuated the deficits.	(Fekete et al., 1986)
Passive avoidance	Long-Evans rat	ZT time; aging	12:12h DL, ZT13 or 23	24h or 6 weeks	In the early dark phase, performance was better than in the late dark phase tested at 6 weeks, but not 24 h, later in the aged, but not young, group	(Winocur and Hasher, 1999)
Fear conditioning	C57BL/6J mouse	ZT time	12:12h LD; ZT2 or 14	Every 24-h up to 5 days	Dark phase groups performed better than light ones.	(Valentinuzzi et al., 2001)
Fear conditioning (tone-cued)	C-57/6J; C-3H; mouse	ZT time	12:12h LD or DL or DD, ZT3 or 15	N.A.	Light phase groups performed better than dark ones.	(Chaudhury and Colwell, 2002)
Fear conditioning	C57Bl/6 mouse	ZT time and time cues	12:12h LD, ZT4-8 or 16-20 (tested at the same or different times)	24h, 36h, or 48h	ZT4 groups had better performances than others.	(Eckel-Mahan et al., 2008b)
Fear conditioning (tone-cued)	Long-Evans rat	Every +3h shift for 6 days(one session) or 4 sessions	12:12h LD, ZT3	24h	No effects of phase shifts.	(Craig and McDonald, 2008)
Fear conditioning	C57Bl/6 mouse	Phase shift before or after the LD	12:12h LD, ZT3 or 6	1 to 7days	The phase-shifted group had impaired performance.	(Loh et al., 2010)
Fear conditioning	C57Bl/6 mouse	SCN lesions	12:12h LD	2weeks	SCN lesions impaired performance.	(Phan et al., 2011)
Fear conditioning (extinction)	Sprague-Dawley rat	ZT time	12:12h LD, ZT4 or 16	24h	Facilitated extinction more in the dark phase group than in the light one.	(Woodruff et al., 2015b)
Fear conditioning	C57BL/6J mouse	+12h shift or bifurcated 6:6h LD	12;12h LD, ZT12 or 24	12h or 24h	The Phase-shifted group showed impaired retrieval, while the bifurcated group showed impaired learning.	(Harrison et al., 2017)
Fear conditioning	C57Bl6/J mouse	ZT time; ITI; Bmal1 deletion	12:12h LD→DD, CT4 or 16	30min, 42h or 54h	The CT4 group had better long-term retrieval than CT16; Bmal1 deletion impaired performance.	(Price and Obrietan, 2018)
Fear conditioning (extinction)	Sprague-Dawley rat	ZT time; Per1/2 knockdown in PFC	12:12h LD, ZT4 or 16	24h	Facilitated extinction in the dark phase group; PFC Per1/2 knockdown eliminated this facilitation.	(Woodruff et al., 2018)
Fear conditioning (extinction)	C57Bl/6J mouse	+8h shifted across 1 or 2 days	12:12h LD, ZT13	24h or 48h	The phase-shifted group showed impaired fear of extinction.	(Clark et al., 2020)
Fear conditioning	Wistar rat	ZT time; DD	12:12h LD or DD, ZT		The DD group showed facilitated fear extinction, specifically during the light phase.	(Asadian et al., 2022)
Fear conditioning	Siberian hamster	−3 h shift	16:8h LD, ZT16-20	48h	The phase-shifted group showed impaired fear memory.	(Steiger et al., 2022)
MWM	Long-Evans rat	+3h shift	12:12h LD, ZT11	2 or 10 days (probe)	There were no effects on learning; the phase-shifted group showed impaired retention.	(Devan et al., 2001)
MWM (delayed non-match-to-sample)	Long-Evans rat	ZT time and time cues; aging	12:12h DL, ZT13 or 23	0-80sec	The early dark phase group performed better than the late dark phase group in the old animals, whereas the reverse was true for the young.	(Winocur and Hasher, 2004)
MWM	Wistar rat	ZT time	12:12h LD, ZT3 or 15	30 min	Dark phase groups performed better than light ones.	(Valentinuzzi et al., 2004a)
MWM	C57BL/6 mouse	LL	12:12h LD or LL	N.A. (no probe)	The LL group showed impaired learning more than the LD one.	(Fujioka et al., 2011)
MWM	Diurnal grass rat	ZT time	12:12h LD, ZT4 or 16	24h	There were no effects on learning; the light phase groups performed better than the dark ones.	(Martin-Fairey and Nunez, 2014a)
MWM	ICR mouse	Abnormal LD	12:12h, 3:5h or 5:3h LD	N.A.	Abnormal LD impaired the performance.	(Li et al., 2017)
MWM	Wistar Kyoto and SHR rat	LD disturbance	12:12h LD, ZT1-2	24h (probe)	The LD disturbance impaired the performance.	(Wang et al., 2020)
MWM	Wistar rat	LL or DD; fluoxetine	12:12h LD→ LL or DD, ZT0-3	24h	The LL group showed impaired performance, while fluoxetine reversed the deficits.	(Sharma et al., 2021)
MWM	Long-Evans rat	12:9h LD for 6 days	12:12h LD→12:9h LD, ZT10.5	18 days (probe)	The 21-day group showed impaired memory, but not learning.	(Deibel et al., 2022)
MWM	Wistar rat	ZT time; DD	12:12h LD or DD, ZT		The DD group showed impaired spatial memory.	(Asadian et al., 2022)
MWM	Non-Tg and 3xTg-AD mice	ZT time; 3xTg-AD	12:12h LD, ZT4 or 16	24h (probe)	The light group learned faster than the dark one, while the dark group had better reversal learning than the light one. AD impaired probe and reversal memory.	(Carvalho da Silva et al., 2022)
NOR	Siberian hamster	ZT time; arrhythmic circadian; PTZ (1 mg/kg, i.p. daily)	16:8h LD, ZT3, 7, 11, 15, 19, 23	60min	Early light groups had impaired performance in controls; the arrhythmic group showed impaired performance; PTZ facilitated performance	(Ruby et al., 2008)
NOR	Wistar rat	ZT time	12:12h LD, ZT2, 12, 14 or 20	Min or 24h	No effects in time-of-day.	(Takahashi et al., 2013)
NOR	Djungarian hamster	ZT time; arrhythmic circadian	14:10h LD, ZT4, 7, 13, 16, 19	1h	ZT13, 16, and 19 groups performed better than ZT4 and 7 groups, and arrhythmic circadian impaired performance.	(Müller et al., 2015)
NOR	Djungarian hamster	Arrhythmic circadian	14:10h LD	1h	Arrhythmic circadian impaired performance	(Weinert et al., 2016)
NOR	C57BL/6 mouse	ZT time; DD; SON lesions	12:12h LD, ZT0, 4, 8, 12, 16, 20	24h	Early light and SCN lesioned groups showed impaired performance.	(Shimizu et al., 2016)
NOR & object-displacement recognition	C57BL/6J mouse	ZT time; LL	12:12h LD, ZT6 or 18; 12:12h LD→LL, CT6 or 18	5min or 24h	The light phase group performed better than the dark phase one.	(Tam et al., 2017)
NOR	B6/C3H mouse	Ts65Dn; SCN lesions	12:12h LD, ZT6	24h	SCN lesions did not affect memory in controls; Ts65Dn mice had impaired performance, while SCN lesions alleviated it in Ts65Dn mice.	(Chuluun et al., 2020)
NOR (one object during the learning and removal in the test)	C57BL/6N Crl mouse	ZT time	12:12h LD or DD, ZT3.5 or 15.5	N.A.	The light phase groups performed better than the dark ones in learning, but not memory.	(McCauley et al., 2020)
NOR	C57BL/6 mouse	ZT time; dim light (ZT12-16)	12:12h LD, ZT2 or 14	12h	The light phase group performed better than dark phase one; dim light exposure reversed this pattern.	(Tam et al., 2021)
NOR	Wistar rat	LL or DD; fluoxetine	12:12h LD→ LL or DD, ZT0-3	24h	The LL group showed impaired memory, while fluoxetine reversed the deficits.	(Sharma et al., 2021)
NOR	C57BL/6J mouse	Cry1, Cry2 KO, AAV-Cry1	12:12h LD→DD, ZT20-22	24h	AAV-Cry1 expression in the SCN attenuated the deficits of Cry1-Cry2 KO.	(Maywood et al., 2021)
NOR	C57BL/6 mouse	ZT time	12:12h LD, ZT0, 6, 12, 18	12h or 24h	No effects of time of day (the dark phase groups seemed to perform better in complex object recognition).	(Gessner et al., 2022)
