Biochemical Mechanism of the Mammalian Circadian Clock

Abstract

Circadian rhythms, regulated by core clock proteins, coordinate physiological functions with daily environmental fluctuations across organisms, from bacteria to humans. The circadian clock interacts with various biological processes, and its disruption is associated with numerous human diseases, including sleep disorders, metabolic syndrome, and potentially, cancer. In mammals, the circadian clock is driven by cell-autonomous transcription–translation feedback loops (TTFLs), in which CLOCK and BMAL1 act as transcriptional activators, while PER and CRY serve as transcriptional repressors. During the early repression phase, the CRY–PER–CK1 complex binds to CLOCK–BMAL1, displacing it from target promoters. In the late repression phase, in the absence of PER, CRY1 alone inhibits CLOCK–BMAL1 activity by blocking the recruitment of transcriptional coactivators. Biochemical and structural studies have highlighted the essential roles of protein–protein interactions, protein–DNA interactions, and post-translational modifications in regulating the molecular clock. In this Review, we summarize the molecular mechanisms that govern the circadian clock and focus on the coordination of protein–protein interactions and post-translational modifications, underscoring the importance of the circadian clock in disease progression and treatment strategies.

Introduction

Circadian rhythms are approximately 24-hour oscillations in biochemical, behavioral, and physiological processes that are synchronized to the solar light–dark cycle [1]. The circadian clock, a conserved mechanism across species, enables organisms to anticipate regular environmental changes such as food availability, temperature, and predation [2]. In mammals, the circadian system consists of a hierarchical network of oscillators operating at the cellular, tissue, and organismal levels [3,4]. The suprachiasmatic nucleus (SCN), located in the hypothalamus, serves as the central circadian pacemaker. It receives light signals from photosensitive retinal ganglion cells via the retinohypothalamic tract [5]. The SCN master clock, entrained by external light–dark cues, synchronizes local clocks in peripheral tissues through neuronal and endocrine signaling pathways [6]. In parallel, peripheral clocks can also be entrained by other stimuli, such as feeding–fasting cycles, which may cause misalignment between the central and peripheral clocks [7,8].

In mammals, the circadian clock of 24 hours periodicity is governed by time-delayed transcription–translation feedback loops (TTFLs). It has been argued on theoretical grounds that the ca. 24-hour periodicity of circadian oscillation requires at least a single delay feedback loop of one quarter (~ 6 hours) of the period [9]. In the mammalian circadian clock, this is achieved by molecular switches that ultimately lead to the alteration of biological pathways (network switches). The molecular switches in the circadian clock include cooperativity, antagonistic enzymes, phosphorylation, acetylation, positive feedback, and sequestration [9]. In the positive arm of the TTFL, CLOCK (circadian locomotor output cycles kaput) and BMAL1 (brain and muscle ARNT-like protein 1) proteins form a heterodimer that binds to E-box sequences to activate the transcription of target genes, including Per (Period) and Cry (Cryptochrome). In the negative arm of the TTFL, PER and CRY proteins form a repressive complex with CK1 (casein kinase 1) and physically interact with the CLOCK–BMAL1 complex to inhibit their own expression [10–12]. The transcription activity of CLOCK–BMAL1 is repressed via two distinct mechanisms: PER-dependent “displacement” repression and PER-independent “blocking” repression [12,13]. In a secondary feedback loop, the CLOCK–BMAL1 complex induces the expression of nuclear receptors REV-ERBs and retinoic acid receptor-related orphan receptors (RORs). These receptors compete for binding to ROR/REV-ERB response elements (ROREs) in the promoters of Bmal1 and Clock, thereby regulating their transcription [14]. Recent biochemical and genetic studies have greatly advanced our understanding of the role of post-translational modifications and protein-protein interactions in modulating clock protein function [15–17].

