Metabolic Dysfunctions following Chronic Oral Corticosterone are modified by Adolescence and Sex in Mice

Abstract

Adolescence is a period of development in which shifts in responses to glucocorticoids is well-documented. Obesity and metabolic syndrome are substantial health issues whose rates continue to rise in both adult and adolescent populations. Though many interacting factors contribute to these dysfunctions, how these shifts in glucocorticoid responses may be related remain unknown. Using a model of oral corticosterone (CORT) exposure in male and female mice, we demonstrate differential responses during adolescence (30-58 days of age) or adulthood (70-98 day of age) in endpoints relevant to metabolic function. Our data indicate that CORT resulted in significant weight gain in adult- and adolescent-exposed females and adult-exposed males, but not adolescent-exposed males. Despite this difference, all animals treated with high levels of CORT showed significant increases in white adipose tissue, indicating a dissociation between weight gain and adiposity in adolescent-treated males. Similarly, all experimental groups showed significant increases in plasma insulin, leptin, and triglyceride levels, further suggesting potential disconnects between overt weight gain, and underlying metabolic dysregulation. Finally, we found age- and dose-dependent changes in the expression of hepatic genes important in glucocorticoid receptor and lipid regulation, which showed different patterns in males and females. Thus, altered transcriptional pathways in the liver might be contributing differentially to the similar metabolic phenotype observed among these experimental groups. We also show that despite little CORT-induced changes in the hypothalamic levels of orexin-A and NPY, we found that food and fluid intake were elevated in adolescent-treated males and females. These data indicate chronic exposure to elevated glucocorticoid levels results in metabolic dysfunction in both males and females, which can be further modulated by developmental stage.

Introduction

Glucocorticoid hormones are essential in maintaining metabolic and homeostatic functions, acting on various tissues, including the brain and liver [1]. However, chronic exposure to elevated levels of glucocorticoids, due to long-term exposure to stress or endocrine disorders like Cushing’s syndrome, can lead to significant metabolic dysfunctions, including obesity and metabolic syndrome [2, 3]. Given the prevalence of these metabolic disorders and their associated morbidities like cardiovascular disease and type 2 diabetes mellitus [4-7], it is imperative to better understand the relationship between increased glucocorticoids and altered metabolism.

Chronically exposing mice to elevated glucocorticoid levels by providing them with exogenous corticosterone (CORT) via their drinking water leads to all of the fundamental hallmarks of metabolic syndrome [8]. Specifically, adult male and female mice exposed to this non-invasive treatment show increased food intake and abdominal fat, hyperinsulinemia, hyperlipidemia, and glucose dysregulation [8-12]. Thus, this preclinical model provides a powerful experimental platform for a more mechanistic understanding of how altered levels of glucocorticoids contribute to metabolic dysfunction.

Our laboratory and others have documented how steroid hormones have very different effects at particular stages of development, such as puberty and adolescence [13, 14]. Along these lines, we have recently shown that adolescent male mice chronically exposed to oral CORT also develop signs of metabolic syndrome, including increased adiposity and hyperinsulinemia [15]. Though male mice respond to these treatments when they occur during either adolescence or adulthood, the magnitude and direction of the effects vary. For instance, adult-exposed males respond to CORT treatments with increases in body weight and adiposity, while adolescent-exposed males show increased adiposity but no change in overall body weight [15]. Furthermore, males exposed during adolescence show greater reductions in bone density compared to males exposed in adulthood [15]. Thus, these data indicate that the stage of development can significantly modulate the effects of chronic CORT exposure on metabolic outcomes, at least in males. A clear gap in our current understanding is how these treatments impact female metabolic function and whether the mechanisms involved might differ depending on both sex and age.

Given these gaps, and in an effort to extend this model by investigating peripheral and central factors contributing to CORT-mediated metabolic dysfunctions, two experiments were conducted. In the first experiment, male and female C57BL/6N mice were exposed to oral CORT during adolescence (30-58 days of age) or young adulthood (70-98 days of age) during which we assessed effects on somatic and hormonal measures. Since we hypothesized these observed effects may be in part mediated by changes in the liver, we assessed hepatic transcriptional changes in the glucocorticoid receptor (Nr3c1), as well as the transcriptional repressor Hes1, which plays a central role in the regulation of glucocorticoid-induced gene expression [16]. We also probed changes in the genes for plasma lipoprotein lipase (Lpl), which is important in fatty acid metabolism in the liver [17], Srebf1, a transcription factor for the low-density lipoprotein receptor [18], and Nsdhl, a gene that codes for an enzyme important for cholesterol biosynthesis [19]. In the second experiment, with identical experimental groups, the trajectory of the somatic changes and feeding and drinking behavior were examined. Furthermore, at the end of treatment, we assessed hypothalamic neuropeptide levels, namely orexin-A and neuropeptide Y (NPY), as these signals are known to be modulated by CORT and involved in metabolism and feeding [20-24].

Materials and Methods

Experimental animals

Adult (63 days of age) and pre-adolescent (23 days of age) male and female C57BL/6N mice were obtained from Charles River Laboratories (Wilmington, MA) and allowed to acclimate for one week prior to the start of the experiments. Mice were housed 2-3 per cage in polycarbonate cages (28 x 17 x 12 cm) with Bed-o’Cobs 1/4-inch bedding. Prior to treatment, mice were given ad libitum access to water and standard rodent chow (Lab Diet #5012; PMI Nutritional International LLC, Brentwood, MO, USA). The mice were housed in a 12-h light-dark schedule (lights on at 0800 h) and the animal room temperature was maintained at 21±2 °C. Except for weekly cage maintenance and body weight monitoring, mice were minimally handled for the duration of the experiments. All procedures were in accordance with the Institutional Animal Care and Use Committee (IACUC) of Columbia University.

