Circadian Rhythms and Mood Disorders: Time to See the Light

Summary:

The importance of time is ever prevalent in our world and disruptions to the normal light/dark and sleep/wake cycle have now become the norm rather than the exception for a large part of it. All mood disorders, including seasonal affective disorder (SAD), major depressive disorder (MDD), and bipolar disorder (BD), are strongly associated with abnormal sleep and circadian rhythms in a variety of physiological processes. Environmental disruptions to normal sleep/wake patterns, light/dark changes and seasonal changes can precipitate episodes. Moreover, treatments that target the circadian system have proven to be therapeutic in certain cases. This review will summarize much of our current knowledge of how these disorders associate with specific circadian phenotypes, as well as the neuronal mechanisms that link the circadian clock with mood regulation. We also discuss what has been learned from therapies that target circadian rhythms and how we may use current knowledge to develop more individually designed treatments.

In Brief

In Dollish et al., they review the clinical literature and provide mechanistic insights into how circadian rhythms contribute to mood disorders. They also describe chronotherapeutic treatments and provide hypotheses for testing in future studies of both mechanism and novel treatments.

Introduction:

Mood disorders including SAD, MDD, BD, all have abnormalities in sleep/wake patterns as a core symptom, and in fact this is one of the primary diagnostic criteria used clinically. Moreover, acute changes to the sleep/wake cycle can precipitate mood episodes.¹ There are now hundreds of studies which find an association between having a late chronotype (i.e. preference for the evening over morning activities) and developing a mood disorder.^2,3 For example, one study in the Netherlands of nearly 2000 people found a significant association between late chronotype and depression/anxiety (P = .004) even when adjusting for sociodemographic, somatic health, and sleep-related factors which was largely driven by those with depression.⁴ Furthermore, a very large study in the United Kingdom (433,268 adults) comparing definite evening type to definite morning types in a wide assessment of health related issues found that the associations were strongest for psychological disorders and late chronotype (OR 1.94, 95% CI 1.86–2.02, p = < 0.001) over other health issues, though interestingly having a late chronotype was associated overall with poorer health outcomes and even an increased risk for all causes of mortality.⁵

We also know that chronobiologic treatments aimed at changing circadian rhythms in particular ways can be highly beneficial in the treatment of these disorders.^6,7 Due to the complex etiology of these diseases, however, it is difficult to say whether mood disorders are directly caused by circadian rhythm disruptions, or how much of the association between chronotype, circadian rhythms, sleep and mood disorders is due to desynchronization from the environment versus genetic or other abnormalities in the core molecular pacemaker.

Circadian clocks evolved to calculate and keep track of the amount of time that has passed for the daily rotation of the earth. Without this development, organisms would not be able to anticipate the daily sunrise, develop reliable wake and rest periods, or control metabolic demands in an efficient way.^8,9 Humans share evolutionary conservation of this ancient time keeping mechanism¹⁰ and every organ, and in fact nearly every cell in the body, is tuned to this daily cycle. This allows the body to maximize energy production during active periods and conserve energy, as well as restore and repair cellular functions, during rest periods. The clock is also crucial in timing immune signaling to protect organisms from harm without creating internal immune problems.¹¹ Thus, when this internal mechanism of time keeping is thrown off (which occurs frequently in our modern world), it can contribute to or even cause a plethora of physiological consequences, including diseases like sleep disorders, heart disease, metabolic disorders, cancer, neurodegenerative disease and psychiatric disorders.^9,10,12,13

Circadian rhythms are centrally controlled in the brain by the suprachiasmatic nucleus (SCN) located at the base of the hypothalamus. The SCN core receives direct projections from the retina through cells which contain the photoreceptor, melanopsin, which allows the SCN to respond directly to light, the primary entrainer of circadian rhythms (Figure 1).¹⁴ An important aspect of circadian rhythms is that even in the absence of light they will continue to run with a period very close to 24 hrs and this is controlled largely by the shell of the SCN which receives and gives feedback to the core to regulate rhythms over the light/dark cycle.¹⁵ Without light or other entraining stimuli, rhythms will “free run” such that they drift ever so slightly every day leading in humans typically to later and later wake and sleep timing.¹⁶ Entrainment by light and other factors is critical for the proper functioning of all biological processes as it ensures rhythms are timed to the proper phase of the solar light/dark cycle as well as with each other. For example, when one travels to a new time zone, they typically experience some form of “jet lag” which is a condition in which the internal pacemaker in the SCN becomes out of synch with the environment and needs time to entrain to the new light/dark cycle. Jet lag is characterized by daytime fatigue, mood changes, headaches, inability to concentrate and digestive problems.¹⁷ Eastward travel produces a phase advance in rhythms while westward travel produces a phase delay, changing the timing of rhythmic peaks and troughs.¹⁸ It typically takes one day for every hour traveled to entrain to the new light/dark cycle, and the response to the phase advance of eastward travel is typically harder to adjust to compared to westward travel.¹⁷ In addition to entrained or non-entrained (e.g. free running) rhythms, a person’s chronotype or lifestyle choices can lead to either highly irregular sleep/wake patterns, or a phase advance or delay compared to normally entrained rhythms such that they wake up earlier or later than what is typical (Figure 2). In extreme cases, persistent advanced or delayed rhythms are classified as advanced or delayed sleep phase disorder in which people have a difficult time with school, work or social activities due to the rigid timing of their endogenous rhythms.¹⁹

Figure 1. The Retinohypothalamic circuit responds to light from the environment and uses it to entrain internal rhythms.

Light information is directly projected to the SCN through the retinal hypothalamic tract. Starting in the retina, it travels to ventral core of the SCN in the hypothalamus which communicates with the dorsal shell that houses the endogenous pacemaker. The bidirectional communication between the shell and core together entrain and synchronize peripheral clocks in the brain and body.

Figure 2. Circadian rhythm perturbations as depicted through activity patterns.

Daylight hours (yellow), darkness (blue), activity periods (black bars (1 line = 1 day) 1) Entrained rhythms align with the light/dark cycle, 2) delays are shifted with onset occurring later than observed entrained rhythms, 3) advances are shifted to occur earlier than usual, 4) fragmentation is a breakdown and disorganization of entrained rhythms, and 5)free running conditions are the absence of cues or the inability of rhythms to entrain to external cues.

