biology

Review
Circadian Misalignment and Metabolic Disorders: A Story of
Twisted Clocks
Aurore Woller 1 and Didier Gonze 2, *

1 Department of Molecular Cell Biology, Weizmann Institute of Science, Rehovot 76100, Israel;
apolaire@gmail.com
2 Unité de Chronobiologie Théorique, Faculté des Sciences CP 231, Université Libre de Bruxelles,
Bvd du Triomphe, 1050 Bruxelles, Belgium
* Correspondence: dgonze@ulb.ac.be

Simple Summary: In mammals, many physiological processes follow a 24 h rhythmic pattern. These
rhythms are governed by a complex network of circadian clocks, which perceives external time cues
(notably light and nutrients) and adjusts the timing of metabolic and physiological functions to allow
a proper adaptation of the organism to the daily changes in the environmental conditions. Circadian
rhythms originate at the cellular level through a transcriptional–translational regulatory network
involving a handful of clock genes. In this review, we show how adverse effects caused by ill-timed
feeding or jet lag can lead to a dysregulation of this genetic clockwork, which in turn results in
altered metabolic regulation and possibly in diseases. We also show how computational modeling
can complement experimental observations to understand the design of the clockwork and the onset
of metabolic disorders.

Abstract: Biological clocks are cell-autonomous oscillators that can be entrained by periodic envi-





 ronmental cues. This allows organisms to anticipate predictable daily environmental changes and,
thereby, to partition physiological processes into appropriate phases with respect to these changing
Citation: Woller, A.; Gonze, D.
Circadian Misalignment and
external conditions. Nowadays our 24/7 society challenges this delicate equilibrium. Indeed, many
Metabolic Disorders: A Story of studies suggest that perturbations such as chronic jet lag, ill-timed eating patterns, or shift work in-
Twisted Clocks. Biology 2021, 10, 207. crease the susceptibility to cardiometabolic disorders, diabetes, and cancers. However the underlying
https://doi.org/10.3390/ mechanisms are still poorly understood. A deeper understanding of this complex, dynamic system
biology10030207 requires a global holistic approach for which mathematical modeling can be highly beneficial. In
this review, we summarize several experimental works pertaining to the effect of adverse conditions
Academic Editor: Eleonore Maury on clock gene expression and on physiology, and we show how computational models can bring
interesting insights into the links between circadian misalignment and metabolic diseases.
Received: 30 January 2021
Accepted: 2 March 2021
Keywords: circadian oscillations; gene regulatory network; light; food; conflicting zeitgebers; com-
Published: 10 March 2021
putational modeling

Publisher’s Note: MDPI stays neutral

with regard to jurisdictional claims in
published maps and institutional affil-
1. Introduction
iations.
In mammals a large range of physiological and metabolic processes display a 24 h
rhythmic pattern. These processes are controlled by an elaborate circadian system con-
sisting of a finely tuned network of clocks throughout the body coupled together by
Copyright: © 2021 by the authors.
periodically released hormones, nutrient sensors, and neuronal connections. Besides con-
Licensee MDPI, Basel, Switzerland.
trolling the sleep–wake cycle, this system appears to be optimized to ensure metabolic
This article is an open access article
homeostasis (by allowing anticipation of food uptake), an effective immune response, and
distributed under the terms and synchronized cell division [1,2]. It relies on the alignment of the light–dark cycle and the
conditions of the Creative Commons feeding–fasting pattern and on a precise genetic clockwork. Dysregulation of this system,
Attribution (CC BY) license (https:// through genetic mutations or through alteration of the timing of zeitgebers (jet lag, shift
creativecommons.org/licenses/by/ work, ill-timed food intake, etc.) can lead to metabolic syndromes and pathologies, includ-
4.0/). ing obesity, diabetes, auto-immunity, and tumorigenesis and cancer [2,3]. By “zeitgeber”

Biology 2021, 10, 207. https://doi.org/10.3390/biology10030207 https://www.mdpi.com/journal/biology

Biology 2021, 10, 207 2 of 15

we refer to any periodic input that entrains the circadian clock, with light and food being
external zeitgebers, while hormones and neuroal projections are internal zeitgebers.
Understanding the cogs and wheels of this multi-oscillatory system is challenging
because it involves the integration of organ-specific genetic control, metabolic regulation,
and hormone-mediated inter-organ coupling. Experimentally, the logic of this regulatory
system can be probed by genetic studies and by artificial environmental perturbations
(e.g., conflicting zeitgebers) and by looking at the consequences on gene expression and
on physiology. Experimental studies can be complemented by mathematical modeling
investigations. Theoretical models aim at providing a rational explanation for observed
dynamic responses and allow the integration of disparate experimental observations.
We review here several studies that contributed to unraveling the complex interplay
between the genetic clocks and the physiology and show how mathematical models
help to interpret the observations and to provide guidelines for future investigations.
Particular attention is given to the importance of the timing of clock gene expression for
proper coordination of metabolism and to the alteration of this timing due to conflicting
environmental signals. We first describe the global circadian organization and the genetic
circuitry underlying circadian oscillations, which is responsible for the proper timing
of gene expression in physiological conditions. We then show, through a selection of
experimental and theoretical papers, how the phase relationships between the clock gene
expression are altered under ill-timed feeding or in response to a jet lag and how, in turn,
those “twisted clocks” impact clock-controlled cellular processes.

2. Global Organization of the Circadian System

The circadian system consists of a fine-tuned network of clocks throughout the
body [4]. The suprachiasmatic nucleus (SCN) of the hypothalamus is regarded as the
circadian pacemaker [5]. The SCN is composed of two clusters of about 20,000 densely con-
nected neurons in rodents (and about 100,000 neurons in humans). Each neuron expresses
a dozen clock genes, which form a transcriptional regulatory network able to generate circa-
dian oscillations, which persist in constant environmental conditions. Although oscillations
observed in isolated cells are noisy and present large cell–cell variability, coupling through
the circadian release of neuropeptides allows a robust synchronization of the cells [6,7].
The SCN receives light signals from the retina through the retinohypothalamic tract (RHT).
Information on the light/dark alternation is thereby transmitted to the hypothalamus
and, in normal conditions, entrains the SCN circadian pacemaker. The SCN dictates daily
changes in body temperature, determines food intake, controls sleep–wake cycles [8] and,
through the control of hormonal release, influences the rhythms in peripheral organs [9–11].
For example, by stimulating the secretion of cortisol—the main glucocorticoid (GC) in
humans—by the adrenal glands via the hypothalamic–pituitary–adrenal axis, the SCN
affects insulin signaling and insulin secretion in the pancreas [12] as well as gluconeogene-
sis in the liver [13]. This level of control adds to the nutrient-induced secretion of insulin
and to the SCN input via the vagus nerve. The SCN also controls the nocturnal release of
melatonin by the pineal gland. This hormone may target the SCN but appears to also be
involved in the regulation of glucose metabolism [14].
Peripheral clocks are present in many organs including the liver, pancreas, gut,
adipocytes, or muscles. They all possess autonomous clocks, which were initially viewed
as slave oscillators predominantly controlled by the SCN pacemaker [15]. This hierarchical
organization of circadian clocks was supported by the observation that the rhythmicity in
peripheral organs is lost upon SCN lesion and by SCN transplantation experiments, which
showed that host hamster acquired the period of activity from the SCN donors [16,17].
The SCN was then seen as a synchronizing center to keep peripheral clocks in phase [18].
This master–slave view has been challenged by several observations. First, the production
of hormones such as melatonin or cortisol does not adjust to the rhythm of the grafted
SCN in hamsters [19]. Second, the liver was shown to strongly respond to food intake.
An inversion of the food cycle was indeed shown to invert the rhythm of the liver [20,21],
Biology 2021, 10, 207 3 of 15