Experimental design and tissue collections

Two experiments were conducted in both male and female mice. Experiment 1 assessed the effects of chronic oral CORT during adolescence or adulthood on somatic, hormonal, and hepatic measures, while Experiment 2 assessed the effects of chronic oral CORT during adolescence or adulthood on behavioral and neurobiological measures. In Experiment 1, adolescent (30 days of age) and adult (70 days of age) male and female mice were exposed to one of four treatment conditions for 28 days: (i) tap water, (ii) 1% ethanol (EtOH) in tap water, (iii) 1% EtOH in tap water with 25 μg/ml CORT (crystalline; C2505; Sigma, St. Louis, MO), or (iv) 1% EtOH in tap water with 100 μg/ml CORT (n = 4-6 per sex, age, experimental condition). Due to the hydrophobic nature of CORT, it was dissolved in 100% EtOH via sonication, vortexed, and added to tap water to a 1% concentration. These doses of CORT and vehicle were based on previous experiments in male mice [15]. In Experiment 2, adolescent (30 days of age) and adult (70 days of age) male and female mice were exposed to one of two treatment conditions for 28 days: (i) 1% ethanol (EtOH) in tap water or (ii) 1% EtOH in tap water with 100 μg/ml CORT (n = 8 per sex, age, experimental condition). An initial pilot study in adolescent and adult mice established that the addition of 1% EtOH to the tap water did not significantly affect the behavioral or neurobiological parameters measured in the present studies, and thus to reduce animal numbers, a tap water only control was not included in these experimental designs (Shahanoor and Romeo, unpublished observation). Moreover, as the 100 μg/ml CORT condition resulted in the most robust metabolic phenotype in our previously published work [8, 9, 15] and Experiment 1, only this dose was used in Experiment 2.

In both Experiments, after the 4-week exposure period, these non-fasted mice were killed at 58 (adolescent-exposed) or 98 (adult-exposed) days of age by rapid decapitation (Experiment 1) or transcardial perfusion (Experiment 2) during the light phase of their light-dark cycle (between ~1000-1400 h). During tissue collections, the experimenters were unaware of the specific experimental conditions of the mice. In Experiment 1, body weights and terminal thymus, testes, seminal vesicles, uterine horns, and visceral white adipose tissue (WAT) weights were obtained. Livers were also collected and snap frozen in sterile tubes (Eppendorf, BIOPUR, Hauppauge, NY) in a dry ice-ethanol bath. Trunk blood samples were collected in heparinized tubes, centrifuged at 2500rpm for 15 min at 4°C, and plasma stored at −20°C until assays were performed for CORT, insulin, leptin, and triglycerides (see below). In Experiment 2, body weights and fluid and food intake were measured weekly. At the end of treatment, mice were weighed and then euthanized with an overdose of ketamine and xylazine and perfused with 0.9% heparinized sterile saline followed by 4% paraformaldehyde in 0.1 M phosphate buffer (PFA). After the perfusion, brains were removed from the skull and post-fixed for 24h in 4% PFA at 4°C. The brains were then transferred to a 20% sucrose solution in 0.1 M phosphate buffer (PB) at 4°C for 48h. Brains were sectioned on a cryostat in the coronal plane at 35μm and stored at −20°C in cryoprotectant solution (1 part ethylene glycol to 1 part 20% sucrose). These sections were later processed for orexin-A or NYP immunohistochemistry (see below), as these two neuropeptides influence metabolic function and food intake [20, 25].

Radioimmunoassay (RIA) and Enzyme-linked Immunosorbent Assays (ELISAs)

Plasma CORT levels were measured using a commercially available RIA kit (MP Biomedicals; Solon, OH; catalog # 07-120102). Plasma insulin and leptin levels were measured using commercially available ELISA kits (EMD Millipore Corporation, Billerica, MA; catalog #EZRMI-13K and EZML-82K, respectively), while plasma triglycerides levels were measured using a commercially available colorimetric assay kit (Cayman Chemical, Ann Arbor, MI; catalog #10010303). All assays were performed as indicated by the supplier; however, for the insulin ELISA, samples from all of the 100 μg/ml CORT treatment groups were diluted 1:10 in assay buffer to ensure the samples were within the detectable range of the standard curve. Thus, insulin levels obtained from these groups were multiplied by 10. For all assays, samples were run in duplicate, and values were averaged. The intra-assay coefficient of variation (CV) and lower limit of detectability for each assay was as follows: CORT = 11.2% and 19.28 ng/ml, insulin = 9.1% and 0.2 ng/ml, leptin = 8.4% and 0.23 ng/ml, and triglycerides = 5.4% and 3.12 mg/dl.

RT-qPCR

To determine hepatic gene expression, RT-qPCR was used as previously described [26-28]. Briefly, total RNA was extracted using RNeasy Micro kits (Qiagen; Valencia, CA). cDNA synthesis was performed with MultiScribe^™ MuLV reverse-transcriptase following the High Capacity cDNA Synthesis Kit protocol (Life Technologies Inc; Frederick, MD). Real-time PCR assays were performed using TaqMan (Table 1). Samples were run in triplicate on a Life Technologies/Applied Biosystems Viia7 real-time PCR machine with 20μl reaction volume. Samples were then compared using the ΔΔCT method of relative quantification [29], with RPS18 used to normalize between biological replicates.

Table 1.

Gene symbol, gene name, and TaqMan Assay ID for genes investigated.