NLR	C57BL/6J mouse	ZT time; sex	12:12h LD→DD, CT4 or 16	30min	The dark phase group performed better than the light one in males, while the opposite effect was found in females.	(Goode et al., 2022)
Active avoidance	Sprague-Dawley rat	ZT time; ACTH	12:12h LD, ZT1.5-2.5 or 9-10	Variable ratio: 90 sec	The late light phase group performed better than the early light phase one; ACTH facilitated performance	(Pagano and Lovely, 1972)
Active avoidance (extinction)	Rat	ZT time; ITI	12:12h LD, ZT0-4, 7-10, 12-15, 18-21	15min, 6h, 12h, 18h, 24h	15min, 12h, and 24h groups had stronger resistance to extinction than the others.	(Holloway and Sturgis, 1976)
Active avoidance (extinction)	Rat	ZT time; pinealectomy	12;12h LD, ZT3 or 15	60sec	The dark phase group learned faster than the light one.	(Catalá et al., 1985)
Shock avoidance operant task	Sprague-Dawley rat	ZT time	12:12h LD or DL, ZT2, 6 or 10, 14, 18, 22	N.A.	Dark phase groups performed better in shock avoidance, but not responses, than light ones.	(Ghiselli and Patton, 1976)
CPA (foot shock)	Syrian hamster	ITI	14:10h LD,	24, 48, 32, or 40h	24 and 48h groups were more effective than the 32 and 40h groups.	(Cain et al., 2004a)
CPA (foot shock)	Syrian hamster	ZT time and time cues	14:10h LD, ZT4 or 13	24h	Non-matched LD cues impaired performance	(Cain et al., 2008)
CPA (foot shock)	Syrian hamster	ZT time and time cues; SCN lesions	14:10h LD, ZT2 or 11	24h	Non-matched LD cues impaired performance: SCN lesions did not affect it.	(Cain et al., 2012)
CPP (wheel running)	Syrian hamster	ZT time	14:10h LD, ZT4 or 13; LL, CT4 or 13	24h	The effects were paired with LD cues.	(Ralph et al., 2002)
CPP (food)	Rat	Time cue	12:12h LD, CT2 or 11 (tested at the same or different times)	24h	No effects of time cues.	(McDonald et al., 2002)
CPP (wheel running)	Syrian hamster	ZT time; SCN lesions	14:10h LD, ZT3 or 14	24h	The effects were paired with LD cues, and SCN was not needed.	(Ko et al., 2003)
CPP (food)	Wistar and Long-Evans rats	ZT time and time cues	12:12h LD, ZT3 or 11 (training at ZT11)	24h	No effects were observed in the Long-Evan rats; non-matched LD cues affected the Wistar rats.	(Cain et al., 2004b)
CPP	Syrian hamster	+6h shift every 3 days for 25 days	14:10h LD, ZT18	8 days	The phase-shifted group showed impaired performance.	(Gibson et al., 2010)
CPP (food)	Long-Evans rat	Every +3h shift for 6 days→LD 10days, repeated 4 times	12:12h LD	24h	The phase-shifted group showed impaired performance.	(McDonald et al., 2013)
CPP (morphine)	Wistar rat	ZT time and time cue	12:12h LD, ZT1-2 or 11-12	24h	The late-light phase group showed a stronger preference than the early-light ones.	(Khaksari et al., 2022)
Radial-arm maze	Sprague-Dawley rat	DL; theophylline (15mg/kg, i.p. daily)	12:12h LD or DL	7 or 14 days	The dark phase group performed better than the light one; theophylline facilitated performance only in the light phase.	(Hauber and Bareiß, 2001)
Radial-arm maze	C3H/HeN mouse	ZT time; Per1 depletion	12:12h LD, ZT2 or 14	24h	The light phase group performed better than the dark ones; Per1 depletion reversed this pattern.	(Rawashdeh et al., 2014)
T-maze operant task	Wistar rat	ZT time; SCN ablation	12:12h LD→DD, ZT3-4, or 10-11	Variable ratio 10 sec	Intact learning without LD cues and SCN.	(Mistlberger et al., 1996)
Spontaneous T-maze	Siberian hamster	ZT time; arrhythmic circadian; PTZ (0.3 or 1 mg/kg, i.p. daily)	16:8h LD, ZT3, 7, 11, 15, 19, 23	N.A.	Late light and dark phase groups performed better than early light ones; PTZ improved performanc e in the arrhythmic circadian group.	(Ruby et al., 2013)
Spontaneous T-maze	Siberian hamster	Arrhythmic circadian; SCN lesions	16:8h LD,	N.A.	Arrhythmic circadian lesions, but not SCN lesions, impaired performance.	(Fernandez et al., 2014)
Spontaneous T-maze	C57BL/6J mouse	ZT time; diet	12:12h LD, ZT2 or 14	N.A.	The dark phase group performed better than the light one; high fat and sucrose diets impaired it.	(Davis et al., 2020)
Spontaneous T-maze	C57BL/6J mouse	ZT time; diet	12:12h LD, ZT2 or 14	N.A.	The dark phase group performed better than the light one; a high-fat diet impaired it.	(Davis et al., 2021)
Spontaneous T-maze	C57BL/6J mouse	ZT time; diet	12:12h LD, ZT2 or 14	N.A.	The dark phase group performed better than the light one; a high-fat diet impaired it.	(Davis et al., 2021)
Spontaneous Y-maze	Wistar rat	LL or DD; fluoxetine	12:12h LD→ LL or DD, ZT0-3	N.A.	LL group showed impaired memory, while fluoxetine reversed the deficits.	(Sharma et al., 2021)
Spontaneous T-maze	C57BL/6J mouse	ZT time; Tg-SwDI	12:12h LD, ZT2 or 14	N.A.	The dark phase group performed better than the light one; the Tg-SwDI mouse had impaired performance.	(Fusilier et al., 2021)
6-point alley T-maze	C57Bl/6 mouse	ZT time	12:12h LD, ZT2, 6, 10, 14, 18, 22	N.A.	Dark phase groups performed better than light ones.	(Hoffmann and Balschun, 1992)
Spontaneous Y-maze	C57BL/6J mouse	SCN lesions	12:12h LD	N.A.	SCN lesions did not affect performance.	(Mulder et al., 2014)
Sustained attention operant task	Sprague-Dawley rat	ZT time	12:12h LD, ZT4 or 16	N.A.	The dark phase group performed better than the light one.	(Gritton et al., 2012b)
Sustained attention operant task	Sprague-Dawley rat	ZT time; SCN lesions	12:12h LD, ZT 4 or 16	N.A.	The dark phase group performed better than the light one; SCN lesions impaired performance at ZT4.	(Gritton et al., 2013)
Barnes maze	Mouse	ZT time; Bmal1 forebrain deletion	12:12h LD, ZT4 or 16	17days (probe)	Bmal1 deletion led to learning and recall deficits at both times.	(Price et al., 2016)
Barnes maze	C57BL/6J mouse	ZT time; elF4E or MNK depletion	12:12h LD, ZT6 or 18	24h (probe)	Dark phase groups performed better than light ones; elF4E and MNK depletions affected learning but not memory.	(Liu et al., 2022)
Social recognition	Djungarian hamster	Arrhythmic circadian; social ranking	14:10h LD	2min or 24h	The social subordinate, but not dominant, animals had intact 24-h memory; arrhythmic circadian impairment impaired the memory	(Weinert et al., 2016)
Social recognition	C57BL/6N mouse	ZT time; time cues; Bmal1 deletion	12:12h LD, ZT4 or 10	6h or 24h	ZT10 showed impaired retrieval of weak social memory in WT; Bmal1 deletion impaired memory at both times.	(Hasegawa et al., 2019)
Licking latency task	Rat	ITI	12:12h LD, ZT1.5-4 or 5.5-8	15min, 1-h or every 6h up to 36h	12h, 24h, and 36h groups performed better than others.	(Wansley and Holloway, 1975)
Circadian time-place three-arms task	C57BL/6J mouse	ZT time; Cry depletion	12:12h LD, ZT2-3.5, 5-6.5 or 8-9.5	3h	Animals were able to learn specific time-place rules, but the Cry gene knockout impaired it.	(Van der Zee et al., 2008)

