Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2021 Jan 1.
Published in final edited form as: J Endocrinol. 2019 Oct;243(1):27–42. doi: 10.1530/JOE-19-0230

The glucocorticoid receptor in osteoprogenitors regulates bone mass and marrow fat

Jessica L Pierce 1, Ke-Hong Ding 2, Jianrui Xu 2, Anuj K Sharma 1, Kanglun Yu 1, Natalia del Mazo Arbona 1, Zuleika Rodríguez-Santos 1, Paul J Bernard 3, Wendy B Bollag 1,4,5,6, Maribeth H Johnson 2, Mark W Hamrick 1,5, Dana L Begun 7, Xing-Ming Shi 2, Carlos M Isales 2,5,8, Meghan E McGee-Lawrence 1,5
PMCID: PMC6938567  NIHMSID: NIHMS1537244  PMID: 31370004

Abstract

Excess fat within bone marrow is associated with lower bone density. Metabolic stressors such as chronic caloric restriction (CR) can exacerbate marrow adiposity, and increased glucocorticoid signaling and adrenergic signaling are implicated in this phenotype. The current study tested the role of glucocorticoid signaling in CR-induced stress by conditionally deleting the glucocorticoid receptor (GR) in bone marrow osteoprogenitors (Osx1-Cre) of mice subjected to CR and ad libitum diets. Conditional knockout of the GR (GR-CKO) reduced cortical and trabecular bone mass as compared to wildtype (WT) mice under both ad libitum feeding and CR conditions. No interaction was detected between genotype and diet, suggesting that the GR is not required for CR-induced skeletal changes. The lower bone mass in GR-CKO mice, and the further decrease in bone by CR, resulted from suppressed bone formation. Interestingly, treatment with the β-adrenergic receptor antagonist propranolol mildly but selectively improved metrics of cortical bone mass in GR-CKO mice during CR, suggesting interaction between adrenergic and glucocorticoid signaling pathways that affects cortical bone. GR-CKO mice dramatically increased marrow fat under both ad libitum and CR-fed conditions, and surprisingly propranolol treatment was unable to rescue CR-induced marrow fat in either WT or GR-CKO mice. Additionally, serum corticosterone levels were selectively elevated in GR-CKO mice with CR, suggesting the possibility of bone-hypothalamus-pituitary-adrenal crosstalk during metabolic stress. This work highlights the complexities of glucocorticoid and β-adrenergic signaling in stress-induced changes in bone mass, and the importance of GR function in suppressing marrow adipogenesis while maintaining healthy bone mass.

Keywords: skeleton, marrow adiposity, osteoprogenitor, osteoblast, nutrition, metabolic stress, corticosterone, sympathetic tone, β-adrenergic

INTRODUCTION

The bone marrow niche houses stem cell populations that give rise to blood, immune, bone, and fat cells within the skeletal microenvironment. The fate of bone marrow progenitor cells can be influenced by hormones and nutrient signals from the circulation, establishing a relationship between whole-body physiology and bone health (Li et al., 2016). A vital component of the bone marrow microenvironment is the bone marrow stromal cell (BMSC), which is a multipotent progenitor capable of differentiating into osteogenic and adipogenic cell populations within bone (Baker et al., 2015). The differentiation fate of the BMSCs can be influenced by genetic predispositions, dietary and lifestyle factors, stress and injury responses, and epigenetics. Osteoblastogenesis is influenced by pathways including Wnt/β-catenin, transforming growth factor-β (TGF-β), bone morphogenetic proteins (BMPs), and glucocorticoid signaling, among others (Kang et al., 2007, Bennett et al., 2007, Day et al., 2005, Zhang et al., 2012). There is a delicate balance between these mechanisms, and interruptions to the homeostasis of the bone marrow microenvironment can dramatically shift BMSC differentiation patterns at the expense of tissue function and bone health.

Recent studies have uncovered an intricate relationship between caloric intake and bone. For example, chronic caloric deficit induces high bone marrow adipogenesis accompanied by osteopenia in humans and mice (Fazeli et al., 2013, Singhal and Bredella, 2018, Devlin et al., 2010, Periyasamy-Thandavan et al., 2015). It has also been reported that a low-calorie diet reduces bone mass in part by decreasing cortical bone mass while largely sparing trabecular bone in mice (Hamrick et al., 2008). There is evidence that the phenotype produced by caloric restriction (CR) is mediated in part by glucocorticoid signaling (Cawthorn et al., 2016). The metabolic stress of caloric restriction can stimulate counter-regulatory hormone secretion, including glucocorticoids, that may interfere with osteoblast function and drive marrow adipose tissue expansion (Cawthorn et al., 2016, Devlin et al., 2010). The enzyme 11β-hydroxysteroid dehydrogenase type 1 (Hsd11b1)—which converts inactive glucocorticoids to their active forms in the periphery—also mediates upregulated glucocorticoid metabolism throughout the body and locally within bone as a result of caloric restriction (Gathercole et al., 2013, Tomlinson et al., 2004). Additionally, it has been reported that in vitro pharmacological inhibition of Hsd11b1 activity protects osteoblasts from glucocorticoid-induced cellular dysfunction (Wu et al., 2013). These data suggest a potential role for aberrant glucocorticoid signaling in the CR-induced bone phenotype, but the mechanistic role of modulators downstream of Hsd11b1 remain largely unknown.

Activated glucocorticoids bind the glucocorticoid receptor (GR) within target tissues and drive downstream stress-response signaling cascades (Almeida et al., 2011). Once bound to glucocorticoids, the GR translocates to the nucleus to bind glucocorticoid response elements (Nixon et al., 2013, Lim et al., 2015) or associate with transcription factors that act on genomic regulatory elements (Hua et al., 2016a, Hua et al., 2016b). While physiological levels of endogenous glucocorticoids are necessary for osteoblastogenesis and bone formation (Zhou et al., 2013), an excess of circulating glucocorticoids can compromise bone mass by inhibiting osteoblast differentiation and inducing osteoblast and osteocyte apoptosis (Rauch et al., 2010, O’Brien et al., 2004, Weinstein, 2012, Weinstein et al., 1998). Downstream glucocorticoid signaling may also drive marrow adipogenesis, as it has been reported that glucocorticoids induce adipocyte commitment and contribute to the fatty infiltration of soft tissues (e.g., liver and skeletal muscle) as well as bone (Hamrick et al., 2016, Canalis et al., 2004, Kuo et al., 2013, Papanastasiou et al., 2017). These findings further support a role for the GR in mediating the bone loss and increased marrow fat caused by CR and suggest that the GR is a key in the response of bone to this facet of metabolic stress.

Sympathetic tone may also be involved in glucocorticoid-induced marrow adiposity (Elefteriou, 2005). Osteoblasts express β-adrenergic receptors (β-AR) that, when activated, initiate signaling cascades that result in decreased osteoblast number and activity as well as loss of bone mass. Importantly, the sensitivity of osteoblast to adrenergic signaling is also increased by glucocorticoid treatment (Ma et al., 2011, Bonnet et al., 2008). β-adrenergic receptor activity also stimulates adipogenesis and mediates CR-induced marrow adiposity, whereas treatment with the β-adrenergic antagonist propranolol in a chronic caloric restriction model partially rescues the trabecular bone loss and marrow fat phenotype in mice and rats (Baek et al., 2014b, Baek and Bloomfield, 2012). Moreover, propranolol protects against the loss of bone mass associated with a high calorie diet (Baek et al., 2014a). Despite these findings, there is still a knowledge gap regarding the functional role of the GR in bone marrow lipid storage in the caloric restriction model when β-blockade occurs.

The current study sought to test the role of the glucocorticoid receptor in caloric restriction-induced changes in bone and marrow fat. Within this model, we also sought to deduce the effect of sympathetic tone on the phenotype of metabolically stressed GR-insufficient bone through the use of the β-blocker propranolol.

