The circadian clock is located above the suprachiasmatic tract, receiving light information from the retina to entrain the SCN activity with LD cycles. The SCN is formed by two bilateral nuclei of ~20,000 neurons distributed into two major subregions: a ventrolateral region, receiving most retinal afferents expressing vasoactive intestinal peptide (VIP), and gamma-aminobutyric acid (GABA), and a dorsomedial region expressing arginine-vasopresin peptide, GABA, and receiving/projecting afferents/efferents from/to other hypothalamic nuclei (7). Most of the dorsomedial neurons exhibit circadian electrical activity through self-sustained transcription-translation feedback loops (TTFL) involving clock-gene expression and its post-translational regulation (8). The transcriptional factor CLOCK and ARNTL, and ARNTL2 form heterodimers which binds to E-boxes in the promoters of their own repressors, Period (Per1-3) and Cryptochrome (Cry1-2), and of the nuclear hormone receptors Rev-erb (α and β), and Ror (α, β, and γ). Two additional loops control Bmal1 expression, by REV-ERBα repression, and RORα activation through RRE elements in its promoter (9). The TTFL molecular machinery is functional in almost all cells in peripheral tissues and is driven by direct autonomic and/or endocrine outputs of the SCN, as well as by indirect blood-borne metabolites temporally gated by the SCN (3).

Clock-controlled mechanisms allow anticipatory temporal organization of biological functions according to predictable changes (e.g., the daily LD cycle) and, just as importantly, a temporal segregation between conflicting or incompatible processes (e.g., feeding and sleeping). Thus, daily photic entrainment of the SCN clock, i.e., to adjust its period with that of the LD cycle, is a key organizer for such temporal homeostasis. Light stimuli synchronize the SCN through glutamatergic signaling by means of the retinohypothalamic tract (4). Ventrolateral neurons respond to photic stimulation during the subjective night in nocturnal rodents by increasing the expression of clock genes of the Period (Per) family, per1 and per2 (10–13). In turn, signaling from the retinorecipient ventrolateral neurons to the dorsomedial SCN synchronize self-sustained clock-gene oscillations. Rhythmic intercellular coupling allows the SCN to act as a tissue pacemaker with coherent circadian outputs. Such coupling occurs by neuropeptide signaling, such as VIP (through VPAC receptors), GRP-mediated communication (14, 15), and NO and GABA neurotransmission (16–20). Uncoupling the SCN neuronal network of gene and neurotransmitter activity by untimed circadian lighting generates dampened SCN outputs controlling behavior and physiology. The SCN acts as the hypothalamic link between the retina and oscillators in peripheral organs, entraining those organs to the LD cycle.

The SCN is a central timing structure that organizes circadian oscillators controlling metabolism, primarily by maintaining behavioral rhythms of activity/rest and feeding/fasting, through both synaptic (21) and diffusible factors (22) that couple the SCN with neuroendocrine and neural outputs. SCN neurons exhibit an endogenous rhythm of electrical activity linked to the molecular oscillator (23), driving rhythmic neurotransmitter release at their efferents (24) even under constant dark conditions (25). Many brain areas serve as effectors of the circadian clock, including projections to different targets such as the paraventricular nucleus of the hypothalamus (PVN) and other non-hypothalamic areas that gate physiologically relevant sensory information (26). One of the major outputs of the SCN is directed toward the control of hormonal secretions, by means of (1) direct contact with neuroendocrine neurons at the PVN, controlling the release of corticotropin-releasing hormone (CRH) and stimulating the adenohypophysis for the release of adrenocorticotropic hormone (ACTH); (2) targeting intermediate neurons, e.g., neurons of the medial preoptic nucleus, the dorsomedial hypothalamic nucleus (DMH), or the sub-PVN; (3) stimulating autonomic PVN projections to intervene into autonomic nervous system (ANS) circuits that affects endocrine organs; and finally (4) by influencing its own feedback (24). In addition, CBT rhythm is a relevant SCN output which can temporize peripheral tissues (27). Figure 1 outlines the integration of circadian energy metabolism, by the main brain structures regulating behavioral, endocrine, and physiological outputs, and the main pathways for peripheral interactions and feedbacks.