Abbreviations: ZT: zeitgeber time; CT: circadian time; LD or DL: light-dark cycle; DD: dark-dark cycle; MWM: Morris water maze; N.A. not applicable or within seconds; CTA: conditioned taste aversion; CPP: conditioned place preference; CPA: conditioned place avoidance; ACTH: adrenocorticotropic hormone; DGAVP: desglycinamide⁻⁹-(Arg⁸)-vasopressin; PTZ: pentylenetetrazol; SHR: spontaneously hypertensive rat; PFC; prefrontal cortex; Tg-SwDI: the human APP gene with K670N/M671L, E693Q and D694N mutations; 3xTg AD: the TauP301L, APPSwe, and γ-secretase (PS1M146V); NOR, novel object recognition with 24 h training-testing interval; TMh, time memory to heat exposure; cTPL, circadian time place memory; CFC, contextual fear conditioning; NLR, novel location recognition; CR, conditioned reflex; Ts65Dn: segmental trisomic 16 on the B6/C3h background

Circadian rhythms regulate mating behavior in many mammals (Antle and Silver, 2016). Male rodents prioritize sexual behavior during the dark phase (Mahoney and Smale, 2005), and female rodents are most receptive to sexual behavior during this time. Sensitivity and perception also exhibit daily rhythms, with high pain thresholds often observed during the dark phase, when nocturnal animals are most likely to encounter painful stimuli (Palada et al., 2020). There is also a relationship between attention and the time of day (Paolone et al., 2012). The time of day can influence drug-seeking behavior and behavioral sensitivity to drugs, which are influenced by the circadian system. Additionally, drugs of abuse can provide feedback to regulate circadian rhythms (Sleipness et al., 2005, 2007; Uz et al., 2005; Webb, 2017). Similarly, nocturnal rodents are more active at night, and aggressive behavior shows predictable patterns of daily occurrence (Todd and Machado, 2019). Natural variation in maternal behavioral expression in laboratory rodents favors increased pup contact and pup-directed behavior during the light phase (Jensen Peña and Champagne, 2013). These processes can significantly impact sleep, cognition, hormone secretion, appetite, and digestion.

