Circadian Rhythms and the Gastrointestinal Tract: Relationship to Metabolism and Gut Hormones

Abstract

Circadian rhythms are 24-hour biological rhythms within organisms that have developed over evolutionary time due to predefined environmental changes, mainly the light-dark cycle. Interestingly, metabolic tissues, which are largely responsible for establishing diurnal metabolic homeostasis, have been found to express cell-autonomous clocks that are entrained by food intake. Disruption of the circadian system, as seen in individuals who conduct shift work, confers significant risk for the development of metabolic diseases such as type 2 diabetes and obesity. The gastrointestinal (GI) tract is the first point of contact for ingested nutrients and is thus an essential organ system for metabolic control. This review will focus on the circadian function of the GI tract with a particular emphasis on its role in metabolism through regulation of gut hormone release. First, the circadian molecular clock as well as the organization of the mammalian circadian system is introduced. Next, a brief overview of the structure of the gut as well as the circadian regulation of key functions important in establishing metabolic homeostasis is discussed. Particularly, the focus of the review is centered around secretion of gut hormones; however, other functions of the gut such as barrier integrity and intestinal immunity, as well as digestion and absorption, all of which have relevance to metabolic control will be considered. Finally, we provide insight into the effects of circadian disruption on GI function and discuss chronotherapeutic intervention strategies for mitigating associated metabolic dysfunction.

diurnal, GI hormones, intestine, metabolic homeostasis

Circadian rhythms are internally generated 24-hour rhythms that are critical to establishing organismal homeostasis throughout the day-night cycle (1). Virtually all biological processes are regulated by the circadian clock, and extensive studies in animal models have definitively linked circadian disruption to the development of a wide range of diseases (2). In line with this evidence, shift work, of which an estimated 16% (https://www.bls.gov/news.release/flex2.nr0.htm) of the population partakes, has been associated with many diseases including those related to metabolic dysfunction. A meta-analysis including more than 300 000 subjects found that shift work was associated with increased risk of overweight, obesity, and type 2 diabetes (T2D) (3, 4). Interestingly, metabolic tissues including the pancreatic islets, liver, adipose tissue, skeletal muscle, and the gastrointestinal (GI) tract have all been shown to express cell-autonomous clocks that act together to coordinate diurnal metabolic homeostasis (5-11). The GI tract, which is the focus of this review, has a unique role in the “metabolic clock” because it is situated to be the first point of nutrient contact upon food intake. Although the main functions of the gut are classically thought to be digestion and absorption of nutrients, this integrated series of distinct tissues is also critical in reducing metabolic inflammation through the maintenance of barrier integrity and via the presence of an active immune system. Furthermore, the GI tract is the largest endocrine organ in the body, responsible for secreting a number of gut hormones that are essential to facilitating adequate pre- and postprandial responses. Herein, we focus on discussing the role for circadian biology in regulating the function of the GI tract and how this organ is a major determinant of diurnal metabolic homeostasis.

Circadian Rhythms

The term “circadian” is derived from the Latin words “circa” (about) and “diem” (day), indicating that circadian rhythms have a period of approximately 24 hours or 1 full day on Earth. These rhythms, which have developed over evolutionary time due to constant predefined environmental changes, provide the ability for all known biological organisms to anticipate and adequately respond to external cues, thereby promoting organismal fitness (1). Within organisms, circadian rhythmicity is generated by clock genes, which have been shown to be autonomously expressed in all nucleated cells of the body.