1. Molecular mechanism of the mammalian circadian clock

1.1. Transcriptional activators: BMAL1 and CLOCK

CLOCK and BMAL1 are members of the basic helix–loop–helix (bHLH) transcription factor family. Both proteins contain an N-terminal bHLH domain, two tandem PAS (Per–Arnt–Sim) domains (PAS-A and PAS-B), and a transactivation domain (TAD) at the C-terminus (Figure 1). The CLOCK–BMAL1 complex has been proposed to function like a pioneer transcription factor by binding to nucleosomes and promoting rhythmic chromatin remodeling [18]. Three domains in both CLOCK and BMAL1, the bHLH, PAS-A, and PAS-B domains, mediate dimerization by forming three distinct protein interfaces [19]. The bHLH dimer of CLOCK–BMAL1 binds to the major groove of E-box DNA motifs, and mutations in the bHLH domain can destabilize the heterodimer and abolish its transactivation activity [19,20]. Interactions between the PAS domains and histones help anchor the CLOCK–BMAL1 complex to nucleosomes [20,21]. The interaction between the CLOCK PAS-B domain and CRY1 is critical for maintaining the repressed state of CLOCK–BMAL1 [22]. Additionally, the CLOCK TAD interacts with circadian protein (CIPC) to suppress CLOCK–BMAL1 transcriptional activity [23,24]. The BMAL1 TAD plays a central role in regulating CLOCK–BMAL1 activity through interactions with transcriptional coactivators such as CBP/p300, as well as with repressors like CRY [25–27]. Mutations that disrupt interactions between BMAL1 and CLOCK, or between the CLOCK–BMAL1 complex and its interacting partners, alter CLOCK-BMAL1 transcriptional activity, underscoring the importance of protein–protein interactions in circadian clock regulation [19,20,22].

Figure 1. Structure of core clock proteins.

CLOCK and BMAL1 are members of the basic helix-loop-helix (bHLH) transcription factor family. Each contains a bHLH domain, two tandem PAS (PER-ARNT-SIM) domains, and a C-terminal transactivation domain (TAD). CRY proteins, members of the photolyase/cryptochrome protein family, consist of a photolyase homology region (PHR) and a C-terminal tail. PER proteins contain two N-terminal PAS domains, a casein kinase-binding domain (CKBD), and a C-terminal CRY-binding domain (CBD).

The transcriptional activity of the CLOCK–BMAL1 complex is tightly regulated by post-translational modifications. It has been reported that CLOCK possesses intrinsic histone acetyltransferase activity, enabling it to acetylate both histones and its dimerization partner BMAL1 at lysine 537, thereby promoting interaction with CRY1 and leading to transcriptional repression [28,29]. In contrast, other evidence indicates that lysine acetyltransferase TIP60, rather than CLOCK, acetylates this lysine of BMAL1, which recruits the BRD4–P-TEFb complex to the E-box promoter [30]. In mouse liver, BMAL1 undergoes rhythmic acetylation, and its deacetylation is mediated by the NAD⁺-dependent deacetylase SIRT1 [29,31]. Multiple kinases, including CK2α, S6K1, AKT, GSK3β, JNK, and ERK, are also involved in the post-translational regulation of CLOCK and BMAL1 [32]. A recent study demonstrated that phosphorylation of CLOCK at Ser38 and Ser42, and of BMAL1 at Ser78, potentially mediated by CK1, within the bHLH domains alters the DNA-binding activity of the CLOCK–BMAL1 complex, underscoring the importance of post-translational modifications in regulating its transcriptional activity [33].

1.2. The CRY-PER-CK1 complex in the early repression phase

The transcriptional activation of Per1/2 and Cry1/2 by the CLOCK–BMAL1 complex leads to the accumulation of PER and CRY proteins in the cytoplasm (Figure 2). PER proteins contain two tandem PAS domains, a casein kinase-binding domain (CKBD), and a CRY-binding domain (CBD) at their C-terminus (Figure 1). They function as scaffolds for CK1 in the circadian feedback mechanism [34]. The PAS domains of PER mediate both homo- and heterodimeric interactions between PER proteins [35,36]. PER1 and PER2 interact with CK1 through the conserved CKBD [37,38]. Interaction between PER and CRY proteins via the PER CBD facilitates the nuclear translocation of the CRY–PER–CK1 complex [11,12,39]. Although the specific acetyltransferase responsible for PER2 acetylation at lysine 680 remains unknown, SIRT1-mediated deacetylation has been shown to regulate PER2 stability and nuclear localization [40,41]. Acetylation of PER2 may reduce its heterodimerization with CRY1 in the nucleus, implying that acetylation regulates PER2 activity by modulating its interactions with other proteins [41]. Multisite phosphorylation of PER occurs throughout the circadian cycle [42]. Although the overall role of phosphorylation remains incompletely understood, two phosphorylation events critical for PER2 stability have been extensively studied: phosphorylation of the familial advanced sleep phase (FASP) region and the degron [43]. CK1-mediated phosphorylation of the degron, located near the tandem PAS domains of PER2, facilitates recruitment of the E3 ubiquitin ligase β-TrCP (beta-transducin repeat-containing protein) and promotes proteasomal degradation of PER [38,44,45]. In contrast, CK1-dependent phosphorylation of the FASP region stabilizes PER proteins by preventing degron phosphorylation [15,46]. A recent study further demonstrated that the phosphorylated FASP region of PER2 directly interacts with and inhibits CK1δ activity [46].