Gene Symbol	Gene Name	TaqMan Assay ID
Hes1	Hes Family Transcription Factor 1	Mm01342805_m1
Lpl	Lipoprotein Lipase	Mm00434764_m1
Nr3c1	Glucocorticoid Receptor	Mm00433832_m1
Nsdhl	NAD(P) Dependent Steroid Dehydrogenase	Mm00477897_m1
Rps18	Ribosomal Protein S18	Mm026017777_g1
Srebf1	Sterol Regulatory Element Binding Transcription Factor 1	Mm00550338_m1

Orexin-A and NPY Immunohistochemistry

In Experiment 2, 4-5 anatomically matched sections, each separated by 105 μm, through either the lateral hypothalamus (LH) or arcuate nucleus (Arc; corresponding to plates 39-42 or plates 41-45, respectively, in a standard mouse atlas; [30] were processed for either orexin-A or NPY. For orexin-A or NPY, all sections were processed together. The sections were first washed in 0.1 M PB followed by a 5 min incubation in 0.3% H₂O₂ and washed with 0.1 M PB with 0.1% Triton-X-100 (PBT). Sections were incubated for 1 h in 2% normal goat serum (NGS), and then in either rabbit anti-Orexin-A (1:16,000; AB6214; Abcam, Cambridge, MA, USA) or rabbit anti-NPY (1:16,000; T-4070; Peninsula Laboratories International, Inc., San Carolos, CA, USA) for 24 h at 4°C. Processing tissue in the absence of these primary antibodies resulted in no detectable immunostaining. Sections were then washed in PBT and incubated in goat anti-rabbit secondary (1:200; Vector Laboratories, Burlingame, CA, USA) and then exposed to Avidin-Biotin Complex (ABC; 1:250; Vectastain ABC Kit, Vector Laboratories) for 1 h at room temperature. The tissue was then washed in 0.1 M phosphate buffer saline (PBS) and incubated in 3,3’diaminobenzadine (DAB) in a 3 M sodium acetate buffer containing 0.05% H₂O₂ for 5 min followed by washes in PBS. The tissue was mounted on Fischer Brand Plus slides (Fischer Scientific, Pittsburg, PA, USA), dried, and exposed to 70%, 95%, and 100% ethanol, followed by xylenes, and covered by a coverslip using DPX (06552, Sigma-Aldrich).

Microscopy and Image Analysis

The number of orexin-A cells were quantified in the LH, while the relative optical staining density of NPY fibers were measured in the Arc. Orexin-A-positive cells in the LH were counted using a light microscope (Zeiss 200 M, Axiovert) by placing a rectangular ocular grid of 124,266 μm² superimposed on top of the images. Bilateral counts for each stained section were averaged and data were expressed as the average number of orexin-A-positive cells per 124,266 μm². NPY-positive fibers in the ARC were analyzed with ImageJ using a region of interest (ROI) to quantify the relative optical density (ROD). The RODs from all stained sections of each animal were averaged and data are expressed as the average ROD for NPY-positive cells.

Statistical analyses

It is important to note that although the experimental designs for both the males and females were identical, analyses were not conducted between the sexes, as the male and female data were collected from different cohorts of mice, with the males collected approximately two months before the females. Furthermore, in Experiment 1, t-tests were conducted between the tap water and 1% EtOH in tap water control groups. As no significant differences were found between these two groups on any of the measures we assessed, these control groups were combined into a single 0 μg/ml CORT control group.

For Experiments 1 and 2, terminal measures, such as tissue weights, plasma hormone concentrations, and gene expression levels in the liver (Experiment 1) or neurobiological measures (Experiment 2), were analyzed with between subjects two-way ANOVAs (age of treatment x dose). In Experiment 2, body weights and food and fluid intake were analyzed within a treatment age and sex with repeated measures two-way ANOVAs (dose x day of treatment). Significant main effects and interactions were further analyzed using the Bonferroni and Tukey’s honestly significant difference post-hoc tests. Data are reported as the mean ± SEM and differences were considered significant at P < 0.05. All statistical analyses were performed using GraphPad PRISM, version 8.2.1 (GraphPad Software Inc., San Diego, CA).

In Experiment 1, one adolescent-treated male and one adolescent-treated female in the 100 μg/ml CORT group and one adult-treated male in the 25 μg/ml CORT group did not yield sufficient quantities of plasma, and thus, were not included in any of these analyses. In Experiment 2, one adolescent-treated female in the 0 μg/ml treatment group was excluded from all analyses due to poor tissue fixation quality.

Results

Experiment 1

CORT-Induced changes in Weight Gain and Adiposity: Age of Exposure Matters in Males, but not Females

Males:

There was a significant interaction between the dose of CORT and the age of exposure on the amount of weight gained (F (2, 32) = 3.62, P <0.05). Overall, weight gain was higher in the more rapidly growing adolescent groups. Adults exposed to 100 μg/ml of CORT gained significantly more weight than the adults exposed to either 0 or 25 μg/ml of CORT, while in adolescent-treated males no differences were observed in weight gain between the CORT doses (Figure 1A). Notably, WAT levels and WAT weights expressed as % body weight in males were found to show significant stepwise increases with increasing doses of CORT, independent of the age of exposure (F (2,32) = 29.90 and 48.13, respectively, P <0.05; Figures 1B and 1C, respectively). These data indicate that the age of exposure influences the amount of CORT-induced weight gain in males, with only the adults showing effects of CORT exposure on body weight, while visceral adiposity increases to similar degrees in males treated with CORT during either adolescence or adulthood.

Figure 1.

Mean (± SEM) body weight gained (g), visceral white adipose tissue (WAT; mg), and WAT percent (%) body weight in adolescent- and adult-exposed males (A, B and C, respectively) and adolescent- and adult-exposed females (C, D, and E, respectively) treated with 0, 25, or 100 μg/ml of oral CORT for four weeks. Exposure during adolescence was from 30-58 days of age, while exposure during adulthood was from 70-98 days of age. Asterisks indicate significant differences between doses within an age of exposure, while the “#” indicates a significant difference between the ages of exposure at that dose. To signify a main effect of dose within a sex, bars that share a letter are not significantly different from one another.

