The human body has an internal biological clock that is constantly running.
Our well-being is dependent on the function of that clock.
New research from the University of Minnesota Medical School found a little stress can make the circadian clock run better and faster.
Research in the past several decades has found that our body has evolved a set of machinery, called the circadian clock, that internally drives rhythms in almost every cell.
The activities of the circadian clock are influenced by various signals in the cells.
In a recent study published in Neuron, Ruifeng Cao, MD, Ph.D., Assistant Professor in the Department of Biomedical Sciences at the Medical School, and a team of seven laboratories in the U.S. and Canada focused on the crosstalk between cellular stress signals and the circadian clock.
Cells respond to various stress signals by activating a signal transduction cascade that is centered on the protein eIF2α, which is a pivotal factor that orchestrates protein synthesis in cells.
Cao and his team found that in one’s central brain clock, stress leads to rhythmic phosphorylation of eIF2α, which promotes production of the ATF4 protein.
The ATF4 protein activates the Per2 gene, which ultimately makes the clock tick faster.
They concluded that this mechanism is necessary to maintain a robust clock, and therefore, that stress signals influence the speed and robustness of the circadian clock.
It has been known that the circadian clock gets broken in many diseases, but the reason for it has been unclear.
Cao’s finding may provide insight into this unanswered question, as it is the first connection between two fundamental processes in cells: stress response and circadian timekeeping.
One explanation could be that stress responses frequently go awry in diseased conditions, which may, in turn, mess up the clock.
“The next step is a more thorough and larger scale study on the crosstalk between the cellular stress network and the circadian clock,” said Cao.
“Hopefully our work can lead to discovering medicine that can manage the stress level and regulate the clock function in disease to keep people healthier.”
The ability to anticipate daily changes in the environment conveys an evolutionary advantage to most species on earth.
Therefore, organisms ranging from plants to higher mammals have developed endogenous circadian clocks that allow them to estimate the time of day.
In the absence of external time cues these clocks free-run with a period close to 24 h. In order to compensate discrepancies between this intrinsic period and the environmental cycle, circadian clocks entrain to external Zeitgebers (from German time giver), with light being the most potent one.
Unlike most biochemical systems, the period of the circadian clock is temperature compensated, a feature that is especially important for poikilothermic species (Buhr and Takahashi, 2013).
In mammals, the circadian clock is based on a molecular oscillator present in virtually every cell of the body. It is built from transcriptional-translational feedback loops (TTLs) generating self-sustained oscillations even on the cellular level.
The clock’s core TTL is composed of the genes brain and muscle arnt-like 1 (Bmal1), circadian locomotor output cycles kaput (Clock), cryptochrome (Cry) 1/2 and period (Per) 1–3. BMAL1 and CLOCK proteins form the positive limb of this core TTL.
They belong to the family of basic helix-loop-helix transcription factors and act as heterodimers binding to E-box regulatory elements within the promoters of Cry and Per genes (Fustin et al., 2009, Gekakis et al., 1998, Hogenesch et al., 1998, Yoo et al., 2004), activating transcription of Per and Cry. PER and CRY proteins constitute the negative feedback limb of the circadian core TTL.
Over the course of the day they accumulate in the cytoplasm and form complexes that translocate back into the nucleus where they inhibit BMAL1/CLOCK-mediated transcription (Kume et al., 1999, Zheng et al., 2001). Before the next cycle can start, the BMAL1/CLOCK heterodimer has to be reactivated. This is achieved by proteasomal degradation of the PER and CRY repressor complex. PER1 and PER2 are subject to phosphorylation mediated by casein kinases 1 delta and epsilon.
This phosphorylation mark leads to their ubiquitination and subsequent degradation by the ubiquitin proteasome system (Camacho et al., 2001, Eide et al., 2005). Similarly, adenosine monophosphate-activated protein kinase (AMPK) and glycogen synthase kinase 3 beta (GSK3β) phosphorylate CRY1 and CRY2, respectively (Harada et al., 2005, Lamia et al., 2009), so that they are ubiquitinated and degraded. Decreasing levels of CRY and PER terminate the repression of BMAL1/CLOCK-mediated transcriptional activation so that the clock can move to the next cycle.
Besides this core loop, there are accessory feedback loops and additional levels of regulation to stabilize the molecular oscillations and mediate additional fine-tuning. The most prominent accessory TTL consists of reverse erythroblastoma (Rev-Erbα/β) and retinoic acid receptor-related orphan receptor (RORα-γ) that also contain E-boxes within their promoter regions.