The SCN controls several outputs driving diurnal rhythms in metabolic hormones to regulate energy homeostasis. A main SCN–CRH–ACTH output drives a rise in adrenal glucocorticoids just before activity onset. This promotes arousal and alertness by enhancing liver gluconeogenesis (from amino acids and fatty acids), promoting release of liver glucose to the blood, and increasing its uptake in the brain and muscles. Adrenal glucocorticoids have been implicated as a peripheral humoral cue for the entrainment of oscillators such as the liver (28). Glucocorticoids can also feedback to either the SCN or its outputs driving behavioral re-entrainment to light in a model of jet-lag (29). The systemic levels of the main metabolic hormones, insulin, glucagon, and adrenalin also exhibit robust circadian rhythms controlled by the SCN (1). The circadian rhythm of glucose has a direct drive from the SCN through a sympathetic pathway that connects with the liver (30). The SCN increases insulin sensitivity through multiple pathways, resulting in an increase in glucose uptake in muscle, and simultaneously causes increased hepatic glucose production (31, 32). In addition, the SCN exerts a direct control on cardiac and glucose metabolism (24). Circadian and LD-controlled pineal melatonin release is driven by SCN output through a multisynaptic pathway relaying in the sympathetic superior cervical ganglion (33, 34), regulating metabolism in photoperiodic species (see below).

Peripheral and central clocks are normally held in adaptive synchrony by the SCN. In particular, the central clock regulates rhythmic feeding behavior, which in turn produces rhythmic food intake generating circadian-metabolic signals in the periphery. Endocrine and autonomic SCN outputs to peripheral tissues regulating nutrients intake, processing, mobilization, usage for fuel and/or storage provide a circadian control on global energy metabolism. While the circadian activity/rest rhythm is entrained to the LD cycle, the feeding/fasting rhythm appears to affect the timing and robustness of circadian rhythms in metabolic organs. As a general rule, peripheral oscillators (like the liver) respond rapidly to cycles of food availability, while the light cycle resets the SCN more quickly than does food (35, 36). Restricting food intake during the light, resting phase in nocturnal rodents under LD, in addition to affect CBT and glucocorticoid amplitudes, completely reverses clock-gene expression in the liver, kidney, heart, and pancreas, but do not affect SCN function (37). Indeed, the expression of genes controlling peripheral clocks can be regulated directly by the circadian clock, by systemic signals from periodic food ingestion, or by a combination of both mechanisms (38, 39). While the circadian component of this regulation optimizes the anticipation of predictable fluctuations in the organ environment, the responsiveness to systemic cues adds a degree of flexibility to the system that is necessary to adapt to unpredictable changes in the environment (e.g., in times of caloric intake restriction, and/or temporal niche switching, see below).

Peripheral clocks are strongly affected by cell metabolism, the physiological and metabolic state of the surrounding tissue, and serum-borne signals. Several metabolic signals such as NADH/NAD+, NADPH/NADP+, ATP/AMP, oxygen and its reactive species, carbon monoxide, or glucose, directly affect transcriptional activity of circadian proteins (40). Peripheral oscillators such as the liver clock are additionally coupled to cellular and general physiology through energy and nutrient sensing systems, such as AMP-activated protein kinase (AMPK), PPARGC1A (also known as PGC1), metabolic feedback loops involving metabolites, such as polyamines, and nuclear hormone receptor signaling pathways.

Peripheral oscillators not only adjust their clock-genes rhythms to the master circadian clock by a mix of direct and indirect signals from the SCN but also can be differentially altered in phase by behavior (e.g., activity, or access to food) or light. While the liver clock adjusts its phase preferentially to feeding time, adrenal glands strongly respond to light input Ishida (41, 42). Animals receiving food six times a day lose their rhythm in white adipose tissue in seven of nine tested oscillatory metabolic/adipokine genes, but not in clock genes. By contrast, abolishing the daily corticosterone peak also rendered the peripheral clock-genes arrhythmic (43).