Females:

Independent of age of exposure, females exposed to 100 μg/ml of CORT gained significantly more weight, WAT, and WAT % body weight than those exposed to either the 0 or 25 μg/ml doses (F (2,33) = 58.72, 127.20, and 173.1, respectively P <0.05, Figures 1D, 1E, and 1F). Given the similar changes in these somatic parameters in females treated with CORT during adolescence or adulthood, with no age-dependent dissociation between CORT-induced weight gain and adiposity, these data demonstrate that the relationship between CORT-induced changes in body weight and WAT weight are only modified by age of exposure in males.

Hormonal and Metabolic Measures: CORT-induced changes in Males and Females

Corticosterone:

Independent of the age of exposure, both male and female mice treated with the 100 μg/ml dose of CORT had significantly higher circulating levels of plasma CORT compared to mice exposed to either the 0 or 25 μg/ml doses (F (2,32) = 20.49 and (2,33) =16.19, respectively, P <0.05; Figure 2A and 2B).

Figure 2.

Mean (± SEM) plasma corticosterone (ng/ml) levels in adolescent- or adult exposed males (A) and females (B) treated with 0, 25, or 100 μg/ml of oral CORT for four weeks. Exposure during adolescence was from 30-58 days of age, while exposure during adulthood was from 70-98 days of age. To signify a main effect of dose within a sex, bars that share a letter are not significantly different from one another.

Insulin:

A significant interaction between dose and age of exposure was noted in males (F (2,32) = 6.87, P <, 0.05), such that exposure to the 100 μg/ml dose of CORT significantly increased plasma insulin, but this increase was greater in the adolescent-treated compared to the adult-treated mice (Figure 3A). In females, there was only a main effect of dose (F (2,33) = 14.48, P <, 0.05), with the 100 μg/ml dose of CORT significantly increasing plasma insulin levels in both the adolescent- and adult-treated females (Figure 3D). It is important to note that none of these animals were fasted prior to this terminal blood sample.

Figure 3.

Mean (± SEM) plasma insulin (ng/ml), leptin (ng/ml), and triglyceride (mg/dl) levels in adolescent- or adult exposed males (A, B, and C, respectively) and females (D, E, and F, respectively) treated with 0, 25, or 100 μg/ml of oral CORT for four weeks. Exposure during adolescence was from 30-58 days of age, while exposure during adulthood was from 70-98 days of age. Asterisks indicate significant differences between doses within an age of exposure, while the “#” indicates a significant difference between the ages of exposure at that dose. To signify a main effect of dose within a sex, bars that share a letter are not significantly different from one another.

Leptin:

In males, a dose dependent effect on plasma leptin was detected, with those exposed to the 100 μg/ml dose of CORT showing higher leptin levels than those exposed to the 0 or 25 μg/ml doses, and males exposed to the 25 μg/ml dose showing greater leptin levels than the 0 μg/ml dose ((F (2,32) = 114.2 P < 0.05; Figure 3B). In females, a significant interaction between dose of CORT and age of exposure on plasma leptin levels was observed (F (2,33) = 9.51, P < 0.05), such that the 100 μg/ml dose of corticosterone increased plasma leptin levels significantly higher than the other doses, and this increase was greatest in the adolescent-compared to adult-exposed females (Figure 3E).

Triglycerides:

In both male and female mice, independent of the age of exposure, plasma triglycerides were significantly increased by CORT treatment. In particular, plasma triglycerides were significantly elevated in adolescent- and adult-treated male and female mice exposed to the 100 μg/ml dose of CORT compared to those exposed to either the 0 or 25 μg/ml doses (F (2,32) = 6.39 and (2,33) = 11.38, respectively, P <0.05, Figure 3C and 3F).

Taken together, these data indicate that despite the similar levels of CORT induced by these treatments in both males and females (Figure 2), peripherally circulating metabolic signals, including insulin and leptin, display different patterns depending on the age of exposure, as well as the on the sex of the individual.

CORT-induced changes in Gene Expression in Liver: Age of Exposure and Dose of CORT modulate Gene Expression differently in Males and Females

Nr3c1:

Nr3c1 expression in the liver of males revealed a significant interaction between dose of CORT and age of exposure (F (2,32) = 3.84, P <0.05), such that mice exposed during adolescence to the 0 or 25 μg/ml doses of CORT had higher levels of Nr3c1 expression than their adult-exposed counterparts at those doses. However, both the adolescent- and adult-exposed mice receiving the 100 μg/ml dose of corticosterone had similar lower levels of Nr3c1 expression (Figure 4A). In females, no main effects or interaction between dose of CORT and age of exposure were detected on Nr3c1 expression (Figure 4F).

Figure 4.

Mean (± SEM) Nr3c1, Hes1, Lpl, Serbf1, and Nsdhl relative expression in the liver of adolescent- or adult exposed males (A, B, C, D, and E, respectively) and females (F, G, Η, I, and J, respectively) treated with 0, 25, or 100 μg/ml of oral CORT for four weeks. Exposure during adolescence was from 30-58 days of age, while exposure during adulthood was from 70-98 days of age. “#” indicates a significant difference between the ages of exposure at that dose. To signify a main effect of dose within a sex, bars that share a letter are not significantly different from one another, while “†” indicates a significant main effect of age of exposure within a sex.

