Intact GR dimerization is critical for restraining plasma ACTH levels during chronic psychosocial stress

Abstract

Male C57BL/6N mice exposed to the chronic subordinate colony housing (CSC; 19 days) paradigm, a preclinically validated model of chronic psychosocial stress, are characterized by unaffected basal morning plasma corticosterone (CORT) concentrations despite adrenal and pituitary hyperplasia and increased adrenocorticotropic hormone (ACTH) plasma concentrations, compared with single-housed control (SHC) mice. However, as CSC mice are still able to show an increased CORT secretion towards novel heterotypic stressors, these effects might reflect an adaptation rather than a functional breakdown of general hypothalamus-pituitary-adrenal (HPA) axis functionality. In the present study we used male mice of a genetically modified mouse line, to investigate whether genetically-driven ACTH overexpression compromises adaptational processes occurring at the level of the adrenals during CSC exposure. Experimental mice carried a point mutation in the DNA binding domain of the glucocorticoid (GC) receptor (GR), attenuating dimerization of GR (GR^dim), resulting in a congenially compromised negative feedback inhibition at the level of the pituitary. In line with previous studies, CSC mice in both the wild type (WT; GR^+/+) and GR^dim group developed adrenal enlargement. Moreover, compared with respective SHC and WT mice, CSC GR^dim mice show increased basal morning plasma ACTH and CORT concentrations. Quantitative polymerase chain reaction (qPCR) analysis revealed neither a genotype effect, nor a CSC effect on pituitary mRNA expression of the ACTH precursor proopiomelanocortin (POMC). Finally, CSC increased anxiety-related behavior, active coping and splenocyte in vitro (re)activity in both WT and GR^dim mice, while a CSC-induced increase in adrenal lipid vesicles and splenic GC resistance was detectable only in WT mice. Of note, lipopolysaccharide (LPS)-stimulated splenocytes of GR^dim mice were resistant to the inhibitory effects of CORT. Together our findings support the hypothesis that pituitary ACTH protein concentration is negatively controlled by GR dimerization under conditions of chronic psychosocial stress, while POMC gene transcription is not dependent on intact GR dimerization under both basal and chronic stress conditions. Finally, our data suggest that adrenal adaptations during chronic psychosocial stress (i.e., ACTH desensitization), aiming at the prevention of prolonged hypercorticism, are protective only to a certain threshold of plasma ACTH levels.

Keywords: Chronic psychosocial stress; Chronic subordinate colony housing (CSC); Glucocorticoid receptor (GR) dimerization; Negative feedback.

Fig. 1

Experimental timeline After arrival at the experimental room (Day −28), all mice were housed singly for four weeks. At day 1, experimental wild type (WT; GR^+/+) and GR^dim (GR^dim/dim) homozygous mice were weighed and assigned to either the single-housed control (SHC) or the chronic subordinate colony housing (CSC) group. In order to induce chronic psychosocial stress, a group of four CSC mice (two WT and two GR^dim mice per cage) was housed together with a dominant male CD-1 mouse for 19 consecutive days. To avoid habituation, CSC mice were exposed to a novel dominant male CD-1 mouse on days 8 and 15. All animals were tested for anxiety-related behavior in the open-field/novel object (OF/NO) test on day 15 and for depressive-like/coping behavior in the forced swim test (FST) on day 19 of CSC, before being euthanized in the morning of day 20. Afterwards, the severity of bite wounds was assessed for bite score analysis. Trunk blood was collected for the assessment of plasma corticosterone (CORT) and adrenocorticotropic hormone (ACTH). The adrenal glands, the pituitary as well as the spleen were removed and weight. Adrenal glands were frozen for the assessment of the adrenal lipid content by Oil Rred-O staining. The pituitary was frozen for quantitative polymerase chain reaction (qPCR) analysis of proopiomelanocortin (POMC). Moreover, splenocytes were isolated for an in vitro glucocorticoid (GC) sensitivity assay. Experimental timeline was partly created with BioRender.com.

Fig. 2

Effects on anxiety-related and depressive-like/coping behavior Chronic subordinate colony housing (CSC; dark-grey bars; triangles) increased anxiety-related behavior during open-field (OF; Fig. 2A) and novel-object (NO; Fig. 2E) exposure in both wild type (WT; GR^+/+; unhatched bars) and GR^dim (GR^dim/dim; hatched bars) mice. In detail, CSC vs. single-housed control (SHC; light-grey bars, circles) mice showed an increased time in corners during NO (Fig. 2G) exposure as well as a decreased distance moved during NO exposure (Fig. 2F), respectively. GR^dim but not WT CSC mice showed an increased time in corners during OF exposure (Fig. 2C) and a decreased number of entries into the inner zone of the OF during OF exploration (Fig. 2D) as well as a decreased number of NO explorations (Fig. 2H) when compared to respective GR^dim SHC mice. The general locomotion (distance moved during OF exploration) was comparable between all experimental groups (Fig. 2B). Representative track- and time visualizations are shown in (Fig. 2I) and (Fig. 2K), respectively. To assess the effects of CSC and an attenuation of GR dimerization on depressive-like/coping behavior, all mice were exposed to the FST (Fig. 2L) in the morning of day 19 of CSC exposure. In WT and GR^dim mice, CSC decreased time of floating (Fig. 2M) and increased the time of swimming (Fig. 2N) during FST exposure, respectively. Moreover, CSC in WT mice decreased the time of climbing during FST exposure (Fig. 2O). Data are presented as bar graphs (mean +SEM) with individual dots. *P ≤ 0.05, **P ≤ 0.01, ***P ≤ 0.001 vs. respective SHC. Illustrations (Fig. 2A, 2E and 2L) were created with BioRender.com.