4.3. The role of biological sex in the circadian clock

Biological sex can impact multiple aspects of physiological functioning, such as circadian rhythm and clock regulation (Krizo and Mintz, 2015). In most laboratory rodent species, the estrous cycle is a 4-5-day rhythm with cyclical changes in vaginal cytology and reproductive hormones (McQuillan et al., 2019). Circulating estrogen and progesterone levels in female rodents alter circadian rhythms of physiology and behavior (Nakamura et al., 2008a). Estrogen directly influences the circadian rhythms of Per2 expression in the rat uterus (Nakamura et al., 2008b). Circadian rhythms in humans, rodents, and some non-human primates change during adolescence and puberty (Logan et al., 2018). Sleep-wake cycles and melatonin rhythms begin to phase delay with the onset of puberty (Roenneberg et al., 2004) and coincide with sexual maturity (Hagenauer and Lee, 2013). Sex-specific structural and functional variations are present in brain regions pertinent to circadian systems. For example, the sexually dimorphic role of the circadian clock genes in the stratum has been demonstrated in regulating alcohol-drinking behavior. Deletion of Bmal1 from medium spiny neurons of the striatum leads to increased alcohol intake in male mice but decreased intake in females (de Zavalia et al., 2023; de Zavalia et al., 2021). Also, sex-dependent characteristics, such as sexually dimorphic physiology, are linked to circadian rhythm disorders in male-dominant brain disorders (Alachkar et al., 2022). Female mice respond to light pulses with more significant phase shifts than male mice (Kuljis et al., 2013), but genetic ablation of ERα in female mice results in an amplified phase-shifting response to light (Blattner and Mahoney, 2013). Similar results have been found for testosterone, with gonadectomized male mice exhibiting more significant phase-shifting responses than non-gonadectomized male mice (Karatsoreos et al., 2011). The SCN is sexually dimorphic and enriched with estrogen receptors (Hatcher et al., 2020), but male mice have significantly higher SCN activation during the day (Bailey and Silver, 2014). In most studies, the gross anatomy of the SCN is similar in females and males, but there are sex differences in general SCN morphology. The shape of the human SCN differs by gender, with the anteroposterior axis 40% longer in women and the mediolateral axis 35% more spherical in men (Fernández-Guasti et al., 2000). The developmental patterns of VIP and AVP neurons in the SCN also show sex differences in mice (Carmona-Alcocer et al., 2023). AVP signaling resets SCN molecular rhythms in a sexually dimorphic manner (Rohr et al., 2022). Androgen, estrogen, and progesterone receptors are all expressed in the human SCN, with men having more androgen receptors and women having more ESR1 (Fernández-Guasti et al., 2000). In ovariectomized female mice, estrogen receptor subtype 1 (ESR1) and ESR2 agonists can mimic the period-shortening effects of estradiol, with ESR1 agonists effective at lower doses (Royston et al., 2014). In contrast, androgen activation does not affect the period in rodents (McGinnis et al., 2007). Executing the influence of biological sex on circadian rhythm is essential for multiple domains, such as sleep medicine, chronobiology, and customized healthcare.

5. Circadian neurogenetics and brain disorders

Circadian rhythm disruption is associated with an increased risk of developing brain diseases. Circadian disruption can occur due to various factors, such as shift work, jet lag, irregular sleep schedules, exposure to artificial light at night, and certain medical conditions. The high prevalence of circadian rhythm disruption in neurodegenerative and psychiatric illnesses provides evidence for the brain’s susceptibility to clock dysfunction and its direct involvement in disease. Different diseases vary in severity, medications differ in specificity and sensitivity, and the instruments used to measure circadian rhythm disruptions in clinical settings also vary, making it challenging to characterize and understand these disruptions. Sleep disorders and circadian disruptions are linked to several neurodevelopmental disorders, including attention deficit hyperactivity disorder (ADHD), Alzheimer’s disease (AD), schizophrenia, and autism spectrum disorders (ASDs) (Logan and McClung, 2019). The investigation of molecular clock mechanisms in psychiatric diseases is gaining momentum, as evidence suggests that misalignment between the endogenous circadian rhythm and the sleep-wake cycle may contribute to various psychiatric disorders. Much of the data relating circadian disruption to human brain illnesses is correlational, given the challenges in establishing causal relationships in human studies. Thus, animal models are frequently used to study circadian disruption in the pathogenesis of disease. Obtaining such knowledge is important to advance our understanding of the disease’s pathogenic mechanisms and develop new therapeutic strategies.

5.1. Human sleep disorders associated with clock gene mutations

Rare gene variants have been reported to cause sleep conditions characterized by extreme sleep duration or timing (Weedon et al., 2022). Familial advanced sleep phase syndrome (FASPS) is a sleep disorder in which a person’s sleep-wake cycle is earlier than the typical sleep-wake cycle (Ashbrook et al., 2020). FASPS is a familial disorder that appears to be isolated as a single gene with an autosomal dominant pattern of inheritance (Reid et al., 2001). A variation in human sleep behavior can be attributed to a missense mutation in the human PER2 gene, which alters the circadian period (Toh et al., 2001). A human PER3 variant causes a circadian phenotype and is associated with a seasonal mood trait (Zhang et al., 2016). A mutation in the human CRY2 gene produces an advanced sleep phase (Hirano et al., 2016), while TIM mutations alter the phase response and lead to an advanced sleep phase as well (Kurien et al., 2019). The stability of PER2 is regulated by casein kinase (CK)1-dependent phosphorylation of residues with FASPS (Shanware et al., 2011). Another study has identified a missense mutation (T44A) in the human CKIdelta gene in the population of FASPS (Xu et al., 2005). Transgenic Drosophila carrying the human CKIdelta-T44A mutation showed a lengthened circadian period, but transgenic mice carrying the same mutation have a shorter circadian period.

Familial delayed sleep phase syndrome (FDSPS) is a delay in a person’s circadian rhythm. FDSPS is associated with a dominant coding variation in the human circadian clock gene CRY1 (Patke et al., 2017). The CRY1 variant is a gain-of-function mutation that causes lengthening of the period of molecular circadian rhythms. Structural polymorphisms of the human PER3 gene have been related to the pathogenesis of FDSPS (Archer et al., 2003; Ebisawa et al., 2001). The human CLOCK gene variation has also been associated with FDSPS (Ebisawa, 2007). Functional alterations induced by missense variants in the human CKIɛ gene have been observed and exhibit an inverse correlation with circadian rhythm sleep disorders (Takano et al., 2004).