Figure 2. Transcription/translation feedback loops (TTFLs) model.

Model illustrating the dual repression mechanisms in the mammalian molecular clockwork. The transcriptional activators CLOCK and BMAL1 form an activator complex that binds to E-box elements in promoters to induce the transcription of target genes, including Per, Cry, Rorα, and Rev-erbα/β (Activation). During the early repression phase, CRY, PER, and CK1δ form a repressor complex that translocates into the nucleus. The recruitment of CK1δ via CRY and PER to CLOCK-BMAL1 leads to phosphorylation of CLOCK, causing the release of CLOCK-BMAL1 from E-box DNA (displacement-type repression). In the late repression phase, following PER degradation, free CRY1 binds to the CLOCK-BMAL1 complex to block its transcriptional activity without disrupting DNA binding (blocking-type repression). The decline in nuclear CRY1 levels marks the end of the repression phase and the resumption of CLOCK-BMAL1 transcriptional activity. Note that the dashed bracket indicates that the interaction between the repressor and activator complexes may be transient and unstable. Additionally, the nuclear receptors RORα and Rev-erbα bind to ROR/REV-ERB response elements (RORE) in the Bmal1 and Clock genes to regulate their expression (secondary circadian loop).

Mammalian CRY proteins belong to the photolyase/cryptochrome protein family but lack photolyase activity [47]. Although the CRY genes (Cry1 and Cry2) are highly conserved in sequence, they have redundant and distinct functions. Cry1 and Cry2 mutant mice exhibit shortened and lengthened behavioral rhythms compared with wild-type mice, respectively [48,49]. Cry1/2 double-mutant mice are arrhythmic in constant darkness, indicating that CRYs play a central role in the circadian clock system of mammals [49,50]. CRY consists of a photolyase homology region (PHR) and an intrinsically disordered C-terminal tail (Figure 1). The PHR domain comprises a primary pocket (known as the FAD (Flavin Adenine Dinucleotide)-binding pocket in photolyase), a secondary pocket (known as the MTHF (5, 10-methenyltetrahydrofolate) -binding domain in photolyase), and a coiled-coil (CC) helix. Two sites within the CRY PHR domain interact with the CLOCK–BMAL1 complex to repress its transcriptional activity: the CC helix directly binds the BMAL1 TAD, and the secondary pocket docks onto the PAS domain core of CLOCK–BMAL1 [22,27,51,52]. The PHR domain of CRY also binds to PER2 via a ~100 amino acid long c-terminal CBD [39,53]. A recent study revealed that the CRY C-terminal tail interacts directly with the PHR domain to regulate CRY1’s affinity for CLOCK–BMAL1 [54]. Similar to PER, CRY protein stability is also regulated by phosphorylation [55]. The nutrient-responsive AMP-activated protein kinase (AMPK) promotes CRY1 degradation by phosphorylating Serine 71, which enhances its interaction with the E3 ubiquitin ligase FBXL3 [55]. In contrast, phosphorylation of CRY1 at Serine 588 increases its stability and prevents FBXL3-mediated degradation [56]. Whether phosphorylation at Ser588 influences the interaction between the CRY1 tail and the PHR domain remains to be determined.

Biochemical analyses have shown that CRY1 interacts with the CLOCK–BMAL1–E-box complex independently of PER [57]. Although PER can interfere with the association between CRY1 and the CLOCK–BMAL1–E-box complex, it does not affect the binding of CLOCK–BMAL1 to the E-box in vitro [57]. However, in vivo data indicate that the nuclear entry of PER leads to the removal of both CRY1 and CLOCK–BMAL1 from the E-box [58,59]. This discrepancy between in vitro and in vivo observations suggests that the function of PER may be modulated by its interacting partners. Electrophoretic mobility shift assays have demonstrated that CK1 can directly dissociate CLOCK–BMAL1 from the E-box [12]. Furthermore, in vivo evidence supports that CK1 is required for the PER-dependent removal of the CRY1–CLOCK–BMAL1 complex from the E-box, highlighting the essential role of CK1 in the PER-mediated displacement of activator complexes from target promoters [12].