Multiple circadian clocks drive local rhythms of cellular metabolism, synchronized by cues from the SCN as well as by other inputs. Autonomous rhythms in the mediobasal hypothalamus might contribute to metabolic homeostasis (44–49). Moreover, the DMH might play an important role as a component of the SCN-independent food entrainable oscillator (50–53). Importantly, regulation of body weight involves the integration of multiple signals in a neural network of several interconnected hypothalamic areas, including the SCN, the arcuate nucleus (AN), the ventromedial hypothalamic nucleus (VMH), and the PVN, whose interactions control the main three factors that underlie energy balance: appetite and food intake, deposition of fat, and energy expenditure (2, 54).

The functions of the liver are rhythmically regulated; an analysis of transcriptions within the liver shows two peaks corresponding to the dusk and dawn oscillators, and this may reflect the differences in physiological requirements such as energy demand (55–58). The liver orchestrates rhythms in energy homeostasis, i.e., levels of glucose transporters, the glucagon receptor, and enzymes regulating the metabolism of hexose sugars were observed with peak phase of expression in accordance with the rhythm of feeding. In addition, enzymes related to lipid homeostasis such as glycerol 3-phosphate pathway (e.g., glycerol-3-phosphate acyltransferase, 1-acyl-glycerol-3-phosphate acyl-transferase, and lipins), which regulates glycerol and lipid metabolism and triglyceride accumulation, are expressed in a circadian manner (59).

The liver circadian clock acts by buffering the fluctuations of blood glucose levels originated by behavioral and feeding/fasting cycles in a circadian fashion. Specific Bmal1 deletion in the liver reduced basal glycemia in mice during the fasting phase, mainly by altering liver glucose export (60), insulin and glucagon secretion (61, 62), and glucose production and uptake (32, 63). In addition, many clock-genes and clock-controlled genes have multiple roles in glucose metabolism. Cryptochromes regulate hepatic gluconeogenesis through interaction with G protein-coupled receptors, inhibiting cAMP storage and increasing transcription of CREB-regulated gluconeogenic genes. In addition, overexpression of Cry1 in liver affects glucose levels in diabetic mice (64). Positive and negative regulators of the core clock oscillator control circadian regulation of the glucose metabolism. Multiple signaling pathways converge on rate-limiting enzymes of glucose anabolism and catabolism, such as phosphoenolpyruvate carboxykinase or pyruvate kinase (65). As mentioned earlier, the liver oscillator is strongly coupled to feeding/fasting cycles, and its insulin-dependent fluctuations, in addition to being regulated at the level of Per1-2 genes (66).

The chronic absence of LD cycles is an abnormal condition for most organisms, except those living in extreme latitudes during winter such as the arctic ground squirrel (132) or the reindeer (133). Reindeers suppress their daily activity/rest pattern being active for most of the 24 h during the polar summer day in which constant lighting is present. On the other hand, the constant darkness observed during polar winter day diminished activity bouts and also generates arrhythmicity in reindeer species living at higher latitudes. During the polar Summer day, when constant light occurs, the daily rest/activity patterns of reindeers are suppressed to permit near 24 h active behavior. By contrast, during the polar Winter, constant darkness reduces the activity bouts of reindeers and can lead to arrhythmicity in reindeer living at higher latitudes (but LD rhythms are kept at lower latitudes). Weak circadian (but not photic) control was found for reindeer melatonin levels, as well as for fibroblasts transduced with Bmal1:luc or Per2:luc reporters (134), and is proposed to be a rule particular to polar resident arctic mammals. Still, there are no data on energy metabolism for these species living in such extreme conditions.