Natural short sleep is a rare genetic trait in which individuals sleep less than average hours without daytime sleepiness or other sleep deprivation consequences. A rare mutation in the human β1AR gene (ADRB1) has been found to be associated with natural short sleep (Shi et al., 2019), and mutations in the GRM1 gene have been identified as contributors to the natural short sleep trait (Shi et al., 2021). Short sleep duration has also been reported to be caused by mutations in the DEC2/BHLHE41 gene (He et al., 2009). Mutations in the NPSR1 gene reduce sleep duration while preserving memory consolidation (Xing et al., 2019). Treatment for this condition may involve modifying sleep habits, such as maintaining a regular sleep schedule and avoiding exposure to bright light in the evening.

5.2. Clock gene disruption and other brain disorders

A mood disorder is a mental health condition that primarily affects emotional state. Circadian rhythm disturbances are common in individuals with mood disorders (Vadnie and McClung, 2017). Disruptions to the circadian rhythm, such as those caused by irregular sleep patterns, shift work, or jet lag, can lead to a range of negative health outcomes, including depression and bipolar disorder (Logan and McClung, 2019; Walker et al., 2020). Kumari and colleagues have shown that alterations in the light-dark cycle in diurnal and nocturnal rodents have different effects on the endocrine, inflammatory, and antioxidant systems associated with depressive-like behavior. Constant light adversely affects both species, but constant darkness primarily negatively affects diurnal squirrels (Kumari et al., 2021). Sleep disorders in major depression disorder (MDD) include increased sleep latency, decreased latency to the first REM sleep episode, and early morning wakeup, although little is known about the role of circadian clock genes in MDD. In individuals with depression, circadian dysfunction may manifest as sleep disturbances, including insomnia or hypersomnia, and changes in appetite and energy levels (Crouse et al., 2021; Germain and Kupfer, 2008). Similarly, individuals with bipolar disorder may experience disruptions to their sleep-wake cycles during episodes of mania or depression (McClung, 2013; Singla et al., 2022). Evening scores are higher in bipolar disorder (BPD) and schizophrenia/schizoaffective patients than in controls. This finding appears to be connected to age in BPD patients (i.e., younger BPD patients are more intense “evening types”), whereas schizophrenia/schizoaffective people tend to show stronger eveningness at all ages. Being identified as an “evening type” could explain some of the sleep difficulties described by BPD patients, as well as the increased severity of BPD, as demonstrated by a younger age at treatment initiation, a higher incidence of self-reported fast mood swings, and rapid-cycling mood changes (Küng et al., 2019). Research suggests that interventions aiming to regulate circadian rhythms, such as chronotherapy, may be effective in treating mood disorders (Ruan et al., 2021). In addition, maintaining regular sleep patterns and avoiding disruptions in circadian rhythms may help prevent the onset or worsening of these diseases. The circadian pattern of the human brain in MDD identifies different genes that may be involved in important circadian events, including sleep/wake cycles and metabolism (Li et al., 2013). Diurnal variation and sex differences in hippocampal physiology are associated with neurological disorders and present differently in men and women (Goode et al., 2022).

Schizophrenia is a chronic mental illness characterized by positive symptoms (distortions of normal function, such as delusions and hallucinations), negative symptoms (absence of normal behaviors, such as anhedonia, asocialness, and blunted affect), and cognitive deficits. Some studies have shown disruptions of circadian rhythms in people with schizophrenia, including disturbances in the sleep-wake cycle (Kaskie et al., 2017), changes in the timing of melatonin secretion (Duan et al., 2021; Monteleone et al., 1997), and altered activity patterns (Qi et al., 2022; Zhang et al., 2021). These disturbances are thought to contribute to the cognitive and emotional symptoms of schizophrenia, such as attention and memory deficits, as well as mood disorders (Mansour et al., 2006). The dorsolateral prefrontal cortex (PFC), known to be crucial for schizophrenia’s fundamental symptoms (Smucny et al., 2022), is believed to have molecular cycles that are disrupted by circadian rhythms. Research has shown that the evening chronotype may increase the risk of psychiatric problems (Zou et al., 2022). Overall, while the relationship between circadian dysfunction and schizophrenia is still being studied, there is growing evidence that disruptions in internal clocks may play a role in its development and progression.

Neurodevelopmental disorders affect the development of the nervous system, leading to abnormal brain function that may affect emotion, memory, self-control, and learning ability. One of the most common neurodevelopmental disorders of childhood, ADHD, is characterized by inattentiveness, impulsivity, and hyperactivity and is associated with high rates of co-occurring sleep problems and circadian alterations (Philipsen et al., 2006). Reductions in sleep quality, delays in the circadian phase, and evening preference are consistently reported in children and adults with ADHD (Coogan and McGowan, 2017), and these may be correlated with the severity of ADHD symptoms (Rybak et al., 2007). Abnormal melatonin rhythms are also reported in children with ADHD (Coogan and McGowan, 2017). In adults with ADHD, a loss of circadian gene expression rhythms in the oral mucosa accompanies delays in cortisol rhythm and reduced amplitude melatonin rhythms (Baird et al., 2012). Advancing melatonin rhythms using bright light therapy improves ADHD symptoms of hyperactivity and impulsivity in adults, independent of changes in sleep time, sleep efficiency, wake time, or waking during sleep (Fargason et al., 2017). There is growing evidence to suggest that individuals with ASD may also experience circadian dysfunction (Lorsung et al., 2021). It is well-recognized that individuals with ASD are more likely to have sleep problems compared to the general population, such as difficulty falling and staying asleep, and waking up prematurely (Koo et al., 2021; Lorsung et al., 2021). This may be due to an imbalance in melatonin secretion, which is regulated by the circadian rhythm system. Sleep problems are very common in children with ASD (Cortesi et al., 2010; Richdale and Schreck, 2009), although identifying underlying circadian rhythm abnormalities has proven challenging (Takase et al., 1998). The most consistent correlation found in prepubertal children, pubertal adolescents, and young adults with ASD is a decrease in melatonin levels at night (Kulman et al., 2000; Tordjman et al., 2005). It is hypothesized that this decrease in melatonin during early development leads to the accumulation of oxidative stress, which is harmful to the developing nervous system and increases the risk of ASD (Jin et al., 2018). Studies have shown that melatonin levels are often disrupted in people with ASD, which can lead to sleep problems and impaired cognitive function (Wu et al., 2020). Additionally, some research suggests that individuals with ASD may have altered patterns of structured activity (Okamoto et al., 2018; Postema et al., 2019). This can contribute to circadian dysfunction and further impact sleep, mood, and behavior in people with ASD. Several studies have investigated the use of timed light or melatonin supplements to help regulate circadian rhythms and improve sleep in people with ASD (Cheng et al., 2021; Fatemeh et al., 2022). Addressing circadian dysfunction in patients with ASD may be an important focus of intervention.