The CK1 family comprises a group of closely related serine/threonine kinases, including CK1α, β, δ, ε, γ1, γ2, and γ3. Among them, CK1δ and CK1ε phosphorylate core clock proteins and regulate their function [15]. A point mutation in CK1ε (Arg178→Cys), known as the tau mutation, results in a short-period circadian phenotype in mice and hamsters, emphasizing the critical role of CK1ε in circadian timing [60–62]. Mechanistically, this mutation remodels CK1 substrate selectivity on PER2 to enhance degron phosphorylation and promotes degradation of PER2 [63,64]. Biochemical studies have shown that CK1 associates with the PER–CRY complex via CKBD located in the C-terminus of PER [12,65]. Within the nucleus, the binding of the PER–CRY–CK1 repression complex to CLOCK–BMAL1 leads to phosphorylation of CLOCK–BMAL1, promoting its dissociation from E-box DNA elements [12,33]. Although an ~1.9 MDa (megadalton) supercomplex composed of the activator (CLOCK–BMAL1) and the repressor (CRY–PER–CK1) in association with a number of other proteins has been detected in nuclear extracts using blue-native acrylamide/agarose gel electrophoresis (BN-APAGE) [10], recent sedimentation analyses have identified separate activator and repressor complexes in association with different sets of proteins from those in the ~ 1.9 MDa complex, but not the supercomplex [12,65]. This discrepancy may arise because clock proteins, particularly the multidomain mosaic PER proteins, engage in interactions with partners involved in noncanonical clock or non-clock functions [65]. As a result, many interacting partners of PER and CRY are likely “fellow travelers rather than conjugal partners” and are not filtered out by BN-APAGE [9,66]. Sedimentation results suggest that the interaction between the activator and repressor complexes is unstable [12,65]. This instability may be attributed to CK1-mediated phosphorylation of CLOCK, which reduces its affinity for the repressor complex (Figure 2) [12,65]. Notably, there is evidence that PER-dependent displacement of CLOCK–BMAL1 not only represses gene activation but can also lead to gene de-repression [59]. For example, PER-mediated reduction of CLOCK–BMAL1 binding at the Cry1 promoter results in the activation of Cry1 expression [59]. Interestingly, a recent study points to the dual role of the PER2-CRY1 complex: this complex not only acts as a component of the negative feedback arm but also redistributes CLOCK:BMAL1 to new target sites [67].

1.3. The CRY1-CLOCK-BMAL1 ternary complex in the late repression phase

CRY1 plays a distinct role among the repressive clock proteins. It is predominantly localized in the nucleus throughout the circadian cycle [13,68]. Degradation of hyperphosphorylated PER proteins reduces the nuclear PER:CRY ratio, thereby releasing free CRY1 in the nucleus [68]. CRY1 can bind stably to the CLOCK–BMAL1–E-box complex independently of PER [57]. ChIP-seq data show that CRY1 co-occupies BMAL1/CLOCK binding sites even in the absence of CRY2 and PER proteins [69]. CRY1 inhibits CLOCK–BMAL1 activity by binding to the PAS-B domain of CLOCK, which facilitates sequestration of the BMAL1 TAD, thereby preventing its interaction with transcriptional coactivators [27]. In contrast to the PER–CRY–CK1 repressor complex, CRY1 alone binds to the CLOCK–BMAL1 complex during the late repression phase, inducing a “poised” state without disrupting DNA binding of the activator complex [69]. As CRY1 levels decline, CLOCK–BMAL1 gradually regains transcriptional activity (Figure 2).

