The glucocorticoid receptor (GR) is expressed throughout the cardiovascular system and glucocorticoids (GC) are known to influence cardiovascular processes ranging from angiogenesis and vascular tone to cardiomyocyte hypertrophy and inflammation. Genetic variation in the human GR gene that associates with relative glucocorticoid resistance is also linked to hypertension and increased risk of cardiovascular disease. Mice with global GR haploinsufficiency (GR+/-) are similarly glucocorticoid resistant, with increased hypothalamic-pituitary-adrenal (HPA) axis activity and elevated blood pressure in adulthood. Previous work from the laboratory has demonstrated that the GR is essential for normal growth and maturation of the foetal heart in late gestation and in vitro studies show that GC can alter cardiomyocyte function and induce cardiomyocyte hypertrophy. I hypothesised that reduced GR density during development would have consequences for cardiovascular function and disease risk in adulthood and that cardiovascular GR signalling is important for postnatal growth of the heart, as well as physiological and pathological cardiac remodeling in adulthood. I tested this hypothesis in GR+/- mice with global alteration in GR density as well as in SMGRKO mice, with deletion of GR in cardiomyocytes and vascular smooth muscle. To investigate the association between GC resistance and cardiovascular disease risk, I have characterised the cardiac phenotype of GR+/- mice, basally and following physiological and pathological cardiac remodeling induced by a swim training programme and Angiotensin II treatment, respectively. Survival to weaning was reduced by 35% in GR+/- mice compared with wild-type (WT) littermates. Ultrasound analysis revealed impairment of systolic cardiac function in utero (E17.5) and at postnatal day (P) 2. However, by P7 cardiac function had normalised in surviving GR+/- mice and remained equivalent to WT littermates in adulthood. Heart weight and morphology were normal in GR+/- mice in adulthood but cardiomyocyte cross sectional area was reduced, in combination with an increase in nuclei per unit area implying an increased number of cardiomyocytes. This could arise from a delay in the developmental transition from hyperplasic to hypertrophic growth of cardiomyocytes and suggests that GR+/- mice may have a reduced ability to respond to the increased cardiac workload at birth and during the early postnatal period. Further cardiac challenge may be posed by the elevated blood pressure, compensatory increase in HPA axis activity and aldosterone levels previously reported in GR+/- mice. Adaptation to pathological cardiac challenge was assessed in adult GR+/- mice and WT littermates in response to AngII treatment, which has a direct hypertrophic effect on cardiomyocytes and, at higher doses, elevates blood pressure. GR+/- and WT mice showed an equivalent, dose-dependent increase in cardiomyocyte hypertrophy and cardiac fibrosis in response to AngII, as well as similar alterations in expression of Ca2+ handling genes. Functionally, these changes to the myocardium resulted in matched reductions in ejection fraction in GR+/- and WT mice. In contrast, when cardiac hypertrophy was induced by the physiological challenge of swim training, normal cardiac function was maintained in both GR+/- mice and WT controls. The physiological cardiac hypertrophy induced by swim training was not associated with cardiac fibrosis or pathological changes to left ventricle (LV) gene expression profiles. GR+/- mice have elevated HPA axis activity at baseline and swim training increased adrenal gland weight to a greater extent in GR+/- mice suggesting that raised GC levels due to compensatory HPA activation in GR+/- mice, may mask the role of GR in cardiac remodeling. To remove the effects of compensatory HPA axis activation and to achieve a greater degree of GR deficiency in the cardiovascular system, homozygous SMGRKO mice were investigated. Similar to GR+/- mice, survival to weaning compared with control littermates was impaired, by 46% and 65% in males and females respectively. Doppler measurements of transmitral inflow and transaortic outflow of blood showed a detrimental increase in the myocardial performance index (MPI), a load-independent measure of combined systolic and diastolic function. This was due to prolongation of the isovolumetric contraction time, indicating impairment of the initial LV contractile phase. Heart/body weight ratio was increased in both and male and female SMGRKO mice. Interestingly, cross