Hes1:

Hes1 expression showed significant main effects of both dose of CORT and age of exposure were found in males (F (2,32) = 5.86 and (1,32) = 8.43, respectively, P < 0.05). In the context of dose, the 25 μg/ml group was higher than the 0 and 100 μg/ml groups, which were not different from each other, and for age of exposure, mice treated during adolescence had higher expression levels of Hes1 than those treated during adulthood (Figure 4B). Though a significant interaction was found between dose of corticosterone and age of exposure on Hes1 expression levels in liver of females (F (2,32) = 4.84, P < 0.05), post-hoc tests did not reveal any significant differences between the groups (Figure 4G).

Lpl:

Lpl expression showed a significant interaction between dose and age of exposure in males (F (2,32)= 10.49, P < 0.05), but only a significant main effect of the dose of CORT in females (F (2, 32) = 12.01, P < 0.05). For the interaction in males, though none of the doses of CORT in the adult-exposed groups affected Lpl expression, the adolescent-exposed mice exhibited decreased Lpl expression with increasing doses of CORT. Moreover, Lpl expression was higher in the adolescent-exposed males in the 0 μg/ml group compared to the adult-exposed males in the 0 μg/ml group (Figure 4C). In females, the main effect of the dose of CORT was such that the 100 μg/ml treated females had lower Lpl expression than both the 0 and 25 μg/ml dose groups, which were not different from each other (Figure 4H).

Serbf1:

Serbf1 expression in the liver of males was significantly altered significant with main effects of dose of CORT and age of exposure (F (2,32) = 3.96 and (1,32) = 10.70, respectively, P <0.05). For the marginal main effect of dose, males exposed to the 25 μg/ml dose had higher Serbf1 expression than the males treated with either the 0 or 100 μg/ml dose, while in the context of age of exposure, the males treated during adolescence had higher Serbf1 expression than the males treated during adulthood (Figure 4D). In females, no main effects or interaction between dose of CORT and age of exposure were detected on Serbf1 expression (Figure 4I).

Nsdhl:

Finally, for Nsdhl expression in the liver, main effects of both the dose of CORT and age of exposure were found in males (F (2,32) = 24.44 and (1,32) = 12.21, respectively, P < 0.05), while only a significant main effect of dose was found in females (F (2, 32) = 6.48). For the main effects in males, mice treated with the 0 or 25 μg/ml dose of CORT had greater Nsdhl expression than the mice treated with the 100 μg/ml dose, while the main effect of age of exposure was such that males exposed during adolescence had higher Nsdhl expression than males exposed during adulthood (Figure 4E). In the context of the main effect of dose of CORT in females, mice exposed to the 100 μg/ml dose had lower Nsdhl expression than both the 0 and 25 μg/ml dose groups, which were not different from each other (Figure 4J).

Overall, these data indicate key genes important in metabolic regulation show different patterns of expression that are dependent on both the dose of CORT received and the age when the treatment was administered. Furthermore, these changes in hepatic gene expression appear to be modified by sex of the individual, as the effects of the dose of CORT and the age of exposure show different patterns in the males and females for each of these genes measured.

CORT-induced changes in Immune and Reproductive Tissues

Thymus:

Thymus weights only showed a main effect of dose of CORT in both males and females (F (2,32) = 25.35 and (2,33) = 40.25, respectively, P <0.05; Table 2). Males receiving either the 25 or 100 μg/ml doses of CORT had significantly lighter thymus weights compared to the 0 μg/ml group, independent of age of exposure. In females, only mice receiving the 100 μg/ml dose of CORT had lighter thymus weights than either the 0 or 25 μg/ml groups (Table 2). These data show that the resulting elevated CORT levels produced by these treatments were sufficient to initiate thymic involution, and that its sensitivity appears to be different in males and females.

Table 2.

Mean (+/− SEM) thymus (mg), paired testis (mg), seminal vesicle (SV; mg), and uterine weights (mg) in adolescent- or adult-exposed males and females treated with 0, 25, or 100 μg/ml of oral corticosterone (CORT) for four weeks. Exposure during adolescence (Adol) was from 30-58 days of age, while exposure during adulthood (Adult) was from 70-98 days of age. Numbers that share a letter within a measure and sex are not significantly different from one another, while “†” indicates a significant main effect of age of exposure within a sex.

Group	Thymus	Testes	SVs	Uterus
Male Adol 0	51.1±3.5a	176.7±6.3	152.8±9.2	---
Male Adol 25	27.1±5.6b	174.5±9.9	166.6±15.3	---
Male Adol 100	22.2±6.6b	159.4±6.3	168.7±17.6	---
Male Adult 0	48.7±3.0a	187.7±7.7†	209.9±25.1†	---
Male Adult 25	30.3±4.3b	194.9±7.1†	248.0±20.8†	---
Male Adult 100	22.6±2.9b	184.0±3.5†	214.0±19.5†	---
Female Adol 0	77.7±5.1a	---	---	107.1±9.5
Female Adol 25	58.2±4.0a	---	---	116.1±13.5
Female Adol 100	34.2±5.8b	---	---	101.8±24.2
Female Adult 0	60.1±4.3a	---	---	84.6±8.0
Female Adult 25	56.2±4.3a	---	---	78.8±12.1
Female Adult 100	20.8±3.9b	---	---	100.6±14.1

Reproductive Tissues:

In males, paired testis and seminal vesicle weights were unaffected by the dose of CORT received. However, males exposed to CORT in adulthood had heavier testes and seminal vesicles compared to the younger adult mice exposed to these CORT treatments during adolescence (F (1,32) = 9.38 and 15.01, respectively, P <0.05; Table 2). In females, there were no significant main effects of dose of CORT or age of exposure on uterine weight, nor an interaction between these two variables (Table 2). Thus, these data indicate no obvious alterations in these tissues important in reproductive function; and therefore, suggest these chronic CORT treatments during adolescent development were not severe enough to grossly delay pubertal maturation.