While the SCN is the master pacemaker that conducts the orchestra of rhythms across the brain and body, a core molecular clock resides in nearly every cell. This core mechanism in mammals lies in a transcription-translation feedback loop between circadian genes and their products (Figure 3). In humans, the Circadian Locomotor Output Cycles Kaput (CLOCK) and Aryl hydrocarbon receptor nuclear translocator like protein (ARNTL also known as BMAL1) genes encode basic helix-loop-helix transcription factors which heterodimerize and bind to DNA E-boxes in a variety of genes.^20,21 Among the many genes controlled by CLOCK:ARNTL, are the Period (PER) and Cryptochrome (CRY) genes. Once translated, PER and CRY proteins form a stable dimer which then can reenter the nucleus to inhibit their own transcription.²⁰ In addition to this primary loop, there is a secondary stabilizing feedback loop involving the REV-ERBɑ and Retinoic acid-related orphan receptor alpha (RORɑ) proteins which regulate ARNTL expression to affect the timing of the primary loop. In addition, several other proteins contribute to the timing and speed of these molecular rhythms which lead to diurnal rhythms in gene expression of thousands of other clock-controlled genes whose identity can be vastly different depending on the cell type.²²

Figure 3. Molecular machinery of the circadian clock.

Circadian locomotor output cycles kaput (CLOCK) and brain and muscle Arnt-like protein 1 (BMAL1; a.k.a. ARNTL) heterodimerize and bind to Enhancer Box (E-box) elements to activate clock-controlled genes (CCGs) transcription, including the Period (Per) and Cryptochrome (Cry) genes. PER and CRY proteins dimerize and translocate back to the nucleus to inhibit CLOCK and BMAL1 activity, forming a negative loop, which cycles every 24 hours. In a secondary loop, CLOCK and BMAL1 proteins regulate the nuclear hormone receptors expression, Rev-erb and Ror, which inhibit or activate Bmal1 transcription, respectively.

This review will summarize current knowledge of the interwoven nature of circadian rhythms and mood disorders. In addition, we will explore possible circadian mechanisms and pathology of SAD, MDD, and BD, as well as current and experimental treatment options that add to the growing literature budding from interest in this field.

Seasonal Affective Disorder (SAD)

Although canonically called seasonal affective disorder (SAD) the current DSM-5 diagnosis is Major Depressive Disorder with a Seasonal Pattern and this is a common mood disorder. According to the American Psychiatric Association about 5% of adults in the United States experience SAD. Seasonal patterns of mood episodes are also experienced by approximately 25% of individuals with BD²³. The diagnostic criteria for SAD can be similar to MDD^24,25 and more closely overlaps with symptoms seen in atypical depression such as hyperphagia, hypersomnia, and fatigue/lethargy as opposed to symptoms of melancholic or typical depression.²⁶ Though SAD is laregely recognized by the greater psychiatric community, one study by Traffanstedt et al. (2016)²⁷ failed to find differences in the Patient Health Questionnaire-8 Depression Scale in a cross sectional survey of U.S. adults that varied with latitude or season, suggesting possibly that seaonal depression as a disorder may not be so clear cut, however this study is limited in that it is cross sectional self report, it only examines one point in time for each individual’s current symptoms, and combines a variety of depression-related symptoms together, some of which are not consistent with the greater litterature on seasonal depression.²⁸ The feature that most distinguishes SAD from MDD is that depressive symptoms last roughly 40% of the year and symptom on and off set is timed to correspond with longer or shorter daylight hours indicating a seasonal change.^24,25 Atypical depression also has diurnal features with symptoms sometimes worsening in the evening; however those with SAD often see a worsening of symptoms in the morning suggesting, along with other differences such as the response to timed light therapy, that although these diseases are similar they are indeed separate disorders.^26,29,30 Typically, SAD occurs during the transition between fall to winter months where the rate of daylight change from long photoperiod to short photoperiods is the highest. However, SAD is not limited to the winter months, and also occurs during the summer months in about 10% of people with SAD and the symptoms of summer SAD often include insomnia and less appetite.³¹ The length and severity of symptom onset is individual and can be measured using the Seasonal Pattern Assessment Questionnaire (SPAQ) by looking at the Global Seasonality Score (GSS) which indicates the presense of SAD and its severity with anything over an 11 indicating moderate to severe SAD.²⁵ Chronotype, gender, and age also play a role in developing or showing symptoms of SAD with a strong sex bias towards women who are 4x more likely to develop the disease. In a study by Höller et al they found that people under the age of 60 and those with evening chronotypes had a higher seasonality score on the SPAQ.³² Interestingly latitude (living in the north) weakly associates with SAD, however the rate of incidence of SAD generally increases as one moves further from the equator.^33,34 A manuscript comparing two studies of environmental factors that could contribute to SAD by Young et. al. (1997)³⁵ found in the first study which examined data from 5 different locations, that latitude and time of year contributed to SAD, but the second study examining data collected over 7 years from 1 location failed to find evidence of daily hours of sunshine or temperature that relate to SAD suggesting that photoperiod is related to the onset of SAD as opposed to total hours of sunlight. However, a study by Murray and Hay (1997)³⁶ failed to find photoperiod specificity to SAD symptomology. However this study, although robust with 526 female participants from the Australian Twin Registry, only looked at a single location (Australia) and not locations across a broad latitude which is important in making this assessment.

Genetics seem to play a role in both developing SAD as well as seasonal symptom onset. Single nucleotide polymorphisms (SNPs) in the clock genes ARNTL, CLOCK, and PER3 have been shown to contribute to developing depression, anxiety and are particularly associated with seasonal depression.³⁷ SNPs in CRY1 and CRY2, have also been linked to seasonal depression in a longitudinal study.³⁸ In addition, Ho et. al. (2018) conducted a genome-wide association study of SAD and found that intronic variant rs139459337 in the zinc finger protein, ZBTB20, was the SNP most strongly associated with SAD, and individuals with this SNP had reduced RNA expression of ZBTB20 in the temporal cortex.³⁹ ZBTB20 is important in the regulation of circadian activity output and of the 330 known human targets of this gene, a significant enrichment of these targets are found in SAD genetic association signals, further suggesting that ZBTB20 may be centrally important in seasonal mood regulation.^39,40 However, it is important to note that most of these studies have been rather small candidate gene studies with the largest, aforementioned GWAS study consisting of only 1380 cases, so larger studies will need to be done to assess potential genetic associations.