MATERIALS AND METHODS

Animals and diet administration

All experiments followed NIH guidelines and were approved by the Institutional Animal Care and Use Committee at Augusta University. As described previously, GR was inactivated through Cre recombinase (Cre)-mediated excision of exons 1C and 2 of the GR gene (Figure 1A). This modification deletes 50% of the mature GR protein, encompassing the translation start site and the tau1 transactivation domain, and has previously been shown to effectively disrupt GR action (Brewer et al., 2003, Kim et al., 2006). Osteoprogenitor-specific inactivation of GR was accomplished through Osx1 promoter-driven Cre expression as described previously (Rodda and McMahon, 2006, McGee-Lawrence et al., 2016) to generate GR conditional knockout mice (GR-CKO; GRfl/fl:Osx1-Cre+) and control littermates (GR-WT; GRfl/fl:Osx1-Cre-) for this study (n=7 to 8 mice/group; Figure 1A). While inclusion of an additional “Cre+ control” (i.e., GR+/+:Osx1-Cre+ for this study) is preferred for knockout models generated with the Osx1-Cre line, due to the presence of a mild, early skeletal phenotype arising from the Osx1-Cre itself, in pilot studies of 6-month old male mice, we observed that marrow fat was elevated specifically in GRfl/fl:Osx1-Cre+ mice, whereas there were no differences observed between GRfl/flOsx1-Cre- (i.e., “Cre- control”) mice and GR+/+:Osx1-Cre+ (i.e., “Cre+ control”) mice (Supplemental Figure 1). Furthermore, the skeletal phenotype of Osx1-Cre+ control mice is reported to resolve by 12 weeks (~3 months) of age, and all studies reported here were only initiated at 6 months of age (Davey et al., 2012). Consequently, we focused the current study on GR conditional knockout mice (GR-CKO; GRfl/fl:Osx1-Cre+) and Cre- control littermates (GR-WT; GRfl/fl:Osx1-Cre-). While the Osx1-Cre used does possess a doxycycline-suppressible Tet-off element, the mice used here were not treated with doxycycline to suppress Cre because studies were initiated after the Osx1-Cre skeletal phenotype was expected to resolve. Consequently, mice used for these studies experienced constitutive expression of Cre in Osx1-expressing cells throughout their development. Female mice were specifically selected for study because in pilot work we observed similar trends of elevated marrow adiposity in both male and female GR-CKO mice (Supplemental Figure 1), but trends were much more pronounced in females, reducing the number of animals needed for study. Mice were genotyped by PCR amplification of tail biopsy DNA isolates with appropriate negative and positive controls. Mice were housed in standard rodent cages on a 12 hour light / 12 hour dark schedule. At 6 months of age, female GR-WT and GR-CKO mice were randomly assigned to either ad libitum (AL) or caloric restriction (CR) feeding for 9 weeks (Figure 1B). Caloric intake was determined by the average weight of food pellets consumed daily, and caloric restriction was defined by a given fraction of the calculated average caloric intake provided to individual AL-fed mice daily. Mice subjected to CR were provided a 10% calorie deficit for one week followed by a 25% calorie deficit for eight weeks. A subset of CR mice received drinking water with 0.01% propranolol (PRO, 0.1 mg/ml; Sigma-Aldrich #P0884) for the duration of the experiment, whereas other mice received only normal water (Vehicle, VEH); drinking water was available to both groups ad libitum. This dosage of propranolol was chosen to match a previous study, where administration of ~0.1 mg/ml propranolol in the drinking water (equating to a dosage of approximately 6 mg propranolol per kg body mass) was sufficient to prevent accumulation of bone marrow adipocytes with caloric restriction in female rats (Baek and Bloomfield, 2012). Calcein injections (10 mg/kg) were administered at 5 and 2 days before sacrifice for measurements of cortical bone mineralization. Mice were sacrificed via carbon dioxide inhalation, and were not fasted the evening prior to sacrifice.

Figure 1. Confirmation of GR knockdown in GR-CKO mice.

Figure 1.

Open in a new tab

(A) Model for glucocorticoid receptor (GR) conditional deletion with Osterix1-Cre (Osx1-Cre) to generate an osteoprogenitor-specific knockout model. (B) Timeline for ad libitum (AL), caloric restriction (CR), and CR + propranolol (PRO) treatment conditions. (C) Expression of GR and Cre RNA was measured by qPCR in BMSC subjected to osteogenic culture for 7 days. n=4 biological replicates per group, *p<0.05 vs. GR-CKO WT for each primer pair. D) GR protein expression was detected via immunohistochemistry in the tibia; the inset label on each image indicates which antibody was used (GR: glucocorticoid receptor, IgG: mouse IgG isotype control). GR-CKO mice demonstrated reduced intensity of GR staining in the growth plate and in cortical bone (e.g., osteocytes). Images are representative of staining patterns observed across the entire group for each genotype (n=7 animals per group); each row of images is from a different mouse. Growth plate images were captured with a 20× objective, whereas cortical bone images were captured with a 40× objective, as indicated.

Body composition and bone mineralization

Whole body dual energy x-ray absorptiometry (DXA; Kubtec Digimus, KUB Technologies, Milford, CT) was performed on isoflurane-anesthetized mice two days before sacrifice. Bone mineral content (BMC) and bone mineral density (BMD) for whole mouse and femur mid-diaphysis regions of interest were calculated using the manufacturer’s software. All DXA analyses excluded heads and ear tags. Percent fat mass was calculated as the fraction of the total body pixels interpreted by the manufacturer’s software as fat tissue via densitometry. Final body mass measurements were recorded 5 days before sacrifice.

Isolation of murine BMSC, cell culture, and analysis of gene expression

Primary murine bone marrow stromal cells (BMSC) from the AL-fed groups were harvested from one femur and two humerii via bone marrow flush with basal medium composed of minimum essential media (MEM)-α, 20% fetal bovine serum (Atlanta Biologicals, Flowery Branch, GA), 1% non-essential amino acids (Gibco #11140–050), and 1% antibiotic/antimycotic (Gibco #15240–062) as previously described (McGee-Lawrence et al., 2016). BMSC-derived osteoblasts were generated via culture in osteogenic induction medium (basal medium + 50 μg/mL ascorbic acid, 10 mM β-glycerophosphate, and 10−7 M dexamethasone), which began at seeding (Day 0), and were maintained under osteogenic conditions until harvest after 7 days in culture. Media changes for all cells in culture were performed every three days until harvest. Total RNA was isolated from the cultures in TRIzol (Invitrogen) following the manufacturer’s instructions. RNA quality was assessed by absorbance at 260 and 280 nm (Nanodrop, Thermo Scientific). RNA (1000 ng) was reverse-transcribed from four biological replicates per genotype (where each replicate represented an independent well of a 6-well plate) into cDNA using a Superscript III cDNA synthesis kit (Superscript III First Strand synthesis (#18080–051), Life Technologies) and a programmable thermal cycler (Bio-Rad). Expression levels of mRNA for the GR (official gene name: nuclear receptor subfamily 3, group C; Nr3c1) were quantified by subjecting cDNA to real-time PCR amplification (37.5 ng cDNA per 15 μl reaction volume reaction, run in triplicate) using a Bio-Rad CFX Connect system, SYBR green reagent (Quanta Biosciences #95054–500) and primers designed to bind within the floxed region of the GR Exon 2. Expression levels of Cre mRNA were also assessed as previously described (McGee-Lawrence et al., 2013, McGee-Lawrence et al., 2016). Gene expression levels were quantified using the comparative threshold cycle (2−ΔΔCt) method (McGee-Lawrence et al., 2016). Glyceraldehyde-3-phosphate dehydrogenase (GAPDH) was used as the internal control (housekeeping gene) for normalization. Primer sequences are shown in Table 1.

Table 1.

Primer sequences used for PCR analyses of mRNA expression. Note that “GR” refers to the gene Nr3c1, the murine glucocorticoid receptor.

Primer pair name Target / amplicon size Forward primer, 5’−3’ Reverse primer, 5’−3’
GR primer pair 1 GR Exon 2, 293 bp CTCCCCCTGGTAGAGACGAA TCCCATGGACAGTGAAACGG
GR primer pair 2 GR Exon 2, 174 bp GGGGAATGACTTGGGCTACC AACTCCTTCTCTGTCGGGGT
GR primer pair 3 GR Exon 2, 78 bp CTCAATAGGTCGACCAGCCG CTTCTCTGTCGGGGTAGCAC
GR primer pair 4 GR Exon 2, 150 bp CATGGCGTGAGTACCTCTGG TCCAGACCCTTGGCACCTAT
Cre Cre, 200 bp ACCAGCCAGCTATCAACTCG TTACATTGGTCCAGCCACC
Gapdh Gapdh, 59 bp GGGAAGCCCATCACCATCT GCCTCACCCCATTTGATGTT
Open in a new tab

Micro-computed tomography, histology, histomorphometry, and immunohistochemistry

Femora and tibiae were harvested at sacrifice and fixed in 10% neutral buffered formalin for 24 hours followed by storage in 70% EtOH. Femora were scanned by micro-computed tomography (microCT, 10.5 μm isotropic voxel size; Scanco VivaCT 40,) with settings of 70 kV and 300 ms for analysis of cortical and trabecular bone architecture. The cortical bone region of interest was equal to 5% of total bone length, with a proximal starting point at 50% of total bone length. The trabecular bone region of interest was equal to 7% of total bone length, with a proximal starting point at 82% of total bone length. Architectural parameters (e.g., cortical bone area, trabecular number and spacing, trabecular volume) were calculated with the manufacturer’s software.

Following microCT analysis, femora were embedded in methyl methacrylate (MMA) for microtome sectioning and mounting (5 μm sections; Leica Biosystems, Germany). Calcein-labeled sections were imaged on a fluorescence microscope (20× objective, Olympus IX70; Olympus Life Science, Waltham, MA) with a digital camera attachment (Qicam; QImaging) for dynamic histomorphometry measurements in the distal femoral metaphyseal trabecular bone with Bioquant Osteo. Mineralizing surface (MS/BS, %) was quantified as 1/2 single labeled surface + double labeled surface normalized to bone surface. Mineral apposition rate (MAR, μm/day) was quantified as inter-label width divided by time between calcein injections. Bone formation rate (BFR/BS, μm/day) was quantified as the product of MAR and MS/BS. Tibiae were decalcified for two weeks in 15% EDTA followed by paraffin embedding and microtome sectioning. Sections for each tibia were stained with either hematoxylin and eosin (H&E; Sigma #HHS16, #H110316) or a commercially available tartrate-resistant acid phosphatase kit (TRAP; Sigma-Aldrich #386A) and mounted using a xylene-based mounting medium. Images were captured with 10× (H&E) and 20× (TRAP) objectives on a brightfield microscope (Olympus IX70). Bone marrow adiposity was measured from the H&E stained sections as adipocyte ghost area normalized to marrow area (Ad.Ar/Ma.Ar, %) and adipocyte density (N.Ad/Ma.Ar, #/mm2). Osteoclast number (N.Oc/BS, #/mm) and osteoclast surface (Oc.S/BS, %) were quantified from the TRAP-stained sections in the proximal tibia metaphyseal trabecular bone using Bioquant Osteo software. All histological data were collected in a blinded fashion.

