The Functional and Clinical Significance of the 24-Hour Rhythm of Circulating Glucocorticoids

Abstract

Adrenal glucocorticoids are major modulators of multiple functions, including energy metabolism, stress responses, immunity, and cognition. The endogenous secretion of glucocorticoids is normally characterized by a prominent and robust circadian (around 24 hours) oscillation, with a daily peak around the time of the habitual sleep-wake transition and minimal levels in the evening and early part of the night. It has long been recognized that this 24-hour rhythm partly reflects the activity of a master circadian pacemaker located in the suprachiasmatic nucleus of the hypothalamus. In the past decade, secondary circadian clocks based on the same molecular machinery as the central master pacemaker were found in other brain areas as well as in most peripheral tissues, including the adrenal glands. Evidence is rapidly accumulating to indicate that misalignment between central and peripheral clocks has a host of adverse effects. The robust rhythm in circulating glucocorticoid levels has been recognized as a major internal synchronizer of the circadian system. The present review examines the scientific foundation of these novel advances and their implications for health and disease prevention and treatment.

Figure 1.

Schematic conceptual representation of the mammalian circadian system as it was understood circa 1990. A circadian pacemaker located in the SCN of the hypothalamus was widely thought to be the only self-sustained 24-hour clock and to act as master pacemaker for the entire organism. The molecular mechanism generating the self-sustained central circadian signal was not known. The light-dark cycle had been recognized as the main external synchronizer (zeitgeber) of the master pacemaker, but inputs from social cues and physical activity had also been identified. A few peripheral tissues—including the adrenals—had been found to generate damped oscillations that require input from the master pacemaker to be sustained (“slave” oscillators). Most peripheral tissues were considered not capable of generating self-sustained circadian oscillations and were thought to be passively entrained by the master pacemaker (“passive slaves”). The 24-hour rhythm of circulating GCs was seen as an “overt” rhythm reflecting central circadian timing, with inputs modifying its wave shape elicited by external and internal stimuli. [Redrawn and modified from E. Van Cauter and J. Aschoff: Endocrine and other biological rhythms. In: DeGroot LJ, ed. Endocrinology. Vol 3. WB Saunders; Philadelphia, PA: 1989; 2658–2705 (19), with permission.]

Figure 2.

Illustration of the mammalian circadian system as conceptualized circa 2010. The molecular mechanism generating a self-sustained circadian oscillation in SCN neurons is a complex transcriptional-translational feedback loop comprising core transcriptional activators BMAL1/CLOCK and two sets of repressors PER and CRY. The core transcriptional activators BMAL1/CLOCK regulate numerous genes, referred to as “clock-controlled genes” (CCGs).The same molecular machinery has been found in other central tissues as well as in nearly all peripheral tissues examined so far. The light-dark cycle is the main external synchronizer of the central circadian pacemaker (via the retino-hypothalamic tract [RHT]), but other external stimuli can affect the phase and amplitude of peripheral oscillators. In particular, the timing of food intake has been recognized as an important external synchronizer for circadian time-keeping in peripheral organs involved in the control of energy metabolism. The master clock in the SCN serves to synchronize central and peripheral oscillators to optimize the function of the organism relative to the 24-hour periodicities in the environment. Signals from peripheral tissues can affect the phase and amplitude of the central pacemaker. [Elements of this figure were published in M. Garaulet and J. A. Madrid: Chronobiological aspects of nutrition, metabolic syndrome and obesity. Adv Drug Deliv Rev. 2010;62(9–10):967–978 (20), with permission. © Elsevier.]

Figure 3.

Schematic representation of the pathways involved in the internal synchronization of the central suprachiasmatic (SCN) clock with the peripheral oscillators in the human circadian system. Both direct neural signals (transmitted by the autonomous nervous system, represented by black arrows) and indirect hormonal signals are involved. The 24-hour rhythms of circulating melatonin (released by the pineal gland, represented by a purple circle) and cortisol (released by the adrenals, represented by a blue circle) are considered as primarily controlled by the central SCN clock. In the schematic representations of the melatonin and GC profiles, the black bars represent the sleep/dark period. The blue and purple arrows symbolize, respectively, the synchronizing effects of the GC and melatonin rhythms. Because of the ubiquity of GRs in the entire organism, the 24-hour rhythm of circulating GCs plays a major role in synchronizing central and peripheral clocks. [Modified and redrawn from P. Pevet and E. Challet: Melatonin: both master clock output and internal time-giver in the circadian clocks network. J Physiol Paris. 2011;105(4–6):170–182 (99), with permission. © Elsevier.]

Figure 4.