Genetics also determine a person’s chronotype⁴¹ which can increase their susceptibility to developing SAD. A study looking at 1539 adolescents found that when evaluated with the Morning-Eveningness Questionnaire and the SPAQ, those with evening preference had significantly higher scores on the SPAQ than those with intermediate and morning chronotypes.^42,43 Furthermore, a 3-year study evaluating 244 individuals found a positive association between lower mood and eveningness.^44,45 This could be due to the fact that evening chronotypes often wake up later in the day compared to the rest of the population, which would reduce exposure to daytime light, especially in the winter months when individuals are more likely to develop SAD. However, chronotype does not entirely predict SAD and reduced exposure to light during the day does not track with all cases of SAD, since summer variations of SAD occur in a small subset of SAD suffers. Or another possibility is that evening types are already biologically out of phase from the environment, and have rhythms that are strongly set, making adaptation to seasonal changes challenging.

Potential mechanisms linking circadian rhythms and SAD

Although SAD is highly prevalent, its etiology is not fully understood. Animal studies point to an important role for hypothalamic and midbrain dopamine neuron populations in SAD. In a landmark study, adult rats exposed to a short-day photoperiod experienced what appears to be “neurotransmitter switching” in the hypothalamus in which somatostatin containing, GABAergic interneurons became dopamine neurons, and blockade or ablation of these new dopamine neurons increased anxiety and depression-like behavior.⁴⁶ Long-day photoperiods produced the opposite effect. These results are intriguing in that they suggest that monoaminergic systems in the brain can be fundamentally altered simply by changing the photoperiod, and this can result in changes in mood-related behavior. Interestingly, a recent study by Jameson et al. (2023) also found that photoperiod length can modulate dopamine signaling in the nucleus accumbens (NAc) core and this is sex specific. Both dopamine release and uptake in the NAc were increased in female mice raised in a long photoperiod versus a short photoperiod however, there were no differences in male mice.⁴⁷ This is particularly interesting given the higher prevalence of SAD in females compared to males.

Mice generally find light aversive as they are nocturnal, making some results such as those stated above difficult to compare to humans, so investigators have instead used the diurnal grass rat as model with greater face validity. Indeed, grass rats housed in 8:16 photoperiods mimicking winter develop an increase in depressive-like behavior.⁴⁸ These studies also found that the neuropeptide orexin helps mediate light induced changes in mood and anxiety in these animals. As opposed to mice, in grass rats the number of hypothalamic dopamine neurons were reduced in a short photoperiod, and numbers of dopamine neurons can be altered with treatment with an orexin 1 receptor antagonist, suggesting that orexin is involved in mediating these changes in dopamine neurons. Another study by Costello et. al. (2023) found that administering light to diurnal grass rats under a protocol similar to that used in bright light therapy (BLT) induced higher wakefulness, increased nighttime sleep quality and improved rhythm entrainment, similar to what is observed in humans.⁴⁹ Moreover, this treatment led to changes in prepro-orexin and orexin receptor expression which was brain region and sex specific. Taken together there is strong evidence that photoperiod changes directly impact dopamine signaling, potentially via orexin and that this contributes to seasonal mood changes. However, it is important to note that photoperiod induced changes in dopamine signaling are different in diurnal versus nocturnal species as well as brain region and sex specific.

Cortisol rhythms are widely known to be synchronized to the external light-dark cycle as seen by a nadir in the early evening and progressively increasing till it peaks just before waking along with a spike in serotonin before gradually tapering off again.⁵⁰ Individuals with SAD have a cortisol awakening response (CAR) in the summer that does not differ from healthy populations, but during the winter months, where SAD patients report increased feelings of stress, depression and anxiety, the CAR is attenuated compared to healthy individuals.^51,52 However, this finding does not necessarily indicate that SAD is a hypocortisolemic condition. In a systematic review of 13 papers, those that used heterogeneous methods and samples were unable to consistently replicate the attenuated CAR response in SAD patients.⁵³ Thus further studies are needed to understand if perhaps there is a subgroup of subjects in which the winter CAR response is attenuated.

In long photoperiods, cells in the caudal SCN are desynchronized with rostral cells having a bi-modal pattern of activity with one activity peak at dawn and the other at dusk.⁵⁴ Control of seasonal neural activity and behaviors also seems to be somewhat localized to the pars tuberalis (PT) a tubular sheath that wraps around the pituitary gland and may act in tandem with the SCN to produce seasonal timekeeping and entrainment signals and output in mammals.⁵⁵ It might be that similar to the pineal gland which houses the “night clock” via melatonin secretion, that the pars tuberalis is the home of the “seasonal clock” with its own unique seasonal timekeeping system and zeitgebers. SCN plasticity occurs in response to adjustments in daylength. Long day photoperiods indicative of summer lead to an increase in the duration of neuronal activity periods at the population level in hamster SCN slice compared to short days.⁵⁶ In addition to population activity, the phase distribution between specific types of SCN neurons is also important for seasonal entrainment.⁵⁷ Short days result in a relatively synchronized firing pattern between these SCN subpopulations while long days result in the activation of these subpopulations at different phases of the day/night cycle.⁵⁷ It is possible that individuals with SAD have an SCN that cannot properly adapt to these different seasonal light cues. It is also possible that people with SAD differ in the sensitivity to light. Light and photoperiod entrainment can be measured using a post-illumination pupil response (PIPR). In a study by Roecklein et. al. (2013) subjects with SAD or control subjects were exposed to either red or blue light in the fall/winter.⁵⁸ They found that the SAD group had a reduced PIPR and lower overall change relative to controls, specifically to blue light. They propose that this reduced PIPR response may be linked to a specific melanopsin gene OPN4 variant, suggesting that this circadian photoreceptor is less sensitive to light input.

Importantly, while there is evidence as described above that mood-related responses to light are regulated via the pathway from the eye to the SCN and then other brain networks, work from Samar Hattar’s group has found that there are direct projections from the eye to the perihabenula region, and that depression-related responses to changes in the light/dark cycle in mice occur via this pathway which is completely independent of the SCN.^59,60 Communication between the perihabenular region and the lateral habenula may modulate dopaminergic activity in response to photoperiod changes as described above, resulting in reduced dopaminergic transmission.⁶¹ Therefore, it is unclear how much of the mood related responses to light at night, or a shortened photoperiod are regulated by the SCN versus projections from the eye to the perihabenular region.