Serial sections from the decalcified tibiae were immunohistochemically stained to detect the localization of the GR, using methodology as previously described (McGee-Lawrence et al., 2013, Hagan et al., 2019). Briefly, paraffin sections were deparaffinized in xylene and rehydrated through graded alcohols to water. Tissue sections were heated in a citrate buffer solution for 10 minutes for antigen retrieval followed by incubation for 2 hours at room temperature with a Glucocorticoid Receptor monoclonal antibody (BuGR2; 1:50 dilution, #MA1–510, Invitrogen) or mouse IgG isotype control (1:50 dilution, I-2000; Vector Laboratories). The sections were then incubated in chromogens and were detected with a polyvalent secondary HRP kit (Abcam, ab93697) and 3,3′-diaminobenzidine (DAB, 10 minutes incubation; Sigma-Aldrich, D3939) as per manufacturer’s instructions. Images were captured using a microscope (Olympus IX-70) and digital camera (QImaging QICam and QCapture software).

Enzyme-Linked Immunosorbent Assays (ELISAs)

Serum samples were obtained by cardiac puncture at sacrifice followed by centrifugation in serum separator tubes (BD Microtainer #365967, Franklin Lakes, NJ) and stored at −80°C until analysis. A mouse TRAcP-5b ELISA (MouseTrap™; #SB-TR103, Immunodiagnostic Systems, Gaithersburg, MD) was used to measure the circulating bone resorption marker tartrate-resistant acid phosphatase 5b. Additional aliquots of serum were analyzed in a rat/mouse procollagen I intact N-terminal (P1NP) enzyme immunoassay (#AC-33F1, Immunodiagnostic Systems, Gaithersburg, MD) for quantification of bone formation markers. Serum glucocorticoids were measured using a corticosterone ELISA (#ab108821, Abcam, Cambridge, MA) using a 100-fold dilution of mouse serum and provided protocols. All ELISA plates were measured with a microplate reader (BioTek Synergy HT, Winooski, VT) using Gen5 software and readings at 405 (P1NP) and 450 nm (TRAcP-5b, Corticosterone) wavelengths, as recommended for each assay. Serum data were normalized to calibrants and standards provided in respective ELISA kits.

Statistical analyses

After checking for normality and appropriate distribution, the effects of diet were compared between WT and GR-CKO mice by two-factor (2 genotype × 2 diet) analysis of variance (ANOVA) with interaction using JMP Pro 14 statistical software (SAS Institute, USA). The effects of propranolol intervention on the caloric restriction diet background were compared between WT and GR-CKO mice with two-factor ANOVA (2 genotype × 2 treatment) with interaction. Histomorphometric measurements of osteoclasts (N.Oc/BS and Oc.S/BS) were rank-transformed prior to analysis to stabilize the variance across groups and to reduce the influence of outliers. For gene expression analyses, GR-WT and GR-CKO groups were compared via t-tests. Statistical significance was defined by 95% confidence (p<0.05). Box plots in figures show median, quartiles, and outlier fences for each dataset, where outlier fences represent first quartile - 1.5*(interquartile range) and third quartile + 1.5*(interquartile range).

RESULTS

Confirmation of GR knockdown in GRfl/fl:Osx1-Cre+ mice.

BMSC and skeletal tissue were harvested from the GR-CKO and GR-WT mice on ad libitum diets to assess the efficiency of GR knockdown in the GR-CKO mice. BMSC from the GR-CKO mice subjected to osteogenic culture demonstrated an approximately 90% knockdown of GR mRNA levels as compared to cells from GR-WT mice (Figure 1C). These same BMSC cultures expressed high levels of Cre mRNA (+55 fold signal as compared to GR-WT cultures). To confirm these results at the protein level in bone tissue, we performed immunohistochemistry against the GR. These studies demonstrated robust knockdown of the GR in Osx1-Cre-targeted regions including the pre-hypertrophic chondrocytes and osteo-lineage cells (Figure 1D). These studies confirmed that the GR-CKO mice were deficient in skeletal expression of the glucocorticoid receptor.

GR deletion in Osx-expressing cells reduced bone mass under ad libitum and caloric restriction feeding conditions.

GR-WT and GR-CKO mice were subjected to ad libitum or caloric restriction feeding conditions for 9 weeks. Body mass was reduced by both GR-deficiency (−10.6%, p=0.0002) and by caloric restriction (−17.6%, p<0.0001), but there was not a significant interaction between these factors (p=0.453), meaning that CR decreased body mass similarly in GR-WT and GR-CKO mice (Figure 2A). Percent fat mass was not different between groups (data not shown; p>0.203), perhaps because female mice resist loss of fat mass during caloric restriction (Cawthorn et al., 2016, Li et al., 2010b, Shi et al., 2007). While no differences were observed in whole body bone mineral density or bone mineral content between groups (p>0.516, data not shown), both genotype and CR diet administration negatively impacted cortical bone mineralization in the femur as measured by DXA (Figure 2B–D). Specifically, femur BMD and BMC were reduced by GR-deficiency (−5.3% and −7.6% respectively; p<0.004), and by CR (−10.4% and −13.9% respectively, p<0.0001) with no interaction effect (p>0.535, Figure 2C–D). Cortical bone geometry was similarly affected; for example, cortical bone thickness was reduced in GR-CKO mice (−7.6%, p=0.0001) and by CR (−15.9%, p<0.0001), but no interaction effects were observed (p>0.736; Figure 2E and Table 2). Taken together, these data support the idea that both GR-deficiency and caloric restriction are deleterious for cortical bone, but CR effects are not exacerbated in GR-CKO mice.

Figure 2. GR-CKO mice are smaller and have less bone mass than GR-WT mice.

Figure 2.

Open in a new tab

(A) Body mass of animals at the conclusion of study (n=7 to 8 per group) and (B) representative images of body composition dual x-ray absorptiometry scans for GR-WT and GR-CKO mice (white=bone mass, blue=lean mass, green=fat mass). Larger white boxes on representative images outline area of exclusion from analysis. Smaller red boxes outline area of analysis for (C) femur bone mineral density (BMD) and (D) femur bone mineral content (BMC). Micro-computed tomography analyses for femur (E) cortical bone thickness and (F) trabecular bone separation are shown quantitatively. Boxes show median, quartiles, and outlier fences for each dataset. Each data point represents one mouse. Statistical analyses for comparing the influence of diet (blue text) and propranolol intervention (red text) between GR-WT and GR-CKO mice are shown beneath each graph.

Table 2.

Micro-computed tomography (microCT) analysis of bone architecture, comparing the effects of diet (Ad libitum vs. caloric restriction) and genotype (WT vs. GR-CKO). Parameter means and standard deviations are presented. P-values for variables and interactions assessed by two-factor analysis of variance (ANOVA) are shown.

Parameter Mean ± SD pgenotype pdiet pgenotype × diet interaction
Ct. B.Ar (mm2) 0.751 ± 0.046 0.681 ± 0.030 0.647 ± 0.055 0.560 ± 0.043 <0.0001 <0.0001 0.613
Ct. T.Ar (mm2) 1.542 ± 0.106 1.459 ± 0.093 1.516 ± 0.142 1.349 ± 0.069 0.004 0.094 0.286
Ct. B.Ar/T.Ar 0.488 ± 0.017 0.468 ± 0.015 0.427 ± 0.016 0.415 ± 0.020 0.021 <0.0001 0.527
Ct. Imax (mm4) 0.180 ± 0.026 0.161 ± 0.019 0.163 ± 0.030 0.123 ± 0.018 0.002 0.005 0.240
Ct. Imin (mm4) 0.104 ± 0.012 0.089 ± 0.010 0.089 ± 0.014 0.070 ± 0.007 0.0003 0.0005 0.690
Ct.Th (mm) 0.194 ± 0.007 0.180 ± 0.004 0.163 ± 0.008 0.152 ± 0.010 0.0001 <0.0001 0.736
Ct. T.BMD (mg/cm3) 1030.2 ± 19.9 1008.1 ± 14.2 1016.3 ± 25.5 979.8 ± 38.55 0.007 0.044 0.474
Tb. BV/TV 0.012 ± 0.007 0.007 ± 0.004 0.013 ± 0.005 0.011 ± 0.006 0.130 0.237 0.371
Tb.Th (mm) 0.040 ± 0.006 0.041 ± 0.005 0.037 ± 0.005 0.036 ± 0.006 0.924 0.068 0.721
Tb.N (1/mm) 2.110 ± 0.182 1.573 ± 0.271 2.250 ± 0.371 1.997 ± 0.348 0.002 0.020 0.221
Tb.Sp (mm) 0.480 ± 0.043 0.668 ± 0.122 0.459 ± 0.079 0.517 ± 0.080 0.0007 0.012 0.052
WT AL CKO AL WT CR CKO CR
Open in a new tab

With regards to trabecular bone, although trabecular bone volume fraction was not different between groups (p>0.130), subtle but statistically significant differences were observed in the trabecular bone properties Tb.N and Tb.Sp between groups. Trabecular number was decreased by GR-deficiency (−17.5%, p=0.002) and increased by caloric restriction (+14.9%, p=0.020), whereas trabecular separation was increased by GR-deficiency (+25.2%, p=0.0007) but decreased by caloric restriction (−14.7%, p=0.012) (Figure 2F). Taken together, these data suggest that loss of GR in Osx-expressing cells is mildly detrimental for trabecular bone, but caloric restriction minimally impacts trabecular bone properties, as previously reported (Hamrick et al., 2008).

Propranolol treatment differentially affected cortical bone mass in GR-WT and GR-CKO mice subjected to CR.

Propranolol treatment had little to no impact on body mass, whole body DXA measurements, and trabecular bone in the calorically restricted GR-WT and GR-CKO mice, except for stimulating a mild decrease in trabecular bone thickness (−11.2%, p=0.023) that was comparable between GR-WT and GR-CKO mice (Figure 1C, 1F, Table 3). In cortical bone, however, interaction effects between genotype and drug administration were observed for several properties including cortical bone moment of inertia, cortical bone area, and femur BMC (Figure 1E–G, Table 3). These data suggest that propranolol treatment differentially affected cortical bone in the calorically restricted GR-WT and GR-CKO mice, with a mild detrimental effect on cortical bone in GR-WT mice but a mild stimulatory effect on cortical bone mass in GR-CKO mice.