Protocols of chronic phase-shifting of the 24-h LD cycle (i.e., chronic jet-lag, CJL), or LD schedules with periods different from 24 h (i.e., T-cycles) that affect steady-state entrainment of the SCN clock, demonstrate to be a disruptive condition generating the internal desynchronization of the whole circadian system. It ranges from behavior (177, 178), ventrolateral-dorsomedial uncoupling of Per1 and Bmal1 SCN expression (177), to peripheral regulatory outputs as CBT and sleep stages (179) and melatonin secretion (180). Thus, it is not surprising that this chronic misalignment of the circadian clock with the photic cycles leads to metabolic alterations (see Figure 2 and Table 1). Exposing mice to chronic 6-h advances of the LD cycles every 2 days (i.e., CJL by advances) induces weight gain, increases body fat, plasma triglycerides, and adipocyte size, yet without a concomitant increase in food consumption (118). Interestingly, these effects were not observed when applying a similar CJL delaying schedule, which allowed behavioral entrainment. Similar CJL protocols consisting of two 6-h advances on a weekly basis, or of 8-h advances every 2 days also reported disrupted genes expression in the liver (152, 153). Some groups have tested protocols of alternating advances and delays of the LD cycle in rats and found increases in total body weight and adipose tissue along with alterations in metabolic hormones and liver function (156).

Rats exposed to T-cycles of similar length to their endogenous period (T25 and T26) show decreased body weight, food consumption, and less weight gain/food grams when compared with those exposed to shorter T-cycles (181). This experiment provides an interesting hypothesis of circadian-metabolic resonance, in which light cycles very similar to the circadian period efficiently couple SCN-peripheral entrainment to nutrient intake/delivery with energy output. In agreement, mice exposed to shortened T-cycles of 20 h display high body weight, increased leptin, insulin, and insulin/glucose ratio (160). Together, the studies using CJL and T-cycle protocols indicate that chronically forcing the clock to entrain by advances is the worse condition for the body energy metabolism and weight regulation.

Although social cues also act as relevant zeitgebers, the human circadian clock responds in a canonical way to light, similar to other diurnal mammals. It exhibits the main sensitivity to light in the late biological day/early biological night, delaying up to 3 h the clock phase, a phase advance region (~2-h shifts) in the early biological day, small phase shifts during the middle of the biological day, and a transition point toward the end of the biological night (182). Importantly, this timed response to light is of functional significance for the entrainment of the human clock by natural daylight, shifting its phase (mid-sleep time on free days) to the time of dawn at different latitudes (183). It could be generalized that human rhythms, such as activity/rest and melatonin secretion, become significantly earlier and shift to the natural crepuscules when exposed to strong (“natural”) zeitgebers composed of bright daylight and absence of artificial nightlight (184, 185). Indeed, light pollution in urban areas, i.e., the artificial lighting during the biological night, might have induced a shift of human activity toward the night and is emerging as a negative factor for health increasing the risk for metabolic disorders (186). Altered exposure to the natural LD cycle by attenuated exposure to daylight within buildings or increased exposure to artificial nightlight is one of the mechanisms leading to insufficient sleep (187). This is the so-called social jet-lag, i.e., the situation in which the body’s biological clock and the actual sleep schedules do not match, and which can be quantified as the difference between mid-sleep time on work days and mid-sleep time on free days (188). Poor (self-reported) lighting exposure, together with metabolic factors (changes in weight and appetite) of the seasonal variations in mood, is a key to the MS in the Finland population (189). In addition, irregular activity-rest patterns with higher interdaily stability was also associated in older (81.6 ± 7.5 years) adults with MS (190), although photic effects on the daily patterns were not determined in this study.

Light intensity of ~200 lux is able to shift the human circadian clock (191) and even very dim light can decrease plasma melatonin levels (192), which indicates that subjects under typical indoor illumination can receive enough irradiance at night to affect their circadian system. Higher levels of LAN exposure (>3 lux) in elder people, resulted in higher body weight and body mass index (BMI), elevated triglyceride, and LDL-cholesterol levels independent of demographic and socioeconomic parameters, and without changing melatonin levels (193). Using satellite images of night time illumination, and clinical data on countries worldwide, a statistical significant association has been found between the exposure to LAN and being overweight (194). A similar study in Korean urban areas also found this association together with delayed and disrupted sleep (195). Taking together, the data from Wright et al. (184), de la Iglesia et al. (185), and Roenneberg and Merrow (188), it could be hypothesized that humans living in urban areas with access to electrical light likely experience LAN, delaying its biological clock, and therefore generating social jet-lag. Remarkably, a large-scale epidemiological study has demonstrated that social jet-lag is a key factor increasing BMI and obesity, independently of sleep duration (196). Greater social jet-lag associated positively with triglycerides, fasting insulin, insulin resistance, waist circumference, and BMI, and negatively with HDL cholesterol (197).

