HPA axis clock — suprachiasmatic pacemaker, ACTH-driven adrenal cortisol pulses, and how chronic exogenous glucocorticoid therapy suppresses the rhythm
Deep in the anterior hypothalamus, roughly 20,000 neurons form the suprachiasmatic nucleus (SCN) — the body's master circadian pacemaker. Through direct retinal input and an intricate molecular clock, the SCN imposes a ~24-hour rhythm on nearly every downstream physiological system, including the hypothalamic-pituitary-adrenal (HPA) axis that governs cortisol secretion.
Intrinsically photosensitive retinal ganglion cells (ipRGCs) containing melanopsin detect ambient light — independent of the rods and cones used for vision — and project directly to the SCN via the retinohypothalamic tract. This daily light signal resets an otherwise slightly-longer-than-24-hour intrinsic oscillator generated by the CLOCK/BMAL1 transcription-translation feedback loop, in which PER and CRY proteins accumulate, repress their own transcription, and are degraded on a ~24h cycle.
Each SCN neuron carries its own molecular clock, but gap junctions and neuropeptide signaling (VIP, AVP, GRP) synchronize the population into a coherent, robust rhythm. Ablation of the SCN in animal models abolishes circadian cortisol rhythmicity almost entirely, confirming it as the indispensable pacemaker.
The SCN does not directly control the adrenal gland. Instead it relays timing information through a multisynaptic pathway to the paraventricular nucleus (PVN) of the hypothalamus, where corticotropin-releasing hormone (CRH) neurons reside. This relay uses both direct GABAergic/glutamatergic projections and an indirect route through the dorsomedial hypothalamus and autonomic nervous system, including a polysynaptic pathway to the adrenal cortex itself that modulates its sensitivity to ACTH independent of the pituitary.
The net effect is a CRH neuron population whose excitability is gated by time of day: excitability rises through the night and peaks in the hours before habitual wake time, setting up the pre-dawn hormonal surge that follows.
Shift work and jet lag desynchronize the SCN's light-driven phase from the behavioral sleep/wake cycle, producing a blunted or phase-shifted cortisol rhythm that is associated with metabolic syndrome, impaired glucose tolerance, and increased cardiovascular risk over years of exposure.
Cortisol is not secreted at a constant level — it is one of the most tightly time-structured hormones in the body, changing more than three-fold between its morning peak and midnight trough. This rhythm anticipates the metabolic demands of the wake/sleep cycle: cortisol mobilizes glucose and primes cardiovascular tone before waking, then permits immune and reparative processes to dominate overnight when levels are lowest.
Because diagnostic cortisol testing (e.g. for Cushing syndrome or adrenal insufficiency) depends on comparing measured values to the expected time-of-day norm, understanding this rhythm is essential — a "normal" total cortisol value drawn at 4 PM would be frankly abnormal if drawn at 8 AM.
In the hours before habitual waking, hypothalamic CRH neurons begin an accelerating burst of activity that drives the anterior pituitary to release adrenocorticotropic hormone (ACTH). This surge — culminating in the well-characterized cortisol awakening response (CAR) — is one of the most reproducible neuroendocrine events in human physiology.
Corticotropin-releasing hormone, a 41-amino-acid peptide secreted from PVN neurons into the hypophyseal portal circulation, binds CRH receptor type 1 (CRHR1) on anterior pituitary corticotrophs. This triggers cleavage of the large precursor protein proopiomelanocortin (POMC) into ACTH (and several other peptides including β-endorphin), which is released into systemic circulation in discrete pulses.
Arginine vasopressin (AVP), co-secreted with CRH from PVN neurons especially during stress, potentiates CRH's effect on corticotrophs synergistically — the two peptides together produce a substantially larger ACTH pulse than either alone.
Within 30–45 minutes of waking, cortisol rises roughly 50% above the pre-wake trough — a distinct, superimposed spike on top of the underlying circadian rise that is triggered by the act of waking itself (light exposure, postural change, cognitive arousal) rather than by clock time alone. The CAR is measured clinically by sampling saliva at 0, 15, 30, and 45 minutes after waking.
A blunted CAR is associated with burnout, depression, and chronic fatigue, while an exaggerated CAR is linked to anticipatory anxiety about the day ahead. The CAR is superimposed on, but mechanistically distinct from, the broader diurnal cortisol rhythm — it persists even when subjects wake at unusual clock times, showing it is gated by the sleep-wake transition itself.