The mammalian molecular clock

The core mammalian circadian molecular machinery is comprised of the basic helix-loop-helix transcription factors, circadian locomotor output cycles kaput (CLOCK) along with its paralog neuronal PAS domain protein 2 and brain and muscle ARNT-like 1 (BMAL1, encoded by ARNTL), which heterodimerize and bind to the E-box promoter elements of the Period 1, 2, and 3 (PER1/2/3) and Cryptochrome 1 and 2 (CRY1/2) genes (Fig. 1). CRY and PER then form a complex which feedback-inhibits ARNTL and CLOCK expression, thereby forming a transcriptional/translational autoregulatory feedback loop. The casein kinases 1δ and -ε (CK1δ/ε) and AMP-activated protein kinase phosphorylate the PER and CRY proteins, respectively, causing their destabilization and subsequent degradation. Furthermore, the F-box/leucine rich-repeat protein 3 protein acts in complement with CK1δ/ε by mediating the ubiquitination and subsequent degradation of CRY1/2. The rate of PER and CRY degradation is ultimately responsible for establishing the 24-hour circadian period (1). In addition to this core loop, the BMAL1/CLOCK heterodimer also induces expression of the nuclear receptors, REV-ERBα (encoded by NR1D1) and REV-ERBβ as well as retinoic acid related-orphan receptor α and γ (RORα and -γ) (12, 13). These proteins function in opposing directions, whereby REV-ERBα/β repress ARNTL transcription while RORα/γ act to stimulate ARNTL expression (12, 13). Importantly, all of these clock genes are involved in transcriptional regulation of what are termed clock-controlled genes, thereby establishing a circadian rhythm to key cellular processes. Furthermore, it has also now been elucidated that clock genes are important in establishing epigenetic modification signatures that determine DNA accessibility for the transcriptional machinery, which is ultimately responsible for circadian expression of many genes (14-16). As a whole, it has been demonstrated that approximately 43% of all protein-coding genes display a circadian rhythm in their expression in 1 or more tissues (17) because of direct binding of the clock machinery or through the actions of other rhythmically expressed clock-controlled gene proteins.

Figure 1.

The core mammalian molecular clock. The core biological clock is comprised of a positive arm in which BMAL1 and CLOCK heterodimerize to induce expression of the negative arm consisting of PER and CRY, which then feed back to inhibit Bmal1 and Clock. Additionally, BMAL1/CLOCK activate Rev-Erbα/β and RORα/γ, which have opposing roles in regulation of Bmal1, whereby Rev-Erbα/β stimulates and RORα/γ inhibits Bma1l expression. The rate of PER and CRY degradation, mediated by CK1δ/ε and F-box/leucine rich-repeat protein 3, generates the 24-hour circadian period. Finally, transcriptional regulation of clock-controlled genes (CCGs) by the core clock establishes a circadian rhythm in a wide range of cellular functions. BMAL, brain and muscle ARNT-like 1; CK1, casein kinase 1; CLOCK, circadian locomotor output cycles kaput; CRY, cryptochrome; PER, period.

Organization of the mammalian circadian system

Although it is now well-established that the core clock is expressed in all nucleated cells of the body, it was originally described within the suprachiasmatic nuclei (SCN) of the hypothalamus (18). The SCN, which has come to be termed the “master clock,” is mainly regulated by light, which is the strongest known clock zeitgeber (ZT; German for “time giver”). Light entrainment of the SCN occurs when light is sensed by the melanopsin-containing retinal ganglion cells and the signal is relayed to the SCN via the retinohypothalamic tract (19) (Fig. 2). The SCN then sends out both humoral and nonhumoral signals to synchronize peripheral tissues (18). Coordination of peripheral tissue rhythmicity is thought to be carried out through innervation of both the sympathetic and parasympathetic arms of the autonomic nervous system by the SCN (18). Furthermore, hormonal, temperature, and behavioral outputs controlled by the SCN all are essential to maintaining circadian alignment (18). Another key pathway by which the master clock aligns peripheral clocks is through regulation of the activity/rest cycle, which largely determines the feeding/fasting cycle, downstream release of feeding-related hormones, and, subsequently, the production and disposition of metabolites (20).

Figure 2.

Circadian functions of gastrointestinal tract play a key role in metabolic control. Light, the strongest ZT, entrains the SCN which is the master clock. The SCN sends out signals through the ANS, as well as hormonal, temperature, and behavioral outputs to synchronize peripheral tissues. However, food intake can directly synchronize the GI tract through FEOs that are independent of the SCN. Circadian rhythms in key GI functions such as nutrient digestion and absorption, barrier integrity and immunity, as well as enteroendocrine hormone secretion all feed into establishing diurnal metabolic homeostasis. ANS, autonomic nervous system; FEO, food-entrainable oscillator; GI, gastrointestinal; SCN, suprachiasmatic nuclei; ZT, zeitgeber.