Table 3.

Micro-computed tomography (microCT) analysis of bone architecture, comparing the effects of drug (vehicle vs. propranolol) and genotype (WT vs. GR-CKO). Parameter means and standard deviations are presented. P-values for variables and interactions assessed by two-factor analysis of variance (ANOVA) are shown.

Parameter Mean ± SD Pgenotype Pdrug Pgenotype × drug interaction
Ct. B.Ar (mm2) 0.647 ± 0.055 0.560 ± 0.043 0.607 ± 0.068 0.603 ± 0.043 0.025 0.941 0.041
Ct. T.Ar (mm2) 1.516 ± 0.142 1.349 ± 0.069 1.485 ± 0.099 1.442 ± 0.078 0.006 0.394 0.095
Ct. B.Ar/T.Ar 0.427 ± 0.016 0.415 ± 0.020 0.408 ± 0.025 0.418 ± 0.012 0.902 0.244 0.127
Ct. Imax (mm4) 0.163 ± 0.030 0.123 ± 0.018 0.151 ± 0.025 0.141 ± 0.017 0.004 0.745 0.079
Ct. Imin (mm4) 0.089 ± 0.014 0.070 ± 0.007 0.083 ± 0.013 0.082 ± 0.010 0.022 0.538 0.042
Ct.Th (mm) 0.163 ± 0.008 0.152 ± 0.010 0.154 ± 0.014 0.157 ± 0.008 0.270 0.588 0.051
Ct. T.BMD (mg/cm3) 1016.3 ± 25.5 979.8 ± 38.5 1000.7 ± 20.8 983.9 ± 11.8 0.008 0.546 0.301
Tb. BV/TV 0.013 ± 0.005 0.011 ± 0.006 0.021 ± 0.009 0.011 ± 0.004 0.018 0.111 0.065
Tb.Th (mm) 0.036 ± 0.005 0.036 ± 0.006 0.034 ± 0.003 0.030 ± 0.005 0.229 0.023 0.371
Tb.N (1/mm) 2.250 ± 0.371 1.997 ± 0.348 2.505 ± 0.385 1.991 ± 0.191 0.003 0.306 0.282
Tb.Sp (mm) 0.459 ± 0.079 0.517 ± 0.080 0.410 ± 0.063 0.510 ± 0.045 0.003 0.263 0.396
WT CR VEH CKO CR VEH WT CR PRO CKO CR PRO
Open in a new tab

Bone formation was impaired by caloric restriction and GR-CKO in vivo

To determine how bone remodeling activity contributed to the skeletal phenotypes observed with GR-CKO and caloric restriction, ELISAs were performed on sera isolated from the mice to quantify markers of bone formation and resorption. Circulating markers were verified in a site-specific fashion through histomorphometric measurements of metaphyseal trabecular bone in the proximal tibia (TRAP staining) and distal femur (calcein measurements). Bone formation activity was suppressed by both GR-deficiency and by caloric restriction, as seen by changes in serum P1NP levels (−41.3% in GR-CKO vs. GR-WT, −42.8% in GR vs. AL) and trabecular bone dynamic histomorphometry (Figure 3A–D). No interaction effects were observed, suggesting that CR suppressed bone formation similarly between GR-WT and GR-CKO mice (p>0.184). Histomorphometric measurements of trabecular bone osteoclasts suggested a similar pattern of decreased bone resorption with both GR-deficiency and caloric restriction, with no interaction effects (Figure 3E–G). In contrast, serum TRAcP5b levels, which serve as a circulating marker of osteoclast number throughout the entire body (Alatalo et al., 2003, Halleen et al., 2002), were suppressed by GR-deficiency (−41.0% in GR-CKO vs. GR-WT, p=0.006) but were not impacted by caloric restriction in this study (Figure 3H). Taken together, these data support the assertion that the reduced bone mass observed with GR-insufficiency and with caloric restriction could be attributed to decreased bone formation activity, rather than high levels of bone resorption. Propranolol treatment had little impact on bone remodeling activity, except for a mild suppression of serum P1NP levels that was similar between genotypes (−24.5%, p=0.014; Figure 3).

Figure 3. GR conditional deletion and caloric restriction suppress bone remodeling.

Figure 3.

Open in a new tab

(A) Representative images of calcein-labeled bone surface from each group. Arrows in representative images highlight areas of active mineralization (original magnification: 10× objective). (B) Mineralizing surface (MS/BS) and (C) bone formation rate (BFR/BS) for each condition. (D) Serum levels of procollagen type I N-terminal (P1NP, ng/mL), a circulating marker of bone formation activity. (E) Representative images of tartrate-resistant acid phosphatase (TRAP) staining on tibiae (original magnification: 20× objective). (F) Quantification of TRAP-positive osteoclast surface and (G) osteoclast number (N.Oc) normalized to trabecular bone surface. (H) Enzyme-linked immunosorbent assay (ELISA) measurement of serum concentrations of TRAcP-5b, a circulating marker of osteoclast number. Boxes on each graph show median, quartiles, and outlier fences for each dataset. Each data point represents one mouse. Statistical analyses for comparing the influence of diet (blue text) and propranolol intervention (red text) between GR-WT and GR-CKO mice are shown beneath each graph.

GR-insufficient mice accumulated more marrow fat than GR-WT mice in vivo

Previous work demonstrated that chronic caloric restriction induced marrow fat accumulation in wildtype mice, and that β-blockade with propranolol partially rescued this CR-induced marrow fat phenotype (Cawthorn et al., 2016, Devlin et al., 2010, Periyasamy-Thandavan et al., 2015, Baek et al., 2014b). To determine whether the glucocorticoid receptor is a mechanistic factor in caloric restriction-induced marrow adiposity in vivo, marrow adipogenesis was measured histologically in long bones harvested from the mice. Interestingly, these studies demonstrated that GR-deficiency in osteoprogenitors drove an increase in marrow adiposity, as both marrow adipocyte density and area fraction were greater in GR-CKO as compared to GR-WT mice (+169% adipocyte density, +139% marrow adipocyte fraction, p<0.003; Figure 4). To the best of our knowledge, this phenotype has not been previously reported in mechanistic investigations of the role of the GR in bone biology (Liu et al., 2016). Consistent with previous studies, caloric restriction was also observed to drive an increase in marrow adiposity (+96% adipocyte density, +396% marrow adipocyte fraction, p<0.003; Figure 4). No interaction effects were observed between genotype and diet, suggesting that the GR is not required for CR-induced marrow adipogenesis. Interestingly, although propranolol was previously reported to ameliorate bone marrow adiposity induced by caloric restriction (Baek and Bloomfield, 2012, Baek et al., 2014b), we observed no effect of propranolol on marrow adiposity in our studies even in WT mice (p>0.591, Figure 4). These data suggest that β-adrenergic signaling might not always be a critical component of CR-induced marrow adipogenesis.

Figure 4. GR-CKO mice have more marrow fat than GR-WT mice.

Figure 4.

Open in a new tab

(A) Representative H&E-stained tibiae (original magnification: 20× objective) showing “adipose ghosts” indicative of marrow fat. (B) Adipocyte area fraction (Ad.Ar/Ma.Ar) and (C) adipocyte density (# adipocytes per mm2 marrow area) were quantified for each sample. Boxes show median, quartiles, and outlier fences for each dataset. Each data point represents one mouse. Statistical analyses for comparing the influence of diet (blue text) and propranolol intervention (red text) between GR-WT and GR-CKO mice are shown beneath each graph.

GR-CKO mice have high serum glucocorticoids under CR, which is rescued by propranolol treatment.

A canonical physiological response to metabolic stressors such as caloric restriction is the upregulated secretion of glucocorticoids—such as corticosterone in mice—from the adrenal cortex into the circulation (Cawthorn et al., 2016, Teich et al., 2016, Chapman et al., 2013). To determine the glucocorticoid response of GR-WT and GR-CKO mice in response to caloric restriction and propranolol treatment, serum samples from the mice were analyzed by a corticosterone enzyme-linked immunosorbent assay (ELISA). Under ad libitum conditions, there were negligible differences in circulating corticosterone between GR-WT and GR-CKO mice (Figure 5). Surprisingly, caloric restriction differentially impacted corticosterone levels in the serum between GR-CKO and GR-WT mice, significantly increasing circulating corticosterone only in the GR-CKO animals (genotype × diet interaction p-value = 0.012; Figure 5). Propranolol treatment differentially affected corticosterone levels in GR-WT and GR-CKO mice during caloric restriction, suppressing corticosterone levels in GR-CKO mice while having no impact on GR-WT mice (genotype × drug interaction p-value = 0.001; Figure 5). Together, these data suggest an aberrant stress response of GR-insufficient animals to caloric restriction that is sensitive to β-blockade, and a potential role for bone-specific GR activity in regulating whole-body glucocorticoid signaling.

Figure 5. Circulating corticosterone levels are elevated by GR-CKO and caloric restriction.

Figure 5.

Open in a new tab

ELISA measurement at the conclusion of the study of the active glucocorticoid corticosterone (ng/mL) in serum harvested from GR-WT and GR-CKO mice subjected to AL, CR, or CR+Prop treatment conditions. Boxes show median, quartiles, and outlier fences for each dataset. Each data point represents one mouse. Statistical analyses for comparing the influence of diet (blue text) and propranolol intervention (red text) between GR-WT and GR-CKO mice are shown beneath the graph.

