Adrenal rather than central dysfunction limits HPA axis recovery after chronic glucocorticoid treatment in male mice
Data files
Jul 09, 2026 version files 51.07 KB
-
README.md
8.65 KB
-
Supporting_data_raw.xlsx
42.42 KB
Abstract
Glucocorticoid-induced adrenal insufficiency (GIAI) can persist for months after discontinuation of chronic corticosteroid therapy, placing patients at risk for life-threatening adrenal crises. This prolonged suppression has been attributed primarily to delayed restoration of hypothalamic–pituitary signaling based on indirect measures of central axis activity. To identify the rate-limiting site of hypothalamic-pituitary-adrenal (HPA) axis recovery, we evaluated the timing of functional and histologic recovery at each node of the axis following 8 weeks of dexamethasone (DEX) treatment in adult, male mice. DEX administration fully suppressed HPA axis activity. Unexpectedly, within one week of DEX withdrawal, hypothalamic Crh mRNA and plasma ACTH rebounded above control levels, whereas corticosterone (CORT) remained suppressed for an additional seven weeks. DEX-treated adrenals were markedly atrophic and contained large clusters of lipid-filled macrophages. Even after adjusting for macrophage content, CORT secretion was disproportionately low relative to the remaining adrenocortical cell mass despite supraphysiologic ACTH stimulation. The adrenal is thus the principal site of post-withdrawal GIAI, involving adrenocortical cell loss and a superimposed defect in steroidogenesis. We next tested whether preserving adrenal trophic signaling during glucocorticoid exposure could prevent GIAI. Adrenal function recovered more slowly in mice treated with DEX and daily cosyntropin (a synthetic ACTH analog) compared to those treated with DEX alone. In contrast, mice with non-suppressible endogenous ACTH due to targeted hypothalamic deletion of the glucocorticoid receptor maintained normal adrenal architecture and steroidogenic capacity despite prolonged DEX treatment. Pharmacologic treatments that mimic sustained trophic signaling to the adrenal during chronic glucocorticoid treatment may thus prevent GIAI.
Dataset DOI: 10.5061/dryad.1g1jwsvd0
Description of the data and file structure
This dataset contains the raw numerical data underlying all quantitative figures in a study characterizing the site and mechanisms of glucocorticoid-induced adrenal insufficiency (GIAI) in a murine model. Adult male C57BL/6J mice were treated with dexamethasone (DEX) via drinking water for 8 weeks, followed by serial assessment of hypothalamic-pituitary-adrenal (HPA) axis function and adrenal histology at 1–2 week intervals through 8 weeks post-withdrawal. Outcome measures included hypothalamic Crh mRNA expression (quantified by RNAScope densitometry), plasma ACTH and corticosterone concentrations following insulin-induced hypoglycemia and cosyntropin stimulation (measured by ELISA), adrenal gland weights, morphometric quantification of adrenocortical and adrenal macrophage area (by fluorescence microscopy and QuPath image analysis), steroidogenic pathway gene expression per zona fasciculata cell (by multiplexed fluorescent RNAScope and QuPath puncta counting), and adrenocortical proliferation and zona glomerulosa thickness. Two intervention paradigms were additionally studied: pharmacologic co-treatment with daily cosyntropin during DEX exposure, and a genetic model of sustained endogenous ACTH secretion during DEX treatment (Sim1Cre:GRfl/fl mice, or HGRKO).
Files and variables
File: Supporting_data_raw.xlsx
Description:
This file contains 34 worksheets. Each worksheet corresponds to one quantitative panel of a main or supplemental figure in the manuscript. Worksheet names indicate the figure panel (e.g., "Fig. 1B" contains the raw data plotted in Figure 1B). Each column represents one experimental group and each row represents one biological replicate (individual mouse).