In addition to light as a ZT for the master clock, rodents that are fed once per day and housed under conditions of constant darkness exhibit increased activity, body temperature, and corticosterone levels before mealtime, suggesting the existence of food-entrainable oscillators (FEOs) (21). Intriguingly, the anticipatory behaviors in response to food persist in animals with lesioned SCN, indicating the presence of FEOs that can act independently of the light-entrained SCN (22). In line with evidence for the existence of peripheral FEOs, metabolic tissues such as the pancreatic islets, liver, adipose tissue, skeletal muscle, and GI tract have all been shown to have cell-autonomous rhythms that are largely entrained by nutrient intake (5-11). These metabolic tissues are responsible for coordinating the digestion, absorption, and utilization of nutrients throughout the 24-hour day; hence, they collectively form what has become known as the “peripheral metabolic clock” (23). Because the GI tract is the first point of nutrient entry into the body, its proper function throughout the 24-hour fasting/feeding cycle is therefore an essential determinant of circadian metabolic homeostasis.

Circadian Regulation of Gastrointestinal Function

Consisting of several integrated layers, including the muscularis and the submucosa and mucosa, the GI epithelium is juxtaposed to the intestinal lumen and therefore constitutes the first layer of contact with ingested nutrients. Comprised of multiple cell types, the epithelium is a key determinant of metabolic homeostasis through the digestive, absorptive, and defensive functions of the gut, as well as secreting more than 100 peptide hormones (24). Importantly, the intestinal epithelium has been shown to exhibit circadian clock gene expression, which is largely thought to be entrained by nutrient intake (10). Interestingly, vagal innervation of the gut does not seem to be a major ZT for the GI clock because vagotomy does not affect gastric and intestinal clock gene expression (10). It is currently unknown whether exercise, a known ZT for peripheral tissues such as skeletal muscle (1), affects the rhythmicity of the gut. However, the intestinal microbiome, has been shown to exhibit circadian fluctuations in both its composition and function (25, 26). Furthermore, subsequent work established the microbiota as a key player in regulating circadian rhythmicity within the intestinal epithelium (25-29). The role for the microbiome as a regulator of circadian metabolic homeostasis has recently been reviewed by Alvarez and colleagues (30). We discuss the circadian regulation of key GI epithelial functions that are essential to metabolic control in the following section.

Digestion and absorption

A critical function of the GI tract is to ensure adequate digestion and absorption of nutrients upon food intake. Transit time of nutrients is dependent on gut motility, which has been shown to be under the regulation of the circadian clock (31). Circadian Per1/2 mutant mice as well as humans with polymorphisms in the CLOCK and PER3 genes, exhibit disrupted gut motility resulting, most notably, in constipation/diarrhea (32, 33). Because many GI hormones are secreted in response to the presence of luminal nutrients, transit time plays an important role in their release and, ultimately, metabolic homeostasis. Importantly, digestion and absorption of all 3 macronutrients, carbohydrates, lipids, and proteins, is clock controlled to ensure synchrony with food availability. Thus, expression of the disaccharidase enzymes involved in carbohydrate digestion is increased during the active/feeding phase of the 24-hour day in anticipation of food intake (34). Similarly, enzymes involved in protein and lipid digestion display diurnal patterns peaking during the feeding period (34). Furthermore, subsequent nutrient absorption is under circadian regulation, driven by increased expression of hexose, peptide (for example, H(+)/peptide cotransporter 1), as well as triglyceride and cholesterol transporters (for example, apolipoprotein B and the intestinal microsomal triglyceride transfer protein) within the intestine during the feeding period (34-39). Interestingly, rhythmic expression of some of these transporters, such as the sodium-dependent glucose transporter 1 (40), may also have important implications for establishing rhythms in gut hormone secretion because glucose is a known secretagogue for the incretin hormones discussed in the following section. In accordance with these findings, disrupted feeding rhythms, such as increased food intake during the dark/inactive phase in individuals who conduct shift-work, results in disrupted postprandial responses in terms of hormone secretion and overall metabolic control (41).