Cortisol awakening response magnitude is one of the most widely used non-invasive biomarkers of HPA axis reactivity in psychoneuroendocrinology research, sampled via at-home salivary swabs rather than venipuncture.
ACTH is not released continuously but in roughly 15–18 discrete pulses per 24 hours, with pulse amplitude — not frequency — carrying most of the circadian information. Pre-dawn pulses are large and closely spaced (every 60–90 minutes); afternoon and evening pulses become smaller and more widely spaced.
Each ACTH pulse reaches the adrenal cortex within roughly a minute via systemic circulation, and the adrenal cortisol response lags the ACTH pulse by about 10–15 minutes — the time required for cholesterol mobilization and steroidogenic enzyme flux to convert the ACTH signal into secreted steroid hormone.
ACTH reaching the adrenal cortex binds melanocortin-2 receptors (MC2R) on zona fasciculata cells, triggering a cAMP/PKA cascade that mobilizes cholesterol into mitochondria and drives the steroidogenic enzyme cascade converting cholesterol to cortisol. The result is a series of discrete secretory bursts whose amplitude — not just circulating level — tracks the time of day.
MC2R activation raises intracellular cAMP, activating protein kinase A (PKA), which phosphorylates steroidogenic acute regulatory protein (StAR). StAR shuttles cholesterol across the mitochondrial membrane to CYP11A1 (cholesterol side-chain cleavage enzyme), the rate-limiting step of all steroidogenesis. The resulting pregnenolone is sequentially converted by 3β-HSD, CYP17A1, and CYP21A2 to 11-deoxycortisol, then by CYP11B1 in the mitochondria to cortisol.
Because cortisol is not stored in significant quantity — unlike catecholamines in the adrenal medulla — nearly every molecule of cortisol released reflects synthesis triggered within the preceding 10-20 minutes, making adrenal cortisol output an almost real-time readout of pituitary ACTH pulses.
Cortisol's actions extend far beyond the classic "stress hormone" label:
• Metabolism: stimulates hepatic gluconeogenesis, promotes lipolysis in adipose tissue, and induces mild insulin resistance in skeletal muscle — mobilizing energy substrates ahead of the waking, active period • Immune modulation: at physiological circadian levels, cortisol shapes leukocyte trafficking (redistributing lymphocytes to marrow and skin overnight) and dampens pro-inflammatory cytokine transcription via glucocorticoid receptor binding to NF-κB • Cardiovascular tone: potentiates catecholamine vasoconstriction, contributing to the well-documented early-morning surge in blood pressure and cardiovascular events • Central nervous system: modulates memory consolidation, arousal, and mood via mineralocorticoid and glucocorticoid receptors in the hippocampus and amygdala
These effects are only appropriate at the correct time of day — the same cortisol level that primes morning wakefulness would be inappropriately catabolic and immunosuppressive if sustained continuously.
Roughly 90% of circulating cortisol is bound to corticosteroid-binding globulin (CBG) and albumin; only the free fraction (~5–10%) is biologically active, which is why free salivary cortisol is often preferred over total serum cortisol for circadian rhythm assessment.
Cortisol secretion is best described mathematically as a series of ultradian pulses (roughly hourly) whose amplitude is modulated by a slower circadian envelope. This two-timescale architecture means a single blood draw can be misleading — cortisol can vary two- to three-fold within a single hour depending on where in the pulse cycle the sample was taken, even at a fixed time of day.
This pulsatility is not noise — target tissues appear to respond differently to pulsatile versus continuous glucocorticoid receptor occupancy, with pulsatile exposure producing more efficient transcriptional cycling of glucocorticoid-responsive genes than sustained exposure at an equivalent average concentration.
As plasma cortisol rises through the morning, it acts back on its own regulators — the hypothalamus and pituitary — through glucocorticoid receptor (GR) binding that suppresses further CRH and ACTH release. This closed-loop feedback is the primary mechanism producing the steady decline of cortisol across the afternoon and evening toward the midnight nadir.
Glucocorticoid negative feedback operates on at least two timescales. Fast feedback occurs within minutes via non-genomic glucocorticoid receptor signaling at the hypothalamus and pituitary, rapidly damping CRH and ACTH pulse amplitude in proportion to the rate of rise of cortisol. Delayed (genomic) feedback occurs over hours, as GR activation alters transcription of POMC and CRH genes themselves, exerting a slower, more sustained brake on the axis.
Hippocampal and prefrontal cortical glucocorticoid and mineralocorticoid receptors add a third, higher-order layer of feedback, modulating hypothalamic drive based on cognitive and emotional appraisal of the environment — the neuroanatomical substrate through which chronic psychological stress can override or blunt the normal circadian pattern.