DISCUSSION

Chronic caloric restriction interrupts the metabolic homeostasis of bone. Glucocorticoid signaling, which is upregulated by stress and inflammation, is implicated in decreasing bone mass and mediating fat accumulation in the marrow. The current study sought to test the role of glucocorticoids in CR-induced marrow adipogenesis and bone turnover through conditional deletion of the glucocorticoid receptor in osteoprogenitors, revealing that loss of GR function in osteoprogenitors reduced bone mass and impaired bone formation as measured by histomorphometry and serum P1NP levels. The low bone mass phenotype of the GR-insufficient animals highlights the importance of physiological levels of glucocorticoid signaling and GR function for proper osteogenesis. Our results suggest that sympathetic tone has complex roles in bone maintenance during caloric restriction. Propranolol-treated GR-CKO mice were slightly better protected against the CR-induced loss of femur cortical bone than WT littermates, but showed no such improvement in trabecular bone mass. It is plausible that CR-induced cortical bone loss is promoted by activation of the β2AR, which can be upregulated by circulating glucocorticoids and leptin (Ma et al., 2011, Baek and Bloomfield, 2012). Further work is needed to better understand these stress-response mechanisms in bone.

The GR-CKO model used in this study conditionally deleted the transactivation domain of the GR in Osterix1-expressing osteoprogenitor cells, rendering the GR inactive through dimerization but not completely deleted by Cre-LoxP recombination. Previous work using a Runx2-Cre-mediated GR conditional deletion model confirmed that the monomeric GR is still functional and can attenuate osteoblast function when activated (Rauch et al., 2010). Moreover, the monomeric GR induces genetic transrepression via downstream genomic elements—described as negative glucocorticoid response elements (nGREs)—and the recruitment of mediating transcription factors such as AP-1, NF-κB, and nuclear receptor corepressors 1 and 2 (NCoR1 and SMRT) (Hudson et al., 2013, Lim et al., 2015, Oakley and Cidlowski, 2013). As an example, there is evidence that activation of an nGRE upstream of a Runx2 binding site on the osteocalcin promoter is a mechanism by which glucocorticoids can repress the Egr2/Krox20 enhancer that is required for osteoblast function and trabecular bone formation (Leclerc et al., 2005). A GR dimerization mutation in mice also showed that the monomeric GR can inhibit bone resorption by downregulating retinoic acid receptor-dependent RANKL production (Conaway et al., 2011). These findings raise the possibility that the upregulated serum corticosterone resulting from CR in GR-CKO animals could act via the monomeric GR to induce bone loss through impaired osteogenesis rather than increased resorption. Additional work will be necessary to determine the activity of the GR in our model and its potential interaction with transcription factors and genomic elements during metabolic stress responses. Moreover, whether this mechanism contributes to CR-induced marrow adiposity in GR-CKO mice remains in question.

We were surprised to observe that a caloric restriction-induced increase in bone marrow adiposity was not prevented by propranolol administration in either GR-WT or GR-CKO mice. This may be due in part to the dosage of propranolol used to modulate sympathetic tone in this study. We selected a dose of 0.1 mg/ml propranolol administered in drinking water based on a previous study demonstrating efficacy of this dosage in limiting marrow adipogenesis during caloric restriction (Baek and Bloomfield, 2012). However, we recognize that several other investigations into the role of sympathetic tone in regulation of bone and marrow adiposity have used higher dosages of approximately 0.5 mg/mL propranolol in drinking water (Baek et al., 2014a, Baek et al., 2014b, Motyl et al., 2013, Takeda et al., 2002). Accordingly, the limited impact of propranolol on bone and marrow fat in the current study may have been due to incomplete blockade of adrenergic signaling. However, it is important to note that propranolol administration did induce several biological changes in the current study, including blunting the up-regulation of serum corticosterone in GR-CKO mice subjected to caloric restriction (Figure 5), and inducing a mild decrease in both circulating P1NP levels and trabecular bone thickness similarly between genotypes (Figure 3D and Table 3). Based on our data, we hypothesize that β-adrenergic signaling might not always be a critical component of CR-induced marrow adipogenesis, but further studies are needed to thoroughly test this hypothesis.

Loss of GR function in the murine caloric restriction model presented here was not protective against excess marrow adipogenesis. Recent work in rats demonstrated that treatment with the GR antagonist mifepristone was protective against adiposity rebound in rats following the completion of chronic exercise and caloric restriction conditions (Teich et al., 2016). In this study, however, GR conditional deletion did not protect against CR-driven marrow fat, and instead resulted in a high marrow fat phenotype even under ad libitum feeding conditions. This is somewhat surprising, as recent work suggested that the GR augments adipose tissue formation, as GR-deficient mouse embryonic fibroblasts were impaired in their ability to form adipocytes both in vitro and in vivo (Bauerle et al., 2018). Our results suggest alternative mechanisms by which marrow fat accumulates in this osteoprogenitor-specific GR-deficient model. One possibility is that systemic rather than local glucocorticoids in the GR-CKO mice bind the GR to drive its activation of the master adipogenic transcription factor, peroxisome proliferator-activated receptor γ (PPARγ), via known interactions of the GR with PPARγ-transcription regulating CCAAT/enhancer-binging proteins (C/EBPs) (Grontved et al., 2013, Zhao et al., 2013, Li et al., 2013). However, little is known regarding whether the monomeric GR interacts with C/EBPs or the potential for C/EBP-mediated transrepression (rather than activation) to stimulate BMSC-derived adipogenesis. Another possibility is that the binding and activation of the mineralocorticoid receptor (MR) is involved in the phenotype, as corticosterone also binds this receptor (Li et al., 2005); indeed, MR has high affinity for glucocorticoids and in the absence of the inactivating enzyme, 11β-hydroxysteroid dehydrogenase-2 may be occupied by these steroids. Moreover, the MR has implications in facilitating adipogenesis, and adipocyte-specific MR activity is associated with metabolic syndrome and glucocorticoid-induced lipid accumulation (Nguyen Dinh Cat et al., 2016, Hirata et al., 2009, Caprio et al., 2011, Hoppmann et al., 2010). These findings also highlight a potential limitation to the GR conditional deletion model used in the current study, as Osx1-Cre fate mapping studies have confirmed expression of this Cre reporter in bone marrow adipocytes (Liu et al., 2013); therefore, at present, it is not possible to discriminate between whether the enhanced adiposity in the GR-CKO mice is a progenitor-driven effect or an adipocyte-driven effect. However, it is important to note that several studies have addressed the role of the GR in adiponectin-expressing adipocytes, and bone marrow adipocytes are an important source of adiponectin during caloric restriction (Cawthorn et al., 2014). Interestingly, the majority of these reports suggest that GR-deletion in adipocytes does not independently promote adipocyte expansion or hypertrophy (Shen et al., 2017, Desarzens and Faresse, 2016, Bose et al., 2016) and may even decrease fat mass (Mueller et al., 2017). Thus if the primary effect of Osx1-Cre on marrow adiposity resulted from directly targeting bone marrow adipocytes, it would be logical to expect that marrow fat should be unaffected or decreased in GR-CKO animals. In contrast, the current study found robust increases in marrow adiposity in the GR-CKO animals under both ad-libitum and caloric restriction feeding conditions. A remaining question is the role of glucocorticoid-activated MR in BMSC-derived osteoblasts and adipocytes—including the role of MR-driven marrow adipogenesis and mechanistic interactions of the GR and MR in BMSC. With both of these possible mechanisms, however, it is important to note that we only observed an increase in circulating corticosterone levels in the GR-CKO mice under conditions of caloric restriction. As we only measured serum corticosterone at the time of sacrifice, and glucocorticoids demonstrate substantial circadian variation in their circulating levels (Halberg et al., 1959, Gong et al., 2015), it is possible that we may have missed the optimal time point at which to compare corticosterone levels across different groups. In future studies, it will be important to conduct a thorough time course of blood sampling to observe how circulating GC levels are impacted by loss of GR in the skeleton throughout periods of activity and dormancy.

It is also important to consider the complex role of β-adrenergic signaling in marrow adipogenesis, as it has been demonstrated that β-AR antagonism with propranolol can positively regulate BMSC adipogenesis as well as C/EBP and PPARγ expression, while the β-AR agonist isoproterenol inhibits these processes (Li et al., 2010a). As previously mentioned, however, β-adrenergic signaling during caloric restriction shows differing roles in terms of facilitating bone loss and marrow fat accumulation (Baek et al., 2014b, Baek and Bloomfield, 2012). It was also recently reported that marrow adipose tissue has a blunted response to adrenergic signaling as compared to other fat depots like white adipose tissue (Scheller et al., 2019). This further demonstrates the complexities of inflammatory and stress responses in bone-resident cell types and the potential for canonical signaling pathways to be pro- or anti-adipogenic depending on mechanistic crosstalk that has yet to be elucidated in the CR stress model.