Biology 2021, 10, x FOR PEER REVIEW 3 of 15

leading to a (partial) decoupling of this peripheral clock from the SCN. Third, these pe-
SCN in hamsters [19]. Second, the liver was shown to strongly respond to food intake. An
ripheral clocks feed back to the SCN. For example, leptin, an appetite suppressor, which
inversion of the food cycle was indeed shown to invert the rhythm of the liver [20,21],
is secreted
leading by adipocytes,
to a (partial) decouplingmay target
of this the SCN
peripheral clockvia
from the
thearcuate nucleus
SCN. Third, these(ARC)
pe- as demon-
strated by the disruption of the circadian rhythm in food intake
ripheral clocks feed back to the SCN. For example, leptin, an appetite suppressor, which induced by ablation of
leptin-receptor-expressing neurons in the ARC [22].
is secreted by adipocytes, may target the SCN via the arcuate nucleus (ARC) as demon-
stratedWhereas
by the disruption
the SCNofmainly
the circadian
senses rhythm
light,inperipheral
food intakeclocks
induced by ablation
receive inputsof from nutrient
leptin-receptor-expressing neurons in the ARC [22].
sensors, including SIRT1 (Sirtuin 1) [23,24], AMPK [25], hormones [18,26], and neuronal
Whereas the SCN mainly senses light, peripheral clocks receive inputs from nutrient
cues [27]. Peripheral clocks are thus subject to multiple zeitgebers, mutually coupled, and
sensors, including SIRT1 (Sirtuin 1) [23,24], AMPK [25], hormones [18,26], and neuronal
able to feed back to the SCN pacemaker. Moreover, central and peripheral clocks differ in
cues [27]. Peripheral clocks are thus subject to multiple zeitgebers, mutually coupled, and
their sensitivity
able to feed back to the forSCN
different zeitgebers,
pacemaker. Moreover,andcentral
mostandlikely the peripheral
peripheral clocks
clocks differ in themselves
also differ in their sensitivity for different zeitgebers. For example,
their sensitivity for different zeitgebers, and most likely the peripheral clocks themselves the nutrient sensor
AMPK
also differdifferentially affects
in their sensitivity for different peripheral
different zeitgebers. Forclocks [28].
example, theAs a consequence,
nutrient sensor changing
AMPK differentially affects different peripheral clocks [28]. As a consequence,
the timing of one zeitgeber can be sufficient to disorganize the circadian system, since suchchanging
the timing
a shift of one
will have zeitgeber
differentcanconsequences
be sufficient to fordisorganize the circadian
the different clocks.system, since
Altering zeitgebers thus
such a shift will have different consequences for the different clocks. Altering zeitgebers
has effects on the global circadian organization, both at the genetic and at the organ levels,
thus has effects on the global circadian organization, both at the genetic and at the organ
with with
levels, potential consequences
potential consequences on onphysiology
physiology and behavior
and behavior (Figure
(Figure 1A). 1A). Understanding
Understand-
ing the working of this system and the emergence of pathologies thus requires system- system-level
the working of this system and the emergence of pathologies thus requires
investigations
level investigations that
thatintegrate bothgenetic
integrate both geneticandand tissue
tissue levels.
levels.

Figure
Figure 1. (A)
1. (A) TheThe circadian
circadian systemsystem
consistsconsists of aofnetwork
of a network coupled of coupled
clocks, clocks,
receives receives
multiple inputs multiple inputs
(zeitgebers),
(zeitgebers),andand
controls manymany
controls aspectsaspects
of physiology and metabolism.
of physiology Alteration of the
and metabolism. zeitgebersof the zeitgebers
Alteration
(e.g., via circadian misalignment or jet lag) can lead to metabolic syndromes and pathologies. Of note,
central and peripheral clocks differ by the nature and/or sensitivity of the different zeitgebers. We
can also distinguish external zeitgebers (such as light and food, denoted by an asterisk) and internal
zeitgebers (hormones and neural projections involved in the internal coupling between the clocks).
(B) Scheme of the genetic regulatory network underlying circadian gene expression. The scheme
focuses on the core genes/proteins which are circadianly expressed. Clock, which is constitutively
expressed, is not represented.
Biology 2021, 10, 207 4 of 15

3. Architecture of the Circadian Gene Regulatory Network

The molecular mechanism of the circadian clock in mammals relies on multiple inter-
locked transcription/translation feedback loops, where several clock genes and regulatory
proteins play key roles (see [29] or [30] for comprehensive reviews) (Figure 1B). At the core
of this genetic regulatory network, a negative feedback loop is centered on the PER and
CRY proteins, which repress the expression of their own genes, Per (period) and Cry (cryp-
tochromes). There are actually three homologs of Per genes (named Per1, Per2, and Per3) and
two Cry homologs (Cry1 and Cry2). PER and CRY proteins heterodimerize in the cytoplasm
and translocate into the nucleus where they interact and thereby inhibit their transcriptional
factor CLOCK:BMAL1. The latter is a heterodimeric complex between CLOCK, encoded by
the constitutively expressed Clock (circadian locomotor output cycles kaput) gene, and BMAL1,
encoded by the circadian expression of Bmal1 (brain and muscle ARNT-like 1) gene. PER and
CRY proteins are degraded through ubiquitin-dependent pathways, leading to a release of
CLOCK:BMAL1, which can then launch a new circadian cycle.
Kinases and phosphatases regulate the phosphorylation state and thereby the localiza-
tion and stability of the clock proteins [29–31]. These multiple phosphorylations contribute
to create a delay necessary to produce long-period oscillations and to the fine-tuning of the
circadian oscillator [32]. Impaired phosphorylation, due to mutations either in the phos-
phorylation site of the PER proteins or in their kinases/phosphatases has been associated
with advanced/delayed sleep phase syndrome in humans [33] as well as with perturbed
feeding rhythms and the development of obesity in mice [34].
Two secondary feedback loops are mediated through transcriptional activation of
Bmal1 by the retinoid-related orphan receptors (RORα,β,γ) [35] and repression by Rev-
Erbα/β [36] (Figure 1B). These interlocked regulatory loops contribute to produce a robust
rhythm in Bmal1 transcription. Whereas knockout of Bmal1 leads to arrhythmicity [37],
rhythmic change in BMAL1 abundance is however not required to drive circadian oscilla-
tions [38]. The idea that interlocked feedback loops promote and amplify genetic oscilla-
tions was also supported by mathematical modeling [39,40]. Moreover, the ROR/Rev-Erb
feedback loops induce a delay in Cry1 expression, which is crucial for proper circadian
timing [29,41]. Next to ROR and Rev-Erb, the transcription factors DBP (D-box binding
protein) and E4BP4 (E4-binding Protein 4) further participate in circadian gene regulation
through binding to D boxes present in the promoter of Per and RevErb genes [42]. Inter-
estingly, these factors also play an important role in metabolism and immune response
regulation [43,44].
This intricate gene regulatory network operates at the single-cell level both in the
SCN and in peripheral tissues [45]. In the SCN, there is, however, an additional level of
organization. Neurons are indeed strongly coupled together via the circadian production
of neurotransmitters and of their receptors [46]. Supported by theoretical models [47,48], it
was suggested that this intercellular coupling contributes to generate a robust, overt rhythm
from otherwise weak and possibly damped single-cell SCN oscillators [49]. Intercellular
coupling appears weak or absent in peripheral organs. Cellular oscillators in these organs,
thus, appear mainly under the influence of external cues, which may play the role of
global synchronizers.
Light signals perceived by the retina interact with the SCN clock by inducing the
expression of Per1 and Per2, as well as that of Cry1 and Cry2 genes [50–52]. The effect
of light is mediated by CREB, a Ca2+ /cAMP response element-binding protein [53–55].
Phosphorylation of CREB can be induced by light or glutamate and is required for the
entrainment of the circadian oscillations by light–dark cycles [53]. Peripheral clocks do
not perceive light but are targets of hormonal and neuronal signals coming from the SCN
or other peripheral organs. These signals modify clock gene expression. For example,
glucocorticoid signaling is known to act on the transcription of Per [18] and of RevErbα
genes [56].
The CLOCK:BMAL1 complex not only activates the transcription of the Per and
Cry genes but also drives the expression of numerous clock-controlled output genes [57].
Biology 2021, 10, 207 5 of 15

By recruiting histone-modifying enzymes (through methylation and acetylation), this

complex is responsible for chromatin modification, allowing or preventing gene expres-
sion [58,59]. BMAL1 targets were highly enriched in genes involved in carbohydrate and
lipid metabolism, as well as in genes coding for transcription factors, in particular nuclear
receptors [60]. Thus, by stimulating the expression of the PAR-bZIP transcription factors,
CLOCK:BMAL1 also controls the expression of many additional genes [61]. Besides its key
role in the circadian clock, the transcriptional repressor RevErbα also controls the expres-
sion of many genes involved in carbohydrate and lipid metabolism, in the inflammation
response, and in hormone signaling [62].

4. Circadian Timing of Clock Genes and Clock-Controlled Genes in

Physiological Conditions
The above gene regulatory network contains all the necessary elements to produce
autonomous 24 h oscillations. In particular, the core negative transcriptional feedback,
the delay (due to phosphorylations and transport), and several sources of nonlinearity
(resulting from transcriptional repression, multisite phosphorylation, and/or complex
formation) are key ingredients for the oscillatory mechanism. Moreover, computational
simulation shows that these oscillations can readily be entrained by light–dark cycles
through the periodic forcing of light-controlled kinetic parameters [39,63–65].
The occurrence of oscillations is however not the only requirement for a functional
clock. The timing of the oscillations (i.e., the phase of each component) is of course critical
to guarantee a proper adaptation to the periodic lighting and feeding conditions. Despite
the conservation of the genetic machinery in most tissues and a relatively well-conserved
entrainment phase of the different clock components [57,66,67], at least under healthy
conditions, the phase of numerous clock-controlled genes appears tissue-specific [57,68].
It should also be stressed that the number of oscillatory genes strongly depends on the
organ, with the liver ranked at the top with more than 3000 oscillatory genes, and that
the number of genes displaying oscillations in several organs is very low [57]. Thus, the
circadian scheduling of physiology is highly organ specific.
It is possible that the multiplicity of feedback loops and apparent redundancy of core
clock genes—not absolutely required to generate entrainable circadian oscillations—arose
from the need to precisely and robustly set the phase of each component. Gene-specific
regulation of post-translational processes may further contribute to adjust the phase of
expression of core genes [32]. This complexity also offers more flexibility to the clock,
which can be optimized for specific functions in each organ. The difference in the timing of
the expression of clock genes and clock-controlled genes across organs can be explained,
at least partly, by differences in kinetic parameters and by the combinatorial presence
and interplay of multiple regulatory binding sites [69,70]. It is also likely that additional
factors such as tissue-specific clock inputs and differentially expressed regulatory genes
also contribute to this variability.
The computational studies mentioned above, together with other theoretical works,
support the idea that the clock has evolved to precisely adjust gene expression to cyclical
changes in the zeitgebers under natural conditions. As we will see below, the phase of
the clock and clock-controlled genes can, however, be perturbed under unusual living
conditions and in particular under conflicting time cues such as ill-timed feeding, with
consequences on the physiology.