Fig. 3

Effects on HPA axis parameters When compared to respective single-housed control (SHC; light-grey bars; circles) mice, chronic subordinate colony housing (CSC; dark-grey bars; triangles) resulted in an increased relative adrenal weight (Fig. 3A) in both wild type (WT; GR^+/+; unhatched bars) and GR^dim (GR^dim/dim; hatched bars) mice. Relative adrenal weight of both GR^dim SHC and CSC mice was decreased when compared to respective WT mice. Relative pituitary weight (Fig. 3B) was increased in WT but not GR^dim CSC vs. SHC mice and decreased in GR^dim vs. WT CSC mice. Plasma morning corticosterone (CORT; Fig. 3C) and adrenocorticotropic hormone (ACTH; Fig. 3D) concentrations were increased in GR^dim CSC compared to GR^dim SHC as well as WT CSC mice, respectively. Adrenal lipid content represented by the ratio of Oil Red-O-stained adrenal area to total adrenal area in sections of the right adrenal gland (Fig. 3E) as well as adrenal lipid content per adrenal weight (Fig. 3F) was increased in WT but not GR^dim CSC vs. SHC mice and decreased in both GR^dim vs. WT SHC and CSC mice, respectively. Representative Oil Red-O staining images of each experimental group are shown in (Fig. 3G). Data are presented as bar graphs (mean +SEM) with individual dots. *P# *P ≤ 0.05, **P ≤ 0.01, ***P ≤ 0.001 vs. respective SHC. ^#P ≤ 0.05, ^##P ≤ 0.01, ^###P ≤ 0.001 vs. respective WT.

Fig. 4

Effects on relative pituitary POMC mRNA expression The relative pituitary proopiomelanocortin (POMC) mRNA expression levels (Fig. 4A) did not differ between wild type (WT; GR^+/+; unhatched bars) and GR^dim (GR^dim/dim; hatched bars) mice exposed to chronic subordinate colony housing (CSC; dark-grey bars; triangles) or single-housing (SHC; light-grey bars; circles). As housekeeping gene, ribosomal proteins (RPL) was used. Data are presented as bar graphs (mean +SEM) with individual dots.

Fig. 5

Effects on spleen weight and functional splenic in vitro GC sensitivity When compared to respective single-housed control (SHC; light-grey bars; circles) mice, chronic subordinate colony housing (CSC; dark-grey bars; triangles) increased relative spleen weight (Fig. 5A) in both the wild type (WT; GR^+/+; unhatched bars) and GR^dim (GR^dim/dim; hatched bars) group. Furthermore, relative spleen weight of GR^dim SHC and CSC mice was increased compared to respective WT mice. Both, WT and GR^dim CSC mice had a bite score above 20 (WT: 27.5; GR^dim: 44.2; Fig. 5B). Splenic cell in vitro viability in response to lipopolysaccharide (LPS; Fig. 5C) was increased in all groups. Splenocytes from WT and GR^dim CSC mice showed a higher basal and LPS-induced cell viability compared to respective SHC mice. Basal cell viability was increased in splenocytes from GR^dim vs. WT CSC mice. Delta (LPS - basal) cell viability (0 μM corticosterone (CORT) set to 100%; Fig. 5D) in response to 0.1 μM CORT was reduced in WT SHC and CSC mice when compared to the 0 μM condition. Delta cell viability in response to 0.1 μM CORT was significantly increased in GR^dim SHC and CSC vs. respective WT mice, as well as WT CSC vs. SHC mice. Data are presented as bar graphs (mean +SEM) with individual dots. **P ≤ 0.01, ***P ≤ 0.001 vs. respective SHC. ^#P ≤ 0.05, ^###P ≤ 0.001 vs. respective WT. ^$$$P ≤ 0.001 vs. respective basal or 0 μM condition.

Fig. 6

Summary Figure Upon binding of corticotrophin releasing hormone (CRH) to the CRH-receptor (R) 1 on corticotroph cells of the anterior pituitary, proopiomelanocortin (POMC) gene transcription is initiated. POMC mRNA is subsequently translated into POMC and post-transcriptionally cleaved into inter alia adrenocorticotropic hormone (ACTH). During chronic psychosocial stress exposure, glucocorticoid receptor (GR) dimerization is proposed to inhibit the translation of different enzymes/convertases involved in POMC mRNA translation or post-transcriptional processing into ACTH protein (Harno et al., 2018). Figure was created with BioRender.com.