Molecular feedback loops generating circadian rhythmicity in both central and peripheral tissues. The upper part of the figure illustrates the interactions with GR activity. The basic transcription factors CLOCK (or its analog NPAS2) and BMAL1 heterodimerize and initiate the main positive loop by activation of other clock genes, including three Period genes (Per 1–3) and two Cryptochrome (Cry 1–2) genes. PER and CRY proteins form complexes that translocate to the nucleus where they inhibit their own CLOCK/BMAL1-induced transactivation, defining a main negative loop. CLOCK(NPAS2)/BMAL1 dimers also drive transcription of nuclear receptors of the REV-ERB and ROR families, including Rev-erb α-β (Nr1d1–2) and Ror α-β-γ. In turn, REVERBs and RORs inhibit and activate, respectively, the rhythmic transcription of Bmal1 and Clock. Although GR activation can reset the phase of the clock by regulating Per expression and REVERB activity, the clock machinery modulates GR activity at transcriptional and post-translational levels in multiple tissues, thus gating the regulation of GC target genes in a tissue-specific fashion.

Figure 5.

Representative 24-hour profiles of plasma ACTH and cortisol levels sampled at 15-minute intervals in a healthy young man studied under normal conditions. During scheduled sleep times (shown as a black bar), the sampling catheter was connected to plastic tubing extending into the adjacent room to avoid disturbing the subject. The parallelism of both pulsatile and circadian variations of ACTH and cortisol concentrations is evident.

Figure 6.

Simultaneous profiles of plasma total cortisol (solid line) and saliva free cortisol (dashed line) in two healthy young adults who were each submitted to blood sampling via an indwelling catheter at 20-minute intervals from 9 AM to midnight and provided a saliva sample at the time of each blood sampling. Caloric intake was exclusively in the form of a glucose infusion at a constant rate of 5 g/kg/24 h. Simultaneous levels of plasma cortisol and saliva cortisol were highly correlated in both individuals (r: Pearson correlation coefficient). [Unpublished illustration of data included in A. Guyon et al: Adverse effects of two nights of sleep restriction on the hypothalamic-pituitary-adrenal axis in healthy men. J Clin Endocrinol Metab. 2014;99(8):2861–2868 (117), with permission. © The Endocrine Society.]

Figure 7.

Schematic representation of the control of the circadian rhythmicity of GC release in mammals. Most components of the HPA axis contain circadian oscillators. The circadian secretion of GC is dependent on the rhythmic release of ACTH and a gating process by the adrenal clock. GC secretion is also modulated by nervous signals coming from the PVN of the hypothalamus via sympathetic nervous pathways. ACTH release is controlled by the rhythmic release of CRH and vasopressin from the PVN. Rhythmic activity of the HPA axis is under the control of the master clock in the SCN, reset by ambient light via the retina. The peak of vasopressin release from the SCN to the PVN region occurs during daytime in both nocturnal and diurnal rodents. In nocturnal rats, vasopressin exerts an inhibitory action on the PVN (probably via activation of γ-aminobutyric acid-containing interneurons), thus reducing GC secretion during daytime (green curve). By contrast, in diurnal grass rats, vasopressin stimulates PVN activity (probably via activation of glutamatergic interneurons), thus increasing GC secretion during daytime (red curve). Clock symbols represent self-sustained oscillators.

Figure 8.

Mean profiles of plasma cortisol concentrations measured at 20-minute intervals from eight healthy young men studied over a 53-hour period including a night of nocturnal sleep (black bar), 28 hours of continuous wakefulness in a semirecumbent position with a night of total sleep deprivation (red bar), and an 8-hour period of daytime recovery sleep (blue bar). Caloric intake was exclusively in the form of an iv glucose infusion at a constant rate. The persistence of the circadian rhythmicity of plasma cortisol despite these drastic manipulations of the light-dark cycle, sleep-wake cycle, and feeding schedule is evident. Note that the absence of wake-sleep and sleep-wake transitions during the night of total sleep deprivation results in a slight dampening of the amplitude of the circadian variation. [Redrawn from E. Van Cauter et al: Modulation of glucose regulation and insulin secretion by circadian rhythmicity and sleep. J Clin Invest. 1991;88(3):934–942 (144), with permission. © American Society for Clinical Investigation.]

Figure 9.

Mean profiles of plasma cortisol concentrations measured at 20-minute intervals from eight healthy young men who each participated in a randomized clinical trial comparing the impact of either two or three identical carbohydrate-rich meals presented at fixed clock times. A clear short-term increase in cortisol concentrations occurs after each meal but does not affect the overall wave shape of the circadian profile. Thus, acute changes in feeding schedules that have a major effect on peripheral circadian oscillators in metabolic tissues do not readily desynchronize the GC rhythm from the central circadian pacemaker, consistent with its role as a robust internal synchronizing signal. [Adapted from E. Van Cauter et al: Circadian modulation of glucose and insulin responses to meals: relationship to cortisol rhythm. Am J Physiol. 1992;262(4 Pt 1):E467–E475 (146), with permission. © American Physiological Society.]

Figure 10.