5. Ill-Timed Feeding Pattern and Twisted Clocks

The circadian system allows the organisms to adapt to normal environmental and
feeding conditions by anticipating daily changes and, thus, helping them to remain in a
homeostatic, healthy state. The proper working of the circadian system, however, relies
on the alignment between the environmental lighting conditions and feeding time. By
“alignment” we mean that the timings of these external cues are such that the circadian
clock system can anticipate these changes and, thereby, guarantee an optimal adaptation.
For example, by regulating the transcription of genes involved in insulin secretion, the
Biology 2021, 10, 207 6 of 15

pancreas prevents an excessive accumulation of blood glucose [71]. The liver also plays a
critical role in the maintenance of glucose homeostasis through the circadian regulation of
gluconeogenesis [13,72]. There is an increasing amount of evidence showing that circadian
disturbances lead to metabolic syndromes and pathologies such as diabetes in humans.
These disturbances include shift work, wrong eating timing, sleep loss, and chronic jet
lag [73–78].
To understand how metabolic syndromes and diseases occur under perturbed condi-
tions, it is necessary to investigate the impact of environmental factors and their timing
on the clock gene expression and, subsequently, on the physiology. It was, for example,
observed that perturbing the metabolism through a high-fat diet or through ill-timed
feeding results in tissue and gene-specific alteration of the expression levels of clock and
clock-controlled genes, hormone production, and activity [79–82].
Bray et al. [80] measured clock gene expression in various organs in mice with day-
and night-restricted feeding (liver, epididymal fat, gastrocnemius muscle, and heart). They
report dramatic phase shifts, in the range of 6 to 11 h, in the expression of clock genes
in the liver of mice fed during the day vs. those fed at night. Compared to the liver,
restricted feeding had markedly lower and less consistent effects on the phases of clock
gene oscillations in epididymal fat, muscle, and heart. In these organs, average phase shifts
were around 3 to 7 h. Furthermore, the amplitude of gene expression oscillations were
often significantly dampened in these peripheral tissues. Bray et al. [80] also recorded the
expression of numerous metabolic genes. Consistently with the behavior of clock genes’
expression, metabolic genes’ expression appears to be strongly rhythmic and shifted by
12 h in the liver and more damped and only partially shifted in other organs.
These observations are in line with the results of Opperhuizen et al. [83], who also
showed that ill-timed feeding results in gene-specific phase shifts of clock and metabolic
genes’ rhythms in the liver and in a loss of rhythmicity in the muscle in rats. Similarly,
de Goede et al. [84] highlight differences in the gene expression in muscle and in brown
adipose tissues upon time-restricted feeding in rats (Figure 2). These data support the
hypothesis that time-restricted feeding leads to a desynchrony of peripheral organs from the
SCN pacemaker and that the changes in clock gene expression are gene specific. Indeed, in
a given organ, not all clock genes are affected in the same way, and the phase relationships
between the genes are not conserved. In the following, we refer to these dysregulated
clocks as “twisted clocks”.
In an earlier study, Bur et al. [85] compared the expression of clock genes in the liver
and in the pituitary gland at different photoperiods and upon various feeding regimes
in mice. The pituitary was chosen because it conveys the SCN signal to peripheral or-
gans through periodic hormone release while being able to generate rhythmic expression
of circadian genes independently of the SCN. The pituitary also controls the release of
glucocorticoids (GCs) by the adrenal gland and, thereby, impacts peripheral clocks and
metabolism. Bur and colleagues found that under the 12:12 LD cycle, the expression of
clock genes was very similar in the liver and in the pituitary gland. Compared to the
12:12 LD cycle, oscillations under a long day (LD 16:8) appear delayed in a similar way in
both organs. In contrast, under a short day (LD 8:16), the two organs exhibit different and
gene-specific phase shifts (with respect to the 12:12 LD case). Bur et al. [85] also found that
daytime-restricted feeding (compared to nighttime-restricted) reversed the rhythmicity of
clock gene expression in the liver, whereas it strongly dampened the oscillations in the pi-
tuitary gland. These authors further noticed that, in the absence of GCs (adrenalectomized
mice), the inversion of gene expression in the liver clock following the transition from
day-restricted feeding to night-restricted feeding is faster (consistent with the hypothesis
that the GCs acts as an inhibitor of phase shift), but that the daily expression of clock genes
is not affected by the feeding timing. In contrast, GC removal does not alter the response of
the pituitary clockwork to daytime-restricted feeding. The pituitary clock appears resilient
to adrenalectomy. From the complex, intermediate response of the pituitary, the authors
concluded that this gland integrates both light and food signals, contrarily to the SCN,
 pituitary gland. These authors further noticed that, in the absence of GCs (adrenalecto-
mized mice), the inversion of gene expression in the liver clock following the transition
from day-restricted feeding to night-restricted feeding is faster (consistent with the hy-
pothesis that the GCs acts as an inhibitor of phase shift), but that the daily expression of
clock genes is not affected by the feeding timing. In contrast, GC removal does not alter
Biology 2021, 10, 207 7 of 15
the response of the pituitary clockwork to daytime-restricted feeding. The pituitary clock
appears resilient to adrenalectomy. From the complex, intermediate response of the pitu-
itary, the authors concluded that this gland integrates both light and food signals, contra-
which
rily to primarily
the SCN, adjusts to the lightadjusts
which primarily signal, to
and
theperipheral organs
light signal, and(such as the organs
peripheral liver), which
(such
are mainly
as the entrained
liver), bymainly
which are food. entrained by food.

Figure2.
Figure 2. Experimentally
Experimentally measured
measured clock
clock gene
gene expression
expression in
in different
different organs
organs in
in control
control (ad
(ad libitum)
libitum) and
and daytime-restricted
daytime-restricted
feeding (RT) conditions in rodents. Data for the suprachiasmatic nucleus (SCN) and liver are from [81]. DataData
feeding (RT) conditions in rodents. Data for the suprachiasmatic nucleus (SCN) and liver are from [81]. for muscle
for muscle and
brown adipocytes (BATs) are from [84]. Note that for muscle and BATs, the data at t = 24 h are replicated fromfrom
and brown adipocytes (BATs) are from [84]. Note that for muscle and BATs, the data at t = 24 h are replicated datadata
at t
at t = 0. The circle plots on the right show the phase of the peak of gene expression. The small size of the arrows for muscle
= 0. The circle plots on the right show the phase of the peak of gene expression. The small size of the arrows for muscle
and BATs under RT reflects the fact that the rhythmicity in these conditions is strongly altered. This figure highlights the
and BATs under RT reflects the fact that the rhythmicity in these conditions is strongly altered. This figure highlights the
organ-specific phase relationships between the expression profile of core clock genes and the gene-specific alteration of
organ-specific
the expressionphase relationships
profile under RT. between the expression profile of core clock genes and the gene-specific alteration of the
expression profile under RT.
The genetic
The genetic and
and metabolic
metabolic consequences
consequences ofof ill-timed
ill-timed feeding
feeding was
was further
further investigated
investigated
by Mukherji et al. [81,82]. These authors found that transferring mice from nighttime
by Mukherji et al. [81,82]. These authors found that transferring mice from nighttime to to
daytime feeding shifted the phase of metabolic markers (levels of blood glucose,
daytime feeding shifted the phase of metabolic markers (levels of blood glucose, insulin, insulin,
free fatty
free fatty acids,
acids, etc.)
etc.) and
andthat
thatclock
clockgene
geneexpression
expressionininperipheral
peripheral organs
organswas
wasinverted, in
inverted,
in contrast to the SCN pacemaker, which remains unaffected by the feeding schedule
(Figure 2). Due to this desynchrony of the circadian system, animals developed metabolic
syndrome such as hypoinsulinemia, hypertriglyceridemia, and hyperglycemia, which, in
the long run, translate into pathologies including obesity and diabetes.

