Psychopharmacology

, Volume 218, Issue 1, pp 303–312

A measure of glucocorticoid load provided by DNA methylation of Fkbp5 in mice

Authors

  • Richard S. Lee
    • Department of Psychiatry and Behavioral SciencesJohns Hopkins University School of Medicine
  • Kellie L. K. Tamashiro
    • Department of Psychiatry and Behavioral SciencesJohns Hopkins University School of Medicine
  • Xiaoju Yang
    • Department of MedicineJohns Hopkins University School of Medicine
  • Ryan H. Purcell
    • Department of Psychiatry and Behavioral SciencesJohns Hopkins University School of Medicine
  • Yuqing Huo
    • Department of Psychiatry and Behavioral SciencesJohns Hopkins University School of Medicine
  • Michael Rongione
    • Department of Psychiatry and Behavioral SciencesJohns Hopkins University School of Medicine
    • Department of Psychiatry and Behavioral SciencesJohns Hopkins University School of Medicine
    • Center for EpigeneticsJohns Hopkins University School of Medicine
  • Gary S. Wand
    • Department of MedicineJohns Hopkins University School of Medicine
Original Investigation

DOI: 10.1007/s00213-011-2307-3

Cite this article as:
Lee, R.S., Tamashiro, K.L.K., Yang, X. et al. Psychopharmacology (2011) 218: 303. doi:10.1007/s00213-011-2307-3
  • 457 Views

Abstract

Rationale

Given the contribution of cortisol dysregulation to neuropsychiatric and metabolic disorders, it is important to be able to accurately compute glucocorticoid burden, a measure of allostatic load. One major problem in calculating cortisol burden is that existing measures reflect cortisol exposure over a short duration and have not been proven to reliably quantify cortisol burden over weeks or months.

Method

We treated two cohorts of mice with corticosterone in the drinking water and determined the relationship between serial plasma corticosterone levels drawn over 4 weeks and the whole-blood DNA methylation (DNAm) changes in a specific glucocorticoid-sensitive gene, Fkbp5, determined at the end of the treatment period.

Results

We observed that the percent reduction in DNAm in the intron 1 region of Fkbp5 determined from a single blood draw strongly reflected average glucocorticoid burden generated weekly during the prior month of glucocorticoid exposure. There were also strong correlations in DNAm with glucocorticoid-induced end organ changes in spleen weight and visceral fat. We tested a subset of these animals for anxiety-like behavior in the elevated plus maze and found that DNAm in the blood also has predictive value in determining the behavioral consequences of glucocorticoid exposure.

Conclusion

A whole-blood assessment of Fkbp5 gene methylation is a biomarker that integrates 4 weeks of glucocorticoid exposure and may be a useful measure in states of excess exposure. It will be important to determine if Fkbp5 DNAm changes can also be a biomarker of glucocorticoid burden during chronic social stress.

Keywords

DNA methylationFkbp5Glucocorticoid burdenCorticosteroneEpigeneticsAllostasis

Supplementary material

213_2011_2307_MOESM1_ESM.ppt (156 kb)
Suppl Fig. 1Regression analysis of mean plasma corticosterone levels vs. DNAm in Expt. 2. a A strong correlation was observed between blood DNAm of Fkbp5 intron 1, CpG position 1 from the second experiment and the mean corticosterone levels calculated from 25 daily blood draws. Similar results were obtained for CpG position 2 (b) (PPT 156 kb)
213_2011_2307_MOESM2_ESM.ppt (275 kb)
Suppl Fig. 2Organ weights and regression analysis of DNAm vs. spleen mass and percent visceral fat. a Thymus and adrenal glands were weighed for corticosterone-treated (varying shades of gray and black for 100, 75, 50, and 25 μg/ml) and vehicle-treated (white bars) mice. The absence of data for thymic mass of corticosterone-treated mice for doses >50 μg/ml reflects a complete atrophy of the organ after 4 week of treatment with corticosterone. Asterisks (*) indicate differences that are statistically significant (P < 0.05). A strong correlation was observed between the DNAm of Fkbp5 intron 1, CpG position 1 vs. spleen mass (b), and DNAm vs. percentage of visceral fat (c). Similar results were observed for CpG position 2 vs. spleen mass (d) and percentage of visceral fat (e). All of the organ weights are expressed as weight per gram total body weight of the mice (PPT 275 kb)

Copyright information

© Springer-Verlag 2011