Experiment 2

Dynamics of CORT-induced changes in Body Weight: Age of Exposure Matters in Males, but not Females

Males:

In males, a significant interaction between dose of CORT and day of exposure in the adolescent-exposed males (F (4, 56) =36.08, P <0.05), such that males exposed to the 100 μg/ml dose of CORT during adolescence weighed significantly less than the 0 μg/ml adolescent-exposed males at day 7 (Figure 5A). There were no other significant differences in body weight between the 0 and 100 μg/ml groups throughout the rest of the experiment in adolescent-exposed males. In the adult-exposed males, the repeated measures ANOVA also revealed a significant interaction between dose and day of exposure (F (4, 56) = 83.46, P<0.05). However, here adult-exposed males at the 100 μg/ml dose of CORT weighed significantly more than the 0 μg/ml adult-exposed males at days 14, 21, and 28 of treatment (Figure 5B).

Figure 5.

Mean (± SEM) body weight (g) in adolescent-exposed males (A), adult-exposed males (B), adolescent-exposed females (C), and adult-exposed females (D) treated with 0 (open circles) or 100 (open triangles) μg/ml of oral CORT for four weeks. Exposure during adolescence was from 30-58 days of age, while exposure during adulthood was from 70-98 days of age. “*” indicates the100 μg/ml dose of CORT significantly reduced body weight compared to the 0 μg/ml group, while “**” indicates the 100 μg/ml dose of CORT significantly increased body weight compared to the 0 μg/ml group.

Females:

For the analyses conducted on females, significant interactions between dose of CORT and day of exposure in adolescent- and adult-exposed females was observed (F (4, 52) = 37.58 and (4, 56) = 36.92, respectively, P<0.05). In particular, adolescent-exposed females in the 100 μg/ml group weighed significantly more on day 28 than the adolescent-exposed females in the 0 μg/ml group, while the adult-exposed females in the 100 μg/ml group weighed significantly more on days 21 and 28 of treatment compared to the adult-exposed females in the 0 μg/ml group (Figure 5C and 5D).

Thus, males exposed to CORT during adolescence show a transient decrease in body weight, an effect not observed in the adult-exposed males or females treated during either adolescence or adulthood. Consistent with Experiment 1, changes in body weight are modified by the developmental stage in which the exposure to CORT occurs, particularly in males.

Dynamics of CORT-induced changes in Food and Fluid Intake: Age of Exposure Matters in both Males and Females

Males:

Along with repeatedly measuring body weight, we also assessed weekly food and fluid intake across these treatment periods. In the adolescent-exposed males, we found significant interactions between dose and day of exposure (F (3, 18) = 28.35 and 40.43, respectively, P <0.05), such that males exposed to the 100 μg/ml dose of CORT during adolescence showed significantly greater food and fluid intake at days 21 and 28 of treatment compared to males exposed the 0 μg/ml dose of CORT during adolescence (Figure 6A and 6E, respectively).

Figure 6.

Mean (± SEM) food (g) and fluid (ml) intake of adolescent-exposed males (A and E, respectively), adult-exposed males (B and F, respectively), adolescent-exposed females (C and G, respectively), and adult-exposed females (D and H, respectively) treated with 0 (open circles) or 100 (open triangles) μg/ml of oral CORT for four weeks. Exposure during adolescence was from 30-58 days of age, while exposure during adulthood was from 70-98 days of age. “*” indicates the 100 μg/ml dose of CORT significantly increased food or fluid intake compared to the 0 μg/ml group, “†” indicates a significant main effect of CORT treatment within a sex for that measure.

In males exposed during adulthood, there were significant main effects of both dose and day of exposure on food intake (F (1, 18) = 11.01 and (3, 18) = 14.98, respectively, P <0.05). Specifically, independent of the day food intake was measured, males treated with the 100 μg/ml dose of CORT showed greater food intake than the males treated with the 0 μg/ml dose, while independent of treatment, food intake increased throughout the first 21 days of treatment (Figure 6B). There were no significant effects of dose or day of exposure on fluid intake in the adult-exposed males (Figure 6F).

Females:

Similar to adolescent-exposed males in the context of food intake, a significant interaction between dose and day of exposure was found in adolescent-exposed females (F (3, 15) = 6.40, P <0.05). In particular, females exposed to the 100 μg/ml dose of CORT during adolescence had significantly greater food intake on days 21 and 28 of treatment compared to the adolescent-exposed females at the 0 μg/ml dose of CORT (Figure 6C). For fluid intake, there were significant main effects of both dose and day of exposure in the adolescent-exposed females (F (1, 15) = 8.52 and (3, 15) = 8.23, respectively, P <0.05). Specifically, independent of the day fluid intake was measured, females treated with the 100 μg/ml dose of CORT during adolescence showed greater fluid intake than the females treated with the 0 μg/ml dose, while independent of treatment, fluid intake increased after 21 days of treatment (Figure 6G).

Finally, in females treated in adulthood, there was a significant main effect of day of exposure on food intake (F (3, 18) = 63.43, P <0.05), such that independent of dose, food intake increased throughout first 21 days of treatment (Figure 6D). For fluid intake in the adult-exposed females, despite a significant interaction between dose and time (F (3, 18) = 4.04, P <0.05), post-hoc tests did not detect any significant differences between the groups (Figure 6H). There were no significant main effects of dose or day of treatment on fluid intake in the adult-treated females.

These data do not support a direct association between food and fluid intake on the observed changes in body weight in these adolescent- and adult-treated males and females. However, it generally appears that both males and females treated with CORT during adolescence show greater food and fluid intake than their age-matched controls, an effect not observed in either the males or females treated with CORT in adulthood.