5.1. Internal Twist Explained by Mathematical Modeling

A closer look at the time profiles of blood glucose, insulin, and the expression of clock
genes reveals that the rhythms in the pancreas are not simply inverted upon inversion of
Biology 2021, 10, 207 8 of 15

the feeding time [81,82]. There is indeed a gene-specific phase shift in clock gene expression:
whereas Per2 is shifted by 12 h, Bmal1 and RevErbα are shifted by 8 h. Such a twist would
not be expected if the clock would be under the control of a single zeitgeber. In that case,
the timing of the zeitgeber would simply dictate the phase of the clock, and the phase
relationships between each gene would be set by the internal transcriptional–translational
circuitry. Differential phase shifts can however be explained by the mismatch between
two different signals entraining the peripheral clocks, one from the (unaffected) light-
controlled SCN pacemaker and the other from the (inverted) nutrient cues [86]. More
specifically, the conflicting inputs are likely to differently affect distinct core clock genes,
and mathematical modeling has shown that this is sufficient to create an internal twist
between these clock genes. These conflicting inputs could be, on the one hand, the nutrient
sensors AMPK/SIRT1 and hormones (e.g., insulin), which mainly affect the negative limb
of the clock [18,23–26] and on the other hand, cues from the autonomic nervous system,
which seem to strongly affect the Bmal1–RevErbα loop [87]. In turn, this twist will affect
downstream metabolic pathways. Indeed, since insulin secretion is under the control of
both the pancreatic circadian clock and food-dependent blood glucose, its rhythm is not
simply shifted but also reduced in amplitude, as supported by a mathematical model that
incorporates both the nutrient periodic signal and the pancreatic circadian clock, which
receives both the SCN and the nutrient cues (Figure 3) [86]. This altered rhythm in insulin
Biology 2021, 10, x FOR PEER REVIEW 9 of 15
impairs glucose regulation, such that the amplitude of glucose is increased (hyperglycemia),
a model prediction consistent with the observations of Mukherji et al. [82].

Figure3.3. (A)
Figure (A) Scheme
Scheme ofof the
the model
model proposed
proposedbyby Woller
Wollerand
andGonze
Gonze[86].
[86].The
Therelease
releaseofofinsulin
insulinbybythe
thepancreas
pancreasisiscontrolled
controlled
by the pancreatic circadian clock and by the level of circulating glucose, the latter being directly dependent
by the pancreatic circadian clock and by the level of circulating glucose, the latter being directly dependent on feeding. on feeding.
The
The pancreatic clock is controlled both by the light-dependent SCN clock and by the blood glucose. (B) Summary
pancreatic clock is controlled both by the light-dependent SCN clock and by the blood glucose. (B) Summary of the results of the
results obtained
obtained with the with
modelthe model
[86]. [86]. In
In normal normal conditions,
conditions, when lightwhen lightcues
and food andare
food cues the
aligned, are clock
aligned, the clock
optimally optimally
mediates the
mediates the response so that the output (insulin) reaches a high level when needed to store glucose. Under conflicting
response so that the output (insulin) reaches a high level when needed to store glucose. Under conflicting conditions, the
conditions, the expression of clock genes is altered (twisted clock), and this translates into an impaired output (phase
expression of clock genes is altered (twisted clock), and this translates into an impaired output (phase shifted and low level
shifted and low level of insulin rhythm and hyperglycemia).
of insulin rhythm and hyperglycemia).

5.2. Another Conflicting

Thus, once Zeitgeber
challenged Paradigmzeitgebers, the complex architecture of the circa-
by conflicting
dian clock system is subject to twists
In a recent work, Heyde and Oster in gene expression
[88] designed and, consequently,
another to metabolic
conflicting zeitgeber para-
disorders.
digm. Metabolic activity was measured as well as clock gene expression in theand
We may wonder what the physiological advantage of such a complex SCN sensi-
and
tive circadian architecture is. Some insights can be provided by mathematical
in peripheral organs in mice subject to zeitgeber desynchrony. In one series of experi- modeling.
The above
ments, model
they can
apply indeed
a 28 h LDbecycle
modified to compare
(LD 14:14) and athe
24 original system with
h feeding/fasting hypothetical
alternance (ZD
alternative architectures, which either do not incorporate the influence
condition). In a second series of experiments, these conditions are inversed: LD of the SCN on and
12:12 the
pancreatic clock or do not rely on any clock control [86]. Numerical analysis
feeding/fasting 14:14 (inversed ZD or iZD condition). These protocols lead to an alter- of the three
systems predicts that in the absence of light information from the SCN, the pancreatic clock
nance of in-phase and out-of-phase regimes over time. Under ZD, most mice exhibit a
period between 24 and 28 h in their locomotor activity, highlighting a combined effect of
the two periodic cues on the overall circadian rhythmicity. Regarding clock gene expres-
sion, Bmal1 and Per2 profiles in the SCN are affected by the regime: both genes display a
phase shift of about 5 h between the in-phase and out-of-phase regimes, but the amplitude
Biology 2021, 10, 207 9 of 15

(and therefore glucose metabolism) would perfectly adapt to the new feeding schedule
(no twist), but the adaptation of such a system is predicted to take some time because of
the inertia of the local clock. In contrast, in the absence of any clock control, the metabolic
response to large chronic changes in food availability is immediate, without any anticipa-
tion. This suggests that there is a tradeoff between the ability to sense light and nutrient
cues (with anticipation) and the ability to fully adapt to chronic perturbations in these
zeitgebers [86].

5.2. Another Conflicting Zeitgeber Paradigm

In a recent work, Heyde and Oster [88] designed another conflicting zeitgeber paradigm.
Metabolic activity was measured as well as clock gene expression in the SCN and in pe-
ripheral organs in mice subject to zeitgeber desynchrony. In one series of experiments, they
apply a 28 h LD cycle (LD 14:14) and a 24 h feeding/fasting alternance (ZD condition). In a
second series of experiments, these conditions are inversed: LD 12:12 and feeding/fasting
14:14 (inversed ZD or iZD condition). These protocols lead to an alternance of in-phase
and out-of-phase regimes over time. Under ZD, most mice exhibit a period between 24
and 28 h in their locomotor activity, highlighting a combined effect of the two periodic
cues on the overall circadian rhythmicity. Regarding clock gene expression, Bmal1 and
Per2 profiles in the SCN are affected by the regime: both genes display a phase shift of
about 5 h between the in-phase and out-of-phase regimes, but the amplitude of Bmal1
expression appears weaker in antiphase days. The profiles of the clock gene expression in
the liver, white adipocytes, and adrenal gland are locked to the feeding regime and less
affected by the in-phase vs. out-of-phase condition. This is consistent with the fact the
SCN clock genes are directly targeted by light input, whereas in these peripheral organs
the expression of clock genes is mostly influenced by the feeding pattern. The release of
corticosterone—the main GC in rats and mice—by the adrenal gland is slightly shifted and
damped on antiphase days, possibly explained, at least partly, by the reduced expression
of Per2 in this condition. This impaired rhythmicity is expected to affect entrainment of
the peripheral clock (which is known to be mediated by corticosterone). Leptin release,
normally regulated by feeding, appears also to be influenced by the SCN under conflicting
cues. Finally, mice under ZD or iZD undergo cycles of weight gain and loss without overall
weight increase. This supports the hypothesis of a stabilization of peripheral clocks by
the entrained SCN, allowing energy homeostasis, in healthy (in phase) conditions and
a desynchrony in perturbed (out-of-phase) conditions. Mathematical modeling of such
nontrivial dynamics would certainly be helpful to better grasp the internal logic of the
circadian network.

