Identification of human glucocorticoid response markers using integrated multi-omic analysis [Adipose Tissue]

A multi-omic approach in a clinical experimental study identified circulating biomarkers reflecting glucocorticoid exposure. Background: Endogenous glucocorticoids (GC) are mechanistically linked to common diseases and are important as drugs in the treatment of many disorders. There is no marker that can measure and quantify GC action. Our aim was to identify circulating biomarkers of GC action using a clinical experimental study. Methods: In a randomized, crossover, single-blind trial, subjects with primary adrenal insufficiency received intravenous hydrocortisone infusion in a circadian pattern (physiological GC exposure) or isotonic saline (GC withdrawal) over 22 hours. Samples were collected at 7 AM (end of infusion). Integrated multi-omic analysis was used because of the complexity in GC action and the low number of subjects. The transcriptome in peripheral blood mononuclear cells (PBMCs) and adipose tissue, plasma miRNAomic, and serum metabolomics were compared between the interventions. Replication of the plasma miRNA findings was performed in three independent studies. Results: During GC withdrawal, overnight urinary cortisol and cortisone excretion were undetectable. Correlation and hypernetwork analyses identified a transcriptomic profile derived from PBMCs and adipose tissue predictive of GC exposure, and a multi-omic cluster predictive of GC exposure. From the circulating ‘omic data, decreased expression of plasma miR-122-5p was associated with increased GC exposure. This finding was reproduced in three independent studies. Conclusion: We developed a human experimental model for physiological GC exposure and withdrawal. The integrated multi-omic data identified circulating miRNAs and metabolites associated with GC-responsive genes. In independent studies, miR-122-5p was shown to be associated with GC exposure. Background: Endogenous glucocorticoids (GC) are mechanistically linked to common diseases and are important as drugs in the treatment of many disorders. There is no marker that can measure and quantify GC action. Our aim was to identify circulating biomarkers of GC action using a clinical experimental study. Ten subjects (4 women with 3 of them post-menopausal) with a median age of 50 years (range, 25 to 57) and a median disease duration of 23.5 years (range, 1 to 33) completed all aspects of the study. The median daily replacement dose of hydrocortisone (HC) prior to the study was 30 mg (range, 20 to 30) and 9 out of 10 subjects had treatment with fludrocortisone (mineralocorticoid) at a median daily dose of 0.1 mg (range, 0.1 to 0.2). The main time points for sample collection in each intervention were at 9 AM on the first intervention day (“before start”) and at 7 AM on the second intervention day (“morning”) (Fig. 1B). The subjects’ last ordinary oral HC dose was administered the day before admission to the study unit. Infusion of HC mixed with isotonic saline (“GC exposure”) had no effect on systolic and diastolic blood pressure, body weight, serum sodium and potassium, or plasma glucose concentrations compared to the same amount of isotonic saline infusion alone (“GC withdrawal”).Differential expression was conducted on the ‘omic data from the samples collected at 7 AM comparing physiological GC exposure to GC withdrawal