Barrier integrity and immunity

Metabolic inflammation is characterized by low-grade systemic inflammatory responses and is strongly associated with metabolic dysfunction, including insulin resistance and dysglycemia (42). Because the GI tract is continuously exposed to dietary antigens, the maintenance of gut barrier integrity and the ability of the gut to mount adequate immune responses to luminal challenges both play important roles in the prevention of metabolic inflammation. Recent studies have shown that the tight junction proteins, occludin and claudin-1, exhibit circadian patterns of expression within the colon that are driven by rhythmic binding of the BMAL1/CLOCK heterodimer to their respective gene promoters (43). Importantly, the function of the tight junctions to provide a barrier to the luminal contents also demonstrates diurnal fluctuations, and enhanced intestinal permeability is observed in both Clock^Δ19/Δ19 and Per2 mutant mice, as well as in mice on a shifted light/dark cycle, thereby increasing translocation of luminal pro-inflammatory agents and predisposing them to metabolic disease (43, 44). Circadian mutant mice (Clock^Δ19/Δ19, as well as Bmal1-, Per1/2-, and Rev-Erbα knockout mice) have also been shown to exhibit increased susceptibility to dextran sulfate sodium-induced colitis and to demonstrate higher levels of inflammatory markers such as IL-6, interferon γ, and chemokine (C-X-C motif) ligand 2 in colonic biopsies (44-46). These changes are accompanied by reduced crypt cell proliferation and increased apoptosis within the intestinal epithelium. In line with this evidence, mice with an intestinal epithelial-specific knockout of RORα show increased intestinal inflammation and sensitivity to dextran sulfate sodium-induced colitis, accompanied by increased nuclear factor-κB expression in the proliferative crypt stem cells (47). In addition to genetic models of circadian disruption, mice exposed to constant light exhibit increased apoptosis in ileal crypts in response to administration of the cytokine, TNF-α (46).

Circadian disruption induced by phase shift and alternating light schedules has been shown to exacerbate colitis and increase the secretion of proinflammatory cytokines in mice (48, 49). Indeed, the circadian clock is also an important regulator of intestinal immune responses by establishing daily rhythms in immune cell generation, translocation, and function (50). More specifically, there is increased migration of leukocytes to the intestine at the onset of the active period as well as expression of innate immune system components, which confers increased sensitivity to pathogens (39, 51-53). Interestingly, time of day is an important determinant of gut susceptibility to infection, with salmonella administration to mice at ZT4 (early rest period), resulting in higher levels of bacterial colonization and pro-inflammatory gene expression as compared to exposure at ZT16 (early active period) (54). In accordance with this, animals have also been shown to mount greater immune responses at the onset of their active phase when exposed to inflammatory and infectious agents including lipopolysaccharide (LPS) (55, 56) and the listeria-causing bacteria, Listeria monocytogenes (57, 58). Interestingly, the intestinally derived endotoxin LPS, a membrane component of gram-negative bacteria, induces insulin resistance and diabetes following systemic exposure (59). Recent work has also identified LPS as a secretagogue for the incretin hormone glucagon-like peptide-1 (GLP-1; discussed in detail later in terms of its metabolic role), which can act locally to reduce intestinal inflammation through activation of the GLP-1 receptor (GLP-1R) on intraepithelial lymphocytes (60, 61). Further work has also implicated the intestinal immune system as a major determinant of GLP-1 bioavailability, as these cells can sequester secreted GLP-1, thereby preventing it from entering the circulation (62). Disruption of endogenous circadian rhythmicity thus results in dampened intestinal immune function, leading to increased inflammation. Given the recognized role of inflammation in the development of metabolic disease, this circadian-mediated immune disruption of the intestine may ultimately play a major role in metabolic disease development.

Enteroendocrine hormone secretion

Ghrelin.

Ghrelin is one of the few orexigenic gut hormones, mainly released by the gastric X/A cells during the preprandial phase and suppressed by food intake (63). Diurnal rhythms in ghrelin have been reported in both humans and rodent models, with peak levels occurring in the rest/fasting period (64-66), consistent with a reported circadian rhythmicity in hunger (67). Interestingly, ghrelin has also been shown to stimulate growth hormone release (68, 69). GH similarly exhibits a diurnal pattern of expression (70, 71), peaking during the rest/fasting period (72, 73), potentially implicating ghrelin in GH oscillations. Furthermore, murine ghrelin-containing oxyntic cells have been shown to rhythmically express the clock genes Per1/2, with peak expression occurring at ZT18 (middle of the active period in rodents) and which is antiphasic to the rhythm in ghrelin levels, which peaks at ZT6 (middle of rest period) (65). Although the rhythm in ghrelin secretion is lost in Per1/2^-/- mice (65), the exact mechanism by which the circadian clock regulates ghrelin release has yet to be elucidated. Interestingly, ghrelin can also modulate the circadian clock, restoring circadian gene expression via mammalian target of rapamycin signalling in a model of hepatic steatosis induced by high-fat diet feeding (74). Furthermore, ghrelin induces a phase shift in synchronized SCN explants (75), thereby suggesting a bidirectional relationship between the central and metabolic clocks in the timing of food intake.