6. Jet Lag and Gene-Specific Resynchronization Time

After a jet lag, the circadian system does not instantaneously synchronize to the new
time frame. It takes some time for the molecular clock, and hence for the physiological
processes, to adapt to the new schedule. This resynchronization time was shown to be
tissue-specific and to be responsible for the temporary disruption of the overall physio-
logical coordination. Yamazaki et al. [89] have found that during jet lag, Per1 expression
rhythm re-entrains faster in the SCN than in peripheral organs (skeletal muscle, liver, and
lung), consistent with the fact that the LD cycle is the most important zeitgeber for the
SCN. This difference in the resynchronization time also indicates that, while the SCN circa-
dian pacemaker entrains circadian oscillators in the periphery, this inter-organ coupling is
impaired after a jet lag. Similar conclusions were reported for Per2 [90].
By measuring the expression of individual clock genes upon jet lag, Reddy et al. [91]
and Kiessling et al. [92] found that the resynchronization time is gene specific. In the
mouse SCN, in response to a large advance of the light–dark cycle, Per genes were found
to resynchronize much faster than Cry genes, which readapt more gradually, similar
to the activity–rest cycle [91]. Still in the SCN, Kiessling et al. [92] observed that the
resynchronization time of the Per genes is much shorter than that for Bmal1 or RevErb.
 the activity–rest cycle [91]. Still in the SCN, Kiessling et al. [92] observed that the resyn-
chronization time of the Per genes is much shorter than that for Bmal1 or RevErb. Other
transient dynamics are, however, observed in other organs. For example, in the pancreas,
all genes take more time to re-entrain, and RevErbα appears to be slightly faster.
Biology 2021, 10, 207 This prompted the authors to have a closer look at the link between the pacemaker 10 of 15
and the peripheral clocks. Adrenal glucocorticoids (GCs) are able to reset peripheral
clocks [18], and the adrenal circadian clock regulates the rhythmic release of GCs into the
blood [93]. Kiessling et al. [92] discovered that genetic ablation of the adrenal clock accel-
Other transient dynamics are, however, observed in other organs. For example, in the
erated the rate of re-entrainment and that the application of an inhibitor of GC synthesis
pancreas, all genes take more time to re-entrain, and RevErbα appears to be slightly faster.
resulted in either acceleration or deceleration of behavioral adaptation to the new time
This prompted the authors to have a closer look at the link between the pacemaker and
zone,
the depending
peripheral on the
clocks. time at
Adrenal which the GC(GCs)
glucocorticoids inhibitor is administered.
are able They clocks
to reset peripheral concluded[18],
that the adrenal circadian clock, through control of GC rhythms, is
and the adrenal circadian clock regulates the rhythmic release of GCs into the blood [93]. a major regulator of
re-entrainment to jet lag.
Kiessling et al. [92] discovered that genetic ablation of the adrenal clock accelerated the rate
In the same line,
of re-entrainment Onothe
and that et al. [94] reported
application of anthat Per and
inhibitor of GCBmal1 re-entrained
synthesis resultedat in
the new
either
phase at different speeds after the application of a 9 h light pulse
acceleration or deceleration of behavioral adaptation to the new time zone, depending given just before the
subjective night (Figure 4A). Such a dissociation between the Per1
on the time at which the GC inhibitor is administered. They concluded that the adrenal and Bmal1 oscillations
was observed
circadian clock,inthrough
the SCNcontrol
ex vivoofand GCin living mice.
rhythms, is a major regulator of re-entrainment to
A
jet lag. strong heterogeneity in the kinetics of re-entrainment is thus found both at the
tissue
Inand
the at
sameindividual
line, Ono gene levels.
et al. These results
[94] reported suggest
that Per that the
and Bmal1 coordination
re-entrained of clock
at the new
phase at different speeds after the application of a 9 h light pulse given just beforelocal
gene expression is globally disrupted during jet lag, creating a temporary twist of the the
clocks. This
subjective likely
night explains
(Figure 4A).whySuchmetabolic and activity
a dissociation betweenrhythms
the Per1areandtemporarily altered
Bmal1 oscillations
afterobserved
was a jet lag. in the SCN ex vivo and in living mice.

Figure 4. (A) Upon jet lag, not all clock genes resynchronize at the same speed [92,94]. (B) This
transient twist can be explained by the internal desynchrony of the clock genes network [95].

A strong heterogeneity in the kinetics of re-entrainment is thus found both at the

tissue and at individual gene levels. These results suggest that the coordination of clock
gene expression is globally disrupted during jet lag, creating a temporary twist of the local
clocks. This likely explains why metabolic and activity rhythms are temporarily altered
after a jet lag.

Temporary Twist and Internal Decoupling Explained by Mathematical Modeling

Two hypotheses have been put forward to explain the observations: desynchroniza-
tion at the single-cell level or emergence of two groups of cells in which either Per or Bmal1
is predominant [94]. A surrogate data analysis of the experimental time series published
by Kiessling et al. [92] supports the first hypothesis [95]. Schmal et al. [95] designed a
conceptual mathematical model to show how internal dissociation can result from the
decoupling between two autonomous oscillatory circuits (Figure 4B). This model is based
on the hypothesis that Per and Bmal1/Rev intracellular regulatory feedback loops can inde-
pendently generate oscillations. After calibration of the model to reproduce experimentally
measured resynchronization times after a jet lag, they estimated the intrinsic period of each
oscillator as well as its coupling strength. They also successfully reproduced these results
with a minimal model and with a more detailed molecular model. The desynchronization
of the Per and Bmal1/RevErb dynamics is obtained provided that Per and Bmal1/RevErb
submodels can be dissociated. The latter model allows to link the coupling strength to an
Biology 2021, 10, 207 11 of 15

actual regulatory process, namely the strength of inhibition of RevErb transcription by PER
protein. Modeling thus constitutes a valuable tool to interpret experimental data and to
reveal design principles of the molecular architecture. The next step will be to understand
how such an internal twist affects circadian outputs, such as the release of hormones and
the metabolic activity.

7. Concluding Remarks
The observations reviewed in this paper, together with many others, highlight the
importance of considering the circadian regulatory system not as a pure hierarchical system
but rather as an intricate network of interconnected organs, each of them having its own
clock, its own zeitgebers, and its own clock-controlled outputs. There is thus a need to
examine gene expression at the level of each organ and to carefully characterize the profile
of expression of each individual clock gene. Knockout studies are crucial to determine the
role of each gene in the gene regulatory network. It is also important to complement these
findings with studies that mimic natural—but possibly perturbed—conditions such as
ill-time feeding or jet lags. Those conditions can induce gene-specific phase shift (“twist”)
with potential negative consequences on energy homeostasis and physiology.
Since the discovery of the first clock genes, mathematical models for the core regulatory
circuit have been developed. In parallel with the experimental discovery of additional genes
and regulations, these models were updated and gained in complexity [63–65]. Once they
account for the main features of the circadian oscillations and for the behavior of knockout
mutants, these models can be used as a tool to interpret the experimental observations,
which may defy intuition [96]. As exemplified in the present review, the gene-specific phase
shift induced by time-restricted feeding and its consequence for the profile of glucose and
insulin was successfully reproduced by a model that incorporates multiple inputs on the
genetic clock [86]. The gene-specific resynchronization time after a jet lag was convincingly
explained by an internal desynchronization of the gene clockwork [95]. Another recent
theoretical study focused on the durations of the different circadian phases (defined by
the elevated expression of given clock genes) and their alteration upon mutations [97].
Interestingly, this model highlights another possible type of twist. Indeed, the increase of
the duration of the different phases upon mutation was found to be gene dependent: while
the profile of some genes appears relatively insensitive to mutations, other genes appear
highly expressed during a shorter time.
These theoretical works thus allow the exploration of the fine-tuning of the genetic
circadian clocks, which in turn may be responsible for subtle changes in clock-controlled
processes. Models focusing on the circadian control of metabolism already exist [98–100].
The next step will be to incorporate both genetic details and clock-controlled processes into
an integrated model. We expect that in the future, with the accumulation of organ-specific
measurements and high-resolution gene-specific expression recordings, quantitative com-
putational models will play an important role in understanding how the complex circadian
clock system integrates multiple environmental cues to guarantee a proper adaptation of
physiology and metabolism and how misalignment causes diseases. This will open further
perspectives to find new drugs and to establish new chronopharmacological protocols to
treat these diseases.

Funding: This work was supported by the “Fonds National de la Recherche Scientifique” (FNRS,
Belgium) (convention no. T.0015.16 (26021674) and CDR no. J.0076.19). A.W. is supported by the
Feinberg Graduate School from the Weizmann Institute.
Acknowledgments: We would like to thank E. Maury for her invitation to contribute to this
special issue. We also would like to thank E. Maury and the anonymous reviewers for their
constructive suggestions.
Conflicts of Interest: The authors declare no conflict of interest.
Biology 2021, 10, 207 12 of 15