Our current 24-h society requires an increasing number of employees to work nightshifts and, as a result, millions of people worldwide work during the evening or night for a certain period of their life. Most epidemiological studies indicate that LAN is a key factor in shift workers to develop metabolic impairments (198), and the risk of MS increases in health-care personnel working night shifts (199). However, the establishment of a direct link between LAN affecting the circadian clock and metabolic disease in shiftworkers has been prevented by a lack of consistent, quantitative methods for measuring LAN supplying contradictory results [reviewed in Ref. (200)]. Moreover, nightworkers switch their temporal niche by a 180° shift of the activity/rest rhythm, imposing a circadian misalignment in the timing of energy metabolism for a diurnal mammal. The SCN clock tends to entrain to the LD cycle driving endocrine and autonomic outputs for diurnality, as for instance peaking insulin sensitivity at noon in subcutaneous adipose tissue (201). Night work forces the circadian system for repetitive arousals during the night, to sleep during the day, and to eat at abnormal circadian times. To simulate circadian misalignment due to shiftwork in laboratory conditions, imposing an activity/rest pattern different from 24 h during 7 days generates eating and sleeping ~12 h out of phase from habitual times (202). This misalignment systematically decreases leptin, increases glucose despite increased insulin, increases postprandial glucose, completely reverses the daily cortisol rhythm, increases blood pressure, and reduces sleep efficiency. Similar studies also showed decreased insulin sensitivity (without insulin increase) (203) and altered rhythm of energy expenditure (168). This forced desynchronization protocols showed how changing the feeding pattern and timing of metabolic signals for only a week, leads to cardiometabolic alterations and a prediabetic state. In a 3-day study of 12-h simulated shiftwork, the misalignment between the endogenous circadian system and daily environmental/behavioral rhythms with nighttime feeding, affected glucose tolerance, by elevating postprandial glucose level. These changes in glucose metabolism are related to different insulin mechanisms: a decreased pancreatic β-cell function and decreased insulin sensitivity (204). Sleep deprivation and melatonin disruption are also observed in subjects chronically subjected to shiftwork (205), where sleeping less than 6 h daily was associated with weight gain and an increase in BMI (206).

Shiftwork exposure has been hypothesized to increase the risk of chronic diseases, including cancer, cardiovascular disease, MS, and diabetes (198). Several epidemiological studies have evaluated the association between shiftwork and MS. This syndrome is a cluster of risk factors including central obesity, elevated blood pressure, elevated triglycerides, lowered high-density lipoprotein cholesterol, and elevated fasting glucose, which are often seen simultaneously in an individual, and is a risk factor for cardiovascular diseases. Table 2 summarizes information regarding the association between night shiftwork and the different parameters of this syndrome. In addition, these data indicate an association of shiftwork schedules with low concentrations of HDL cholesterol, higher diastolic and/or systolic arterial blood pressure, fasting glucose, insulin, and/or triglycerides. Some of these reports also show an increased risk factor of MS associated with a longer period of time working at night or in rotating schedules (207, 208). Moreover, these associations are stronger in females than in males (207, 209), and indeed it has recently been shown that the risk of obesity in shift workers was not related to these factors in males (210).

A dysregulation in the mechanism of hunger/satiety control and in energy expenditure may be related to the increase of BMI associated with night shift work. First, meal distribution during shiftwork is different. Shift workers tend to have smaller meals for breakfast and lunch but increase their snacking behavior. In the afternoon or nighttime, the frequency of snacking increases, with a tendency for a higher consumption of carbohydrates (199), which is indicative of obesity-related eating behaviors (213, 216). Second, a study including 14 adults in a 6-day simulated shiftwork protocol, showed decreased total daily energy expenditure in nightshift days compared with baseline, driven predominantly by a decrease during the sleep time (217). Third, there was an increase of fat utilization, together with a decrease of carbohydrate and protein, with similar levels of the hunger-stimulating hormone ghrelin and decreased levels of the satiety hormones leptin and peptide YY.

