Cell-free DNA release under psychosocial and physical stress conditions

Abstract

The understanding of mechanisms linking psychological stress to disease risk depend on reliable stress biomarkers. Circulating cell-free DNA (cfDNA) has emerged as a potential biomarker of cellular stress, aging, inflammatory processes, and cell death. Recent studies indicated that psychosocial stress and physical exercise might also influence its release. We compared the effects of acute psychosocial and physical exercise stress on cfDNA release by exposing 20 young, healthy men to both an acute psychosocial laboratory stressor and an acute physical exercise stressor. Venous blood and saliva samples were collected before and after stress exposure. Cell-free DNA was extracted from plasma and quantified by qPCR. Furthermore, cfDNA fragment length was analyzed and cfDNA methylation patterns were assayed across time. In addition, release of stress hormones and subjective stress responses were measured. Results showed a twofold increase of cfDNA after TSST and fivefold increase after exhaustive treadmill exercise, with an overabundance of shorter cfDNA fragments after physical exhaustion. Interestingly, cell-free mitochondrial DNA showed similar increase after both stress paradigms. Furthermore, cfDNA methylation signatures-used here as a marker for diverse cellular origin-were significantly different post stress tests. While DNA methylation decreased immediately after psychosocial stress, it increased after physical stress, suggesting different cellular sources of active DNA release. In summary, our results suggest stimulus and cell-specific regulation of cfDNA release. Whereas the functional role of stress-associated cfDNA release remains elusive, it might serve as a valuable biomarker in molecular stress research as a part of the psychophysiological stress response.

Fig. 1. cfDNA (concentration, methylation, fragment length) and cf-mtDNA concentrations before and after psychosocial and physical stress are shown.

a Changes in cfDNA concentrations (copies/ml plasma) before and after psychosocial and physical stress. Immediately after the TSST, doubling of cfDNA concentration was observed, whereas it increased fivefold after physical stress, peaking 15 min after the cessation of exercise. b Changes in the percentage proportion of the 170 bp cfDNA fragments of purified cfDNA. While minimal increases of the 170 bp fragments could be observed after the TSST, a fivefold percentage increase 15 min after physical exercise was observed. c Percentage cfDNA methylation of a HOXA5 fragment is shown as the average of 9 CpG sites. DNA methylation increased by 7.5% after physical stress, while it significantly decreased by 6.5% after the TSST. d An almost twofold increase of cf-mtDNA directly after both stress conditions is shown. Post-hoc test showed significant difference between the stress paradigms at time point +15 min. Values are reported as means ± SEM. The data were analyzed with repeated measures ANOVAs. In case of significant effects, post-hoc test was carried out to check for group differences (*p ≤ 0.05, ***p ≤ 0.001)

Fig. 2. Hormone levels before and after psychosocial and physical stress conditions.

a–d Progression of stress hormones in plasma and saliva before and after psychosocial and physical stress are shown. a Plasma cortisol is plotted in nmol/l over time. Both stress conditions led to an increase in plasma cortisol with the highest values measured 15 min after cessation of psychosocial and physical stress. Plasma cortisol increased by 107 nmol/l after psychosocial stress and by more than 210 nmol/l after physical exhaustion (main effect time: F_{(1.96, 74.48)}= 25.26, p < 0.001, η² = 0.399; main effect condition: F_(1, 38)= 4.57, p = 0.039, η² = 0.107; time × condition interaction effect: F_{(1.96, 74.48)}= 13.50, p < 0.001, η² = 0.262). b Increased salivary cortisol (in nmol/l) in psychosocial and physical stress conditions. Cortisol levels after psychosocial stress peaked after 15 min, whereas they increased until 30 min after physical stress (main effect time: F_{(1.83, 69.58)}= 17.47, p < 0.001, η² = 0.315; main effect condition: F_(1, 38)= 2.20, p = 0.146, η² = 0.055; time × condition interaction effect: F_{(1.83, 69.58)}= 11.10, p < 0.001, η² = 0.226). Psychosocial and physical stress led to an increase of noradrenaline (c) and adrenaline (d). Immediately after psychosocial stress, the concentration of noradrenaline and adrenaline doubled before dropping back to baseline levels within 15 min. Physical stress led to an 18-fold increase of noradrenaline (c) and adrenaline (d) (noradrenaline: main effect time: F_{(2.21, 83.89)}= 226.57, p < 0.001, η² = 0.856; main effect condition: F_(1, 38)= 152.10, p < 0.001, η² = 0.800; time × condition interaction effect: F_{(2.21, 83.89)}= 104.94, p < 0.001, η² = 0.734; adrenaline: main effect time: F_{(2.21, 73.04) =}103.06, p < 0.001, η² = 0.757; main effect condition: F_(1, 33)= 48.54, p < 0.001, η² = 0.595; time × condition interaction effect: F_{(2.21, 73.04)}= 55.80, p < 0.001, η² = 0.628). Values are reported as means ± SEM. The data were analyzed with repeated measures ANOVAs. In case of significant effects, post-hoc test was carried out to check for group differences (*p ≤ 0.05, **p ≤ 0.01 ***p ≤ 0.001)

Fig. 3

a A significant positive correlation between salivary cortisol increases and cfDNA increases (peak minus baseline) was observed in the physical stress condition. The association between increases in salivary cortisol and cfDNA levels after the TSST followed the same direction but was not statistically significant. b A trend toward significance for a positive correlation between plasma cortisol and cfDNA increase after physical stress and a similar, non-significant relationship for the TSST was observed