Glucocorticoids (GCs) are a class of hormones which activate signal transduction pathways leading to the transcription of a broad range of genes, including those conferring anti-inflammatory effects. In humans, the GC cortisol is produced in the adrenal gland and secreted into systemic circulation in a series of discrete bursts. This discrete release pattern produces significant ultradian rhythms (oscillatory period ~1hr) in plasma cortisol concentration, and circadian variability in the size of the bursts leads to the characteristic circadian rhythms in cortisol levels. It has recently been observed that cells exposed to pulsatile levels of GCs develop a distinct phenotype relative to cells given constant cortisol treatment, even when the total GC exposure is equal (McMaster, Jangani et al. 2011). We applied computational systems biology techniques towards understanding both the mechanistic origins of this observed differential expression and translational implications of the importance of GC pulsatility in vivo.
To mathematically investigate the effect of ultradian GC pulsatility on transcription of GC-responsive genes, we combined a model of rhythmic GC production (Walker, Terry et al. 2010) with a pharmacodynamic model of GC action (Yao, DuBois et al. 2006). The hypothalamic-pituitary-adrenal (HPA) axis is modeled so that the negative feedback loop between GC secretion and adrenocorticotropic hormone (ACTH) production produces oscillations in these hormones. GCs are released from the adrenal gland into circulation. The glucocorticoid receptor (GR) is a cytosolic receptor, so it interacts with GCs that have crossed the cell membrane into the cytoplasm. Then, the activated GR translocates into the nucleus where it regulates gene transcription. This combined model consists of 7 equations and 14 parameters.
The nonlinear binding kinetics between GC and GR can fundamentally drive differential expression between the constant and pulsatile simulations shown in Figure 1. In other words, the peak value of the ultradian rhythm is in the region where GR is becoming saturated and the system gets diminishing returns for increasing GC levels. This type of homeostatic behavior is important in maintaining the responsiveness of the HPA axis by allowing for the rapid production of high levels of GCs and the responsiveness of the GC signaling pathway in peripheral cell, as a high constant level of GC exposure typically leads to tolerance.
The model of ultradian GC production allows for the assessment of the effects of dysregulated HPA axis function. In particular, we explore the implications of altering the feedback loop from GC to ACTH, which is ultimately the driving force behind the oscillatory hormone release. This is done in two different ways, by perturbing (1) the timing of signal transduction from GC to ACTH and (2) the strength of feedback from GC to ACTH.
When the time delay for the effect of GC on ACTH production is decreased, the frequency of ultradian rhythms increases. This means that there is a shorter time window for both the secretion and elimination of GCs, culminating in blunted rhythms. GC levels to rise and fall, which results in lower amplitude ultradian oscillations. (Veldhuis, Iranmanesh et al. 1989) reports that there is a positive relationship between the amount of GC released in each burst and the time between bursts. When the time delay is left constant and the parameter representing the strength of feedback from GC to ACTH is decreased, the ultradian amplitude is lowered without a change in frequency. In both cases, a flatter, chronically elevated hormone profile as the time delay is decreased leads to the loss of the downstream responses to pulsatility described above.
The work presented here has important translational implications. As exogenous GC therapy is applied clinically without considering pulsatility, this modeling work allows for comparing modes of therapeutic delivery of GCs. Under stress, endogenous rhythmicity in hormone release is often diminished or lost, and the downstream transcriptional implications of these changes can be assessed by this model.
Figure 1: Pulsatile production of glucocorticoids (GC) arises from the HPA axis through interactions between ACTH and glucocorticoid receptor (GR) binding in the pituitary, culminating in the release of adrenal GC. Then, the signal propagates from intracellular GC (GC) concentration, to the activated GC-GR complex (DR), and translocation of the complex into the nucleus (DR1) where it regulates the transcription of GC-responsive genes (mRNA). Three simulations are shown: pulsatile (black), constant with the same exposure as pulsatile (red), and constant with the maximum concentration of pulsatile (blue). Gene pulsing is observed in response to ultradian rhythms, and both of the constant GC simulations produce higher levels of gene transcripts.
- McMaster, A., M. Jangani, et al. (2011). "Ultradian cortisol pulsatility encodes a distinct, biologically important signal." PLoS One 6(1): e15766.
- Veldhuis, J. D., A. Iranmanesh, et al. (1989). "Amplitude modulation of a burstlike mode of cortisol secretion subserves the circadian glucocorticoid rhythm." Am J Physiol 257(1 Pt 1): E6-14.
- Walker, J. J., J. R. Terry, et al. (2010). "Origin of ultradian pulsatility in the hypothalamic-pituitary-adrenal axis." Proc Biol Sci 277(1688): 1627-33.
- Yao, Z., D. C. DuBois, et al. (2006). "Modeling circadian rhythms of glucocorticoid receptor and glutamine synthetase expression in rat skeletal muscle." Pharm Res 23(4): 670-9.
See more of this Group/Topical: Food, Pharmaceutical & Bioengineering Division