The BMAL1/CLOCK heterodimer binds to these E-boxes and activates transcription of Rev-Erbs and RORs (Buhr and Takahashi, 2013). In turn, REV-ERB proteins exert a negative feedback, inhibiting Bmal1 transcription (Liu et al., 2008, Triqueneaux et al., 2004). RORs, in contrast, are positive regulators of Bmal1 transcription and compete with REV-ERBs for retinoid orphan receptor response element (RORE) binding sites within the Bmal1 promoter (Akashi and Takumi, 2005). REV-ERBα and β are functionally redundant and are considered to be essential for Bmal1 oscillation. RORs seem to have a modulatory function, but they are dispensable for rhythmic transcription of Bmal1 per se (Liu et al., 2008).
This molecular clock machinery is present in all nucleus-containing cells of an organism. In order to synchronize single-cell oscillators with each other, the mammalian circadian system is organized in a hierarchical manner. A master clock resides in the suprachiasmatic nuclei (SCN) of the hypothalamus (Moore and Eichler, 1972, Ralph et al., 1990, Stephan and Zucker, 1972). The SCN receive light information from melanopsin-expressing cells in the retina via the retino-hypothalamic tract (Provencio et al., 2000). Time information is then passed on to subordinate peripheral tissues via humoral and neuronal signals (Buijs et al., 2003, Liu et al., 2007, Welsh et al., 2004, Yoo et al., 2004). In this way all peripheral and non-SCN tissue clocks are coordinated by the master clock.
Role of glucocorticoids in clock regulation
Among all peripheral oscillators, the adrenal gland holds a special role since the adrenal circadian clock can influence rhythms in other peripheral tissues via rhythmic release of hormones with clock-modulating properties. The adrenal gland is composed of an outer cortex and an inner medulla.
The medulla releases catecholamines (epinephrine and norepinephrine), whereas the cortex secretes mineralocorticoids from the outer zona glomerulosa, glucocorticoids (GCs) from the intermediate zona fasciculata, and sex steroids from the inner zona reticularis. Cortisol and corticosterone, the main GCs in humans and rodents, respectively, display a very robust circadian oscillation with blood levels peaking shortly before the onset of the active phase (i.e. the early morning in humans and the early evening in nocturnal rodents). The circadian GC rhythm is overlaid by strong ultradian pulsatility with a period of around one hour and an amplitude that varies considerably during the day (Windle et al., 1998). GCs are secreted upon adrenocorticotropic hormone (ACTH) binding to melanocortin-2 receptors (MC2R) in the adrenal gland. ACTH itself is secreted from the anterior pituitary upon corticotrophin releasing hormone (CRH) signaling, which stems from the paraventricular nucleus (PVN) of the hypothalamus.
Together, these tissues and factors constitute the hypothalamus-pituitary-adrenal (HPA) axis. Circadian oscillations are detectable for all components (CRH, ACTH, and GCs) (Chrousos, 1998, Girotti et al., 2009). However, it is not clear if rhythmic HPA axis activity is necessary for the circadian rhythm of GC secretion.
On one hand, adrenal rhythms persist after hypophysectomy, when no ACTH is present (Fahrenkrug et al., 2008). On the other hand, ACTH is capable of phase-dependently resetting GC rhythms (Yoder et al., 2014). In addition, the HPA axis gets direct input from the SCN via the paraventricular nuclei of the hypothalamus (Dickmeis et al., 2013, Vrang et al., 1995) and the SCN controls GC rhythms as was shown in SCN-lesioned animals (Moore and Eichler, 1972). The circadian pattern of GC secretion can be abolished by specifically disrupting the circadian clock in the adrenal gland (Oster et al., 2006a, Son et al., 2008), indicating that this peripheral tissue clock finally governs GC secretory patterns.
Glucocorticoids act via mineralocorticoid (MR) and glucocorticoid receptors (GR), type-1 nuclear receptors with broad expression patterns throughout the body except for the SCN. GR signaling can mediate phase resetting of peripheral clocks, pointing at a special role of GC rhythms in the coordination of the organism’s circadian network (Balsalobre et al., 2000).
For instance, microarray analysis of murine liver revealed 100 rhythmic genes whose oscillation was directly dependent on adrenal signals, because rhythmicity of these genes is lost in adrenalectomized animals (Oishi et al., 2005). Finally, in a mouse model of jet lag, which is caused by an abrupt phase shift of light conditions, GC rhythms can modify the kinetics of entrainment to the new time zone (Kiessling et al., 2010).