Incretin hormones.

Although the essential role of insulin in maintaining metabolic homeostasis has been known for more than a century, the regulation of insulin secretion by pancreatic β cells continues to be a focus of investigation. Studies conducted in the first half of the 20th century determined that oral administration of glucose results in enhanced insulin release compared with that induced by an isoglycemic IV glucose load. The concept that intestinal factors released in response to oral glucose can promote glucose-dependent insulin release was termed the “incretin effect” and resulted in identification of the 2 key incretin hormones, glucose-dependent insulinotrophic polypeptide (GIP) and GLP-1, in 1973 and 1987, respectively (76, 77). Released into the circulation following nutrient intake, GIP and GLP-1 account for approximately 50% to 70% of the insulin response to nutrient ingestion (78). However, insulin secretion also exhibits a diurnal pattern, with greater secretion occurring in association with the active/feeding period in rodents and humans (79-81). Furthermore, previous studies have shown that the daily insulin secretory rhythm is more pronounced in response to oral rather than IV feeding, implicating a key role for the incretin hormones in entraining circadian insulin secretion (82).

Although more work has been done to study the circadian secretion of GLP-1, GIP secretion by the proximal K cell does exhibit diurnal fluctuations, with greater responses being observed at ZT16 in rats, immediately before their active period (83). Interestingly, changing the feeding-fasting schedule in these animals results in a parallel shift in peak GIP secretion (83), suggesting that nutrient intake is the main zeitgeber for the circadian rhythm in GIP. A study on obese subjects with either normal glucose tolerance or T2D, before and after biliopancreatic diversion surgery, also showed a rhythm in GIP release, with a reduced amplitude as well as a blunting of the rhythmic GIP response to weight loss in the subjects with diabetes, suggesting that metabolic status also affects circadian GIP secretion (84). However, of note, many studies to date interrogating circadian GIP release are confounded by uncontrolled duration of fasting periods and/or by feeding of meals that differ in caloric content over the day, making interpretation of circadian GIP secretory patterns difficult to interpret. Furthermore, functional expression of the core clock machinery in the GIP-producing K cell has not been demonstrated to date. Finally, although a role for circadian GIP in the overall regulation of diurnal metabolic homeostasis has not been reported, GIP receptor expression does not appear to be rhythmic in the pancreatic β cell (6, 85). However, 1 report has shown that GIP receptor expression oscillates in synchronized murine (m) GLUTag L cells that secrete GLP-1 in response to GIP, consistent with a role for GIP as a regulator of circadian secretion of GLP-1 (86). In summary, although some evidence indicates that GIP secretion follows a circadian pattern, further studies are required to elucidate both the regulation and the metabolic impact of these findings.

GLP-1 is released by enteroendocrine L cells, which are predominantly found in the distal small intestine and colon (87). The notion of GLP-1 secretion being under temporal regulation has been suggested by research in humans, which established that there is a greater GLP-1 response in the morning compared with the afternoon in subjects given identical meals although with different durations of fasting (88). More recent work identified differential GLP-1 secretion based on time of day, which was lost when subjects were exposed to nocturnal light (89). Interestingly, a 3-hour phase delay (induced by subjects following a 27-hour daily schedule with 9-hour sleep and 18-hour wake periods), but not a similar phase-advance (induced by subjects following a 21-hour daily schedule with 7-hour sleep and 14-hour wake periods), was also found to affect the GLP-1 secretory rhythm in humans (90). Carefully controlled studies conducted in rats as well as mice have now established that GLP-1 secretion follows a significant 24-hour secretory pattern in response to identical glucose loads administered after identical fasting periods (86, 91, 92). Interestingly, GLP-1 secretion was found to peak at the onset of the dark period/feeding period in rodents, with trough secretion at the onset of the light or fasting period (86, 91, 92), suggesting that nutrient intake is the major zeitgeber of the L cell. Consistent with this possibility, switching food availability to exclusively the light period results in inversion of the GLP-1 secretory rhythm, to peak at the onset of the light period in rats (91).