References
1. Chaix, A.; Zarrinpar, A.; Panda, S. The circadian coordination of cell biology. J. Cell Biol. 2016, 215, 15–25. [CrossRef]
2. Finger, A.; Dibner, C.; Kramer, A. Coupled network of the circadian clocks: A driving force of rhythmic physiology. FEBS Lett.
2020, 594, 2734–2769. [CrossRef]
3. Maury, E. Off the clock: From circadian disruption to metabolic disease. Int. J. Mol. Sci. 2019, 20, 1597. [CrossRef]
4. Mohawk, J.A.; Green, C.B.; Takahashi, J.S. Central and peripheral circadian clocks in mammals. Annu. Rev. Neurosci. 2012, 35,
445–462. [CrossRef] [PubMed]
5. Hastings, M.H.; Maywood, E.S.; Brancaccio, M. The mammalian circadian timing system and the suprachias-matic nucleus as its
pacemaker. Biology 2019, 8, 13. [CrossRef] [PubMed]
6. Herzog, E.D. Neurons and networks in daily rhythms. Nat. Rev. Neurosci. 2007, 8, 790–802. [CrossRef] [PubMed]
7. Herzog, E.D.; Hermanstyne, T.; Smyllie, N.J.; Hastings, M.H. Regulating the suprachiasmatic nucleus (SCN) cir-cadian clockwork:
Interplay between cell-autonomous and circuit-level mechanisms. Cold Spring Harb. Perspect. Biol. 2017, 9, a027706. [CrossRef]
8. Saper, C.B.; Scammell, T.E.; Lu, J. Hypothalamic regulation of sleep and circadian rhythms. Nat. Cell Biol. 2005, 437,
1257–1263. [CrossRef]
9. Challet, E. The circadian regulation of food intake. Nat. Rev. Endocrinol. 2019, 15, 393–405. [CrossRef]
10. Panda, S. Circadian physiology of metabolism. Science 2016, 354, 1008–1015. [CrossRef]
11. Buijs, F.N.; León-Mercado, L.; Guzmán-Ruiz, M.; Guerrero-Vargas, N.N.; Romo-Nava, F.; Buijs, R.M. The circadian system: A
regulatory feedback network of periphery and brain. Physiology 2016, 31, 170–181. [CrossRef]
12. Stenvers, D.J.; Scheer, F.A.J.L.; Schrauwen, P.; La Fleur, S.E.; Kalsbeek, A. Circadian clocks and insulin resistance. Nat. Rev.
Endocrinol. 2019, 15, 75–89. [CrossRef]
13. Mukherji, A.; Bailey, S.M.; Staels, B.; Baumert, T.F. The circadian clock and liver function in health and disease. J. Hepatol. 2019, 71,
200–211. [CrossRef]
14. Owino, S.; Buonfiglio, D.D.C.; Tchio, C.; Tosini, G. Melatonin signaling a key regulator of glucose homeostasis and energy
metabolism. Front. Endocrinol. 2019, 10, 488. [CrossRef] [PubMed]
15. Saini, C.; Suter, D.M.; Liani, A.; Gos, P.; Schibler, U. The mammalian circadian timing system: Synchronization of peripheral
clocks. Cold Spring Harb. Symp. Quant. Biol. 2011, 76, 39–47. [CrossRef] [PubMed]
16. Ralph, M.R.; Foster, R.G.; Davis, F.C.; Menaker, M. Transplanted suprachiasmatic nucleus determines circadian period. Science
1990, 247, 975–978. [CrossRef] [PubMed]
17. Silver, R.; LeSauter, J.; Tresco, P.A.; Lehman, M.N. A diffusible coupling signal from the transplanted suprachi-asmatic nucleus
controlling circadian locomotor rhythms. Nature 1996, 382, 810–813. [CrossRef] [PubMed]
18. Balsalobre, A.; Brown, S.A.; Marcacci, L.; Tronche, F.; Kellendonk, C.; Reichardt, H.M.; Schütz, G.; Schibler, U. Reset-ting of
circadian time in peripheral tissues by glucocorticoid signaling. Science 2000, 289, 2344–2347. [CrossRef] [PubMed]
19. Meyer-Bernstein, E.L.; Jetton, A.E.; Matsumoto, S.I.; Markuns, J.F.; Lehman, M.N.; Bittman, E.L. Effects of suprachi-
asmatic transplants on circadian rhythms of neuroendocrine function in golden hamsters. Endocrinology 1999, 140, 207–218.
[CrossRef] [PubMed]
20. Damiola, F.; Le Minh, N.; Preitner, N.; Kornmann, B.; Fleury-Olela, F.; Schibler, U. Restricted feeding uncouples circadian oscilla-
tors in peripheral tissues from the central pacemaker in the suprachiasmatic nucleus. Genes Dev. 2000, 14, 2950–2961. [CrossRef]
21. Stokkan, K.-A.; Yamazaki, S.; Tei, H.; Sakaki, Y.; Menaker, M. Entrainment of the circadian clock in the liver by feeding. Science
2001, 291, 490–493. [CrossRef]
22. Li, A.J.; Wiater, M.F.; Oostrom, M.T.; Smith, B.R.; Wang, Q.; Dinh, T.T.; Roberts, B.L.; Jansen, H.T.; Ritter, S. Lep-tin-sensitive
neurons in the arcuate nuclei contribute to endogenous feeding rhythms. Am. J. Physiol. Regul. Integr. Comp. Physiol. 2012, 302,
R1313–R1326. [CrossRef]
23. Asher, G.; Gatfield, D.; Stratmann, M.; Reinke, H.; Dibner, C.; Kreppel, F.; Mostoslavsky, R.; Alt, F.W.; Schibler, U. SIRT1 regulates
circadian clock gene expression through PER2 deacetylation. Cell 2008, 134, 317–328. [CrossRef]
24. Nakahata, Y.; Kaluzova, M.; Grimaldi, B.; Sahar, S.; Hirayama, J.; Chen, D.; Guarente, L.P.; Sassone-Corsi, P. The NAD+-Dependent
deacetylase SIRT1 modulates clock-mediated chromatin remodeling and circadian control. Cell 2008, 134, 329–340. [CrossRef]
25. Lamia, K.A.; Sachdeva, U.M.; Di Tacchio, L.; Williams, E.C.; Alvarez, J.G.; Egan, D.F.; Vasquez, D.S.; Juguilon, H.; Panda, S.;
Shaw, R.J.; et al. AMPK regulates the circadian clock by cryptochrome phosphoryla-tion and degradation. Science 2009, 326,
437–440. [CrossRef]
26. Crosby, P.; Hamnett, R.; Putker, M.; Hoyle, N.P.; Reed, M.; Karam, C.J.; Maywood, E.S.; Stangherlin, A.; Chesham, J.E.; Hayter, E.A.;
et al. Insulin/IGF-1 drives PE-RIOD synthesis to entrain circadian rhythms with feeding time. Cell 2019, 177, 896–909. [CrossRef]
27. Schiaffino, S.; Blaauw, B.; Dyar, K.A. The functional significance of the skeletal muscle clock: Lessons from Bmal1 knockout
models. Skelet. Muscle 2016, 6, 1–9. [CrossRef]
28. Um, J.-H.; Pendergast, J.S.; Springer, D.A.; Foretz, M.; Viollet, B.; Brown, A.; Kim, M.K.; Yamazaki, S.; Chung, J.H. AMPK
Regulates circadian rhythms in a tissue- and isoform-specific manner. PLoS ONE 2011, 6, e18450. [CrossRef]
29. Partch, C.L.; Green, C.B.; Takahashi, J.S. Molecular architecture of the mammalian circadian clock. Trends Cell Biol. 2014, 24,
90–99. [CrossRef]
30. Takahashi, J.S. Transcriptional architecture of the mammalian circadian clock. Nat. Rev. Genet. 2017, 18, 164–179. [CrossRef]
Biology 2021, 10, 207 13 of 15