The relative molecular abundance of proteins is critical for their interactions and functions. A stoichiometric balance between repressors and activators is essential for sustaining circadian oscillations [70]. By monitoring the expression dynamics and spatial distribution of circadian clock proteins in live cells of organotypic SCN slices, a recent study revealed that PER2 abundance, which is tightly regulated by protein synthesis and degradation, is likely the most limiting factor in the formation of the PER2–CRY1 complex [13]. This finding supports the idea that PER degradation controls the transition from displacement-type repression to blocking-type repression (Figure 2). However, the oscillations of PER2 and CRY1 abundance are temporally segregated, with CRY1 peaking approximately 7 hours after PER2 in SCN [13]. Consistently, in mouse liver, PER2 occupancy peaks early in the evening with PER1 and CRY2, whereas CRY1 peaks several hours later, in the late night to early morning [69]. This temporal separation suggests that PER2 and CRY1 may play independent roles in the circadian clock. Additionally, this study observed that CRY1 remains at relatively high levels even when CLOCK–BMAL1 reaches maximal transcriptional activity, implying that CRY1 functions as a repressor throughout the entire SCN TTFL cycle. Notably, the distinct SCN cell populations may exhibit varying relative molecular compositions of TTFL proteins. Cells in the phase-leading dorso-medial margin express a high level of CRY1 but exhibit little PER2 expression [13]. Future studies are required to ascertain whether the dynamics ratios of TTFL proteins in other cell types or organs are comparable to those observed in the SCN.

2. Circadian clock and human health

The circadian clock temporarily coordinates or orchestrates a wide range of essential biological processes, including sleep, metabolism, and DNA repair [71,72]. Accumulating evidence indicates that circadian disruption is associated with numerous diseases, such as metabolic syndrome, diabetes, sleep disorders, and, according to some researchers, cancer (Figure 3) [73]. Understanding the interactions between circadian clock components and disease-relevant pathways is critical for advancing circadian precision medicine [74,75].

Figure 3. Circadian rhythms and human health.

The circadian clock regulates various critical biological processes, including sleep, DNA repair, and metabolism. Disruptions to the circadian rhythm, such as those caused by jet lag, shift work, or mutations in core clock genes, are closely associated with a range of human diseases, including sleep disorders and metabolic syndromes. However, the causal relationship between circadian disruption and cancer remains uncertain. Additionally, the effectiveness of chronotherapy in cancer treatment requires further investigation.

2.1. Circadian clock and sleep disorders

The timing of sleep is strongly influenced by the circadian system. Genetic association studies in humans have revealed that mutations in core clock genes are linked to circadian sleep disorders (Figure 3) [76–79]. Sleep phase syndromes are conditions in which the sleep–wake cycle is either extremely advanced or delayed relative to the normal distribution of sleep timing in the population. Inherited forms of advanced sleep phase syndrome have been associated with mutations in Per2, Cry2, Ck1δ, and Per3 [76,78–80]. Conversely, mutations in Cry1, Per2, and Per3 have been linked to familial delayed sleep phase syndrome [77,81–83].

2.2. Circadian clock and metabolism

The circadian clock is closely linked to metabolic homeostasis. Epidemiological studies have shown that circadian disruption caused by shift work is significantly associated with an increased risk of metabolic diseases (Figure 3) [73,84]. In vivo animal studies provided further evidence that core clock components directly regulate metabolic pathways. For example, CRY1 inhibits hepatic gluconeogenesis by repressing glucocorticoid receptor activity, blocking the cAMP/CREB signaling pathway, and promoting degradation of FOXO1 [85–88]. Disruption or depletion of other clock proteins, including BMAL1 and CLOCK, also leads to metabolic disorders in mice [89–91]. Accumulating evidence suggests that both dietary components and meal timing influence metabolic homeostasis by remodeling the circadian clock [92–94]. A high-fat diet (HFD) disrupts circadian function by impairing CLOCK–BMAL1 chromatin recruitment, altering the expression and rhythmicity of core clock genes, and enhancing the oscillation of non-core transcription factors such as PPARα and SREBP [95–97].

Time-restricted feeding (TRF) is a dietary protocol that limits food intake to a specific time window each day, without altering dietary composition or total caloric intake. TRF enhances nutrient-sensing pathways and circadian oscillations and prevents HFD-induced obesity and metabolic syndrome in mice [98]. Notably, TRF also protects mice with a compromised circadian clock from HFD-induced metabolic disorders. Rhythms in metabolic and nutrient-sensing pathways can be restored by TRF in whole-body Cry1/2 double-knockout mice, as well as in liver-specific Bmal1 and Rev-Erbα/β knockout mice. These findings suggest that imposed feeding–fasting rhythms are sufficient to alleviate key metabolic dysfunctions independently of a functional circadian clock [99]. While these results provide important insights into the role of the clock in metabolic regulation, the underlying mechanisms of the crosstalk between circadian rhythms and metabolism remain largely unexplored.