Neurodegenerative diseases involve progressive deterioration in the structure and function of the central or peripheral nervous systems. In healthy older adults, individuals with AD or PD exhibit significantly reduced melatonin rhythm amplitude and excessive sleepiness, as well as other sleep-wake cycle disorders, such as later sleep episodes (Gros and Videnovic, 2017; Uchida et al., 1996; Videnovic and Golombek, 2017). PD is strongly associated with REM sleep behavior disorder, excessive daytime sleepiness, insomnia, and restless legs syndrome (Barone et al., 2009; Chaudhuri et al., 2006). In fact, more than 80% of people with REM sleep behavior disorder are later diagnosed with PD or dementia (Iranzo et al., 2014; Schenck et al., 2013). Transgenic mice overexpressing α-syn exhibit reduced amplitude and greater fragmentation of locomotor activity rhythms during the active phase, which worsens with age (Kudo et al., 2011). In addition, SCN neuronal firing is reduced in early adulthood and gradually progresses with age, suggesting that the circadian pacemaker is weakened (Kudo et al., 2011) and that dopaminergic regulation of circadian rhythms may be lost through the striatum, midbrain, and SCN circuits (Grippo et al., 2017; Korshunov et al., 2017). In a cohort study of 239 patients with PD, sleep latency was significantly increased, sleep efficiency and REM were reduced along with melatonin levels, and there was a loss of BMAL1 serum expression level (Breen et al., 2014). In another study, circadian melatonin rhythm dysfunction led to excessive daytime sleepiness in PD (Videnovic et al., 2014). Extended-release melatonin is an effective treatment option for PD patients with poor sleep quality (Ahn et al., 2020). Dr. Alois Alzheimer described AD as a type of memory loss or dementia (Cipriani et al., 2011). Memory issues may initially manifest as recurrent or chronological experiences, but over time they can affect various aspects of daily functioning. Neurologists have long recognized the existence of different subtypes of AD, such as posterior cortical and logopedic aphasic variations (Elahi and Miller, 2017). Cognitive impairment and dementia are also associated with sleep and circadian rhythm disruption, and the intensity of sleep disruption is linked to the severity of dementia (Winsky-Sommerer et al., 2019). Bmal1 KO mice exhibit premature aging, neurodegeneration, and a reduced lifespan (Musiek et al., 2013). One study found that even a single night of sleep deprivation led to Aβ accumulation in the right hippocampus and thalamus of the human brain (Shokri-Kojori et al., 2018). However, these increases were negatively correlated with mood but not with a specific AD-associated genotype (APOE genotype). In a triple transgenic mouse model of AD, there was a simultaneous alteration of the diurnal variation of hippocampal synaptic plasticity, memory, and metabolism (Carvalho da Silva et al., 2022). A comparative study demonstrated that AD patients had reduced diurnal activity, increased nocturnal activity, and phase delay in body temperature rhythms, and sunset severity was positively correlated with lower amplitude and more phase delay rhythms (Volicer et al., 2001). A clinical study involving 189 elderly residents in group care facilities showed that regular light exposure and melatonin supplementation improved cognitive symptoms of dementia and reduced aggression (Riemersma-van der Lek et al., 2008). However, further research is needed to fully understand the nature and extent of circadian dysfunction in individuals with neurodegenerative disorders and to develop effective interventions.

6. Chronomedicine or Chronotherapy

Chronomedicine, or circadian medicine, is the branch of medicine that studies the effects of biorhythms and time-dependent processes on health and disease. Chronomedicine can be conceptualized as dealing with the prevention, causation, diagnosis, and treatment of human diseases, with a particular focus on the role that time plays in our physiology, endocrinology, metabolism, and behavior at various organizational levels. Circadian medicine is increasingly regarded as an important category in clinical practice, as basic and translational research demonstrates the fundamental significance of circadian rhythms in human health and disease. The function of the local clock supports the temporal optimization of memory processes, elucidating the potential of circadian therapeutic strategies in preventing and treating memory impairment (Hartsock and Spencer, 2020).

6.1. Food supplements that can influence the human circadian clock

Caffeine is the most widely consumed psychoactive drug in the world. Its wakefulness-inducing effects and ability to disrupt sleep have been well-established (Temple et al., 2017). Caffeine has been found to alter circadian clocks in mammalian cells in vitro, in mice ex vivo, and in vivo (Oike et al., 2011). Caffeine exerts its effects by blocking adenosine receptors in the brain (Reichert et al., 2022). Extracellular adenosine levels in different brain areas in rats were shown to increase during wakefulness and decrease during sleep (Huston et al., 1996). While caffeine does not entrain the biological clock, it can improve daytime alertness with non-24-hour rhythms in blind individuals (St Hilaire and Lockley, 2015). In a clinical study, caffeine was shown to reduce sleep efficiency, sleep duration, slow-wave sleep (SWS), and REM sleep during daytime recovery sleep (Carrier et al., 2009). Epidemiological studies have also demonstrated that caffeine affects the timing of circadian rhythms in humans (Burke et al., 2015; Landolt, 2015; Reichert et al., 2022). CLOCK gene polymorphisms play an important role in determining the response to caffeine therapy in premature neonates with apnea (Guo et al., 2021). Resetting the late timing of ‘night owls’ has been shown to have a positive impact on performance, mental health, and sleep timing in real-world settings (Facer-Childs et al., 2019). Similarly, to caffeine, polyphenols found in green and oolong tea have been found to alleviate circadian rhythm disturbances and improve the microbiome and overall health (Guo et al., 2019; Zhang et al., 2020). Probiotics, which are beneficial bacteria (Matenchuk et al., 2020; Takada et al., 2017), vitamin D, an important nutrient (Fallah et al., 2020; Majid et al., 2018), and magnesium, an essential mineral (Abbasi et al., 2012; van Ooijen and O’Neill, 2016), have also been shown to influence various physiological processes in the body, including the regulation of the circadian clock.