31. Hirano, A.; Fu, Y.-H.; Ptáček, L.J. The intricate dance of post-translational modifications in the rhythm of life. Nat. Struct. Mol.
Biol. 2016, 23, 1053–1060. [CrossRef]
32. St John, P.C.; Hirota, T.; Kay, S.A.; Doyle, F.J., III. Spatiotemporal separation of PER and CRY posttranslational regulation in the
mammalian circadian clock. Proc. Natl. Acad. Sci. USA 2014, 111, 2040–2045. [CrossRef] [PubMed]
33. Toh, K.L.; Jones, C.R.; He, Y.; Eide, E.J.; Hinz, W.A.; Virshup, D.M.; Ptácek, L.J.; Fu, Y.H. An hPer2 phosphorylation site mutation
in familial advanced sleep phase syndrome. Science 2001, 291, 1040–1043. [CrossRef]
34. Liu, Z.; Huang, M.; Wu, X.; Shi, G.; Xing, L.; Dong, Z.; Qu, Z.; Yan, J.; Yang, L.; Panda, S.; et al. PER1 phosphorylation specifies
feeding rhythm in mice. Cell Rep. 2014, 7, 1509–1520. [CrossRef]
35. Sato, T.K.; Panda, S.; Miraglia, L.J.; Reyes, T.M.; Rudic, R.D.; McNamara, P.; Naik, K.A.; FitzGerald, G.A.; Kay, S.A.; Hogenesch,
J.B.; et al. A functional genomics strategy reveals rora as a component of the mammalian circadian clock. Neuron 2004, 43, 527–537.
[CrossRef] [PubMed]
36. Preitner, N.; Damiola, F.; Lopez-Molina, L.; Zakany, J.; Duboule, D.; Albrecht, U.; Schibler, U. The orphan nuclear receptor
REV-ERBalpha controls circadian transcription within the positive limb of the mammalian circa-dian oscillator. Cell 2002, 110,
251–260. [CrossRef]
37. Bunger, M.K.; Wilsbacher, L.D.; Moran, S.M.; Clendenin, C.; Radcliffe, L.A.; Hogen Esch, J.B.; Simon, M.C.; Takahashi, J.S.; Brad-
field, C.A. Mop3 is an essential component of the master circadian pacemaker in mammals. Cell 2020, 103, 1009–1017. [CrossRef]
38. Liu, A.C.; Tran, H.G.; Zhang, E.E.; Priest, A.A.; Welsh, D.K.; Kay, S.A. Redundant function of REV-ERBalpha and beta and
non-essential role for Bmal1 cycling in transcriptional regulation of intracellular circadian rhythms. PLoS Genet. 2008, 4, e1000023.
[CrossRef] [PubMed]
39. Relógio, A.; Westermark, P.O.; Wallach, T.; Schellenberg, K.; Kramer, A.; Herzel, H. Tuning the mammalian circadian clock:
Robust synergy of two loops. PLoS Comput. Biol. 2011, 7, e1002309. [CrossRef]
40. Uriu, K.; Tei, H. Feedback loops interlocked at competitive binding sites amplify and facilitate genetic oscilla-tions. J. Theor. Biol.
2017, 428, 56–64. [CrossRef]
41. Ukai-Tadenuma, M.; Yamada, R.G.; Xu, H.; Ripperger, J.A.; Liu, A.C.; Ueda, H.R. Delay in feedback repression by cryptochrome 1
is required for circadian clock function. Cell 2011, 144, 268–281. [CrossRef] [PubMed]
42. Ueda, H.R.; Hayashi, S.; Chen, W.; Sano, M.; Machida, M.; Shigeyoshi, Y.; Iino, M.; Hashimoto, S. System-level identi-fication of
transcriptional circuits underlying mammalian circadian clocks. Nat Genet. 2005, 37, 187–192. [CrossRef] [PubMed]
43. Male, V.; Nisoli, I.; Gascoyne, D.M.; Brady, H.J. E4BP4: An unexpected player in the immune response. Trends Immunol. 2012, 33,
98–102. [CrossRef]
44. Zhao, Z.; Yin, L.; Wu, F.; Tong, X. Hepatic metabolic regulation by nuclear factor E4bpj. Mol. Endocrinol. 2021, 66, R15–R21.
[CrossRef] [PubMed]
45. Yoo, S.H.; Yamazaki, S.; Lowrey, P.L.; Shimomura, K.; Ko, C.H.; Buhr, E.D.; Siepka, S.M.; Hong, H.K.; Oh, W.J.; Yoo, O.J.; et al.
PERIOD2: Luciferase real-time reporting of circadian dynamics reveals per-sistent circadian oscillations in mouse peripheral
tissues. Proc. Natl. Acad. Sci. USA 2004, 101, 5339–5346. [CrossRef] [PubMed]
46. Aton, S.J.; Colwell, C.S.; Harmar, A.J.; Waschek, J.; Herzog, E.D. Vasoactive intestinal polypeptide mediates circa-dian rhythmicity
and synchrony in mammalian clock neurons. Nat. Neurosci. 2005, 8, 476–483. [CrossRef] [PubMed]
47. Bernard, S.; Gonze, D.; Cajavec, B.; Herzel, H.; Kramer, A. Synchronization-induced rhythmicity of circadian os-cillators in the
suprachiasmatic nucleus. PLoS Comput. Biol. 2007, 3, e68. [CrossRef] [PubMed]
48. Webb, A.B.; Taylor, S.R.; Thoroughman, K.A.; Doyle, F.J., III.; Herzog, E.D. Weakly circadian cells improve resyn-chrony. PLoS
Comput. Biol. 2012, 8, e1002787. [CrossRef]
49. Webb, A.B.; Angelo, N.; Huettner, J.E.; Herzog, E.D. Intrinsic, nondeterministic circadian rhythm generation in identified
mammalian neurons. Proc. Natl. Acad. Sci. USA 2009, 106, 16493–16498. [CrossRef]
50. Shigeyoshi, Y.; Taguchi, K.; Yamamoto, S.; Takekida, S.; Yan, L.; Tei, H.; Moriya, T.; Shibata, S.; Loros, J.J.; Dunlap, J.C.; et al.
Light-induced resetting of a mammalian circadian clock is associated with rapid induction of the mPer1 transcript. Cell 1997, 91,
1043–1053. [CrossRef]
51. Albrecht, U.; Sun, Z.S.; Eichele, G.; Lee, C.C. A differential response of two putative mammalian circadian regulators, mper1 and
mper2, to light. Cell 1997, 91, 1055–1064. [CrossRef]
52. Griffin, E.A., Jr.; Staknis, D.; Weitz, C.J. Light-independent role of CRY1 and CRY2 in the mammalian circadian clock. Science
1999, 286, 768–771. [CrossRef] [PubMed]
53. Gau, D.; Lemberger, T.; Von Gall, C.; Kretz, O.; Le Minh, N.; Gass, P.; Schmid, W.; Schibler, U.; Korf, H.W.; Schütz, G.; et al.
Phosphorylation of CREB Ser142 regulates light-induced phase shifts of the circadian clock. Neuron 2002, 34, 245–253. [CrossRef]
54. Tischkau, S.A.; Mitchell, J.W.; Tyan, S.H.; Buchanan, G.F.; Gillette, M.U. Ca2+ /cAMP response element-binding pro-tein (CREB)-
dependent activation of Per1 is required for light-induced signaling in the suprachiasmatic nu-cleus circadian clock. J. Biol. Chem.
2003, 278, 718–723. [CrossRef] [PubMed]
55. Travnickova-Bendova, Z.; Cermakian, N.; Reppert, S.M.; Sassone-Corsi, P. Bimodal regulation of mPeriod pro-moters by
CREB-dependent signaling and CLOCK/BMAL1 activity. Proc. Natl. Acad. Sci. USA 2002, 99, 7728–7733. [CrossRef] [PubMed]
56. Torra, I.P.; Tsibulsky, V.; Delaunay, F.; Saladin, R.; Laudet, V.; Fruchart, J.C.; Kosykh, V.; Staels, B. Circadian and glu-cocorticoid
regulation of Rev-erba expression in liver. Endocrinology 2000, 141, 3799–3806. [CrossRef] [PubMed]
Biology 2021, 10, 207 14 of 15