At the molecular level, using both human and murine L cell models, cell-autonomous circadian clock gene expression has been established (86, 91). Hence, human NCI-H716 and murine GLUTag L cells exhibit circadian expression of the genes for BMAL1, PER2, and REV-ERBα for 42 to 48 hours after cell synchronization (86, 89, 91). Furthermore, the pattern of GLP-1 secretion in response to known L cell secretagogues also demonstrates a significant circadian rhythm which parallels the rhythm in Bmal1 mRNA and protein expression (86, 91). Transcriptomic (microarray), proteomic (mass spectrometry), and chromatin immunoprecipitation analyses conducted in GLUTag L cells at the peak and trough of GLP-1 secretion subsequently identified several key proteins and pathways that may provide a mechanistic link between the core clock gene Bmal1 and circadian GLP-1 secretion (86, 91, 93). Most notably, and consistent with data on pancreatic β cells (6, 85), microarray analysis revealed significant enrichment at the peak of secretion in pathways related to the SNARE proteins (86), which are required for the exocytosis not only of GLP-1 (86, 93-96), but also insulin (5, 6, 97). Subsequent SNARE target knockdown studies established a key role for the SNARE machinery in regulating circadian GLP-1 release, including several proteins also identified in the β cells, secretagogin and Munc18-1; Munc18-1 was also demonstrated to play a role in vivo, using L-cell-specific knockout mice (86, 93). Similar analyses also identified protein tyrosine phosphatase 4a1, a regulator of Erk1/2 signaling, in the regulation of circadian GLP-1 secretion, in vitro (91).

Further studies conducted in rats at the normal peak and trough of GLP-1 secretion, demonstrated that this rhythm is not affected by the circadian pattern in corticosterone levels, although the amplitude is affected to a small extent (98). On the other hand, feeding rodents an obesogenic Western diet, a known circadian disruptor, results in enhanced GLP-1 secretion (92, 98). The idea that diurnal GLP-1 secretion is affected by obesogenic feeding is also supported by evidence from human studies that indicate that obese individuals lose their GLP-1 secretory rhythm (99). Furthermore, morbidly obese individuals with T2D have altered GLP-1 secretory rhythms as compared with those with normal glucose tolerance (84). To provide mechanistic insight into how metabolic dysfunction results in disrupted rhythmic GLP-1 release, murine GLUTag L cells were exposed to the saturated fatty acid, palmitate, a major component of the obesogenic Western diet (100). This led to disrupted clock gene expression, ultimately impairing mitochondrial function through a NAD⁺-dependent pathway and resulting in inadequate production of sufficient ATP for the stimulation of GLP-1 secretion (100).

The physiological relevance of circadian GLP-1 release has been demonstrated in rodents, whereby administration of identical doses of GLP-1 (in combination with an intravenous glucose load to permit the incretin effect) resulted in time-of-day-dependent insulin secretion, with a greater β-cell response observed at the normal peak time point of both GLP-1 and insulin release (92). Although GLP-1R is not expressed in a circadian manner in β cells, several key downstream mediators, including protein kinase A, phospholipase C, and protein kinase C, as well as key SNARE proteins, do exhibit a circadian rhythm, thereby providing a possible mechanism by which the β cell is more sensitive to GLP-1 at certain times of day (6, 85). Furthermore, the GLP-1R agonist, liraglutide, was found to synchronize circadian gene expression in β cells ex vivo (101). Together, these data suggest that circadian GLP-1 secretion may be important in entraining β-cell insulin secretion and thus overall diurnal metabolic homeostasis. However, further examination of circadian GLP-1 secretion is required to determine whether alterations in the pattern of GLP-1 release contribute to the disruption in circadian insulin secretion that has been observed in T2D (87).

Other gut hormones.

Oxyntomodulin, a satiety hormone that is cosynthesized with GLP-1 in the intestinal L cell, displays a circadian pattern of release with peak release in the middle of the dark/active period (102). Furthermore, diurnal meal-induced rhythms in peptide YY (PYY), another anorexigenic hormone secreted by the L cell, have been reported in humans, with peak levels occurring during the day (103). Furthermore, obese rats display disrupted rhythmic PYY release in association with decreased levels during the light period (104), in line with evidence that a high-fat diet causes circadian disruption (105). Interestingly, 24-hour PYY levels were also found to significantly correlate with resting metabolic rate, a major determinant of energy expenditure and overall metabolic homeostasis (103). Further studies are required to elucidate the role for circadian secretion of PYY in diurnal energy balance.