The pineal hormone melatonin is known to reset the circadian clock in animal and human studies (Carriedo-Diez et al., 2022). Exogenous melatonin normalizes circadian variation in human physiology and strongly impacts general health, especially in the elderly and shift workers (Kunz et al., 2004). Recently, melatonergic agents such as Circadin, Ramelteon (TAK-375), Tasimelteon, PD-6735 (TIK-301), and combined melatonergic/serotonergic drugs such as Agomelatine have become available in the clinic to treat insomnia and depression (Naveed et al., 2022). Tasimelteon entrained non-24-hour sleep-wake disorder in completely blind people, but continued treatment was necessary to maintain these improvements (Lockley et al., 2015). In addition, Tasimelteon had no clinically significant effect on next-day driving performance in healthy adults (Torres et al., 2022). In an animal study, idiopathic REM sleep behavior disorder was associated with altered clock gene expression and delayed melatonin secretion (Weissová et al., 2018). High-dose melatonin increases sleep duration during nighttime and daytime sleep episodes in older adults (Duffy et al., 2022). Young children may be highly sensitive to light one hour before bedtime, suggesting that the home lighting environment and its effect on circadian rhythm timing should be considered as a possible factor in behavioral sleep difficulties (Hartstein et al., 2023). The combination of bright light exposure in the early morning and exogenous melatonin in the evening may provide the greatest phase-shift treatment response (Burke et al., 2013). Yet small changes in ordinary light in the evening can significantly affect plasma melatonin concentrations and the entrainment phase of human circadian pacemakers (Zeitzer et al., 2000). In one clinical study, melatonin upregulated Bmal1 expression in AD patients, but Per1 levels remained unchanged (Delgado-Lara et al., 2020). Furthermore, the time of administration may influence the physiological response to melatonin (Lok et al., 2019).

6.2. Dosing time matters for pharmacokinetics

Pharmacokinetics (PK) deals with the movement of drugs in the body. The circadian clock significantly regulates PK depending on the time of day, a discovery that led to the concept of chronomedicine. Drugs can affect the body’s clock function, and these side effects may complicate their use for certain diseases. Severe alterations in circadian rhythms were observed in mice and human patients receiving chemotherapeutic drugs (Ortiz-Tudela et al., 2014), which can be divided into two categories: (i) patients who exhibited unintentional side effects or non-specific toxicity, leading to clock resetting; and (ii) targeted timing drugs. Growing evidence indicates that the effects of drugs are dependent on circadian timing (Table 2). In a cohort study of 14,840 patients with self-poisoning in Sri Lanka, a close link between the diurnal variation of poisoning and death was highlighted. If the poisoning occurred in the late morning, the likelihood of death was three times higher than if it happened in the late afternoon and evening. This difference does not appear to be explained by treatment but is thought to be influenced by intestinal P. glycoprotein (P-gp) and cytochrome P450 3A4 (CYP3A4) rhythms (Carroll et al., 2012).

Table 2.

Clinical studies on the morning vs. evening schedule of drug administration (≥ 50 subjects).

Drugs	Subjects	Type of Study	Drug-Delivery Time	Conclusions	References
Simvastatin,	172 (33 male, 117 female) random	Double blind, placebo-controlled study	Evening vs Morning	Evening is more effective than morning	(Saito et al., 1991)
Simvastatin,	57 (27 men and 33 women), mean age 66	Randomized controlled trial	Evening vs Morning	Evening is more effective than morning	(Wallace et al., 2003)
Valsartan, an ARBs	90 subjects (30 men and 60 women),	Clinical trial	Morning vs Evening	No difference	(Hermida et al., 2003)
Captopril, an ACEIs	121 (75 males, 46 females)	Prospective, randomized, double-blind, placebo-controlled study	Evening	Restored the diurnal BP rhythm and decreased the elevated night/day BP ratio at bedtime administration	(Qiu et al., 2005)
Phenytoin and carbamazepine	148 () 18-65 age	Comparative study	Evening	Improved the response of diurnally active epileptic patients not responding to standard doses at bedtime administration	(Yegnanarayan et al., 2006)
Telmisartan, ACEIs, and ARBs	215 (114 men and 101 women) (46.4+/−12.0 years)	Randomized, double-blind, placebo-controlled trial	Evening vs Morning	Reduced BP during sleep at bedtime administration	(Hermida et al., 2007)
Prednisone, a corticosteroid	288 (54·6 (11·2) 55·4 (41 men and 247 women),	Randomized, double-blind	Modified vs immediate	Reduced morning stiffness of joints by modified release	(Buttgereit et al., 2008)
Torsemide, a diuretic	113 (44 men and 69 women) (51.7+/−10.6 years)	Randomized, double-blind	Evening vs Morning	More effective in lowering BP in patients with uncomplicated essential hypertension at bedtime	(Hermida et al., 2008b)
Nifedipine, a CCBs	180 (52.5 +/−10.7 years) (86 men and 94 women)	Randomized, double-blind	Evening vs Morning	Significantly reduced the bedtime edema at bedtime administration	(Hermida et al., 2008c)
ACIE, ARB, CCB	250 (136 men and 114 women), 60.1±11.7 years of age	Randomized, open-label, blinded endpoint (PROBE), parallel-group trial	Evening vs Morning	A larger reduction in BP at bedtime administration	(Hermida et al., 2008a)
Vaccination (hepatitis A and influenza A)	164 (men and women)	Clinical study	Morning vs Afternoon	Higher antibody response to both the hepatitis A and influenza A vaccines at morning vaccination	(Phillips et al., 2008)
Olmesartan ACEIs and ARBs	123 (39 men and 84 women) (46.6+/−12.3 years)	Randomized, double-blind	Evening vs Morning	Improved wake/asleep BP ratio at bedtime administration	(Hermida et al., 2009)
Ramipril ACEIs and ARBs	115 (52 men and 63 women) (46.7+/−11.2 years)	Randomized, double-blind	Evening vs Morning	Better nocturnal BP regulation at bedtime	(Hermida and Ayala, 2009)
Amlodipine/valsartan, a CCBs and ARBs	203 (92 men/111 women), 56.7 +/− 12.5 years	Single/Combined	Evening vs Morning	Improved the efficacy of lowering sleep BP and increasing sleep duration at bedtime administration	(Hermida et al., 2010b)
ARBs, ACEIs, CCBs, β-blockers, and diuretics	2156 (1044 men/1112 women),	MAPEC study	Evening vs Morning	The progressive decrease in asleep BP and increase in sleep-time were best achieved at bedtime therapy.	(Hermida et al., 2010c)
Spirapril, an ACEIs	165 (65 men/100 women), 42.5 ± 13.9 [mean ± SD] yrs. of age)	Open-label, parallel-group, blinded-endpoint	Evening vs Morning	Better sleep-time BP regulation at bedtime administration	(Hermida et al., 2010a)
Valsartan/Hydrochlorothiazide, a diuretic, and ARBs	204 (95 men/109 women), (49.7 ± 11.1 years)	An open-label, blinded endpoint	Evening vs Morning	Improved efficacy in lowering asleep BP and increased sleep time at bedtime administration	(Hermida et al.,2011b)
Ezetimibe/simvastatin, statins	171 (87 in the morning and 84 in the evening administration group) > 18 years	Randomized, crossover	Morning vs Evening	Morning administration was not inferior in lowering LDL-C	(Yoon et al., 2011)
Antihypertensive drugs (ARBs e.g., valsartan; the ACEIs e.g., ramipril; and CCBs, e.g., amlodipine, etc.)	661 (396 men/265 women), (59.2 ± 13.5 years)	PROBE trial	Evening vs Morning	Improved BP control and reduced risk of cardiovascular events at bedtime	(Hermida et al., 2011a)
Aspirin, COX inhibitor	350 pregnant women (30.7 ± 5.3 years)	Randomized, double-blind, placebo-controlled trial	Evening vs Morning	BP regulation and pregnancy outcome in high-risk pregnant women at bedtime administration	(Ayala et al., 2013)
Prednisolone, a steroid	350 Primarily female (18–80 years)	Randomized, double-blind, placebo-controlled trial	Evening vs Morning	Better reduction of morning stiffness of joints at bedtime administration	(Buttgereit et al., 2013)
BP lowering drugs	2012 (976 men and 1,036 women), (52.7 ± 13.6 years)	Prospective, PROBE trial	Evening vs Morning	Improved ambulatory BP control and reduced the risk of new-onset diabetes at bedtime administration	(Hermida et al., 2016)
Influenza vaccination	276 adults (65+ age)	Cluster-randomized trial	Morning vs Afternoon	Benefitted influenza antibody response at morning vaccination	(Long et al., 2016)
ARB, ACEI, CCB, β-blocker, and/or diuretic	19084 (10614 men/8470 women), 60.5 ± 13.7 years of age	Multi center, controlled, PROBE study	Evening vs Morning	Improved asleep ABP control and reduced CVD morbidity and mortality at bedtime administration	(Hermida et al., 2019)
Aspirin, an NSAIDs	175 (59 women and 116 men), (59.8 years)	Randomized controlled trial	Evening vs Morning	Reduced platelet aggregation at bedtime administration	(Krasińska et al., 2019)
Ibuprofen, an NSAIDs	70 (men and women), (18-35 years)	A randomized, double-blind, placebo-controlled trial	Morning vs Morning	Sufficient for pain management after surgical interventions at daytime administration	(Tamimi et al., 2022)