While the findings in this study highlight the complex effects of glucocorticoid signaling and GR activity in bone during caloric restriction, there are limitations to acknowledge in the study design that, when addressed in future work, may further clarify mechanisms of interest. For instance, the presented work was completed in only in female mice; this was done because pilot studies suggested similar trends of increased marrow fat accumulation in both male and female GR-CKO mice as compared to WT controls, but trends were more pronounced in females. We consequently focused on females for further study, but this observation raises the question of whether sex differences exist in the response of marrow adiposity to loss of GR-mediated signaling and caloric restriction in bone. In addition, as described above, it is important to acknowledge that the preferred control for the Osx1-Cre, GR+/+:Osx1-Cre+ mice (i.e., “Cre+ WT” mice), was not included here. In pilot studies, presented in the supplemental data, we observed no difference in marrow adiposity between Cre˗ and Cre+ WT animals, but, these pilot data were from male mice at 6 months of age, whereas mice in the dietary studies presented here were female mice that were 6 months old at the onset of studies and 8 months old at sacrifice. It will be important to age- and sex-match additional control groups in future studies of Osx1-Cre: GR-CKO animals in order to draw better conclusions regarding the role of the GR in regulation of bone mass and marrow fat during periods of metabolic stress. Lastly, while we confirmed appreciable reduction in GR expression in bone via immunostaining and in marrow-derived osteoblasts from these mice, we did not perform a comprehensive evaluation of GR expression in potential off-target tissues, such as the brain, that may express both GR and Osterix (Park et al., 2011, Madalena and Lerch, 2017). From our immunostaining analyses, we note that the intensity of GR staining in the general bone marrow area tended to be reduced in the GR-CKO mice, although the impact of this observation on outcome measurements in the current study is not clear. Interpretation of data in the Osx1-Cre : GR-CKO model described here could be strengthened by analyses of GR expression in the whole bone marrow stroma and brain to conclusively establish whether observed effects are mediated directly by loss of GR in bone or influenced via off-target effects in other tissues. Our future work will address the aforementioned limitations in methodology to better understand the role of the GR in the skeletal response to stressors.

In conclusion, the present study suggests that GR function in osteoprogenitors is necessary to maintain bone mass and inhibit marrow fat accumulation that occurs with the stress of long-term caloric restriction. The β-blocker propranolol selectively improved cortical bone mass in GR-CKO mice but had no impact on CR-induced marrow adiposity in either GR-WT or GR-CKO animals. Future work will focus on elucidating the molecular mechanisms driving the osteoporotic phenotype of GR-insufficient bone as well as determining how loss of the GR in osteoprogenitors affects the response of bone to common physiological stressors such as aging.

Supplementary Material

01

Supplemental Figure 1: GR-CKO (GRfl/fl:Osx1-Cre+) mice have more marrow fat than Cre-negative (GRfl/fl:Osx1-Cre-) or Cre-positive (GR+/+:Osx1-Cre+) control mice. Right femora from 6-month-old male mice were histologically prepared for quantitative histomorphometry to analyze marrow adiposity. (A) Adipocyte area fraction (Ad.Ar/Ma.Ar) and (B) adipocyte density (# adipocytes per mm2 marrow area) were quantified for each sample. Each data point represents one mouse. Box plots show median, quartiles, and outlier fences for each dataset, where outlier fences represent first quartile - 1.5*(interquartile range) and third quartile + 1.5*(interquartile range). Results of statistical analyses (one-way ANOVA with Tukey’s HSD post-hoc testing) are shown above each graph, and bars with different superscript letters are significantly (p<0.05) different from one another as shown by post-hoc testing.

02

DECLARATION OF INTEREST, FUNDING, AND ACKNOWLEDGEMENTS

The authors declare no conflicts of interest. The authors would like to acknowledge Dr. Louis Muglia for providing the GR-floxed mouse model. This work was supported by the National Institutes on Aging (grant number P01-AG036675; Project 4) and the American Diabetes Association (grant number 1–16-JDF-062). The contents do not represent the views of the U.S. Department of Veterans Affairs or the United States Government.

