Corticotroph isolation from Pomc-eGFP mice reveals sustained transcriptional dysregulation characterising a mouse model of glucocorticoid-induced suppression of the hypothalamus-pituitary-adrenal axis

Abstract

Glucocorticoids (GC) are prescribed for periods > 3 months to 1%-3% of the UK population; 10%-50% of these patients develop hypothalamus-pituitary-adrenal (HPA) axis suppression, which may last over 6 months and is associated with morbidity and mortality. Recovery of the pituitary and hypothalamus is necessary for recovery of adrenal function. We developed a mouse model of dexamethasone (DEX)-induced HPA axis dysfunction aiming to further explore recovery in the pituitary. Adult male wild-type C57BL6/J or Pomc-eGFP transgenic mice were randomly assigned to receive DEX (approximately 0.4 mg kg^-1 bodyweight day^-1 ) or vehicle via drinking water for 4 weeks following which treatment was withdrawn and tissues were harvested after another 0, 1, and 4 weeks. Corticotrophs were isolated from Pomc-eGFP pituitaries using fluorescence-activated cell sorting, and RNA extracted for RNA-sequencing. DEX treatment suppressed corticosterone production, which remained partially suppressed at least 1 week following DEX withdrawal. In the adrenal, Hsd3b2, Cyp11a1, and Mc2r mRNA levels were significantly reduced at time 0, with Mc2r and Cyp11a1 remaining reduced 1 week following DEX withdrawal. The corticotroph transcriptome was modified by DEX treatment, with some differences between groups persisting 4 weeks following withdrawal. No genes supressed by DEX exhibited ongoing attenuation 1 and 4 weeks following withdrawal, whereas only two genes were upregulated and remained so following withdrawal. A pattern of rebound at 1 and 4 weeks was observed in 14 genes that increased following suppression, and in six genes that were reduced by DEX and then increased. Chronic GC treatment may induce persistent changes in the pituitary that may influence future response to GC treatment or stress.

Keywords: HPA axis; chronic; glucocorticoid; recovery.

FIGURE 1

Schematic of experiments. (A) Experiment 1. C57Bl6/J mice were assigned to receive dexamethasone (DEX) in drinking water (n = 6 mice in three cages) or water alone (n = 6 mice in three cages) for 4 weeks. Control (CTL) mice received only standard drinking water (group A). Following 4 weeks treatment, two animals selected at random from each cage were sacrificed (group B). Following one further week, two further mice from each cage were collected as at time 0 (group C). The last animals were collected 4 weeks after DEX withdrawal (group D). Age matched controls were assigned to group A. WT, wild‐type. (B) Experiment 2. Here, 16 cages of three mice were randomly assigned to four treatment groups, A; receiving water in drinking water for 8 weeks, B; drinking water for 4 weeks and then DEX for 4 weeks, C; water for 3 weeks, DEX for 4 weeks then 1 week with drinking water and D; DEX for 4 weeks then drinking water for 4 weeks. All animals were collected at a single time point at the end of the experiment as above. Animals were reverse lit and collected at the point of waking. (C) Experiment 3 was conducted as Experiment 2, except that animals were not reverse lit and Pomc‐eGFP mice were used. Twelve cages of three animals were assigned randomly to the four groups. At the end of the experiment, tissue was collected as in Eexperiments 1 and 2, but anterior pituitary was dissected and dissociated for FACS isolation of corticotrophs. RNA from isolated corticotrophs underwent RNA‐seq. Numbers to the right refer to numbers of animals/in how many cages