Melatonin in humans is released by the pineal gland in the evening to facilitate the onset of sleep. In multiple mammalian species however, melatonin also plays a much larger role in seasonal processes including breeding patterns, aggression and hibernation.^62–64 Symptoms of SAD include increased fatigue, excessive sleeping, carbohydrate craving, and exaggerated seasonal changes in melatonin brought on by the reduction in day length may contribute, cause or be the result of these perturbances (Figure 4). Indeed, an overproduction of melatonin during the winter is phenotypical in some SAD patients.^65,66 Individuals with SAD tend to also have strongly delayed timing of melatonin secretion, which is exacerbated in the winter, leading to more fatigue during the daytime.⁶⁷ Thus, melatonin is likely to play a key role in many of the symptoms of SAD. However, evidence is increasingly pointing towards SAD being a system wide misalignment issue involving multiple neurotransmitters and modulators. The inability to entrain rhythms can be rectified through better understanding of how the SCN and other networks integrate daily and seasonal light information and disseminate that information to entrain biological rhythms.

Figure 4. Melatonin secretion duration changes as a function of seasonal night length.

The summer melatonin secretion profile (Left) is shorter due to summers longer daylight hours suppressing melatonin secretion. In the winter (Right) melatonin secretion is longer because of the shorter daylight hours. The variation in melatonin secretion duration and amplitude changes are a factor of seasonal daylight presence allowing it to act as both a seasonal and night timekeeper.

Treatments for SAD

Bright light therapy (BLT), is the most widely used treatment for SAD. When exposed to bright broad-spectrum light (~10,000 lux) for 30 minutes, most patients with SAD report symptom amelioration or cessation with dedicated use and engagement with the therapy.⁶ Typically, BLT is prescribed in the early morning, as studies show that it is the most effective timing window as it can shift circadian rhythms forward to better align them with the environment and the majority of patients with SAD tend to be phase delayed.^68,69 However, there are a smaller subset of patients who may experience SAD due to a phase advance, perhaps due to the earlier winter dusk. Differences in BLT timing and chronotype have not been observed⁷⁰; however, differences in individual BLT timing exist and optimization of the prescribed time and “dose” of light is important.⁷¹ An important study from Lewy et al (2006) tested the phase shift hypothesis of SAD treatment through the administration of low-dose melatonin either in the morning or afternoon/evening to induce phase advances or delays.⁶⁶ Their administration protocol varied over the course of the 4 years this study took place over, with daily melatonin totaling 0.225 mg (year 2) or 0.3 mg (years 1, 3, and 4). They found that there is an optimal “therapeutic window” for phase aligning patients based on the timing between their dim light melatonin onset and midpoint of sleep which is optimally 6 hrs. Treating people with appropriately timed melatonin (which was typically at night) significantly correlated with symptom improvement while inappropriately timed melatonin did not.⁶⁶

Another way that BLT may be beneficial in SAD is that it decreases the amount of melatonin being produced. When exposed to light – specifically blue light – melatonin production is halted and is quickly eliminated from the body.⁷² For individuals that have either an increased production of melatonin during the day or a phase shift in melatonin release such that it is still present during the day, BLT will reduce melatonin levels. Selective serotonin reuptake inhibitors (SSRIs) have also been used in the treatment of SAD with promising results, however, it should be noted that further studies of the side-effect “cost-benefit” and replication of findings are still needed and noted in many of the studies.^73–75 Further research is needed to determine if other treatments such as agomelatine, a melatonin receptor agonist and 5-HT2c receptor antagonist, has any efficacy in the treatment of SAD.

Major Depressive Disorder (MDD)

Major depressive disorder is a common and debilitating disorder affecting more than 300 million people worldwide.⁷⁶ In the last several decades, prevalence of MDD has been steadily increasing.⁷⁷ MDD is characterized by depressed mood, loss of interest in pleasurable activities as well as physiological and cognitive symptoms. Insomnia, or oversleeping, fatigue and decreased energy, and changes in appetite are common among patients with MDD. Either insomnia or oversleeping is one of the main diagnostic criteria for MDD, and more than 80% of patients with MDD report sleep disturbances.⁷⁸ Mood fluctuates over the course of 24 hours in healthy individuals; however, diurnal variations in mood are altered in patients with MDD. Compared to healthy controls, depressed patients exhibit pronounced diurnal disturbances in positive and negative affect.⁷⁹ Patients often report the worst symptoms early in the morning, with a subset of patients (about 20%) also experiencing an afternoon slump.⁸⁰ Chronotype might also influence diurnal variations in mood, and MDD patients with an evening chronotype are more likely to experience worsening of their symptoms in the morning and overall increased rumination.⁸¹ Moreover, an evening chronotype is a strong risk factor for MDD and is associated with more frequent and severe depressive episodes³, greater risk for suicidality⁸², and reduced SSRI treatment efficacy.⁸³ Increasing rates of depression worldwide can be linked to modernization of society, with its increased exposure to artificial light, shift work and air travel across multiple time zones.^77,84 People living in areas with the highest levels of outdoor LAN show increased odds of depressive symptoms and suicidal behaviors.⁸⁵

Shift work causes disturbances to circadian rhythms leading to a number of negative health consequences, including sleep disturbances, fatigue, cognitive impairments and increased risk of cardiovascular and metabolic disorders and certain types of cancers.^86–88 A meta-analysis by Lee et al suggests that night shift work is also associated with the increased risk of developing depression.⁸⁹ Moreover, compared to the general population, flight attendants have a higher prevalence of sleep disorders, fatigue, depression and anxiety.⁹⁰ In rodents, exposure to jet lag paradigms worsens depressive-like behaviors, as evidenced in reduced sucrose intake, increased immobility time in the forced swim test, and reduced number of entries to the center zone in the open field test.⁹¹ There is also evidence that long-distance flights can trigger depressive episodes in people with a history of mental illness.^92,93

A study of circadian gene expression in human postmortem brain found that when looking across a population of subjects who died at different times of day, rhythmic patterns of gene expression were much weaker (i.e. lower amplitude) in subjects with MDD and this was particularly true in cortico-limbic regions of the brain.⁹⁴ This was due in part to shifts in timing and disrupted phase relationships between individual genes. Animal studies also indicate that circadian rhythms play a significant role in MDD. Knocking-down Bmal1 specifically in the SCN results in increased depression-like behavior in mice as evident by increased immobility times in the tail suspension test and increased escape latency times in the learned helplessness test.⁹⁵ Furthermore, the severity of chronic stress-induced depression and anxiety-like phenotypes in mice correlates directly with a loss of amplitude of molecular rhythms in the SCN.^95,96 Mice that are susceptible to developing a depression-like syndrome following chronic social defeat stress also have a loss of circadian rhythmicity in core body temperature while mice that are resilient to developing a depression-like phenotype following stress do not.⁹⁷ Recently, a Bmal1 knock-out cynomolgus monkey was developed using CRISPR/Cas9 editing of monkey embryos.⁹⁸ These monkeys had higher nocturnal locomotion and reduced sleep, which was exacerbated when they were housed in constant light. Moreover, these monkeys showed phenotypes consistent with anxiety and depression, including elevated blood cortisol. Transcriptome analysis in blood showed multiple changes in genes associated with inflammatory and stress responses including for example Toll-like receptor 4 (TLR4), which was also found in blood transcriptomes of subjects with MDD and in subjects with sleep deprivation.

