FKBP5 Methylation As a Long-Term Marker for Cortisol Exposure in Healthy Subjects

Presentation Number: SAT 399
Date of Presentation: April 1st, 2017

Britta Winkler*1, Hendrik Lehnert2, Henrik Oster3 and Birgit Harbeck3
1University of Luebeck, Lübeck, GERMANY, 2University of Luebeck, Luebeck, Germany, 3University of Lübeck, Lübeck, Germany



Currently employed hydrocortisone replacement regimens for patients with adrenal failure insufficiently mimic the circadian rhythm of cortisol secretion, resulting in temporary hypo- and hypercortisolism. Robust biomarkers measuring cortisol exposure over weeks or months are lacking.

FKBP5 is a co-chaperone of hsp90 which regulates glucocorticoid receptor sensitivity. Previous studies on mice showed that chronic exposure to glucocorticoids (GC) results in loss of DNA methylation in the introns 1 and 5 of the FKBP5 gene in blood and other tissues. Four weeks of GC exposure reduce FKBP5 methylation in a dose-dependent manner. Moreover, demethylation of the FKBP5promoter was recently shown in human trabecular cells after dexamethasone treatment.

The aim of this pilot study was to evaluate the relevance of CpG methylation within the FKBP5gene as a possible long-term biomarker of cortisol exposure.

Materials and Methods

Twenty-one healthy adults (6 females, 15 males, median age: 24 years) were recruited. A high dose (250 mcg) ACTH stimulation test was performed to evaluate adrenal function in all subjects. Serum cortisol was measured prior to and 1h following the stimulation. In addition, blood samples were taken prior and 24h following ACTH-exposure. FKBP5 methylation in leukocytes was measured using pyrosequencing. The CpG-site analysed was located in the promoter region of FKBP5(chr6: 35,729,046; hg38).


Methylation values ranged from 17.72% to 29,79% (mean = 24.14%) prior to the ACTH-test and 19.30% to 31.45% (mean = 25.73%) following the ACTH-test, meaning no significant change in average methylation was found. Basal FKBP5methylation before ACTH stimulation was negatively correlated with basal serum cortisol levels (r = -0.509; p = 0.04). Furthermore, a negative correlation was found between serum cortisol 1h and methylation 24h following ACTH-exposure (r = -0.588; p = 0.005) as well as between the post ACTH-test methylation status and the delta of serum cortisol (r = -0.626; p = 0.002). In addition, delta methylation and delta serum cortisol were negatively correlated as well (r = -0.537; p = 0.012).


The negative correlation observed between FKBP5 methylation and serum cortisol supports previous studies, showing that FKBP5 methylation is altered by GC exposure. Average FKBP5 promoter methylation remained unchanged. Considering previous studies showing demethylation of FKBP5 after 2-4 weeks of GC stimulation, our findings suggest that FKBP5 methylation could be used as a a robust biomarker that reacts stronger to long term GC exposure than to a singular ACTH stimulus as used in our study. This hypothesis will be evaluated in future studies.


Nothing to Disclose: BW, HL, HO, BH