References

  1. ALATALO SL, PENG Z, JANCKILA AJ, KAIJA H, VIHKO P, VAANANEN HK & HALLEEN JM 2003. A novel immunoassay for the determination of tartrate-resistant acid phospatase 5b from rat serum. J Bone Miner Res, 18. [DOI] [PubMed] [Google Scholar]
  2. ALMEIDA M, HAN L, AMBROGINI E, WEINSTEIN RS & MANOLAGAS SC 2011. Glucocorticoids and tumor necrosis factor alpha increase oxidative stress and suppress Wnt protein signaling in osteoblasts. J Biol Chem, 286, 44326–35. [DOI] [PMC free article] [PubMed] [Google Scholar]
  3. BAEK K & BLOOMFIELD SA 2012. Blocking beta-adrenergic signaling attenuates reductions in circulating leptin, cancellous bone mass, and marrow adiposity seen with dietary energy restriction. J Appl Physiol (1985), 113, 1792–801. [DOI] [PubMed] [Google Scholar]
  4. BAEK K, HWANG HR, PARK H-J, KWON A, QADIR AS & BAEK J-H 2014a. Propranolol, a β-adrenergic antagonist, attenuates the decrease in trabecular bone mass in high calorie diet fed growing mice. BMB Reports, 47, 506–511. [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. BAEK K, PARK H-J, HWANG HR & BAEK J-H 2014b. Propranolol attenuates calorie restriction- and high calorie diet-induced bone marrow adiposity. BMB Reports, 47, 587–592. [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. BAKER N, BOYETTE LB & TUAN RS 2015. Characterization of bone marrow-derived mesenchymal stem cells in aging. Bone, 70, 37–47. [DOI] [PubMed] [Google Scholar]
  7. BAUERLE KT, HUTSON I, SCHELLER EL & HARRIS CA 2018. Glucocorticoid Receptor Signaling Is Not Required for In Vivo Adipogenesis. Endocrinology, 159, 2050–2061. [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. BENNETT CN, OUYANG H, MA YL, ZENG Q, GERIN I, SOUSA KM, LANE TF, KRISHNAN V, HANKENSON KD & MACDOUGALD OA 2007. Wnt10b increases postnatal bone formation by enhancing osteoblast differentiation. J Bone Miner Res, 22, 1924–32. [DOI] [PubMed] [Google Scholar]
  9. BONNET N, PIERROZ DD & FERRARI SL 2008. Adrenergic control of bone remodeling and its implications for the treatment of osteoporosis. J Musculoskelet Neuronal Interact, 8, 94–104. [PubMed] [Google Scholar]
  10. BOSE SK, HUTSON I & HARRIS CA 2016. Hepatic Glucocorticoid Receptor Plays a Greater Role Than Adipose GR in Metabolic Syndrome Despite Renal Compensation. Endocrinology, 157, 4943–4960. [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. BREWER JA, KHOR B, VOGT SK, MUGLIA LM, FUJIWARA H, HAEGELE KE, SLECKMAN BP & MUGLIA LJ 2003. T-cell glucocorticoid receptor is required to suppress COX-2-mediated lethal immune activation. Nat Med, 9, 1318–22. [DOI] [PubMed] [Google Scholar]
  12. CANALIS E, BILEZIKIAN JP, ANGELI A & GIUSTINA A 2004. Perspectives on glucocorticoid-induced osteoporosis. Bone, 34, 593–8. [DOI] [PubMed] [Google Scholar]
  13. CAPRIO M, ANTELMI A, CHETRITE G, MUSCAT A, MAMMI C, MARZOLLA V, FABBRI A, ZENNARO M-C & FÈVE B 2011. Antiadipogenic Effects of the Mineralocorticoid Receptor Antagonist Drospirenone: Potential Implications for the Treatment of Metabolic Syndrome. Endocrinology, 152, 113–125. [DOI] [PubMed] [Google Scholar]
  14. CAWTHORN WP, SCHELLER EL, LEARMAN BS, PARLEE SD, SIMON BR, MORI H, NING X, BREE AJ, SCHELL B, BROOME DT, et al. 2014. Bone marrow adipose tissue is an endocrine organ that contributes to increased circulating adiponectin during caloric restriction. Cell Metab, 20, 368–75. [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. CAWTHORN WP, SCHELLER EL, PARLEE SD, PHAM HA, LEARMAN BS, REDSHAW CM, SULSTON RJ, BURR AA, DAS AK, SIMON BR, et al. 2016. Expansion of Bone Marrow Adipose Tissue During Caloric Restriction Is Associated With Increased Circulating Glucocorticoids and Not With Hypoleptinemia. Endocrinology, 157, 508–21. [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. CHAPMAN KE, COUTINHO AE, ZHANG Z, KIPARI T, SAVILL JS & SECKL JR 2013. Changing glucocorticoid action: 11beta-hydroxysteroid dehydrogenase type 1 in acute and chronic inflammation. J Steroid Biochem Mol Biol, 137, 82–92. [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. CONAWAY HH, PIRHAYATI A, PERSSON E, PETTERSSON U, SVENSSON O, LINDHOLM C, HENNING P, TUCKERMANN J & LERNER UH 2011. Retinoids stimulate periosteal bone resorption by enhancing the protein RANKL, a response inhibited by monomeric glucocorticoid receptor. J Biol Chem, 286, 31425–36. [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. DAVEY RA, CLARKE MV, SASTRA S, SKINNER JP, CHIANG C, ANDERSON PH & ZAJAC JD 2012. Decreased body weight in young Osterix-Cre transgenic mice results in delayed cortical bone expansion and accrual. Transgenic Res, 21, 885–93. [DOI] [PubMed] [Google Scholar]
  19. DAY TF, GUO X, GARRETT-BEAL L & YANG Y 2005. Wnt/beta-catenin signaling in mesenchymal progenitors controls osteoblast and chondrocyte differentiation during vertebrate skeletogenesis. Dev Cell, 8, 739–50. [DOI] [PubMed] [Google Scholar]
  20. DESARZENS S & FARESSE N 2016. Adipocyte glucocorticoid receptor has a minor contribution in adipose tissue growth. J Endocrinol, 230, 1–11. [DOI] [PubMed] [Google Scholar]
  21. DEVLIN MJ, CLOUTIER AM, THOMAS NA, PANUS DA, LOTINUN S, PINZ I, BARON R, ROSEN CJ & BOUXSEIN ML 2010. Caloric restriction leads to high marrow adiposity and low bone mass in growing mice. J Bone Miner Res, 25, 2078–88. [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. ELEFTERIOU F 2005. Neuronal signaling and the regulation of bone remodeling. Cell Mol Life Sci, 62, 2339–49. [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. FAZELI PK, HOROWITZ MC, MACDOUGALD OA, SCHELLER EL, RODEHEFFER MS, ROSEN CJ & KLIBANSKI A 2013. Marrow fat and bone--new perspectives. J Clin Endocrinol Metab, 98, 935–45. [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. GATHERCOLE LL, LAVERY GG, MORGAN SA, COOPER MS, SINCLAIR AJ, TOMLINSON JW & STEWART PM 2013. 11beta-Hydroxysteroid dehydrogenase 1: translational and therapeutic aspects. Endocr Rev, 34, 525–55. [DOI] [PubMed] [Google Scholar]
  25. GONG S, MIAO YL, JIAO GZ, SUN MJ, LI H, LIN J, LUO MJ & TAN JH 2015. Dynamics and correlation of serum cortisol and corticosterone under different physiological or stressful conditions in mice. PLoS One, 10, e0117503. [DOI] [PMC free article] [PubMed] [Google Scholar]
  26. GRONTVED L, JOHN S, BAEK S, LIU Y, BUCKLEY JR, VINSON C, AGUILERA G & HAGER GL 2013. C/EBP maintains chromatin accessibility in liver and facilitates glucocorticoid receptor recruitment to steroid response elements. EMBO J, 32, 1568–83. [DOI] [PMC free article] [PubMed] [Google Scholar]
  27. HAGAN ML, BAHRAINI A, PIERCE JL, BASS SM, YU K, ELSAYED R, ELSALANTY M, JOHNSON MH, MCNEIL A, MCNEIL PL & MCGEE-LAWRENCE ME 2019. Inhibition of Osteocyte Membrane Repair Activity via Dietary Vitamin E Deprivation Impairs Osteocyte Survival. Calcif Tissue Int, 104, 224–234. [DOI] [PMC free article] [PubMed] [Google Scholar]
  28. HALBERG F, ALBRECHT PG & BITTNER JJ 1959. Corticosterone rhythm of mouse adrenal in relation to serum corticosterone and sampling. Am J Physiol, 197, 1083–1085. [DOI] [PubMed] [Google Scholar]
  29. HALLEEN JM, YLIPAHKALA H, ALATALO SL, JANCKILA AJ, HEIKKINEN JE, SUOMINEN H, CHENG S & VAANANEN HK 2002. Serum tartrate-resistant acid phosphatase 5b, but not 5a, correlates with other markers of bone turnover and bone mineral density. Calcif Tissue Int, 71, 20–5. [DOI] [PubMed] [Google Scholar]
  30. HAMRICK MW, DING KH, PONNALA S, FERRARI SL & ISALES CM 2008. Caloric restriction decreases cortical bone mass but spares trabecular bone in the mouse skeleton: implications for the regulation of bone mass by body weight. J Bone Miner Res, 23, 870–8. [DOI] [PubMed] [Google Scholar]
  31. HAMRICK MW, MCGEE-LAWRENCE ME & FRECHETTE DM 2016. Fatty Infiltration of Skeletal Muscle: Mechanisms and Comparisons with Bone Marrow Adiposity. Front Endocrinol (Lausanne), 7, 69. [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. HIRATA A, MAEDA N, HIUGE A, HIBUSE T, FUJITA K, OKADA T, KIHARA S, FUNAHASHI T & SHIMOMURA I 2009. Blockade of mineralocorticoid receptor reverses adipocyte dysfunction and insulin resistance in obese mice. Cardiovasc Res, 84, 164–72. [DOI] [PubMed] [Google Scholar]
  33. HOPPMANN J, PERWITZ N, MEIER B, FASSHAUER M, HADASCHIK D, LEHNERT H & KLEIN J 2010. The balance between gluco- and mineralo-corticoid action critically determines inflammatory adipocyte responses. J Endocrinol, 204, 153–64. [DOI] [PubMed] [Google Scholar]
  34. HUA G, GANTI KP & CHAMBON P 2016a. Glucocorticoid-induced tethered transrepression requires SUMOylation of GR and formation of a SUMO-SMRT/NCoR1-HDAC3 repressing complex. Proc Natl Acad Sci U S A, 113, E635–43. [DOI] [PMC free article] [PubMed] [Google Scholar]
  35. HUA G, PAULEN L & CHAMBON P 2016b. GR SUMOylation and formation of an SUMO-SMRT/NCoR1-HDAC3 repressing complex is mandatory for GC-induced IR nGRE-mediated transrepression. Proc Natl Acad Sci U S A, 113, E626–34. [DOI] [PMC free article] [PubMed] [Google Scholar]
  36. HUDSON WH, YOUN C & ORTLUND EA 2013. The structural basis of direct glucocorticoid-mediated transrepression. Nat Struct Mol Biol, 20, 53–8. [DOI] [PMC free article] [PubMed] [Google Scholar]
  37. KANG S, BENNETT CN, GERIN I, RAPP LA, HANKENSON KD & MACDOUGALD OA 2007. Wnt signaling stimulates osteoblastogenesis of mesenchymal precursors by suppressing CCAAT/enhancer-binding protein alpha and peroxisome proliferator-activated receptor gamma. J Biol Chem, 282, 14515–24. [DOI] [PubMed] [Google Scholar]
  38. KIM HJ, ZHAO H, KITAURA H, BHATTACHARYYA S, BREWER JA, MUGLIA LJ, ROSS FP & TEITELBAUM SL 2006. Glucocorticoids suppress bone formation via the osteoclast. J Clin Invest, 116, 2152–60. [DOI] [PMC free article] [PubMed] [Google Scholar]
  39. KUO T, HARRIS CA & WANG JC 2013. Metabolic functions of glucocorticoid receptor in skeletal muscle. Mol Cell Endocrinol, 380, 79–88. [DOI] [PMC free article] [PubMed] [Google Scholar]
  40. LECLERC N, NOH T, KHOKHAR A, SMITH E & FRENKEL B 2005. Glucocorticoids inhibit osteocalcin transcription in osteoblasts by suppressing Egr2/Krox20-binding enhancer. Arthritis Rheum, 52, 929–39. [DOI] [PubMed] [Google Scholar]