Because each morning cortisol pulse triggers proportional feedback suppression of the next CRH/ACTH pulse, the ultradian pulse amplitude — and therefore the circadian envelope — progressively shrinks across the day even though the SCN's excitatory drive to the PVN does not disappear until evening. By late afternoon, ACTH and cortisol pulses are small and infrequent; by the hours around midnight, both hormones reach their circadian nadir, and glucocorticoid receptor occupancy in the hypothalamus and pituitary falls low enough to permit CRH neuron excitability to begin building again for the next cycle.
This elegant self-limiting design — SCN-driven excitation opposed by cortisol-driven inhibition — produces a highly reproducible sinusoidal-like rhythm without requiring active "off" signaling.
Feedback integrity is central to differentiating causes of cortisol dysregulation. In primary adrenal insufficiency (Addison disease), the adrenal cortex cannot respond to ACTH, so ACTH rises unchecked as feedback inhibition is lost — hence markedly elevated ACTH with low cortisol. In secondary adrenal insufficiency (pituitary or hypothalamic failure, or exogenous suppression), both ACTH and cortisol are low because the drive itself is absent, not the adrenal's capacity to respond.
Dynamic testing — the ACTH stimulation test, low-dose dexamethasone suppression test, and CRH stimulation test — all exploit this feedback architecture, deliberately perturbing the loop and observing the compensatory response to localize where in the axis a defect lies.
Synthetic glucocorticoids such as prednisone and dexamethasone are among the most widely prescribed drugs in medicine — and among the most disruptive to the circadian HPA axis. By occupying the same glucocorticoid receptors used for endogenous negative feedback, chronic steroid therapy silences the hypothalamus and pituitary, causing the adrenal cortex to atrophy from disuse and creating real risk of adrenal crisis if the drug is stopped abruptly.
Synthetic glucocorticoids bind hypothalamic and pituitary glucocorticoid receptors just as endogenous cortisol does, but typically with far greater potency and duration of action. The hypothalamus and pituitary "read" the exogenous drug as if cortisol levels were persistently elevated, and respond by suppressing CRH and ACTH release — even though no endogenous cortisol is actually being produced.
Without ACTH stimulation, zona fasciculata cells lose trophic support: steroidogenic enzyme expression falls, cellular size shrinks, and with prolonged suppression the adrenal cortex undergoes measurable atrophy. This is not a drug side-effect on the adrenal directly — it is the predictable downstream consequence of a healthy, functioning feedback loop being deceived.
Suppression risk scales with dose, duration, potency, and timing of administration — evening or bedtime dosing suppresses the nocturnal rise in CRH drive more than the same dose taken in the morning, because it coincides with the period when the axis is naturally reactivating for the next day.
A patient with a suppressed HPA axis cannot mount an adequate endogenous cortisol response to physiological stress — surgery, infection, trauma — because both the signaling (CRH/ACTH) and the effector (atrophied adrenal cortex) are impaired. If exogenous steroid is withdrawn abruptly in this state, or not adequately increased during acute illness, the result can be adrenal crisis: hypotension, shock, hypoglycemia, and altered mental status that is fatal in up to 6% of unrecognized cases.
Risk rises sharply beyond roughly 3 weeks of therapy at doses above 5–7.5 mg/day prednisone-equivalent, though substantial inter-individual variability exists — some patients suppress within days at high doses, others tolerate months of low-dose therapy with preserved axis function.
Management centers on two complementary strategies. First, gradual tapering: rather than stopping abruptly, dose is reduced incrementally (commonly over weeks to months for long courses), allowing the hypothalamus and pituitary time to regain CRH/ACTH secretory capacity and the adrenal cortex time to regain responsiveness — recovery of the full axis can take anywhere from several weeks to as long as 6–12 months after prolonged high-dose therapy.
Second, stress dosing: patients with known or suspected suppression are instructed to temporarily increase glucocorticoid dose during acute illness, surgery, or trauma, mimicking the normal several-fold physiological cortisol surge that a healthy axis would produce automatically. Formal assessment of axis recovery uses the ACTH (cosyntropin) stimulation test — a robust cortisol rise after synthetic ACTH confirms adequate adrenal reserve before therapy is fully discontinued.
| Product | Indication | Trial Design | Key Result |
|---|---|---|---|
| Hydrocortisone | |||
| Prednisone / Prednisolone | |||
| Methylprednisolone | |||
| Dexamethasone |