There have been recent advances in understanding the phases in which circadian rhythms control xenobiotic metabolism. This circadian coordination of xenobiotic or drug metabolism helps increase the xenobiotic’s water solubility and efficient excretion, mainly through urine and bile (Dallmann et al., 2014; Lévi et al., 2010). The pharmacokinetic properties of all drugs can be affected by circadian variations, as shown by hundreds of compounds in laboratory rodents and humans (Levi and Schibler, 2007; Yang et al., 2015). To some extent, xenobiotics are influenced by circadian rhythmic resting activity patterns, such as mealtimes, general sleep-wake cycles, and physical activity, all of which regulate blood pressure and flow. The extent of influence depends not only on the route of administration and excretion but also on circadian rhythm regulation of gastric pH and gastrointestinal (GI) motility, both of which affect drug absorption. In contrast, blood flow and capillary perfusion affect the absorption of drugs from the GI tract, their distribution to tissues and target organs, and even their excretion through the glomerular filtration rate in the kidneys (Ballesta et al., 2017; Dallmann et al., 2016). The circadian clock in the blood-brain barrier regulates xenobiotic efflux, which may affect the response to drug treatments targeting the brain (Zhang et al., 2018). Our ability to pharmacologically target circadian time to alleviate or treat certain chronic diseases will provide the next advance in chronomedicine.

On the other hand, efforts are underway to develop new drugs that target the molecular clock. Robust circadian clock function is important for well-being, and enhancing the body’s clock function is considered to have the potential to treat several human diseases where circadian clock function is disrupted. The clock modulator nobiletin regulates circadian rhythms and physiology in female mice with AD (Kim et al., 2021). Since nobiletin can activate various clock-controlled metabolic genes involved in insulin signaling and mitochondrial function, including Igf1, Glut1, Insr, Irs1, Ucp2, and Ucp4, it may present a novel intervention target against AD progression. Some new drugs, such as Rev-Erbα agonists (Ercolani et al., 2015), are currently being developed to target circadian time coordination or molecular clocks in different tissues to enhance the robustness of these components and/or alter circadian phases. Timed dosing, meaning that drugs are administered at specific times of the day, is one approach to mitigating these side effects.

7. Conclusions

Elucidating the molecular mechanisms of the biological clock in model organisms has revealed the surprising conservation of genes that regulate human circadian rhythms. Emerging genetic studies on human clock genes suggest that genetic variation in these genes may be associated with neurological and mood disorders. This genetic variation in the human clock genes may also lead to phenotypic variation associated with disease pathways, such as mood disorders. Considering the time of day in chronotherapy can significantly improve the success rate of clinical trials and ultimately improve patient care. The question remains regarding whether clock genes primarily influence circadian clock function, or if they have functions independent of the circadian system. In addition, the circadian clock is cell-autonomous and distributed throughout the body, making therapeutic interventions targeting peripheral and central circadian oscillators crucial. Future explorations in these research areas, coupled with increased public awareness, may allow us to appropriately diagnose and treat human circadian rhythm disorders and improve lifestyle choices that align our physiological systems with the daily clock.

Highlights.

• The cellular circadian clock is endogenously driven by transcription-translation feedback loops.

• The suprachiasmatic nucleus (SCN) is the master circadian pacemaker in mammals.

• Circadian rhythms are found in various extra-SCN brain regions and regulate neurophysiology and behavior.

• Disruption of circadian rhythms and clock gene variations are associated with many human brain diseases.

• The concept of chronomedicine and chronotherapy should be widely applied to promote health and treat diseases.

Data Availability

The article includes all data used in this work.