Circadian rhythms have also been demonstrated in neurotensin, a distal small intestinal N-cell hormone that plays a key role in systemic fat disposition (106). Measured in rats on either a normal light-dark cycle or under constant light, the oscillations in neurotensin persist regardless of time of day, suggesting that intestinal contents are the driver of the observed rhythms (107). However, rhythms in neurotensin were also reported in fasting animals, suggesting the existence of other zeitgebers. Although neurotensin expression has been detected in the GLP-1-expressing L cells (108), the rhythms in these hormones appear to be anti-phasic, with neurotensin peaking at the onset of the light (ZT24) and troughing at the onset of the dark (ZT9) periods (107). Collectively, these findings suggest the existence of a cell-autonomous clock in the N cell, although this remains to be confirmed.

Finally, secretin, a duodenal S-cell hormone that stimulates pancreatic exocrine secretion important for nutrient digestion, has recently been suggested to be involved in circadian control, with secretin-receptor knockout mice exhibiting both a shorter circadian phase and a shorter active period (109). Circadian rhythms have also been observed in plasma levels of gastrin, in the gastrin receptor in the stomach mucosa, and in the resultant intragastric pH (110-112). In addition, Clock^Δ19/Δ19 mutant mice exhibit decreased cholecystokinin levels, an enteroendocrine I cell hormone that is also important for nutrient digestion and absorption (113). CLOCK has also been demonstrated to interact with an E-box in the cholecystokinin promoter (113), further suggesting that this gastrointestinal hormone is under circadian control.

Curiously, expression of both secretin and cholecystokinin has been detected in intestinal L cells (114), raising the same question as for neurotensin with respect to the possible expression of cell-autonomous clocks in the S and I cells. Furthermore, although it is apparent that multiple enteroendocrine hormones demonstrate circadian patterns of secretion, it is currently unknown as to how peptides that are co-expressed (ie, GLP-1, PYY, neurotensin, secretin, and CCK) in L cells can demonstrate differential release patterns based on time-of-day. However, recent studies using single-cell analyses as well as lineage tracing approaches have shown that L cells differ in their patterns of gene expression along the length of both the aboral axis and the crypt-villus axis, suggesting the existence of distinct populations of L cells (114, 115). Elucidating the molecular pathways regulating the diurnal release of enteroendocrine hormones from individual cells will be required to establish how these complex secretory patterns are established.

Circadian Rhythms and Gastrointestinal Dysfunction

Importance of chronobiology in gastrointestinal inflammation

Increasing evidence has suggested that circadian disruption can result in a number of GI pathologies, impairing barrier function and thereby, at least potentially, affecting metabolic homeostasis. Several single nucleotide polymorphisms have been identified in clock genes that are associated with impaired gastric motility (33). Furthermore, single nucleotide polymorphisms in clock genes have also been associated with increased susceptibility to diseases of the GI tract such as inflammatory bowel disease, ulcerative colitis, and Crohn disease, which are all characterized by increased intestinal inflammation (116). A number of studies have also reported that biopsies of colonic mucosa from subjects with ulcerative colitis and Crohn disease demonstrate decreased expression of the core clock genes BMAL1, CLOCK, PER1/2/3, NR1D1, and CRY1/2 (48, 117, 118). In addition to the genetic components of the clock, environmental disruption as in shift workers has also been linked to GI disease (119-121). Along with shift work, irregular eating patterns have also been associated with intestinal inflammation in the form of irritable bowel syndrome (122). Together, this evidence suggests that circadian dysfunction caused by genetic or environmental perturbations results in pathologies related to intestinal inflammation.