2.3. Circadian clock, cancer and DNA repair

The notion that chronic circadian rhythm disruption increases cancer susceptibility has been claimed based on both epidemiological and animal studies [100,101]. Night shift work, which has been reported to cause circadian disruption, was classified as a probable human carcinogen by the International Agency for Research on Cancer in 2019 [102]. Several studies have linked persistent night shift work to an elevated risk of colorectal, breast, and prostate cancers [103–107]. It has also been claimed that mutations in core clock genes are associated with cancer susceptibility, and dysregulation of clock genes is commonly observed in various cancer types [108–111]. However, the causal relationship between carcinogenesis and circadian clock disruption remains debatable based on studies using mice with core clock genes knockouts [74,112]. A study reporting that deletion of Per2 predisposes mice to spontaneous and γ-radiation-induced cancers has not been reproduced, as other studies have shown that knockout of either Per1 or Per2 does not increase tumor susceptibility [113]. Similarly, deletion of both Cry1 and Cry2, which abolishes circadian behavioral rhythms, does not render mice more prone to cancer [114]. Interestingly, the expression of the oncogene c-Myc is decreased in Cry1/2 knockout mice [115]. Furthermore, deletion of Cry1/2 in p53 mutant mice significantly delays tumor development and extends lifespan [116]. In contrast, Cry1/2 mutations have no effect on melanoma incidence or survival in the Ink4a−/−;ras(V12G) mouse model of UV-induced melanoma, suggesting that the antitumorigenic effects of Cry1/2 mutations depend on genetic context (Figure 3) [112]. Therefore, further studies are warranted to clarify the complex relationship between circadian clock disruption and cancer development.

Circadian clock components regulate numerous cancer-related processes, including cell proliferation, apoptosis, metabolic reprogramming, and metastasis, through transcriptional modulation of cell cycle regulators or via direct protein-protein interactions that alter checkpoint protein activity [74,101]. For example, the transcription of the G2 checkpoint kinase WEE1 and the cell cycle inhibitor P21 is controlled by core clock components [117,118]. BMAL1 and CLOCK promote proliferation in hepatocellular carcinoma by regulating Wee1 and p21 levels, thereby preventing apoptosis and cell cycle arrest [119]. Furthermore, the tumor suppressor p53, oncogenes c-MYC and RAS, and the circadian clock are intricately interconnected [74,120]. Transcription of c-Myc is regulated by β-catenin, which is repressed by CLOCK-BMAL1 binding to an E-box within its intron [115]. Additionally, the interaction between CRY2 and c-MYC promotes degradation of the c-MYC protein, reducing its levels [121]. Although many studies highlight tumor-suppressing roles of the circadian clock, an equal number of studies report pro-tumorigenic functions as well. For instance, deletion of Bmal1 leads to increased tumor initiation and accelerated progression in mouse colorectal cancer models [122,123]. Conversely, Bmal1 ablation induces apoptosis and impairs proliferation in murine leukemia stem cells of acute myeloid leukemia and in hepatocellular carcinoma cell lines [119,124]. Disruption of BMAL1 sensitizes invasive breast cancer cells (MDA-MB-231) to cisplatin- and doxorubicin-induced apoptosis, while simultaneously enhancing their invasive potential. This suggests a dual role for BMAL1 in the carcinogenesis process, with both tumor-suppressive and tumor-promoting effects [125,126]. In addition, deletion of CRY genes in p53-mutant mouse skin fibroblasts enhances apoptotic and anti-tumorigenic responses following UV-induced DNA damage [127,128]. Mechanistically, mutations in Cry1 and Cry2 increase DNA damage-induced apoptosis by upregulating the expression of p73, a member of the p53 family [129]. The tumor-promoting role of CRYs is further supported by the extended lifespan of Cry1/2 double-knockout mice lacking p53 compared to p53-deficient mice alone [127]. Overall, these findings suggest that circadian clock genes can function either as oncogenes or tumor suppressors, depending on the cancer type and cellular context.