Potential mechanisms linking circadian rhythms and MDD

Dysregulation of the hypothalamic-pituitary axis (HPA) is a common finding in MDD patients. HPA axis controls production and secretion of glucocorticoid hormone, a potent regulator of all major physiological systems. Glucocorticoids have a strong circadian rhythm in their expression. In healthy subjects, maximal cortisol secretion happens in the morning hours before awakening, and then gradually decreases during the day.⁹⁹ In depressed individuals, however, lower levels of cortisol in the morning and higher evening levels of cortisol have been reported across studies.¹⁰⁰ In addition, a phase advance of the morning cortisol in MDD patients compared to controls has also been reported.¹⁰¹ Furthermore, dysregulated cortisol levels may predispose individuals to subsequent depressive episodes.⁹³ A number of studies suggest that crosstalk between the HPA axis and the immune system mediates behavioral responses to stress^102,103, which is a common trigger in the development of depression. The immune system protects against injuries and infections, however a dysregulated immune system leads to chronic inflammation and contributes to the pathophysiology of psychiatric disorders, including depression.^104–106 The circadian clock is directly involved in regulating the immune system, including production of cytokines¹¹, phagocytosis¹⁰⁷, anti-inflammatory responses¹⁰⁸, as well as the expression of pattern recognition receptors.¹⁰⁹ Disruption in circadian rhythms results in increased levels of pro-inflammatory cytokines, including tumor necrosis factor-alpha (TNF-α).¹¹ In turn, TNF-α has been shown to play an important role in regulating circadian rhythms primarily by interfering with E-box-mediated transcription and thus suppressing the expression of clock genes Per1 and Per2 both in vitro and in vivo.^110,111 Chronic sleep deprivation which can result from disrupted circadian rhythms, leading to an increased inflammatory profile in mice and increased microglial activation and astrocytic phagocytosis.¹¹² On the other hand, microglial dysfunction may also be a contributing factor in the pathophysiology of sleep disorders, potentially resulting in a vicious cycle of sleep and mood disruption.¹¹³

Similar to seasonal depression, MDD may also involve disruptions to monoamine signaling. Rats kept for 6 weeks in constant darkness such that they can’t entrain to light, develop a depressive like phenotype, as evidenced by increased immobility times in forced swim test, which is associated with impairment of the noradrenergic locus coeruleus system.¹¹⁴ In addition, these free running conditions lead to increased apoptosis of noradrenergic, dopaminergic and serotonergic neurons which can be partially rescued by desipramine treatment.¹¹⁵ Interestingly these circadian effects on monoaminergic cells were not associated with increased measures of stress, suggesting that this effect is independent of a general stress response. Circadian genes within monoaminergic-rich regions of the brain also play a role in mood regulation both with and without chronic stress. Chronic social defeat stress significantly reduces expression of Per1 and Per2 in the NAc, and consistent with this, knocking down Per1 and Per2 specifically in the NAc leads to increased anxiety-like behavior, as demonstrated by decreased number of entries into open arms in the elevated plus maze and decreased time spent in the center of an open field, in the absence of stress.^97,116 Treatment with fluoxetine normalizes Per gene expression following stress and reverses social interaction deficit, suggesting Per1 and Per2 gene involvement in depression-like and anxiety-like behaviors. Intriguingly, knocking down Bmal1 in the NAc, leads to reduced susceptibility to helpless behavior in a learned helplessness paradigm.¹¹⁷ Furthermore, a knock-down of Bmal1 specifically in astrocytes in the NAc leads to increased exploratory drive (i.e. less anxiety-related behavior).¹¹⁸ Thus, more work needs to be done to understand the complex roles of individual circadian genes within these brain regions and specific cell types and how all of these various factors contribute to the development of MDD (Figure 5).

Figure 5. Mood Regulation by circadian rhythms and the environment.

There is a complex interplay between circadian rhythms, the environment, sleep and mood regulation. The circadian clock affects multiple brain regions and systems and certain mutations in the circadian genes might make an individual more vulnerable to mood disorders. Environmental factors such as seasonal changes, stress or night shift work can lead to sleep and circadian dysfunction and contribute to mood changes. DA, dopamine; 5-HT, serotonin; NE, norepinephrine; MEL, melatonin; SCN, suprachiasmatic nucleus; HPA, hypothalamus-pituitary-adrenal.

Treatments for MDD

Further evidence for the connection between circadian rhythms and depression comes from the fact that treatments directly targeting the circadian system (such as acute sleep deprivation, bright light therapy and sleep phase advance) can be successful in treating depression. Depressed patients typically exhibit sleep phase delay, with the sleep onset falling between 2 am – and 6 am.^119,120 This pronounced phase delay might be due to either a biological or self-imposed misalignment between the internal biological clock and the sleep-wake cycle¹²¹ and treatments targeting the circadian system and/or cognitive behavioral therapy may help to phase advance and align circadian rhythms.¹²²

One of the most effective chronotherapies used as a rapid acting treatment of depression is acute sleep deprivation, which has been shown to improve depressive symptoms in approximately 40% to 60% of patients^123,124 and repetition of acute sleep deprivation treatment leads to increasingly better outcomes.¹²⁴ However, the effects of an acute treatment are typically not long lasting and symptoms often return after a night of sleep. Sleep deprivation may exert its antidepressant action through the resetting of abnormalities in circadian gene expression.¹²⁵ Interestingly, both sleep deprivation and ketamine, another rapid-acting treatment, result in common transcriptional changes in the anterior cingulate cortex, where a number of genes are downregulated, including Per2 and Npas4.¹²⁵ Interestingly both the acute and lasting response to ketamine in people with MDD can be predicted based on their diurnal activity profiles. Prior to ketamine treatment, a phase advanced activity pattern and lower mesor (the rhythm-adjusted mean) distinguished subsequent responders from nonresponders.¹²⁶ Moreover, on day 1, ketamine nonresponders had a blunted 24 hr amplitude relative to baseline while a higher amplitude following ketamine treatment on day 3 was associated with a persistent clinical response.¹²⁷ These data suggest that ketamine is most effective in those with a particular circadian profile at baseline, and that it has a persistent anti-depressant response in those in which the drug induces higher amplitude rhythms.