While light is the most potent zeitgeber, the availability of non-photic cues such as food and exercise, may alter the internal circadian clock independently. Instead of modifying the environmental cues, researchers force the timing of otherwise naturally entrained behaviors or physiological variables. Two simple behavioral interventions are to manipulate the timing of wake/sleep and to schedule feeding times. These behaviors represent both controlled outputs of the clock and feedback mechanisms or inputs to the system. Because of this, altering the phase of such events in relation to the light cycle (like when humans work night shifts) can have a profound effect on biological rhythms and hence in metabolism (see Table 2).

Sleep deprivation in mice has profound consequences on liver activity from the gene transcription level to metabolism of fatty acids and other metabolites and also alters feeding patterns (218, 219). Instead of simply preventing sleep, researchers have developed a “forced work” protocol to keep rats awake by forcing them to remain active by placing them in constantly rotating running wheels. Under this protocol, rats develop obesity, show altered feeding behavior, and glucose and triglycerides levels, as well as desynchronization of clock-genes expression in the liver (166, 167).

Timing of food intake affects the rhythms of clock-gene expression in the liver (37, 39), the dorsomedial hypothalamus and the paraventricular nuclei, and the pituitary gland tissues, but not in the SCN (220). In consequence, feeding during the “wrong time” (e.g., the resting light phase in nocturnal animals) can internally desynchronize the circadian system, leading to disease. Studies have shown that forcing mice or rats to feed during the light phase leads to increased body weight and food consumption, variations in hormonal and metabolite levels in plasma, fat accumulation, alterations in the liver, and MS (169–171, 220, 221).

The unconstrained capacity to manipulate housing lighting conditions in the laboratory might be considered a prerequisite to test potential therapeutic tools. For that reason, researchers have sought model animals to develop strategies that may either overcome or alleviate the symptoms associated with situations of disruption of the circadian-metabolic system.

Behavioral interventions appear as easy alternatives in this sense and represent translatable potential tools for further research in human populations. The main objective of such protocols is either to reinforce entrainment to the LD or other experimental cycle or to provide a different entraining cue that may reorganize rhythms in a way to avoid possible alterations. Scheduling feeding regimes can be as deleterious when “wrongly” timed as can be beneficial under certain challenging conditions. Many studies have shown that TRF can be beneficial, for example, to counter the effects of obesity-inducing diets (222, 223). But it also can help to combat the consequences of protocols that disrupt the biological rhythms and metabolism.

Time-restricted feeding to a favorable circadian timing, i.e., when caloric intake can manage to balance anabolic and catabolic processes with energy expenditure, represents a therapeutic (and experimental) tool preventing metabolic alterations leading to weight homeostasis alterations (119). Some of the reported benefits in rodents include improved glucose tolerance, reduced triglyceride, reduced cholesterol, reduced systemic inflammation, and improved endurance. Forced-desynchronized mice under CJL can keep control weight values when fed only during the dark phases, with no changes in food intake (118). Rats that gain weight under the previously described “forced work” protocol during the rest phase can be phenotypically rescued by preventing feeding during that period and allowing it during the dark phase (166). TRF can be such a strong key that is able to reinstate the Rev-erbα and Bmal1 (but not Per1-2) gene expression rhythms at the SCN in behaviorally arrhythmic mice under LL (141).