FIGURE 2

Dexamethasone reduces weight and corticosterone production that persists 1 week following treatment withdrawal. (A,B) Adrenal weight The mean weight of both adrenal glands at collection is presented with box and whisker charts from experiment 2 (A) and experiment 1 (B). (C,D) Adrenal weight/bodyweight (D,E) The adrenal weight as a proportion of bodyweight are presented for experiment 2(C) and experiment 1 (D). (E,F) Corticosterone levels. DEX treatment reduced corticosterone levels that were still reduced 1 week after stopping DEX treatment at waking (pm) (experiment 2) (E), but which had returned to normal basal levels (am; rest period) (experiment 1) (F). Corticosterone levels had returned to the level of controls 4 weeks after stopping DEX treatment. (G,H) 11‐dehydrocorticosterone levels. As (E,F). (I) ACTH levels. Data from experiment 2. The dotted box highlights mice with recovered ACTH production, the dashed box shows mice with unrecovered ACTH production; grey circles in this box are mice with Corticosterone levels below controls, blue circles are mice with corticosterone within the range of controls. (J) Relationship between plasma corticosterone and ACTH. DEX treatment (yellow triangles) reduced both plasma ACTH and corticosterone compared to controls (grey circles). One week after stopping DEX (blue squares), there was greater variation between animals with some showing high ACTH levels but lower corticosterone (dotted box) and some showing inappropriate ACTH levels for the lower corticosterone levels (dashed box). Four weeks after stopping DEX, ACTH and corticosterone levels had returned to level of controls (green crosses). Data from experiment 2. Legend. Grey dots and boxes represent control mice (Group A), yellow those who have had 4 weeks of DEX treatment (Group B), blue those one week after withdrawal of DEX (Group C) and green those 4 weeks after treatment withdrawal (Group D). Data analysed by linear mixed model with group as dependent variable and cage as random factor. Tukey‐adjusted post hoc tests compared to control group are indicated above boxes where significant differences were identified. *** p < 0.001, ** p < 0.01, * p < 0.05. n = 6‐12 animals from 3‐4 cages.

FIGURE 3

Dexamethasone (DEX) treatment induces significant and persistent changes in adrenal gene expression. (A) Hypothalamus. DEX treatment did not affect gene expression of Avp, Crh, Nr3C1, or Nr3c2 in whole hypothalamus (data from Experiment 1). (B,C) Adrenal. DEX reduced the adrenal expression of the adrenocorticotrophic hormone (ACTH) receptor (Mc2r), steroidogenic enzymes (Cyp11a1 Hsd3b2 and Star) and increased expression of the glucocorticoid receptor (Nr3c1) and mineralocorticoid receptor (Nr3c2). Data are shown from Experiment 1 (am panels) and Experiment 2 (pm panels). For ease of comparison, control samples from each time point in Experiment 1 have been combined, delta delta Ct (ddCT) values were made in comparison to time‐matched adrenal glands. Grey bars represent control mice (group A); yellow represent those who have had 4 weeks of DEX treatment (group B); blue represent those 1 week after withdrawal of DEX (group C); and green represent those 4 weeks after treatment withdrawal (group D). Data analysed by a linear mixed model with group as dependent variable and cage as random factor. Tukey‐adjusted post‐hoc tests compared to control group are indicated above bars (small asterixis) where significant differences were identified. ***p < 0.001, **p < 0.01, *p < 0.05 (n = 6–12 animals from three or four cages)

FIGURE 4

DEX treatment significantly affects the corticotroph transcriptome with changes evident 4 weeks following treatment withdrawal. (A) eGFP expression. The eGFP expression of the isolated cells is shown. In groups A (control; grey line) and C (1 week post DEX withdrawal; blue line) bimodal populations of cells are apparent. The mean fluorescence of three sorts is shown each comprising the dissociated pituitaries of 2‐3 mice. (B) Heatmap showing transcriptomic analysis of isolated cells. The 151 differentially genes with FDR <0.05 are shown. Relative expression across each row is shown as per the colour code to the right of the panel. Columns indicate each sample comprising cells isolated from 2‐3 pituitaries and are clustered according to gene expression pattern. Samples are clustered by Euclidean distance. (C) Principal component analysis (PCA) plot showing the 1st 2 principal components of PCA analysis the RNA‐seq data from corticotrophs following exposure to DEX. (D) Selected genes from RNA‐seq from classes exhibiting persisting changes following withdrawal of DEX. Log2 fold change data are shown for genes that were significantly affected by treatment and recovery along with the molecular function class to which they belong (see table 1 and 2). Data are the same as those presented in 4B. Legend. Grey lines/bars represent control mice (Group A); yellow those who have had 4 weeks of DEX treatment (Group B); blue those one week after withdrawal of DEX (Group C); and green those 4 weeks after treatment withdrawal (Group D). n = 3 comprising a pool of 2‐3 pituitaries.