As mentioned previously, similar to SAD many individuals with MDD are phase delayed and some patients with MDD have lower levels, or inappropriately timed melatonin, which is crucially important in maintaining appropriately timed sleep-wake rhythms.¹²¹ In clinical settings, 8 weeks of morning BLT alone or in combination with fluoxetine was well tolerated and efficacious in reducing depressive symptoms in patients with MDD as evidenced by score changes on the Montgomery-Asberg Depression Rating Scale.¹²⁸ A potential mechanism for the antidepressive effects of BLT for MDD was recently proposed by Huang and colleagues (2019). They showed that a dedicated retina – ventral lateral geniculate nucleus and intergeniculate leaflet – (vLGN/IGL) lateral habenular (LHb) pathway regulates depressive-like behaviors. Specifically, activation of intrinsically photosensitive retinal ganglion cells (ipRGCs) projecting to vLGN/IGL, activation of vLGN/IGL neurons projecting to LHb, or inhibition of postsynaptic LHb neurons was sufficient to decrease depressive-like behaviors in rodents, tested through sucrose preference, forced swim and shuttle box tests.¹²⁹

It is important to emphasize that while light exposure in the morning can aid in ameliorating depressive symptoms, exposure to light at night can produce negative health outcomes and associated with increased odds for depression.^85,130 A recent study by An et al. (2020) helps explain why light exposure has different effects on mood during the day and night. The researchers discovered a circadian-gated neural pathway from intrinsically photosensitive retinal ganglion cells ipRGCs to the dorsal perihabenular nucleus (dpHb) to the nucleus accumbens (NAc), which preferentially relays light signals at night.¹³¹ In particular they, showed that subpopulation of dpHb neurons projecting to NAc were more excitable at night and preferentially responded to light at night and not daylight. Excitation of these dpHb neurons either optogenetically or chemogenetically led to reduced sucrose preference when administered at night but not during the day.

Studies conducted in both humans and rodents show that treatment with antidepressant medications also leads to a phase advancement of circadian rhythms.^84,132 Fluoxetine, a selective serotonin reuptake inhibitor, can phase advance neuronal firing in the SCN.¹³³ Treatment with citalopram, another SSRI, increases the sensitivity of the circadian system to light in MDD patients, which might explain why bright light therapy is more efficacious when administered together with SSRIs.⁹³ Agomelatine, a high affinity agonist for melatonin MT1 and MT2 receptors and a 5HT2C antagonist, has proven to reduce depressive symptoms in MDD patients¹³⁴ and has been shown to phase-shift circadian rhythms of body temperature and release of hormones and corrects the circadian rhythm disruptions caused by prenatal restraint stress in adult rats, suggesting that it’s mechanism of action as an antidepressant involves modulation of the circadian system.^135,136

Insomnia is one of the common symptoms of MDD, and one of the non-medication approaches to treat insomnia is cognitive behavioral therapy for insomnia (CBT-I). Interestingly, CBT-I has been shown to be effective in reducing the risk for the onset of a depressive episode and in treating depression.^137–139 Intriguingly, a randomized control trial evaluating CBT-I effectiveness in individuals with comorbid depression and insomnia found that CBT-I is particularly effective for individuals with an evening preference.¹⁴⁰ The investigators propose that clinical care for patients with comorbid MDD and insomnia could be improved if circadian preferences of the patients are taken into account.¹⁴⁰

MDD is very heterogeneous as a disorder and the key in future clinical studies will be to measure each individual’s circadian and sleep patterns as well as physiological rhythms such that appropriate chronotherapeutic treatments can be chosen which will produce the best response. A better mechanistic understanding of how chronic stress and other factors interact with the circadian system will also lead to the development of more targeted treatments.

Bipolar Disorder (BD)

Bipolar disorder is a complex disorder characterized by states of mania and depression and mixed states with both features. Between states individuals may experience a relatively stable period termed euthymia. An estimated 2.8% of individuals in the United States are diagnosed with BD and it tends to impact males and females fairly equally with some clinical differences in symptom presentation.¹⁴¹ Over the years, it has become apparent that in addition to the emotion and mood changes associated with these states, cardinal features of manic and depressive episodes involve changes in energy and activity.¹⁴² During mania, individuals are typically full of energy, taking on multiple tasks, rapid speech, and racing thoughts, despite very little sleep. During depressive episodes, individuals typically describe a lack of energy in which they are exhausted by routine tasks, behaviors, movements, speech, and thoughts can be slowed, and they typically oversleep. Sleep and circadian disturbances in BD have been found in multiple studies and indicate that individuals with BD have highly disrupted sleep/wake patterns which are significantly more desynchronized during episodes of mania or depression.¹⁴³ Up to one-third of individuals with BD have a diagnosable circadian rhythm disorder with the most common being delayed-sleep phase disorder.¹⁴⁴ Indeed, most individuals with BD also have an evening chronotype similar to individuals with MDD, SAD when compared to control populations. Moreover, a late chronotype predicts a greater number of depressive symptoms and less manic episodes in BD over a 5 year follow up.¹⁴⁵ A number of studies have identified SNPs in genes that regulate circadian rhythms and have found nominally significant associations between various elements of the core circadian clock and BD. The most commonly found are SNPs in CLOCK, PER3, and ARNTL.¹⁴⁶ ARNTL is of particular interest since it just falls short of genome wide significance in large GWAS studies of BD.¹⁴⁷ In addition, when the circadian system is considered as an entire network as opposed to individual genes, there is a significant association with BD.¹⁴⁸