Abbreviations and variables:
- DN — Dexamethasone-naïve (vehicle-treated control)
- DEX — Dexamethasone
- COS — Cosyntropin (synthetic ACTH[1-24])
- DEX+COS — Combined dexamethasone and daily cosyntropin treatment
- 4D DEX — 4 days of dexamethasone treatment
- 4D DEX+COS — 4 days of combined dexamethasone and cosyntropin treatment
8+0, 8+1, 8+2, etc. — 8 weeks of DEX treatment followed by 0, 1, 2, etc. weeks post-withdrawal - HGRWT — Cre-negative littermate control mice (functionally wild-type)
- HGRKO — Sim1Cre:GRfl/fl mice with selective hypothalamic glucocorticoid receptor deletion
- ACTH — Adrenocorticotropic hormone (pg/mL; measured by two-site sandwich ELISA, ALPCO #21-ACTHU-E01)
- CORT — Corticosterone (µg/dL; measured by competitive ELISA, ALPCO #55-CORMS-E01)
- IIH — Insulin-induced hypoglycemia (stressor used to stimulate HPA axis)
- AU — Arbitrary units (normalized to the dexamethasone-naïve control mean, set to 1.0, unless otherwise specified)
- zF — Zona fasciculata
- zG — Zona glomerulosa
- PVH — Paraventricular nucleus of the hypothalamus
- AM — Adrenal macrophage
Worksheet descriptions:
-
Fig 1B — Crh mRNA expression (AU) in the PVH at DN, 8+0, and 8+1, quantified by chromogenic RNAScope densitometry (Weka Segmentation plugin, Fiji ImageJ) and normalized to DN controls
-
Fig. 1C — IIH-stimulated plasma ACTH (pg/mL) at DN, 8+0, 8+1, 8+2, 8+3, 8+4, 8+6, and 8+8.
-
Fig. 1D — IIH-stimulated plasma CORT (µg/dL) at DN, 8+0, 8+1, 8+2, 8+3, 8+4, 8+6, and 8+8.
-
Fig. 1E — Cosyntropin-stimulated plasma CORT (µg/dL) at DN, 8+0, 8+1, 8+2, 8+3, 8+4, 8+6, and 8+8.
-
Fig. 2A — Adrenal gland weight (mg) at DN, 8+0, 8+1, 8+2, 8+6, and 8+8.
-
Fig. 3B — Pan-cortical area (µm²; streptavidin-positive plus autofluorescent regions) at DN, 8+0, 8+1, 8+2, 8+6, and 8+8, from morphometric analysis of thick floating adrenal sections in QuPath.
-
Fig. 3C — Medullary (tyrosine hydroxylase-positive) area (µm²) at DN, 8+0, 8+1, 8+2, 8+6, and 8+8; used to confirm equivalent sectioning across timepoints.
-
Fig. 3D — Autofluorescent adrenal macrophage (AM) area (µm²) at DN, 8+0, 8+1, 8+2, 8+6, and 8+8.
-
Fig. 3E — Steroidogenic cell type area (µm²) at DN, 8+0, 8+1, 8+2, 8+6, and 8+8; computed by subtracting AM area from pan-cortical area.
-
Fig. 3F — Ratio of cosyntropin-stimulated CORT to steroidogenic cell type area, normalized to DN (AU), at DN, 8+0, 8+1, 8+2, 8+6, and 8+8.
-
Fig. 4F — Mc2r mRNA puncta per zF cell (AU, normalized to DN) at DN, 8+0, and 8+1; quantified by multiplexed fluorescent RNAScope and QuPath subcellular detection.
-
Fig. 4G — Star mRNA puncta per zF cell (AU) at DN, 8+0, and 8+1.
-
Fig. 4H — Cyp11a1 mRNA puncta per zF cell (AU) at DN, 8+0, and 8+1.
-
Fig. 4I — Hsd3b2 mRNA puncta per zF cell (AU) at DN, 8+0, and 8+1.
-
Fig. 4J — Cyp11b1 mRNA puncta per zF cell (AU) at DN, 8+0, and 8+1.
-
Fig. 5B — Percentage of TUNEL-positive adrenocortical cells (%) in DN, 4D DEX, and 4D DEX+COS adrenals; quantified by QuPath positive cell detection.