Mean (+SEM) profiles of plasma cortisol (top panels), insulin secretion rates (ISR, middle panels), and plasma glucose (lower panels) in nine healthy young men each of whom participated in four studies performed in randomized order. In all four studies, endogenous cortisol levels were suppressed by metyrapone administration, and caloric intake was exclusively in the form of a constant glucose infusion. Dark horizontal bars represent the scheduled sleep periods. The daily cortisol elevation was restored by oral administration of hydrocortisone (or placebo—data not illustrated) either at the normal time of the circadian peak (5 am, left panels) or 12 hours out of phase (5 pm, right panels). Vertical arrows show the timing of hydrocortisone administration in each study. Horizontal lines on the ISR and glucose graphs show, respectively, the mean ISR and glucose levels at the time of hydrocortisone ingestion to facilitate the visualization of post-hydrocortisone changes. The initial effect of the hydrocortisone-induced cortisol pulse was a short-term inhibition of insulin secretion without concomitant glucose changes, and the magnitude of this acute effect was similar in the evening and in the morning. At both times of day, starting 4–6 hours after hydrocortisone administration, there was a delayed hyperglycemic effect that was minimal in the morning but much more pronounced in the evening, when it was associated with a robust increase in insulin secretion. [Redrawn from L. Plat et al: Metabolic effects of short-term elevations of plasma cortisol are more pronounced in the evening than in the morning. J Clin Endocrinol Metab. 1999;84(9):3082–3092 (279), with permission. © The Endocrine Society.]

Figure 11.

Association between the misalignment of the cortisol profile and reduced glucose tolerance despite increased insulin levels. Mean (+SEM) 28-hour profiles of plasma cortisol (top), plasma glucose (lower left) and serum insulin (lower right) in 10 healthy adults who participated in a 10-day laboratory “forced desynchrony” protocol, where sleep (shown by the large vertical gray shaded areas) and meals (shown by the small vertical gray shaded areas) were scheduled on a recurring 28-hour cycle, a periodicity out of the range of entrainment of the central circadian pacemaker. Subjects ate four meals (designated as breakfast, B; lunch, L; dinner, D; and snack, S) during each 28-hour “day.” Curves shown in black represent the profiles observed when the sleep-wake and meal schedules were aligned with the 24-hour cycle, whereas curves shown in red represent the profiles observed when the 28-hour “day” was 12 hours out of phase with the normal 24-hour day. Cortisol profiles were inverted during circadian misalignment, demonstrating no adaptation to the 28-hour schedule (as expected from a rhythm controlled by the circadian pacemaker). The postprandial glucose and insulin responses were significantly increased as compared to normal alignment, and the differences were also clinically significant. [Redrawn from F. A. Scheer et al: Adverse metabolic and cardiovascular consequences of circadian misalignment. Proc Natl Acad Sci USA. 2009;106(11):4453–4458 (23), with permission. © National Academy of Sciences.]

Figure 12.

Impact of age on the 24-hour profile of plasma cortisol in healthy nonobese men. Data are shown at each time point as mean + SEM. The two age groups included eight men that were matched for body mass index. Note the higher nadir level, the reduced amplitude of the overall rhythm, and the earlier onset of the early morning rise in older participants. The overall disruption of the 24-hour cortisol profile is likely to contribute to and/or exacerbate age-related metabolic, immune, and cognitive deficits. [Redrawn from E. Van Cauter et al: Age-related changes in slow wave sleep and REM sleep and relationship with growth hormone and cortisol levels in healthy men. JAMA. 2000;284(7):861–868 (313), with permission. © American Medical Association.]

Figure 13.

Comparison of the 24-hour profile of plasma cortisol in normal nonobese adults (left), patients with pituitary-dependent Cushing’s disease (center), and patients with untreated major depression of the unipolar subtype (right). For each condition, a representative example is shown in the top panel and mean (+ SEM) profiles from eight to 10 subjects are shown in the bottom panel. [Redrawn from E. Van Cauter: Physiology and pathology of circadian rhythms. In: Edwards CW, Lincoln DW, eds. Recent Advances in Endocrinology and Metabolism. Vol. 3. Churchill Livingstone, Edinburgh, UK, 1989:109–134 (435), with permission.]

Figure 14.

A, Comparison between 24-hour plasma cortisol profile typically achieved with optimal hydrocortisone replacement (10 mg upon awakening, 5 mg at lunch, and 5 mg at 7 pm [dinner]) in patients with adrenal insufficiency (dotted red lines) and normal cortisol levels from healthy young adults (shaded area) (data source, Refs. 313 and 416). B, Profile of circulating cortisol levels achieved by a single early morning administration of a modified-release oral hydrocortisone preparation (solid line) as compared to the profile resulting from immediate release hydrocortisone administered three times daily (dashed line). [Redrawn from G. Johannsson, et al: Improved cortisol exposure-time profile and outcome in patients with adrenal insufficiency: a prospective randomized trial of a novel hydrocortisone dual-release formulation. J Clin Endocrinol Metab. 2012;97(2):473—481 (416), with permission. © The Endocrine Society.] C, Example of a plasma cortisol profile obtained in a healthy subject in whom endogenous cortisol levels were suppressed by 5 days of metyrapone administration and replaced by sc hydrocortisone replacement via an infusion pump programmed to mimic circadian and pulsatile variations. [Adapted from G. M. Russell, et al: Subcutaneous pulsatile glucocorticoid replacement therapy. Clin Endocrinol (Oxf). 2014;81(2):289–293 (422), with permission. © Blackwell Scientific Publications.]