Seasonal patterns of episodes occur in 15–25% of individuals with BD. In these individuals, manic episodes tend to occur during the spring while depressive episodes tend to occur in the fall/winter.¹⁴⁹ Interestingly, a variant in the PER3 gene has been associated with seasonal patterns in BD.¹⁵⁰ Along with seasonal changes, one of the most common triggers for manic or depressive episodes is changes to normal sleep patterns.¹⁵¹ Irregular and disrupted sleep/wake rhythms associate with and may predict first onset of BD, especially in high-risk individuals.^152,153 Travel induced jet lag can be a significant trigger and interestingly a few studies have suggested that the direction of travel is important with eastward travel associated with increased manic episodes and westward travel associated with more depressive episodes.¹⁸ This is in line with studies that find that the genes that control circadian rhythms tend to run faster (i.e. with less than a 24 hour rhythm) in peripheral samples of subjects experiencing a manic episode, while these same genes run slower (greater than 24 hours) in those that are experiencing a depressive episode, with a return to a more conventional 24 hr period during euthymia.¹²⁰ Interestingly, studies have also found that individuals with BD on average have a hypersensitivity to nocturnal light in that this light at night leads to a greater suppression of melatonin along with a delay in sleep, and some have suggested that this could be a trait marker of BD.¹⁴³ Exposure to excessive caffeine, alcohol and drugs of abuse can also trigger episodes, as can exposure to stressful life events.^154,155 Since these environmental factors can result in inappropriately timed, inefficient and shortened amounts of sleep, disruptions to the sleep/wake cycle could be the common thread that leads to the precipitation of episodes. These observations form the Social Zeitgeber Theory of BD, originally put forward by Ehlers, Frank and Kupfer in 1988, which postulates that “life events disturb social zeitgebers (“time givers”), which in turn disrupt biological rhythms, resulting in affective symptomatology in vulnerable individuals”.²

Animal studies also suggest a prominent role for circadian rhythms in BD. Mice with a mutation (a loss of exon 19; Δ19) in one of the central molecular clock genes, Clock, display a range of phenotypes that are highly reminiscent of mania.¹⁵⁶ This includes hyperactivity, increased impulsivity and risk-taking behavior, increased reward value for multiple drugs and natural rewards, and decreased sleep time.¹⁵⁷ Interestingly, these behavioral phenotypes are most pronounced during the inactive (i.e. light) phase of the mice, suggesting a shift between states during the active and inactive phase. Treatment with the mood-stabilizing medications, lithium and valproic acid, both reverse their behavioral phenotypes.^157,158 Other mice with mutations in circadian genes share several phenotypes with the ClockΔ19 mice. For example, mice lacking Rev-erbα have a similar hyperactivity phenotype as do mice with a mutation in F-box and leucine rich repeat protein 13 (Fbxl3), a core circadian gene.^159,160 Recently a mouse model was described that is termed the chronic unpredictable rhythm disturbances (CURD) model in which mice are subjected to a number of circadian, light and sleep disturbances.¹⁶¹ These manipulations lead to the development of a manic-like phenotype. Interestingly, this is in opposition to the effects of chronic unpredictable mild stress which leads to an overall depression like phenotype. The CURD mice showed behavioral and molecular improvements with both lithium and valproic acid. Thus, there are multiple lines of evidence from both human and animal studies suggesting that circadian rhythms are key to the pathophysiology of BD.^{2,18,120,143,149}

Potential mechanisms linking circadian rhythms and BD

The molecular clock tightly controls cellular energy state, metabolism, protein translational processes, neurotransmission, and cellular growth, all processes implicated in BD.¹⁶² These rhythms are essential, as cells in the brain need to be highly metabolically active when the individual is awake, and then have a period of restoration in which they can rid the cell of misfolded proteins, free radicals and reactive oxygen species that have built up with activity. Core molecular clock proteins bind directly to metabolites that are produced with mitochondrial respiration, acting as a “redox sensor” informing the molecular clock in the nucleus of the metabolic state of the cell.¹⁶³ In turn, the clock proteins directly regulate genes necessary for mitochondrial function and antioxidant function. Mitochondrial fusion and fission and the production of new mitochondria have a diurnal pattern and are dependent on having a functional molecular clock.¹⁶⁴ Moreover, when the molecular clock is disrupted in animal models this leads to a large increase in reactive oxygen species in the brain, widespread gliosis and eventually neurodegeneration.¹⁶⁵ A number of studies of BD have suggested that disruptions to mitochondrial function and oxidative stress are central to the disease¹⁶⁶ thus the direct regulation of these processes by the circadian clock could be vitally important in disease development.

Multiple studies indicate that BD is associated with changes in dopaminergic transmission. Mania is generally associated with higher levels of dopamine while reuptake of dopamine may be greater in bipolar depression, though these studies need further replication.¹⁶⁷ In particular, mania involves excessive goal-directed activity and impulsive decision-making driven by heightened reward seeking. These processes are known to be regulated by cortico-limbic dopaminergic systems. ClockΔ19 mice have increased midbrain dopaminergic activity, increased levels of TH and increased dopamine synthesis which is more pronounced during their inactive phase.^163,168 The manic-like behavior of the ClockΔ19 mice can be recapitulated through chronic, optogenetic stimulation of VTA dopamine neurons in wild type mice¹⁶⁸, but only during the mouse’s inactive phase, suggesting that it is inappropriate timing of increased dopamine that leads to mania and not necessarily just excess dopamine at any time of day. Blocking of TH during the light phase but not the dark phase reverses manic-like behavior.¹⁶⁸ This is in line with the precipitation of mania in response to altered light/dark cycles, as this could lead to surges of dopamine at a time of day when it is not expected. Thus, the circadian system is directly regulating key cellular processes and dopaminergic transmission which could contribute to BD.

Another intriguing possibility lies in the concept of “metabolic jet lag”. BD is often associated with metabolic problems including obesity. Metabolic jet lag refers to “a state of shift in circadian patterns of energy homeostasis, effecting neuroendocrine, immune and adipose tissue function”.¹⁶⁹ It is possible that erratic sleeping and eating patterns displayed by individuals with BD leads to highly disrupted rhythms in microbiome function, insulin signaling, expression of metabolic peptides like ghrelin and leptin, all of which are known to influence mood. Recent interest in the use of a ketogenic diet or time restricted feeding to control BD episodes may derive at least some of its effectiveness through stabilization of these systems though more work needs to be done to fully understand the mechanism and rates of efficacy for BD.¹⁷⁰

Treatments for BD

Based on the Social Zeitgeber Theory, Interpersonal and Social Rhythms Therapy (IPSRT) was created as a treatment for BD.¹⁷¹ IPSRT is a manualized, evidence-based therapy that helps align and stabilize sleep/wake and social rhythms to help prevent the onset of episodes and reduce symptoms. In clinical studies, IPSRT produces therapeutic results that are more rapid and longer lasting than intensive clinical management in terms of occupational functioning in individuals with BD.¹⁷² Interestingly a recent study of SRT therapy on its own modified for telehealth delivery to adolescents and young adults with BD, support its role in reducing symptoms of BD, as well as suicide propensity.¹⁷³