Wehrens et al. measured the effect of a 5-h meal delay on the human circadian system and observed that late meals delay rhythms of plasma glucose and adipose PER2 clock-gene expression, suggesting that meal timing may help to reset the circadian system in cases of shift work (224). A 16-year longitudinal study reported a 27% higher risk of coronary heart disease (measured by BMI, hypertension, hypercholesterolemia, and diabetes) in men who skipped breakfast and a 55% higher risk in men who ate late at night, in comparison with men who eat normally. However, no association was observed between eating frequency (times/day) and risk of coronary heart disease (225). In addition, a recent study using a smartphone app to monitor eating time in healthy adults revealed a daily eating pattern highly variable from day to day, indicating that more than 50% of adults spread their daily caloric intake over 15 h or longer. Interestingly, this report showed that reducing the daily eating duration to 10–11 h for 16 weeks, without decreasing calories or change nutrition quality, induced weight loss and improved sleep (226). This observation, together with previously mentioned data recorded in animal models, suggests an important role of eating patterns in metabolic diseases.

Running-wheel activity has been shown to represent a strong feedback mechanism to the circadian system, and as such represents a potential therapeutic tool for chronodisruption models. Voluntary wheel running enlarges the boundaries of behavioral entrainment to T-cycles (227) and increases the amplitude of SCN electrical activity rhythm (228). It was found that wheel access in mice modulates CBT and heart rate rhythm waveforms (i.e., increase these variables during or after energy demand for exercise) and couples the PER2:luc phase of adrenal and liver oscillators driving cardiometabolic output (229). In this direction, we demonstrated that the availability of a running wheel allowed for correct entrainment and prevented the elevated body weight gain observed in mice under the desynchronizing CJL schedule. We also found that social contact, achieved by housing mice in groups, is a successful intervention to alleviate metabolic symptoms in this model (118).

Of course, pharmacological tools have also been tested. One approach is to try to replace or substitute the action of disrupted hormonal function. For example, melatonin administration during the subjective night prevents MS-like symptoms in rats under LL (147). Another study showed that metabolic alterations derived from CJL protocols were prevented by treatment with estrogens (155). Yet, another type of pharmacological approach that requires further research is the application of drugs with the ability to induce or reinforce entrainment to environmental cycles and its changes.

Protocols with photic alterations in nocturnal rodents (such as LAN and LL) provide evidences to approximate metabolic alterations due to circadian disruptions elicited by nocturnal light pollution in humans. Receiving LAN appears to be a common disrupting factor for both nocturnal and diurnals. The main limitations for these approaches, however, are that temporal niche determines a 180° difference between nocturnal rodents and humans, also regarding the regulation of metabolic variables and energy expenditure for behavioral activity. In addition, nocturnal rodents are more sensitive to nocturnal light than humans. Behavioral and feeding pattern misalignment with lighting schedule seems to be a key factor for most alterations, as discussed below.

Endocrine pathways directly modulated by light can play a role in photic-related metabolic alterations. Melatonin is released by the pineal gland and its nocturnal production is sensitive to light exposure during the circadian night (230). Its roles in metabolism go from participating in insulin production and secretion, synchronizing feeding with the organism’s optimal metabolic state [reviewed in Ref. (108)], and regulation of leptin rhythm (95) photoperiodic resistance (see above). Also, melatonin rhythm disruption correlates with that of adipokines and obesity, and its exogenous administration reverse this phenotype (84). Because of this, it is tempting to speculate that disruption of the melatonin pathway under challenging light conditions (see Table 1) may underlie metabolic dysfunction and obesity. Indeed, LL and LAN protocols could be integrated by the SCN and melatonin signal as a “lengthened” photoperiod, thus encoding a summer-like metabolic state, increasing food intake and adipose deposits (see above). However, there is evidence that animals lacking melatonin still suffer LAN-induced metabolic alterations and loss of weight homeostasis (115, 144), and the link of melatonin with leptin in human obesity is still inconsistent (84). Thus, photic melatonin disruption per se does not appear to be a factor linked to obesity neither in nocturnal rodents nor in humans.