On top of their phase shifting ability, GCs can stabilize peripheral rhythms against external perturbation. Timed food restriction can induce phase shifts in peripheral tissues so that peripheral clocks become uncoupled from the master clock in the SCN that stays tied to the light-dark cycle. The circadian system is more robust against such perturbations when GCs are high (Le Minh et al., 2001). In summary, rhythmic GC secretion is an important timing signal for the coordination of peripheral clocks.
Neurobiology of stress (HPA axis, glucocorticoids, catecholamines)
Besides their role in circadian coordination GCs are important vectors of the stress system. Stress refers to an external or internal challenge that requires an adequate reaction of the organism in order to survive or, in less drastic cases, to avoid pain or discomfort. An elaborate response system has evolved that becomes activated when the organism is exposed to stress. It involves an immediate response via activation of the autonomous nervous system (ANS) and a delayed response via HPA axis-mediated release of GCs (Ulrich-Lai and Herman, 2009).
All sensory systems can collect information about stressful events (e.g. a decrease in blood volume, changes in blood composition, or the encounter of a predator) and forward this information to the brainstem (Ulrich-Lai and Herman, 2009). From here, subsequent activation of the ANS and the HPA axis is regulated. In case of the HPA axis, stress-mediated activation triggers the production and release of GCs in the adrenal gland. Stress signals from the hippocampus, prefrontal cortex or amygdala are transferred to the paraventricular nucleus (PVN) of the hypothalamus to stimulate the secretion of CRH to initiate HPA axis activation.
The fact that GCs need to be newly synthesized after each trigger leads to a certain delay in the final effector response. Therefore, the dynamics are slower (in the range of minutes) compared to ANS activation, which occurs within seconds after stress initiation.
Exposure to a stressor results in an immediate increase of catecholamines via activation of sympathetic preganglionic neurons in the spinal cord (Nygren and Olson, 1977, Westlund et al., 1983). From here, the signal is either transferred to postganglionic neurons projecting to peripheral effector organs where they are translated into the classical fight-or-flight response (e.g. increase of heart rate and blood pressure, vasoconstriction, stimulation of sweat glands, energy mobilization etc.) or preganglionic nerves continue as splanchnic nerves to peripheral effector organs. As such, splanchnic nerves constitute a short-cut to the adrenal medulla where the immediate release of catecholamines is initiated (Holgert et al., 1998). The acute stress response is terminated by reflex parasympathetic activation and negative feedback inhibition from GCs that stop the release of CRH and ACTH from hypothalamic and pituitary cells, respectively (Nader et al., 2010).
Even though the ANS and the HPA axis are two different branches within the general stress system and their dynamics are quite different, both act together to coordinate an appropriate response to stress. For the response to certain stressors, for example, HPA axis activation is supported by noradrenergic and adrenergic projections from the hindbrain to the PVN. As such, lesion experiments have shown that stress-induced GC release can be impaired when these projections are not functional (Ritter et al., 2003). Conversely, cells in the inter-mediolateral column receive input from the PVN, suggesting that signals from the hypothalamus can modify the autonomic stress response (Tucker and Saper, 1985).
Both circadian and stress-mediated aspects play an important role in regulating HPA axis activity and GC levels. In case of catecholamines, it is technically difficult to assess to which extent the circadian clock influences catecholamine levels. However, a circadian oscillation of serum levels of epinephrine and norepinephrine has been described (Dimitrov et al., 2009). At the same time, GC secretion can be influenced by SCN-sympatho-adrenal innervations (Ishida et al., 2005, Ulrich-Lai et al., 2006) and ACTH from the HPA axis can stimulate (nor)epinephrine secretion (Valenta et al., 1986) so that ANS and HPA axis are interconnected also for circadian aspects.
In conclusion, both HPA axis and ANS show aspects of circadian and stress-mediated regulation and they interact on several levels (Fig. 1). In the following, we will highlight how the circadian system and stress response influence each other in rodents. Finally, we propose how this knowledge might be used for translational approaches.
Circadian gating of stress responses
Stress impairs the body via a complex network of interacting signaling cascades regulating the vulnerability to and severity of stress. In principle, one has to distinguish between the acute stress response preparing the body for rapid action and repeated stress inducing broader alterations and adaptations, which are associated with changes in energy metabolism and an elevated risk for psychiatric disorders.