-
Fig. 5C — Cosyntropin-stimulated CORT (µg/dL) over 0–8 weeks post-withdrawal in DEX-treated and DEX+COS-treated mice, with DN reference values. Columns: weeks post-withdrawal (X), individual DEX-treated mice, individual DEX+COS-treated mice, DN reference mice.
-
Fig. 5E — Pan-cortical area normalized to DN (AU) in 8+1 DEX-monotherapy and 8+1 DEX+COS adrenals.
-
Fig. 5F — AM area (µm²) in 8+1 DEX-monotherapy and 8+1 DEX+COS adrenals.
-
Fig. 5G — Steroidogenic cell type area normalized to DN (AU) in 8+1 DEX-monotherapy and 8+1 DEX+COS adrenals.
-
Fig. 6B — Percentage of GR-immunoreactive nuclei (%) in the PVH of DEX-treated HGRWT and HGRKO mice.
-
Fig. 6D — Crh mRNA expression (AU) in the PVH of DEX-treated HGRWT and HGRKO mice, quantified by chromogenic RNAScope densitometry (Weka Segmentation plugin, Fiji ImageJ); not normalized.
-
Fig. 7B — Percentage of TUNEL-positive adrenocortical cells (%) in DN, 4D DEX (WT), and 4D DEX HGRKO adrenals.
-
Fig. 7C — Basal plasma ACTH (pg/mL) in HGRWT and HGRKO mice prior to DEX treatment and on 8 weeks of DEX (8+0).
-
Fig. 7D — Plasma CORT (µg/dL) in HGRWT and HGRKO mice under basal, restraint-stimulated, and cosyntropin-stimulated conditions, both prior to DEX and on 8 weeks of DEX (8+0).
-
Fig. 7E — Adrenal weights (mg) in HGRWT and HGRKO mice on long-term DEX (8+0).
-
Fig. 7F — Pan-cortical area normalized to DN (AU) in HGRWT and HGRKO adrenals at 8+0.
-
Fig. 7G — AM area (µm²) in HGRWT and HGRKO adrenals at 8+0.
-
Fig. 7H — Steroidogenic cell type area normalized to DN (AU) in HGRWT and HGRKO adrenals at 8+0.
-
Fig. 8D — Mc2r mRNA puncta per zF cell (AU, normalized to DN) in DN controls, WT 4D DEX, WT 4D DEX+COS, and HGRKO 4D DEX adrenals.
-
Fig. S2A — Adrenal weight (% of DN mean) and IIH-stimulated CORT (% of DN mean) over the post-withdrawal time course; derived from Figs. 2A and 1D respectively.
-
Fig. S2B — Adrenal weight (% of DN mean) and cosyntropin-stimulated CORT (% of DN mean) over the post-withdrawal time course; derived from Figs. 2A and 1E respectively.
-
Fig. S5C — Zona glomerulosa thickness (µm) in DN, 8+0, and 8+1 adrenals, measured from Dab2 immunofluorescence images.
-
Fig. S5D — Percentage of Ki-67-positive adrenocortical cells (%) in DN, 8+0, and 8+1 adrenals, quantified by QuPath positive cell detection.
Code/software
No custom code or scripts are included with this submission. The raw data file (Supporting_data_raw.xlsx) can be opened with any spreadsheet software capable of reading .xlsx format, including Microsoft Excel, Google Sheets, or the free, open-source application LibreOffice Calc (available at libreoffice.org).
Statistical analyses were performed in GraphPad Prism v10.3.1 (GraphPad Software, Boston, MA; graphpad.com). Prism project files are not included in this deposit, but statistical tests applied to each variable and summary statistics are reported in the manuscript text and figure legends. Assumptions for parametric testing are fully reported in Supplemental Table SIII of the manuscript.
Image quantification was performed using QuPath v0.4.3 (qupath.github.io), a free, open-source digital pathology platform. Hypothalamic Crh mRNA densitometry was performed in Fiji ImageJ (fiji.sc) using the Weka Segmentation plugin. No QuPath or ImageJ project files are included in this deposit, as the underlying quantitative outputs are provided in full in the accompanying spreadsheet.
