Timescales of Human Hair Cortisol Dynamics

Abstract

Cortisol is a major human stress hormone, secreted within minutes of acute stress. Cortisol also has slower patterns of variation: a strong circadian rhythm and a seasonal rhythm. However, longitudinal cortisol dynamics in healthy individuals over timescales of months has rarely been studied. Here, we measured longitudinal cortisol in 55 healthy participants using 12 cm of hair, which provides a retrospective measurement over one year. Individuals showed (non-seasonal) fluctuations averaging about 22% around their baseline. Fourier analysis reveals dominant slow frequencies with periods of months to a year. These frequencies can be explained by a mathematical model of the hormonal cascade that controls cortisol, the HPA axis, when including the slow timescales of tissue turnover of the glands. Measuring these dynamics is important for understanding disorders in which cortisol secretion is impaired over months, such as mood disorders, and to test models of cortisol feedback control.

Keywords: Biological Sciences; Chronobiology; Human Metabolism.

Graphical abstract

Figure 1

Cortisol Secretion Is Governed by the HPA Axis (A) Top panel: cortisol shows circadian changes over the day, adapted from Chan and Debono, (2010). Bottom panel: varying stress inputs can cause fluctuations over months, whose timescale is the subject of this study. (B) Classic model of the HPA axis in which stress inputs cause corticotropin-releasing hormone (CRH) secretion from the hypothalamus, causing the pituitary corticotrophs to secrete adrenocorticoptropic hormone (ACTH), which in turn causes cells in the adrenal cortex to secrete cortisol. Cortisol inhibits the secretion of the upstream hormones. (C) The Karin et al. model, which includes the effect of hormones on gland proliferation (red arrows), which introduces the slow timescale of tissue turnover (weeks-months). (D) The adrenal cortex and pituitary corticotrophs effectively form a negative feedback loop when considering the slow timescale of months. (E) An increase in pituitary corticotroph cells' total mass leads to increased ACTH secretion, which increases adrenal cortex mass (1). This leads to an increase in cortisol, which inhibits ACTH, causing a reduction in corticotroph cells (2). This feedback loop with cell turnover times of few weeks has a resonance frequency with an overall timescale of a year. (F) Schematic overview of this study: longitudinal hair cortisol over 1 year is analyzed using Fourier transform to detect frequencies of fluctuations.

Figure 2

Longitudinal Cortisol Measurements from Human Hair (A) Cortisol time series from 55 participants. Each time series has 6 points corresponding to six 2-cm segments of hair, representing about 1 year of growth. (B) Normalized cortisol for the same participants, after correction for decline along the hair. Highlighted in color are three examples of individual cortisol time series.

Figure 3

Hair Cortisol Shows Fluctuations with a Dominant Period of 1 Year Fourier amplitudes averaged over participants quantify the contribution of each frequency component (1[year⁻¹], 2[year⁻¹], and 3[year⁻¹]) to the cortisol signal (black dots). Error bars (SEM) were calculated by bootstrapping the participants. Shuffled control is shown as gray dots (1,000 repeats), with 97.5% and 2.5% confidence intervals shown in dashed gray lines. The amplitude of the 1 [year⁻¹] frequency is significantly higher than shuffled control, p = 0.004.

Figure 4

HPA Model with Gland Mass Dynamics Is Sufficient to Explain the Year-Scale Fluctuations in Cortisol (A) Textbook HPA model includes a cascade of hormones with negative feedback of cortisol on its upstream hormones (Figure 1B). The Fourier spectrum in response to white noise input is similar to the null model. (B) Karin et al. model with gland mass dynamics (Figure 1C) shows a Fourier spectrum with a dominant [year⁻¹] frequency component. The black dots are the mean Fourier amplitude with SEM error bars. Dashed lines: 2.5% and 97.5% confidence intervals of a shuffled data control. Both simulations and null models were subjected to the decline correction.