Orexin-A and NYP levels are Unaffected by CORT Exposure in Males and Females

There were no significant effects of dose of CORT on the number of orexin-A-positive cells in the lateral hypothalamus of adolescent- or adult-treated males or females (Figure 7). Similarly, for the NPY-positive fiber RODs in the arcuate nucleus, there were also no significant effects of CORT in either the adolescent- or adult-treated males or females (Figure 8). These neurobiological data suggest that despite the influence of CORT treatments on a variety of somatic, hormonal, hepatic and behavioral measures, there appear to be no obvious changes in levels of these hypothalamic neuropeptides in male or female mice treated either during adolescence or in adulthood.

Figure 7.

Mean (± SEM) orexin-A cell number in adolescent- or adult exposed males (A) and females (B) treated with 0 or 100 μg/ml of oral corticosterone for four weeks. Exposure during adolescence was from 30-58 days of age, while exposure during adulthood was from 70-98 days of age. Panel C is a representative photomicrograph of orexin-A positive cells in the LH. Bar = 100 μm.

Figure 8.

Mean (± SEM) NPY fiber relative optical density (ROD) in adolescent- or adult exposed males (A) and females (B) treated with 0 or 100 μg/ml of oral corticosterone for four weeks. Exposure during adolescence was from 30-58 days of age, while exposure during adulthood was from 70-98 days of age. Panel C is a representative photomicrograph of NPY positive fibers in the arcuate nucleus. Bar = 50 μm.

Discussion

Collectively, our results show that chronic exposure to oral CORT can have profound effects on metabolic function that can be further modified by the developmental stage and sex of the individual. In particular, chronic oral CORT exposure resulted in changes in somatic, hormonal, behavioral, and hepatic gene expression measures, many of which were differentially modulated depending on the age of exposure, with different patterns of results in males and females. Conversely, whether the exposure occurred during adolescence or in adulthood, hypothalamic levels of orexin-A and NPY were similarly unaffected by these treatments in both males and females. These data highlight the unique ways that adolescent development and gonadal sex shape metabolic function in the presence of chronic oral CORT exposure and provide the basis for focused mechanistic experiments in the future.

As previously reported in adult male and female mice, we found that chronic oral CORT exposure leads to a metabolic syndrome-like phenotype, including increased weight gain and adiposity, along with hyperinsulinemia, hyperlipidemia, and elevated triglyceride levels [8-10, 15]. Furthermore, we replicated our earlier results in adolescent-exposed males [15], such that initial corticosterone-induced weight loss was followed by weight gain and increased visceral adiposity, as well as increased terminal plasma insulin, leptin, and triglyceride levels. Our present work comprehensively extends these observations, showing that females also develop a metabolic syndrome-like phenotype in response to chronic oral CORT treatment, whether these treatments occur during adolescence or adulthood. It is important to note that the adolescent-exposed females did show some initial weight loss similar to that exhibited by the adolescent-exposed males. However, CORT-treated adolescent females exhibit significant gain weight like their adult counterparts, an effect not observed in CORT-treated adolescent males. These data suggest that sex can modify the trajectory of CORT-induced metabolic dysfunctions, particularly when the CORT exposure is initiated during adolescent development. Though the specific mediators of these different patterns of weight loss and gain in adolescent-exposed males and females, respectively, remain unclear, it is likely that the onset of gonadal hormone secretion that occurs at this time could interact with the exogenous CORT exposure. Future studies are needed to address whether the presence or absence of gonadal hormones is responsible for mediating any of these age- and sex-related differences. It is also interesting to note that females generally appeared to be less sensitive to the 25 μg/ml dose of CORT across many of these measures, while males showed more dose-dependent responses. Similar to the role of gonadal hormones mentioned above, it is unclear what might mediate these differential sensitivities to CORT in females and males, but it is possible that sex differences in GR expression or mediators downstream of GR activation in CORT sensitive target tissues might play a role.

Even though the specific metabolic outcomes we measured show marked differences between the experimental groups, our plasma CORT data indicate that the 100 μg/ml dose significantly increased circulating CORT to similar levels in both adolescent- and adult-exposed males and females. This leads us to conclude that any observed age-or sex-related difference in our outcome measures cannot simply be attributed to unequal/different exposure to CORT. It should also be noted that these plasma CORT levels in the 100 μg/ml groups are physiologically relevant, as they are about 50% less than what has previously been reported in C57BL/6 mice exposed to restraint stress [31]. The plasma insulin, leptin, and triglyceride data, in conjunction with the visceral WAT results, provide further evidence that chronic oral exposure to 100 μg/ml of CORT leads to a metabolic syndrome-like phenotype in both male and female C57BL/6N mice, independent of the age of exposure. In males, it has been shown that cessation of treatment results in recovery from these metabolic dysfunctions, but with residual effects on bone growth in adolescent-treated males, specifically in the femur [15]. Given the somewhat altered trajectory observed in females, it would be interesting to establish whether females would also show similar recovery patterns from these treatments and whether any skeletal perturbations occur in the adolescent-treated females.

The somatic data from Experiment 2 provide replication and extension of the findings from Experiment 1. Specifically, the body weight changes in Experiment 2 replicate the changes we observed in Experiment 1, as well as previous findings [15]. While we noted increases in food intake across our groups, these increases were statistically significant only on the adolescent-exposed males and females and only during the last two weeks of treatment. We also noted significant increases in fluid intake, but only in the adolescent-treated males, again during the last two weeks of treatment. It is important to note that these food and fluid intake data were derived from group-housed mice (i.e., 2-3 mice per cage). Thus, these data do not represent the intake of individual mice, which may account for the incongruence with previous findings showing significant increases in food intake in CORT-treated adult males [8]. Regardless, these results do indicate significant CORT-induced changes in food and fluid consumption patterns, specifically in adolescent-treated mice and in the latter stages of treatment. It is interesting to note that even though CORT exposure increased food intake in adolescent-treated males, body weight and weight gain were similar in 0 the 100 μg/ml treatment groups during the last two weeks of treatment. Conversely, even though adult-exposed males and females showed no significant changes in food intake, these mice exhibited significant weight gain, at least during the last two weeks of treatment. Thus, there does not appear to be a direct association between food intake and changes in body weight. Altered activity levels, however, may be playing a role, as a previous study noted CORT-induced decreases in home cage activity, particularly during the active dark phase of the light-dark cycle [8]. Future studies incorporating activity monitoring would help elucidate this relationship between, caloric intake, activity levels, and changes in body weight.