DNA damage is a major cause of cancer, and defects in DNA repair mechanisms lead to tumorigenesis [130]. Nucleotide excision repair, a fundamental DNA repair pathway, removes bulky DNA lesions caused by UV radiation or chemotherapeutic drugs such as cisplatin. It consists of two main sub-pathways: global genome nucleotide excision repair and transcription-coupled repair (TCR). Mutations in nucleotide excision repair pathway genes can cause xeroderma pigmentosum, a disorder that significantly increases the risk of skin cancer [130–132]. Studies in mice have shown that excision repair activity exhibits circadian rhythmicity in various tissues, including liver, kidney, spleen, brain, and skin (Figure 3) [71,133,134]. Two mechanisms have been proposed to underlie the circadian regulation of nucleotide excision repair activity. First, the expression of XPA, a crucial excision repair factor, is controlled by the circadian clock. Second, the rhythmic repair of transcribed DNA strands is driven by circadian transcription through TCR [74]. Disruption of the circadian clock alters the rhythmic repair of cisplatin-induced DNA damage in the liver and kidney [135]. Notably, cisplatin treatment increases the expression of Per1 in both liver and kidney tissues, suggesting that DNA damage and repair pathways may influence the circadian clock by regulating clock gene expression [67].

Chronotherapy refers to the administration of drugs at specific times of the day, guided by the circadian clock, to maximize efficacy and minimize side effects. An early clinical trial involving a small number of subjects reported a substantial effect of chronochemotherapy in ovarian cancer [136]. However, these findings were not confirmed by a subsequent larger study [137]. Similarly, in a large clinical trial of chronochemotherapy in patients with metastatic colorectal cancer, a minor beneficial effect was observed in men, while a greater harmful effect was seen in women [138,139]. Similarly, a clinical trial in patients with endometrial cancer found no significant benefit from chronochemotherapy [140]. Although current clinical data do not support the broad application of chronochemotherapy for all cancer types, carefully designed regimens may still hold promise for certain cancers (Figure 3) [74]. Therefore, further research is needed to elucidate the complex interplay between DNA damage and repair and the circadian clock, which is essential for the development of effective chronochemotherapy strategies.

3. CRY proteins as targets for circadian drugs

The growing understanding of the biochemical mechanisms governing the circadian clock, along with its strong connection to human health, has driven the development of small-molecule compounds to regulate circadian rhythms. CRY functions as the central component of the repressive arm of the TTFL. During the early repression phase, CRY facilitates the nuclear entry of PER and CK1 and promotes phosphorylation of the CLOCK-BMAL1 complex, leading to displacement of these transcription factors from their target promoters. In the late repression phase, following PER degradation, CRY1 represses CLOCK-BMAL1 activity by sequestering the BMAL1 TAD (Figure 2) [12,27,65]. The PHR domain of CRY proteins contains two binding pockets, the primary and secondary pockets, which present attractive targets for drug screening [141].

The first CRY stabilizer, KL001, was reported in 2012 [142]. This carbazole derivative stabilizes both CRY1 and CRY2 by inhibiting the binding of the E3 ligase FBXL3 to the primary pocket [142,143]. KL001 repressed gluconeogenesis in mouse primary hepatocytes [142]. Because CLOCK–BMAL1 transcriptional activity is essential for glioblastoma stem cell (GSC) growth, KL001, which stabilizes the CRY1 protein, inhibited GSC proliferation by reducing CLOCK–BMAL1 activity [144]. In 2014, KS15 was identified through a cell-based screen targeting E-box-mediated transcriptional activity as an inhibitor of both CRY1 and CRY2 [145]. KS15 binds to the C-terminal region of CRYs and disrupts the interaction between CRY and BMAL1 [146] (Figure 4). Inhibition of CRYs by KS15 reduced the rate of cell growth and increased chemosensitivity in breast cancer cells [147], highlighting CRYs as a promising novel target for cancer treatment.

Figure 4. Small molecules targeting CRY proteins.

CRY proteins have emerged as promising therapeutic targets for the treatment of various diseases. Several small molecules have been developed to modulate CRY protein activity. Among them, KL001 stabilizes both CRY1 and CRY2 by preventing their interaction with the E3 ubiquitin ligase FBXL3. In contrast, KS15 inhibits CRY1 and CRY2 activity by disrupting their interaction with BMAL1. Some compounds demonstrate isoform selectivity: M47, M54, and KL101 specifically target CRY1, while TH301, SHP656, SHP1703, and SHP1705 selectively stabilize CRY2. GSCs, glioblastoma stem cells.