While it is known since the 1970s that the responsiveness of HPA axis is modulated by the time of day (Dunn et al., 1972b, Gallant, 1979, Gibbs, 1970, Kant et al., 1986, Torrellas et al., 1981, Zimmermann and Critchlow, 1967), we are far away from understanding the underlying mechanisms. Disruptions of the circadian clock are associated with altered HPA axis activity and GC concentrations as well as with metabolic impairments and major depression (Albrecht, 2010, Barclay et al., 2012, Leliavski et al., 2014, Mukherjee et al., 2010, Turek et al., 2005). However, only little is known about stressor exposure in clock gene mutant mice. So far, we know that the impact of a clock gene deletion on circulating GCs is depending on which aspect of the clock feedback loop is affected. Mice with a clock gene mutation in the positive limb of the oscillator, Bmal1 or Clock, consistently suffer from hypocortisolism (Leliavski et al., 2014, Turek et al., 2005) while Cry mutations (affecting the negative limb) lead to hypercortisolism (Barclay et al., 2013, Lamia et al., 2011) associated with a reduced genotoxic stress response. However, the deletion of Per2, which also affects the negative limb, results in hypocortisolism (Yang et al., 2009). Further, the deletion of Per leads to increased immobilization, stress-induced grooming, and nociceptive behaviors associated with increased CRH expression in the PVN of those mice (Zhang et al., 2011). Immobilization stress and genotoxic stress, however, affect the body in mechanistically different ways. Responses to swim stress, which is mechanistically comparable with immobilization stress, is reduced in Bmal1 knockout mice leading to the assumption that the manipulation of the positive and negative limb of the core clock feedback loop has opposing effects on stress regulation. However, a general statement about stress responsiveness in clock gene mutants is, due to the small number of studies so far, not possible and more comparable experiments are needed to close this gap. Further, the circadian machinery is not completely disrupted by the deletion of only one of the Cry or Per gene, increasing the difficulty to draw a conclusion on clock impact from those studies.
Interestingly, BMAL1 and CRY do not only have opposing functions in the TTL, both interact with the HPA axis at different levels and affect different aspects of the stress response (Lamia et al., 2011, Leliavski et al., 2014), fine-tuning the responsiveness of the system to stress-related stimuli along the course of the day (Fig. 2). In male mice as well as in female rats, adrenal ACTH sensitivity is elevated during the active phase (Bartlang et al., 2012, Engeland et al., 1977, Leliavski et al., 2014, Oster et al., 2006b). A deletion of the core clock gene Bmal1 leads to time independent and low ACTH sensitivity, which is comparable with that of wildtype animals during the day. This results in a lower depression-like behavior in the forced swim test paradigm (Leliavski et al., 2014). GR sensitivity, which is essential for feedback inhibition of the HPA axis, is also regulated by the circadian system (Lamia et al., 2011). CRY1 and 2, expression of which peaks in the early night, interact with the C-terminus of GR, which is also required for ligand binding. In this way, CRY1 and 2 oppose GR activation. Genetic deletion of both Crys leads to non-oscillating and elevated GC levels due to impaired feedback inhibition (Lamia et al., 2011). In summary, clock mediated increased HPA axis sensitivity during the active phase together with elevated ACTH sensitivity (Leliavski et al., 2014, Oster et al., 2006b) and simultaneous repression of GC-mediated feedback inhibition (Lamia et al., 2011) drive time-of-day dependent responsiveness to stressor exposure.
At the time when the HPA axis is most sensitive to stimulation, physical stressor exposure like hemorrhage (Lilly et al., 2000), hypoglycemia (Kalsbeek et al., 2003), or oxidative stress (Antoch et al., 2005; Fanjul-Moles and Lopez-Riquelme, 2016) yield a greater increase in circulating GCs than at other periods of the day. However, during the inactive phase, when the HPA axis should be less responsive, restraint/immobilization, foot shock, or shaking stress result in a stronger increase in GC and ACTH release, and blood pressure (Bernatova et al., 2002, Bradbury et al., 1991, Gattermann and Weinandy, 1996, Gutierrez-Mariscal et al., 2012, Mathias et al., 2000, Retana-Marquez et al., 2003, Torrellas et al., 1981). More experiments are needed to clarify, how time-of-day affects the responsiveness to stress. Differences in experimental setups (e.g. different water temperatures in the forced swim test (Bachli et al., 2008)) may explain some of the discrepancies between studies (Retana-Marquez et al., 2003). In addition, species (e.g. rats vs. mice) or gender differences (see below) may further play a role.
More information: Salil Saurav Pathak et al, The eIF2α Kinase GCN2 Modulates Period and Rhythmicity of the Circadian Clock by Translational Control of Atf4, Neuron (2019). DOI: 10.1016/j.neuron.2019.08.007
Journal information: Neuron
Provided by University of Minnesota