Diurnal redistribution of human lymphocytes and their temporal associations with salivary cortisol

Abstract

Immune cell trafficking is crucial for surveillance and effector functions of the immune system. Circadian rhythms of the hypothalamic-pituitary-adrenal (HPA) axis and of cortisol have been implicated in circadian redistribution of circulating lymphocytes and granulocytes. However, information regarding the diurnal redistribution of immune cells and their temporal correlations with cortisol is scarce. In this study, we investigated the diurnal redistribution of T, B, and natural killer (NK) cell subsets in relation to the endogenous cortisol rhythm. Saliva and blood samples were collected every 15 min over an 8-h period. Salivary-free cortisol was measured by enzyme-linked immunosorbent assay (ELISA). Surface markers (CD3, CD19, CD8, CD56, CD16, KIR) were measured in whole blood samples by 6-color flow cytometry and cell subsets quantified as a percentage of the total lymphocyte population. To study associations between the diurnal cortisol rhythm and the redistribution of T, B, and NK cells, we calculated cross-correlations with lag periods of 15 min. The salivary cortisol levels showed the typical diurnal variations with a significant morning cortisol awakening response (CAR) peaking around 07:30 h followed by an afternoon nadir. Whereas B cells remained stable throughout the 8 h, T cells (CD3 + CD8+ and CD3 + CD8−) showed a significant positive cross-correlation with cortisol levels when a delay of 30–105 min was taken into account. This was followed by a negative correlation covering a period of 165–285 min after the cortisol peak. Conversely, NK cells showed an initial negative correlation at 45–105 min, followed by a positive correlation at 120–285 min. The major CD56 + CD16+ subset and the CD56 − CD16+ population showed similar temporal correlation profiles. The minor CD56 + CD16− NK cell subset showed no temporal changes. The major NK subset (CD56 + CD16+) contains cells with higher cytolytic activity (KIR+) cells, whereas the single-positive subsets CD56 + CD16− and CD56 − CD16+ are mainly involved in cytokine production. Significant positive correlations were observed in KIR+ subsets coincident with this of NK cells covering a period of 105–300 min after the cortisol peak. In conclusion, our results suggest that cortisol and the HPA axis orchestrate tidal waves of immune cells that are alternatively based toward innate and acquired immune surveillance. (Author correspondence: [email protected])

• Cellular redistribution
• cortisol awakening rise
• cross-correlations
• cytolytic NK cells
• immunity

Acknowledgments

We would like to thank to Sophie Mériaux, Gerome Graas, Karine Mosselmans, and Thelma Velez for their technical assistance. We also acknowledge the collaboration of Wim Ammerlaan and Nicolaas H. C. Brons from the Flow Cytometry Unit and Manon Gantenbein from the Clinical and Epidemiological Investigation Centre (CIEC), Centre de Recherche Public de la Santé, Luxembourg.