Pathological changes in metabolic function can have myriad underpinnings. Our previous work showed significant hepatic changes in response to chronically elevated CORT [9, 32]. Thus, we explored changes in gene expression in the liver following these treatments at both developmental stages, and in both sexes. Given the main manipulation was exposure to CORT, we first investigated changes in the hepatic expression of Nr3c1 (i.e., the glucocorticoid receptor, GR). Our findings revealed a significant interaction of dosage and age of exposure in male mice, an effect not observed in female mice. This finding suggests that the CORT-induced effects in females might not be mediated via changes in GR expression, though changes in protein levels of subcellular localization cannot be ruled out. To further probe transcriptional changes, we investigated changes in Hes1. Hes1 is a transcriptional repressor that can bind to N-boxes in the genome, recruiting chromatin modifiers to control transcriptional activity. Notably, Hes1 appears to play a central role in regulating glucocorticoid-induced gene expression [16], with activation of GR rapidly leading to a silencing of Hes1 activity [16, 33]. The silencing of Hes1 activity then allows for GR-mediated increases of a large number of downstream targets. In our results, Hes1 expression appears higher in adolescent-exposed male mice when compared to adult-exposed males. Remarkably, CORT treatment appears to have the same pattern of effects in both male and female adolescent-treated mice, though these changes did not reach statistical significance. In both adult male and female mice, CORT treatment did not appear to have a marked effect on hepatic Hes1 expression. This pattern of results suggests that the overall difference we observed in males and females may be caused by different transcriptional pathways, at least in the context of GR and downstream gene regulation.

We also probed the expression of hepatic genes, such as Lpl, Nsdhl, and Srebf1, that may contribute more directly to the specific physiologic pathways observed in these phenotypes of weight gain, increased mass of WAT, and increased triglycerides. Plasma lipoprotein lipase (i.e., Lpl/LPL) activity results in increased fatty acids that are taken up by many tissues, including the liver where they give rise to increased lipid droplets [17]. In addition, hepatic Lpl activity contributes to the development of non-alcoholic steatosis [34]. Thus, the significant dose-dependent decrease observed in adolescent males may represent a compensatory response at this developmental stage that is compromised in adult males and females at both ages. The changes in Nsdhl and Srebf1 are also both related to lipid regulation. Serbf1 acts to increase lipogenesis by acting as a transcription factor for low density lipoprotein receptor gene, as well as increasing the activity of other lipogenic genes [18], further adding to systemic lipogenesis. CORT only seemed to affect changes in Srebf1 in males, in a dose-dependent manner, while females did not show any changes in hepatic expression. Relatedly, Nsdhl expression was significantly reduced by CORT in adolescent and adult males, with adolescents overall having higher levels of expression. There was no similar age-related difference observed in females, but CORT did seem to reduce Nsdhl expression in the liver similarly, particularly at the high dose. Together, our initial investigation reveals a complex set of changes in the livers of mice exposed to CORT that show both age- and dose-related interactions, and that females and males have a different pattern of responses. These data further support the notion that there are sex-dependent differences in transcriptional pathways, at least in the liver, that underlie the metabolic dysregulation observed following chronic CORT exposure. Broader-based transcriptomic analyses are needed, as well as a direct comparison between the sexes, to disentangle some of these effects so we can more clearly understand the way hepatic function is affected by high levels of CORT in males and females and during different stages of development.

Despite these age- and sex-dependent effects of chronic oral CORT treatment on our hepatic, somatic, and hormonal measures, we found no difference in the levels of hypothalamic orexin- A and NPY, which are known to modulate metabolism, arousal, and appetite [20, 25, 35-38]. Specifically, we found no effect of oral CORT or age of exposure on orexin-A cell number in the LH or NPY fiber density in the Arc in either males or females. Thus, while chronic oral CORT treatments lead to metabolic changes in males and females, these changes are not accompanied by significant alterations in levels of two integral neuropeptides involved in feeding and metabolism, indicating a dissociation between peripheral and central responses to these CORT treatments. Along these lines, two recent studies revealed that adult mice lacking hypothalamic AgRP or ghrelin receptors still exhibited oral CORT-induced metabolic abnormalities [10, 39]. Therefore, it will be important for future experiments to establish whether neurobiological substrates contribute to these metabolic effects of chronic CORT exposure.

Together, these studies help clarify the effects of chronically elevated CORT levels on somatic, hormonal, hepatic, and neurobiological endpoints typically associated with metabolic function. Moreover, using mice of both sexes and at two unique stages of maturation, these data provide a solid foundation for future studies to investigate the potential mechanisms that mediate the development of obesity and metabolic syndrome, major morbidities that continue to affect a large proportion of the global population, particularly among adolescents [40-42].

Highlights.

• Metabolic syndrome continues to rise in both adolescents and adults

• Using chronic oral CORT we show age- and sex-dependent effects on weight gain

• However, we show a dissociation in weight gain and adiposity in adolescent males

• Across groups we show changes in hepatic genes important in lipid regulation

• CORT leads to metabolic issues in males and females and is modulated by development