Although the high sequence conservation between CRY1 and CRY2 presents a challenge for developing isoform-selective compounds, several molecules have been identified that selectively target either CRY1 or CRY2. Recent studies show that KL101 and TH301 selectively stabilize CRY1 and CRY2, respectively (Figure 4) [148]. Mechanistically, although these compounds interact with the primary pocket, which is nearly identical in CRY1 and CRY2, selectivity arises from differences in the C-terminal disordered regions outside the binding pocket, highlighting the functional importance of the PHR–tail interaction [148].

Using structure-based drug design methods, a recent study identified 171 molecules that can target functional domains of CRY1 [149]. Among them, a small molecule called M47 selectively binds to the primary pocket of CRY1, but not to CRY2. This binding enhanced CRY1 degradation by increasing its ubiquitination, thereby lengthening the circadian period [150]. M47 may stabilize the interaction between CRY1 and FBXL1, promoting CRY1 degradation. Treatment with M47 also enhanced oxaliplatin-induced apoptosis in Ras-transformed p53-null fibroblast cells. Furthermore, in vivo administration of M47 extended the lifespan of p53⁻/⁻ mice, consistent with previous findings that the loss of both CRY proteins prolonged survival in p53⁻/⁻ mice, highlighting the potential of M47 in anticancer therapy (Figure 4) [116,150]. In contrast, M54, another small molecule that also binds to the primary pocket of CRY1, enhanced CRY1 stability by interfering with the CRY1–FBXL1 interaction [151].

SHP656, a derivative of KL001, selectively interacts with and stabilizes CRY2 [152]. Treatment with SHP656 reduced numbers of GSCs and inhibited tumor growth in vivo [144]. SHP1703, the active R-enantiomer of SHP656, similarly decreased GSCs viability [152]. SHP1705, which has shown success in Phase 1 safety trials, exhibits increased selectivity for the CRY2 isoform. In vivo studies demonstrated that SHP1705 treatment delayed glioblastoma tumor growth and prolonged survival in tumor-bearing mice (Figure 4) [153]. Together, these findings demonstrate that targeting circadian clock proteins with small molecules offers promising therapeutic potential for human diseases, including metabolic disorders and cancer.

Concluding remarks

Recent biochemical, genomic, and fluorescence imaging evidence supports the dual repression model of TTFLs [12,13,54,69]. In this model, the transcriptional activity of CLOCK-BMAL1 is intricately regulated by protein-protein interactions and post-translational modifications. In the early repression phase, the CRY-PER-CK1δ repressive complex is essential for displacing CLOCK-BMAL1 from E-box DNA. In the late repression phase, CRY1 alone blocks the recruitment of transcriptional coactivators, thereby inhibiting CLOCK-BMAL1 activity. The transition from early to late repression phase depends on PER degradation and free CRY1 accumulation in the nucleus. Recent in vitro data indicate that CK1δ-mediated phosphorylation of PER2 disrupts the interaction between CRY1 and PER2, potentially contributing to this phase transition [154]. Disruption of circadian rhythms is strongly linked to various human diseases, including sleep, metabolic disorders, and, in a context-dependent manner, carcinogenesis or cancer therapy. Clock proteins have emerged as promising therapeutic targets for these conditions. Further studies are needed to elucidate the mechanisms regulating the circadian clock and to identify new interacting partners of clock proteins in different cellular contexts. These insights are critical for developing novel therapeutic strategies that modulate circadian pathways in a context-specific manner.

Abbreviations:

TTFLs

transcription–translation feedback loops

CLOCK

circadian locomotor output cycles kaput

BMAL1

brain and muscle ARNT-like protein 1

PER

period

CRY

cryptochrome

SCN

suprachiasmatic nucleus

TIP60

60 kDa Tat-interactive protein

RORs

retinoic acid receptor-related orphan receptors

ROREs

ROR/REV-ERB response elements

CK1

Casein kinase 1

bHLH

basic helix–loop–helix

PAS

Per–Arnt–Sim

TAD

transactivation domain

CIPC

circadian protein

FASP

familial advanced sleep phase

CKBD

casein kinase-binding domain

CBD

CRY-binding domain

TRF

time-restricted feeding

HFD

high-fat diet

TCR

transcription-coupled repair

FAD

flavin adenine dinucleotide

MTHF

5, 10-methenyltetrahydrofolate

PHR

photolyase homology region

BN-APAGE

blue-native acrylamide/agarose gel electrophoresis

GSCs

glioblastoma stem cells

β-TrCP

beta-transducin repeat-containing protein