  41. LI H, FONG C, CHEN Y, CAI G & YANG M 2010a. Beta-adrenergic signals regulate adipogenesis of mouse mesenchymal stem cells via cAMP/PKA pathway. Mol Cell Endocrinol, 323, 201–7. [DOI] [PubMed] [Google Scholar]
  42. LI J, LIU X, ZUO B & ZHANG L 2016. The Role of Bone Marrow Microenvironment in Governing the Balance between Osteoblastogenesis and Adipogenesis. Aging Dis, 7, 514–25. [DOI] [PMC free article] [PubMed] [Google Scholar]
  43. LI J, ZHANG N, HUANG X, XU J, FERNANDES JC, DAI K & ZHANG X 2013. Dexamethasone shifts bone marrow stromal cells from osteoblasts to adipocytes by C/EBPalpha promoter methylation. Cell Death Dis, 4, e832. [DOI] [PMC free article] [PubMed] [Google Scholar]
  44. LI X, COPE MB, JOHNSON MS, SMITH DL JR, & NAGY TR 2010b. Mild calorie restriction induces fat accumulation in female C57BL/6J mice. Obesity (Silver Spring), 18, 456–62. [DOI] [PMC free article] [PubMed] [Google Scholar]
  45. LI Y, SUINO K, DAUGHERTY J & XU HE 2005. Structural and biochemical mechanisms for the specificity of hormone binding and coactivator assembly by mineralocorticoid receptor. Mol Cell, 19, 367–80. [DOI] [PubMed] [Google Scholar]
  46. LIM HW, UHLENHAUT NH, RAUCH A, WEINER J, HUBNER S, HUBNER N, WON KJ, LAZAR MA, TUCKERMANN J & STEGER DJ 2015. Genomic redistribution of GR monomers and dimers mediates transcriptional response to exogenous glucocorticoid in vivo. Genome Res, 25, 836–44. [DOI] [PMC free article] [PubMed] [Google Scholar]
  47. LIU P, BAUMGART M, GROTH M, WITTMANN J, JACK HM, PLATZER M, TUCKERMANN JP & BASCHANT U 2016. Dicer ablation in osteoblasts by Runx2 driven cre-loxP recombination affects bone integrity, but not glucocorticoid-induced suppression of bone formation. Sci Rep, 6, 32112. [DOI] [PMC free article] [PubMed] [Google Scholar]
  48. LIU Y, STRECKER S, WANG L, KRONENBERG MS, WANG W, ROWE DW & MAYE P 2013. Osterix-cre labeled progenitor cells contribute to the formation and maintenance of the bone marrow stroma. PLoS One, 8, e71318. [DOI] [PMC free article] [PubMed] [Google Scholar]
  49. MA Y, NYMAN JS, TAO H, MOSS HH, YANG X & ELEFTERIOU F 2011. beta2-Adrenergic receptor signaling in osteoblasts contributes to the catabolic effect of glucocorticoids on bone. Endocrinology, 152, 1412–22. [DOI] [PMC free article] [PubMed] [Google Scholar]
  50. MADALENA KM & LERCH JK 2017. The Effect of Glucocorticoid and Glucocorticoid Receptor Interactions on Brain, Spinal Cord, and Glial Cell Plasticity. Neural Plast, 2017, 8640970. [DOI] [PMC free article] [PubMed] [Google Scholar]
  51. MCGEE-LAWRENCE ME, BRADLEY EW, DUDAKOVIC A, CARLSON SW, RYAN ZC, KUMAR R, DADSETAN M, YASZEMSKI MJ, CHEN Q, AN KN & WESTENDORF JJ 2013. Histone deacetylase 3 is required for maintenance of bone mass during aging. Bone, 52, 296–307. [DOI] [PMC free article] [PubMed] [Google Scholar]
  52. MCGEE-LAWRENCE ME, CARPIO LR, SCHULZE RJ, PIERCE JL, MCNIVEN MA, FARR JN, KHOSLA S, OURSLER MJ & WESTENDORF JJ 2016. Hdac3 Deficiency Increases Marrow Adiposity and Induces Lipid Storage and Glucocorticoid Metabolism in Osteochondroprogenitor Cells. J Bone Miner Res, 31, 116–28. [DOI] [PMC free article] [PubMed] [Google Scholar]
  53. MOTYL KJ, BISHOP KA, DEMAMBRO VE, BORNSTEIN SA, LE P, KAWAI M, LOTINUN S, HOROWITZ MC, BARON R, BOUXSEIN ML & ROSEN CJ 2013. Altered thermogenesis and impaired bone remodeling in Misty mice. J Bone Miner Res, 28, 1885–97. [DOI] [PMC free article] [PubMed] [Google Scholar]
  54. MUELLER KM, HARTMANN K, KALTENECKER D, VETTORAZZI S, BAUER M, MAUSER L, AMANN S, JALL S, FISCHER K, ESTERBAUER H, et al. 2017. Adipocyte Glucocorticoid Receptor Deficiency Attenuates Aging- and HFD-Induced Obesity and Impairs the Feeding-Fasting Transition. Diabetes, 66, 272–286. [DOI] [PubMed] [Google Scholar]
  55. NGUYEN DINH CAT A, ANTUNES TT, CALLERA GE, SANCHEZ A, TSIROPOULOU S, DULAK-LIS MG, ANAGNOSTOPOULOU A, HE Y, MONTEZANO AC, JAISSER F & TOUYZ RM 2016. Adipocyte-specific mineralocorticoid receptor overexpression in mice is associated with metabolic syndrome and vascular dysfunction: role of redox-sensitive PKG-1 and rho kinase. Diabetes, 65, 2392–2403. [DOI] [PubMed] [Google Scholar]
  56. NIXON M, ANDREW R & CHAPMAN KE 2013. It takes two to tango: dimerisation of glucocorticoid receptor and its anti-inflammatory functions. Steroids, 78, 59–68. [DOI] [PubMed] [Google Scholar]
  57. O’BRIEN CA, JIA D, PLOTKIN LI, BELLIDO T, POWERS CC, STEWART SA, MANOLAGAS SC & WEINSTEIN RS 2004. Glucocorticoids act directly on osteoblasts and osteocytes to induce their apoptosis and reduce bone formation and strength. Endocrinology, 145, 1835–41. [DOI] [PubMed] [Google Scholar]
  58. OAKLEY RH & CIDLOWSKI JA 2013. The biology of the glucocorticoid receptor: new signaling mechanisms in health and disease. J Allergy Clin Immunol, 132, 1033–44. [DOI] [PMC free article] [PubMed] [Google Scholar]
  59. PAPANASTASIOU L, FOUNTOULAKIS S & VATALAS IA 2017. Adrenal disorders and non-alcoholic fatty liver disease. Minerva Endocrinol, 42, 151–163. [DOI] [PubMed] [Google Scholar]
  60. PARK JS, BAEK WY, KIM YH & KIM JE 2011. In vivo expression of Osterix in mature granule cells of adult mouse olfactory bulb. Biochem Biophys Res Commun, 407, 842–7. [DOI] [PubMed] [Google Scholar]
  61. PERIYASAMY-THANDAVAN S, HERBERG S, AROUNLEUT P, UPADHYAY S, DUKES A, DAVIS C, JOHNSON M, MCGEE-LAWRENCE M, HAMRICK MW, ISALES CM & HILL WD 2015. Caloric restriction and the adipokine leptin alter the SDF-1 signaling axis in bone marrow and in bone marrow derived mesenchymal stem cells. Mol Cell Endocrinol, 410, 64–72. [DOI] [PMC free article] [PubMed] [Google Scholar]
  62. RAUCH A, SEITZ S, BASCHANT U, SCHILLING AF, ILLING A, STRIDE B, KIRILOV M, MANDIC V, TAKACZ A, SCHMIDT-ULLRICH R, et al. 2010. Glucocorticoids suppress bone formation by attenuating osteoblast differentiation via the monomeric glucocorticoid receptor. Cell Metab, 11, 517–31. [DOI] [PubMed] [Google Scholar]
  63. RODDA SJ & MCMAHON AP 2006. Distinct roles for Hedgehog and canonical Wnt signaling in specification, differentiation and maintenance of osteoblast progenitors. Development, 133, 3231–44. [DOI] [PubMed] [Google Scholar]
  64. SCHELLER EL, KHANDAKER S, LEARMAN BS, CAWTHORN WP, ANDERSON LM, PHAM HA, ROBLES H, WANG Z, LI Z, PARLEE SD, SIMON BR, et al. 2019. Bone marrow adipocytes resist lipolysis and remodeling in response to beta-adrenergic stimulation. Bone, 118, 32–41. [DOI] [PMC free article] [PubMed] [Google Scholar]
  65. SHEN Y, ROH HC, KUMARI M & ROSEN ED 2017. Adipocyte glucocorticoid receptor is important in lipolysis and insulin resistance due to exogenous steroids, but not insulin resistance caused by high fat feeding. Mol Metab, 6, 1150–1160. [DOI] [PMC free article] [PubMed] [Google Scholar]
  66. SHI H, STRADER AD, WOODS SC & SEELEY RJ 2007. Sexually dimorphic responses to fat loss after caloric restriction or surgical lipectomy. Am J Physiol Endocrinol Metab, 293, E316–26. [DOI] [PubMed] [Google Scholar]
  67. SINGHAL V & BREDELLA MA 2018. Marrow adipose tissue imaging in humans. Bone. [DOI] [PMC free article] [PubMed] [Google Scholar]
  68. TAKEDA S, ELEFTERIOU F, LEVASSEUR R, LIU X, ZHAO L, PARKER KL, ARMSTRONG D, DUCY P & KARSENTY G 2002. Leptin regulates bone formation via the sympathetic nervous system. Cell, 111, 305–17. [DOI] [PubMed] [Google Scholar]
  69. TEICH T, DUNFORD EC, PORRAS DP, PIVOVAROV JA, BEAUDRY JL, HUNT H, BELANOFF JK & RIDDELL MC 2016. Glucocorticoid antagonism limits adiposity rebound and glucose intolerance in young male rats following the cessation of daily exercise and caloric restriction. Am J Physiol Endocrinol Metab, 311, E56–68. [DOI] [PMC free article] [PubMed] [Google Scholar]
  70. TOMLINSON JW, WALKER EA, BUJALSKA IJ, DRAPER N, LAVERY GG, COOPER MS, HEWISON M & STEWART PM 2004. 11beta-hydroxysteroid dehydrogenase type 1: a tissue-specific regulator of glucocorticoid response. Endocr Rev, 25, 831–66. [DOI] [PubMed] [Google Scholar]
  71. WEINSTEIN RS 2012. Glucocorticoid-induced osteonecrosis. Endocrine, 41, 183–90. [DOI] [PMC free article] [PubMed] [Google Scholar]
  72. WEINSTEIN RS, JILKA RL, PARFITT AM & MANOLAGAS SC 1998. Inhibition of osteoblastogenesis and promotion of apoptosis of osteoblasts and osteocytes by glucocorticoids. J Clin Invest, 102, 274–282. [DOI] [PMC free article] [PubMed] [Google Scholar]
  73. WU L, QI H, ZHONG Y, LV S, YU J, LIU J, WANG L, BI J, KONG X, DI W, ZHA J, LIU F & DING G 2013. 11β-hydroxysteroid dehydrogenase type 1 selective inhibitor BVT.2733 protects osteoblasts against endogenous glucocorticoid induced dysfunction. Endocr J, 60. [DOI] [PubMed] [Google Scholar]
  74. ZHANG Y, KHAN D, DELLING J & TOBIASCH E 2012. Mechanisms underlying the osteo- and adipo-differentiation of human mesenchymal stem cells. ScientificWorldJournal, 2012, 793823. [DOI] [PMC free article] [PubMed] [Google Scholar]
  75. ZHAO QH, WANG SG, LIU SX, LI JP, ZHANG YX, SUN ZY, FAN QM & TIAN JW 2013. PPARgamma forms a bridge between DNA methylation and histone acetylation at the C/EBPalpha gene promoter to regulate the balance between osteogenesis and adipogenesis of bone marrow stromal cells. FEBS J, 280, 5801–14. [DOI] [PubMed] [Google Scholar]
  76. ZHOU H, COOPER MS & SEIBEL MJ 2013. Endogenous Glucocorticoids and Bone. Bone Res, 1, 107–19. [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

01

Supplemental Figure 1: GR-CKO (GRfl/fl:Osx1-Cre+) mice have more marrow fat than Cre-negative (GRfl/fl:Osx1-Cre-) or Cre-positive (GR+/+:Osx1-Cre+) control mice. Right femora from 6-month-old male mice were histologically prepared for quantitative histomorphometry to analyze marrow adiposity. (A) Adipocyte area fraction (Ad.Ar/Ma.Ar) and (B) adipocyte density (# adipocytes per mm2 marrow area) were quantified for each sample. Each data point represents one mouse. Box plots show median, quartiles, and outlier fences for each dataset, where outlier fences represent first quartile - 1.5*(interquartile range) and third quartile + 1.5*(interquartile range). Results of statistical analyses (one-way ANOVA with Tukey’s HSD post-hoc testing) are shown above each graph, and bars with different superscript letters are significantly (p<0.05) different from one another as shown by post-hoc testing.

NIHMS1537244-supplement-01.tif (1.5MB, tif)
02
NIHMS1537244-supplement-02.docx (27.5KB, docx)

RESOURCES