BD is commonly treated with mood-stabilizing medications that include lithium salts, anticonvulsant drugs such as valproic acid (VPA), and/or second-generation antipsychotic medications. Morning BLT commonly used to treat SAD and certain forms of MDD, has been used to treat bipolar depression, however, it can trigger a switch to mania.⁷ Studies find that bright light therapy given during the afternoon which will not lead to as great a phase advance in the circadian cycle, may be helpful for people with bipolar depression with less risk of triggering a manic episode.¹⁷⁴ Importantly, all of these treatments can have a strong impact on circadian mechanisms. For example, lithium increases the amplitude of molecular rhythms, likely through inhibition of GSK3β, which is known to phosphorylate several members of the molecular clock, altering their stability and nuclear entry.¹⁷⁵ This could act much like IPRST to keep sleep/wake patterns stable, preventing episodes from happening. In a multicenter prospective trial of lithium in 386 subjects with BD, individuals on stable lithium therapy had the least circadian disruptions while individuals with no prior lithium exposure had the most pronounced disruptions.¹⁷⁶ Another well-known effect of lithium is that it lengthens the circadian period, increasing it to greater than 24 hrs.¹⁷⁷ Lithium’s effects on IP3 signaling in addition to Gsk3β may underlie some of the period lengthening effects of lithium.^178–180 Early studies found that lithium responders are those that have an endogenous period less than 24 hours, and this was also generally found during a manic state.¹⁸¹ In agreement with these studies, fibroblasts isolated from individuals with BD had muted circadian responses to lithium if they had a long period, and interestingly these individuals were not prescribed lithium by their physicians at the time of tissue collection, suggesting that they are not lithium responsive.¹⁸² In addition, iPSCs derived into neuronal precursor cells and glutamatergic neurons from lithium responders had a shorter circadian period, and lithium lengthened the period in these cells.¹⁸³ Genetic variation in IP3 signaling may underlie some of these differences in responders and non-responders to lithium.¹⁸⁰ Interestingly, VPA has a similar impact on rhythm amplitude in molecular rhythms in cell culture, but it can have the opposite effect of lithium on period, creating a shorter circadian cycle, and it’s tempting to speculate that measurement of circadian rhythms in BD may help predict treatment response to lithium or VPA.¹⁸⁴ Multiple clinical trials have been performed testing the efficacy of melatonin or melatonin receptor agonists on BD and they have yielded mixed results. Two meta-analyses of these trials conclude that the receptor agonist, ramelteon might prevent relapse into depression in BD, and that the largest efficacy signal detected was for manic symptoms. Indeed, in the two trials assessing manic symptoms during acute mania, adjunctive melatonin had superior treatment effects versus placebo but there was substantial heterogeneity between studies and patient characteristics, and some studies found no significant differences in patient outcomes.^185,186

Conclusions

In conclusion, it seems clear that circadian rhythms and the ability to adapt to photoperiod changes play a key role in mood regulation (Table 1). It is interesting that a consistently found risk factor across multiple studies for these diseases is a late chronotype. However, it is unclear if this association with psychiatric disease is due to the genetic factors that contribute to “eveningness” or a misalignment between internal rhythms and early wake times due to school or work.¹⁸⁷ The recent COVID-19 pandemic and subsequent school and office closings provided a rare opportunity to examine chronotype on its own without environmental pressures for early wakening. Interestingly, a study of over 19,000 adults across 12 different countries found that evening-types had a large increase in sleep delay, particularly on workdays and an increase in sleep duration.¹⁸⁸ However, they also had worse mental health, well-being, and quality of life self-report than other circadian types. These results held up with corrections for age, sex, socio-economic status or duration of confinement during the pandemic. These results suggest that even without the pressures for early wake times, evening types may be more at risk for depression or other mood episodes, particularly when faced with chronic stress. Other studies, however that have examined the impact of the delay of school start times, and flexible work schedules have suggested that environmental shifts to a later wake time have a beneficial effect on sleep and mood in evening chronotypes^189,190 so it’s likely that both a biological vulnerability exists and its effects are exacerbated by social jet lag induced by early morning waking.

Table 1.

Sleep and circadian characteristics that associate with SAD, MDD and BD

	SAD	MDD	BD
Sleep duration	Increased	Increased or decreased	Increased during depression and decreased during mania
Seasonal changes	Defined by seasonal symptoms which typically begin in fall/winter	More prominent in moderate depression with greatest risk in fall/winter. Stronger in women and little to no seasonal variation in psychotic depression	About 25% of cases show increased depression in fall/winter and mania in spring/summer
Chronotype	Strong association with evening chronotype	Evening types have both an increased risk for depression and increased symptom severity	Evening chronotype linked to depression more than mania
Circadian patterns	Mostly phase delayed with difficulty adapting to photoperiod change	Can be phase delayed, advanced, or low amplitude rhythms	Can be phase delayed or advanced, which can fluctuate with mood state. High intra daily variability and difficulty adapting to any change

Another common feature across disorders is that an unstable sleep/wake or a changing light/dark cycle can precipitate episodes or make episodes worse. This suggests that many people with mood disorders may have an inflexible clock that is unable to properly adapt to change. This could be due to a deficiency in entrainment mechanisms in that there is perhaps less sensitivity to morning light, or the SCN is not able to properly respond to cues. Indeed, studies in animal models suggest that an SCN that is out of synch with the rest of the brain leads to worse behavioral outcomes than a total SCN lesion which can actually produce a protective effect against stress.^95,191 It is also possible that particularly in the case of BD that the SCN is too easily thrown into a state of phase advance or phase delay and the intra-daily instability is what creates vulnerability for episodes. More work will need to be done to determine what types of mechanisms drive these episodes in the face of a change to sleep patterns or the environment.

There is also a growing body of evidence to suggest that stabilization and alignment of circadian rhythms is therapeutic for mood disorders. In the case of SAD, rhythms need to be better aligned with the environment which can typically be accomplished through BLT in the morning to shift rhythms forward, but some individuals may benefit from intervention with melatonin or other zeitgebers at a different time of day. For MDD phase advancing of rhythms may also be therapeutic for some individuals but for others, increasing rhythm amplitude through medications like ketamine may also be therapeutic. For BD, daily rhythm stabilization is key, but there is also evidence to suggest that rhythms might need to be delayed during mania and advanced during depression. In addition, the internal phase of each individual during a euthymic state should also be assessed to determine the appropriate chronotherapeutic treatment.¹⁹² Better circadian profiling of individuals and more thorough testing of chronotherapeutic treatments will ultimately guide appropriate personalized treatment options which can augment or replace existing therapies for mood regulation.