Chronotherapy

Chronotherapy refers to the administration of medicine at a certain time of day, thereby taking advantage of circadian rhythms in physiology to maximize therapeutic response(s) and minimize adverse effects. In 1 example of the enteroendocrine system, the dual-incretin (GIP-GLP-1) receptor agonist DA-JC1 has been found to exhibit chronotherapeutic effects. In mice exhibiting β-amyloid-induced circadian disruption, DA-JC1 administration improved circadian rhythmicity by restoring disrupted hippocampal Per2 levels (123). Similarly, oxyntomodulin has been found to reset the liver clock by regulating the transcription of the core clock genes Per1/2 (102). Whether time-of-day administration of GLP-1 will be beneficial in the treatment of T2D remains to be determined; however, given recent interest in the development of L-cell secretagogues (124), there may be therapeutic advantages to enhancing the natural rhythm of GLP-1 secretion. Furthermore, current incretin-based therapies are administered irrespective of time of day and, given that rhythms in GLP-1 secretion drive diurnal patterns in insulin release (91), further research into the therapeutic implications of timed delivery to the maintenance of normal circadian hormone and glycemic profiles is warranted.

In addition to circadian pharmacological interventions, there is mounting evidence for the importance of consuming meals at certain times of day. This notion is based on the idea that mammalian metabolism is primed to intake nutrients during waking hours, resulting in coordinated nutrient digestion, absorption, hormone secretion, and nutrient deposition, thereby optimizing metabolic control. The 2 main approaches centered around timed food intake are intermittent fasting (ie, caloric restriction 2-3 days per week) and time-restricted feeding (ie, limiting nutrient intake to approximately 8 waking hours) (125). Recent meta-analyses found that both intermittent fasting and time-restricted feeding induce significant improvements in glycemic control, insulin resistance, and body weight (126, 127). Although beneficial changes have been found with both approaches, it must be acknowledged that at least some of the improvements in metabolic homeostasis may be a result of reduced caloric intake and subsequent weight loss. However, because key GI processes are involved in all aspects of nutrient handling, including GI motility, gastric emptying, intestinal permeability, and secretion of both digestive enzymes and enteroendocrine hormones, and have all been shown to be affected by time of day, these findings suggest that the beneficial effects of timed food intake may be, at least in part, dependent on the circadian functionality of the GI tract.

Conclusion

The importance of circadian rhythmicity in the maintenance of metabolic homeostasis has been well established, based on both animal models and human data. The GI tract is a key organ system essential to the regulation of metabolism through functions such as nutrient digestion and absorption, ensuring defense from foreign antigens which trigger inflammation, and enteroendocrine hormone secretion. The gut exhibits circadian rhythmicity in all of these parameters, which is likely driven by the pattern in food intake as established by the master clock in the SCN. Future studies on GI circadian rhythms will need to elucidate the relationship between changes in the GI clock and the progression of metabolic disease. Understanding whether a disrupted circadian clock within the gut is a cause or consequence of metabolic diseases such as obesity and T2D may ultimately lead to the development of novel chronotherapeutic strategies for these disorders.

Abbreviations

Abbreviations
 
• BMAL

  brain and muscle ARNT-like 1

 
• CK1

  casein kinase 1

 
• CLOCK

  circadian locomotor output cycles kaput

 
• CRY

  Cryptochrome

 
• FEO

  food-entrainable oscillator

 
• GI

  gastrointestinal

 
• GIP

  glucose-dependent insulinotrophic polypeptide

 
• GLP-1

  glucagon-like peptide-1

 
• GLP-1R

  glucagon-like peptide-1 receptor

 
• LPS

  lipopolysaccharide

 
• PER

  period

 
• PYY

  peptide YY

 
• RORα

  retinoic acid related-orphan receptor α

 
• SCN

  suprachiasmatic nuclei

 
• T2D

  type 2 diabetes

 
• ZT

  zeitgeber

Acknowledgments

Financial Support: S.E.M. was supported by graduate studentships from the Ontario Graduate Scholarship (OGS) and the Banting and Best Diabetes Centre (BBDC), University of Toronto; A.D.B. by an OGS graduate studentship; and P.L.B. by the Canada Research Chairs program. Studies in the Brubaker laboratory are supported by operating grants from the Canadian Institutes of Health Research (PJT-14853 and PJT-15308) and BBDC.

Additional Information

Disclosure Summary: P.L.B. is a consultant to VectivBio AG; A.M., S.E.M., and A.D.B. have no conflicts of interest to declare.

Data Availability

Data sharing is not applicable to this article as no datasets were generated or analyzed during the current study.