57. Zhang, R.; Lahens, N.F.; Balance, H.I.; Hughes, M.E.; Hogenesch, J.B. A circadian gene expression atlas in mammals: Implications
for biology and medicine. Proc. Natl. Acad. Sci. USA 2014, 111, 16219–16224. [CrossRef] [PubMed]
58. Etchegaray, J.-P.; Lee, C.; Wade, P.A.; Reppert, S.M. Rhythmic histone acetylation underlies transcription in the mammalian
circadian clock. Nat. Cell Biol. 2002, 421, 177–182. [CrossRef]
59. Koike, N.; Yoo, S.-H.; Huang, H.-C.; Kumar, V.; Lee, C.; Kim, T.-K.; Takahashi, J.S. Transcriptional architecture and chromatin
landscape of the core circadian clock in mammals. Science 2012, 338, 349–354. [CrossRef]
60. Rey, G.; Cesbron, F.; Rougemont, J.; Reinke, H.; Brunner, M.; Naef, F. Genome-wide and phase-specific DNA-binding rhythms of
BMAL1 control circadian output functions in mouse liver. PLoS Biol. 2011, 9, e1000595. [CrossRef] [PubMed]
61. Fang, B.; Everett, L.J.; Jager, J.; Briggs, E.; Armour, S.M.; Feng, D.; Roy, A.; Gerhart-Hines, Z.; Sun, Z.; Lazar, M.A. Circa-dian
enhancers coordinate multiple phases of rhythmic gene transcription In Vivo. Cell 2014, 159, 1140–1152. [CrossRef] [PubMed]
62. Wang, S.; Li, F.; Lin, Y.; Wu, B. Targeting REV-ERBα for therapeutic purposes: Promises and challenges. Theranostics 2020, 10,
4168–4182. [CrossRef] [PubMed]
63. Leloup, J.-C.; Goldbeter, A. Toward a detailed computational model for the mammalian circadian clock. Proc. Natl. Acad. Sci USA
2003, 100, 7051–7056. [CrossRef]
64. Forger, D.B.; Peskin, C.S. A detailed predictive model of the mammalian circadian clock. Proc. Natl. Acad. Sci. USA 2003, 100,
14806–14811. [CrossRef]
65. Mirsky, H.P.; Liu, A.C.; Welsh, D.K.; Kay, S.A.; Doyle, F.J. A model of the cell-autonomous mammalian circadian clock. Proc. Natl.
Acad. Sci. USA 2009, 106, 11107–11112. [CrossRef]
66. Yamamoto, T.; Nakahata, Y.; Soma, H.; Akashi, M.; Mamine, T.; Takumi, T. Transcriptional oscillation of canonical clock genes in
mouse peripheral tissues. BMC Mol. Biol. 2004, 5, 18. [CrossRef] [PubMed]
67. Liu, S.; Cai, Y.; Sothern, R.B.; Guan, Y.; Chan, P. Chronobiological Analysis of circadian patterns in transcription of seven key
clock genes in six peripheral tissues in mice. Chronobiol. Int. 2007, 24, 793–820. [CrossRef]
68. Storch, K.F.; Lipan, O.; Leykin, I.; Viswanathan, N.; Davis, F.C.; Wong, W.H.; Weitz, C.J. Extensive and divergent cir-cadian gene
expression in liver and heart. Nature 2002, 417, 78–83. [CrossRef] [PubMed]
69. Korenčič, A.; Bordyugov, G.; Košir, R.; Rozman, D.; Goličnik, M.; Herzel, H. The interplay of cis-regulatory ele-ments rules
circadian rhythms in mouse liver. PLoS ONE 2012, 7, e46835. [CrossRef]
70. Korenčič, A.; Košir, R.; Bordyugov, G.; Lehmann, R.; Rozman, D.; Herzel, H. Timing of circadian genes in mamma-lian tissues.
Sci. Rep. 2014, 4, 5782. [CrossRef]
71. Perelis, M.; Marcheva, B.; Ramsey, K.M.; Schipma, M.J.; Hutchison, A.L.; Taguchi, A.; Peek, C.B.; Hong, H.; Huang, W.; Omura, C.;
et al. Pancreatic β cell enhancers regulate rhythmic transcription of genes controlling insulin secretion. Science 2015, 350, 4250.
[CrossRef] [PubMed]
72. Kalsbeek, A.; La Fleur, S.; Fliers, E. Circadian control of glucose metabolism. Mol. Metab. 2014, 3, 372–383. [CrossRef] [PubMed]
73. Kettner, N.M.; Mayo, S.A.; Hua, J.; Lee, C.; Moore, D.D.; Fu, L. Circadian dysfunction induces leptin resistance in mice. Cell Metab.
2015, 22, 448–459. [CrossRef]
74. Dibner, C.; Schibler, U. Circadian timing of metabolism in animal models and humans. J. Internal. Med. 2015, 277, 513–527.
[CrossRef] [PubMed]
75. Oosterman, J.E.; Kalsbeek, A.; La Fleur, S.E.; Belsham, D.D. Impact of nutrients on circadian rhythmicity. Am. J. Physiol. Integr.
Comp. Physiol. 2015, 308, R337–R350. [CrossRef]
76. Qian, J.; Scheer, F.A. Circadian system and glucose metabolism: Implications for physiology and disease. Trends Endocrinol. Metab.
2016, 27, 282–293. [CrossRef]
77. Challet, E. Keeping circadian time with hormones. Diabetes Obes. Metab. 2015, 17, 76–83. [CrossRef]
78. Allada, R.; Bass, J. Circadian mechanisms in medicine. N. Engl. J. Med. 2021, 384, 550–561. [CrossRef]
79. Kohsaka, A.; Laposky, A.D.; Ramsey, K.M.; Estrada, C.; Joshu, C.; Kobayashi, Y.; Turek, F.W.; Bass, J. High-fat diet dis-rupts
behavioral and molecular circadian rhythms in mice. Cell Metab. 2007, 6, 414–421. [CrossRef]
80. Bray, M.S.; Ratcliffe, W.F.; Grenett, M.H.; Brewer, R.A.; Gamble, K.L.; Young, M.E. Quantitative analysis of light-phase restricted
feeding reveals metabolic dyssynchrony in mice. Int. J. Obes. 2013, 37, 843–852. [CrossRef]
81. Mukherji, A.; Kobiita, A.; Chambon, P. Shifting the feeding of mice to the rest phase creates metabolic altera-tions, which, on their
own, shift the peripheral circadian clocks by 12 hour. Proc. Natl. Acad. Sci. USA 2015, 112, E6683–E6690. [CrossRef] [PubMed]
82. Mukherji, A.; Kobiita, A.; Damara, M.; Misra, N.; Meziane, H.; Champy, M.-F.; Chambon, P. Shifting eating to the circadian rest
phase misaligns the peripheral clocks with the master SCN clock and leads to a metabolic syndrome. Proc. Natl. Acad. Sci. USA
2015, 112, E6691–E6698. [CrossRef]
83. Opperhuizen, A.-L.; Wang, D.; Foppen, E.; Jansen, R.; Boudzovitch-Surovtseva, O.; De Vries, J.; Fliers, E.; Kalsbeek, A. Feeding
during the resting phase causes profound changes in physiology and desynchronization between liver and muscle rhythms of
rats. Eur. J. Neurosci. 2016, 44, 2795–2806. [CrossRef] [PubMed]
84. De Goede, P.; Sen, S.; Oosterman, J.E.; Foppen, E.; Jansen, R.; la Fleur, S.E.; Challet, E.; Kalsbeek, A. Differential effects of diet
composition and timing of feeding behavior on rat brown adipose tissue and skeletal muscle periph-eral clocks. Neurobiol. Sleep
Circadian Rhythms 2018, 4, 24–33. [CrossRef] [PubMed]
Biology 2021, 10, 207 15 of 15

85. Bur, I.M.; Zouaoui, S.; Fontanaud, P.; Coutry, N.; Molino, F.; Martin, A.O.; Mollard, P.; Bonnefont, X. The comparison between
circadian oscillators in mouse liver and pituitary gland reveals different integration of feeding and light schedules. PLoS ONE
2010, 5, e15316. [CrossRef] [PubMed]
86. Woller, A.; Gonze, D. Modeling clock-related metabolic syndrome due to conflicting light and food cues. Sci. Rep. 2018, 8, 13641.
[CrossRef] [PubMed]
87. Nakao, R.; Yamamoto, S.; Horikawa, K.; Yasumoto, Y.; Nikawa, T.; Mukai, C.; Oishi, K. Atypical expression of cir-cadian clock
genes in denervated mouse skeletal muscle. Chronobiol. Int. 2015, 32, 486–496. [CrossRef] [PubMed]
88. Heyde, I.; Oster, H. Differentiating external zeitgeber impact on peripheral circadian clock resetting. Sci. Rep. 2019, 9, 1–13. [CrossRef]
89. Yamazaki, S.; Numano, R.; Abe, M.; Hida, A.; Takahashi, R.; Ueda, M.; Block, G.D.; Sakaki, Y.; Menaker, M.; Tei, H. Re-setting
central and peripheral circadian oscillators in transgenic rats. Science 2000, 288, 682–685. [CrossRef]
90. Davidson, A.J.; Castanon-Cervantes, O.; Leise, T.L.; Molyneux, P.C.; Harrington, M.E. Visualizing jet lag in the mouse suprachias-
matic nucleus and peripheral circadian timing system. Eur. J. Neurosci. 2009, 29, 171–180. [CrossRef]
91. Reddy, A.B.; Field, M.D.; Maywood, E.S.; Hastings, M.H. Differential resynchronisation of circadian clock gene expression within
the suprachiasmatic nuclei of mice subjected to experimental jet lag. J. Neurosci. 2002, 22, 7326–7330. [CrossRef] [PubMed]
92. Kiessling, S.; Eichele, G.; Oster, H. Adrenal glucocorticoids have a key role in circadian resynchronization in a mouse model of jet
lag. J. Clin. Investig. 2010, 120, 2600–2609. [CrossRef]
93. Oster, H.; Damerow, S.; Kiessling, S.; Jakubcakova, V.; Abraham, D.; Tian, J.; Hoffmann, M.W.; Eichele, G. The circa-dian
rhythm of glucocorticoids is regulated by a gating mechanism residing in the adrenal cortical clock. Cell Metab. 2006, 4, 163–173.
[CrossRef] [PubMed]
94. Ono, D.; Honma, S.; Nakajima, Y.; Kuroda, S.; Enoki, R.; Honma, K.-I. Dissociation of Per1 and Bmal1 circadian rhythms
in the suprachiasmatic nucleus in parallel with behavioral outputs. Proc. Natl. Acad. Sci. USA 2017, 114, E3699–E3708.
[CrossRef] [PubMed]
95. Schmal, C.; Ono, D.; Myung, J.; Pett, J.P.; Honma, S.; Honma, K.I.; Herzel, H.; Tokuda, I.T. Weak coupling between in-tracellular
feedback loops explains dissociation of clock gene dynamics. PLoS Comput. Biol. 2019, 15, e1007330. [CrossRef] [PubMed]
96. Forger, D.; Gonze, D.; Virshup, D.M.; Welsh, D.K. Beyond intuitive modeling: Combining biophysical models with innovative
experiments to move the circadian clock field forward. J. Biol. Rhythm. 2007, 22, 200–210. [CrossRef]
97. Almeida, S.; Chaves, M.; Delaunay, F. Transcription-based circadian mechanism controls the duration of mo-lecular clock states in
response to signaling inputs. J. Theor. Biol. 2020, 484, 110015. [CrossRef] [PubMed]
98. Woller, A.; Duez, H.; Staels, B.; Lefranc, M. A mathematical model of the liver circadian clock linking feeding and fasting cycles to
clock function. Cell Rep. 2016, 17, 1087–1097. [CrossRef]
99. Bae, S.-A.; Androulakis, I.P. The synergistic role of light-feeding phase relations on entraining robust circadian rhythms in the
periphery. Gene Regul. Syst. Biol. 2017, 11. [CrossRef]
100. Bae, S.-A.; Androulakis, I.P. Mathematical analysis of circadian disruption and metabolic re-entrainment of hepatic gluconeogene-
sis: The intertwining entraining roles of light and feeding. Am. J. Physiol. Metab. 2018, 314, E531–E542. [CrossRef] [PubMed]