Glucocorticoid hormones are key to the regulation of metabolism and are also tightly linked to the circadian system (231), and as such, can also be considered as an underlying cause for metabolic disease in conditions of biological rhythms disruption. Alterations in the 24-h rhythm of glucocorticoid levels may promote circadian misalignment and its adverse health consequences. Multiple studies have shown that the 24-h profile of cortisol concentrations does not adapt rapidly to acute shifts in light–dark, activity–rest, and/or feeding cycles, as occurs in jet lag and shift work rotations, although its amplitude may be reduced under some conditions (232). But the revision of the literature on animal models (67) and human subjects summarized in the previous sections shows that under the different disrupting light conditions, glucocorticoids are impaired, increased, or not affected at all. A study in humans has found that light effects on cortisol are strongly dependent on light intensity, which can explain the seemingly contradictory reports in both humans and laboratory animals (233).

Figure 2 depicts a scheme of the main circadian changes found using protocols of photic desynchronization in nocturnal rodent models (mice and rats), to help the construction of a generalized framework. It seems that photic hormonal pathways alterations per se cannot be considered the sole triggers of metabolic disturbances under disrupting lighting conditions. Photic inputs to the SCN can directly and indirectly affect physiological responses given the presence of multiple pathways regulating peripheral oscillators (see above). Importantly, peripheral synchronization for metabolic organs such as the liver is achieved mainly by SCN entrainment of the feeding/fasting pattern, and derived entraining signals can act indirectly of photic effects (234) through blood-borne nutrients assimilated and/or metabolites (3). In this sense, a desynchronized feeding/fasting pattern with temporal niche driven by LD cycles can be considered as a main factor for most metabolic alterations.

We will now focus on behavioral effects of light disruption that have also metabolic implications: meal time shift, activity levels, and sleep disruption. To distribute feed bouts or meals during the rest phase, tending to the abolishment of the feeding/fasting pattern, results a common feature of most the aforementioned lighting protocols in rodents (dim LAN, LL, or CJL) (see Figure 2). Indeed, most metabolic disruptions are reversed by scheduled feeding (see Strategies to Overcome Metabolic Effects of Circadian Disruption), restoring normal weight homeostasis under CJL, even while gross locomotor activity remains desynchronized with the LD cycle (118). The same can be said of decreased or disrupted locomotor activity levels and rhythms, or even the sleep/wake pattern (235, 236).

Postprandial glycemia is under circadian control (71). In humans, there is a decreased glucose tolerance during the evening and night as result of a reduced insulin secretion, as well as peripheral and hepatic insulin resistance. This pattern is independent of both the sleep/wake and feeding/fasting rhythms (204, 237) but depends on glucocorticoid rhythm (238). Moreover, increased caloric intake when the body is not able to manage it (i.e., to storage, mobilize, or fuel) is a key factor to develop insulin resistance, MS, and obesity (239, 240).

Circadian misalignment due to behavioral and/or photic desynchronization (e.g., shiftwork or CJL) alters endocrine and autonomic pathways set to manage metabolic fuels needed for energy demand (i.e., glucose, fatty acids, or ketone bodies under chronic starvation). Figure 3 schematizes the main disruptions that can be generalized in nocturnal rodents and nightworkers. First, the daily amount of food intake seems to be the same among day and shiftworkers. Wrongly timed circadian feeding leads to postprandial glucose increase, first due to misaligned circadian glucose sensitivity at the pancreas, both in nocturnal rodents (6) and humans (202, 204). Importantly, eating at night during simulated shiftwork reduced total daily energy expenditure during nightshift schedules (217). Thus, impairments in glucose uptake (i.e., a prediabetic state) can be chronically established due to changing the set point for glycemia, leading to further insulin insensitivity, and hyperglycemia. It is conceivable that, starting from reduced nighttime postprandial glucose tolerance, chronic dysregulation in the uptake and/or usage of glucose can lead to further increased usage of fatty acids from triglycerides stores in the liver and adipose tissue. This misbalance between carbohydrates/triglycerides usage can lead to dyslipidemia and hyper-triglyceridemia (i.e., increased free fatty acid levels) found in most animal and human studies described above and increased caloric storage. In addition, these alterations in caloric intake and storage, where also attributed to negative energy balance imposed by sleep debt (217, 241). Thus, eating, and being awake at the wrong time, i.e., when the circadian clock promotes sleep, is a risk factor for MS, weight gain, and obesity.