sectional area was reduced in adult female SMGRKO mice cardiomyocyte, as was found in the GR+/- mice. In contrast, in male SMGRKO mice, cardiomyocyte cross sectional area and nuclei per unit area were equivalent to control littermates at 6 weeks of age, when heart/body weight ratio was already elevated. By 12 weeks of age, cardiomyocyte cross sectional area was greater in male SMGRKO mice than control littermates. In addition, levels of mRNA encoding myosin heavy chain-β, a marker of pathological cardiac hypertrophy, were greater in the LV of male but not female SMGRKO mice at 12 weeks. These findings suggest that cardiomyocyte hyperplasia in early neonatal life, possibly in combination with physiological elongation of cardiomyocytes, may underlie the elevated heart weight in female SMGRKO mice, whereas in male SMGRKO mice there is a transition to potentially pathological hypertrophy of cardiomyocytes. This may occur at puberty, in response to increased androgens, when marked LV growth occurs in males. Histopathology showed LV fibrosis in SMGRKO mice in both sexes, accompanied by elevated levels of mRNA encoding pro-fibrotic and matrix-remodeling genes in the LV. Intriguingly, levels of mRNA encoding the mineralocorticoid receptor (MR) were elevated in both sexes, which may be causal in the development of fibrosis. Indeed, in the LV, levels of mRNA encoding MR were already elevated in 6 week old SMGRKO males, at a time when cardiac collagen levels were only modestly increased. Levels of mRNA encoding the ryanodine receptor, which is essential for cardiac excitation contraction coupling, were reduced in the LV of female, but not male SMGRKO mice. The mechanisms underlying gender differences could be further investigated by comparing neonatal cardiac development in male and female SMGRKO mice. In conclusion, deletion of GR in cardiomyocytes and vascular smooth muscle causes gender specific pathological cardiac remodeling, demonstrating the essential role of cardiovascular GR signalling in cardiac maturation and function. Global GR deficiency alters the trajectory of cardiac development and increases risk of mortality. In surviving GR+/- mice, compensatory adaptations occur in response to the functional impairment seen in utero but subtle cardiac abnormalities remain in adulthood which, together with the elevated blood pressure and GC levels, may contribute to cardiovascular disease risk.

Estrogen and glucocorticoid receptors (ER and GR) are both members of the same subfamily of steroid nuclear receptors, and can both signal classically as ligand-activated transcription factors. However, many responses to estrogen and glucocorticoid exposure occur through the non-classical pathways, which include rapid activation of kinase cascades, activation of membrane-associated receptors, gene regulation through transcription by non-classical transcription factors, and protein regulation by translation and post-translational modification. Male territorial aggression is a hypothalamically-mediated steroid hormone-dependent adaptive behavior in mice. The hypothalamus, which expresses multiple ERs and GRs, is also responsive to estrogen and glucocorticoid treatment at a cellular level. Experiments were conducted to test the effects of estrogen and glucocorticoid interactions on spinogenesis in the ventromedial hypothalamus (VMH) and on male territorial aggression through the resident-intruder paradigm. Studies in male postnatal primary hypothalamic cell cultures demonstrate the expression of classic ERα, the variant ERα-36, and GPR30. PSD-95 protein, a marker for dendritic spines, is increased in response to 12 hours of treatment with the GPR30 agonist G-1 in an ERK/MAPK-dependent manner. Further work in immortalized embryonic hypothalamic cell lines (mHypoE-11 and mHypoE-42) demonstrate non-classical effects of a membrane-limited glucocorticoid on rapid nuclear translocation of the intracellular GR. Additionally, pharmacological inhibition of the ERK/MAPK pathway results in similar GR translocation in the absence of a ligand. Male postnatal primary hypothalamic cell cultures also respond to glucocorticoid exposure with increased 17Î_-E synthesis, suggesting crosstalk between GR signaling and estrogen signaling. Spine density in the gonadally intact adult male VMH decreases following suppression of estrogen synthesis with the aromatase inhibitor letrozole, suggesting estrogen is necessary to maintain spine density. In vivo studies in adult male mice demonstrate that estrogen is necessary to maintain basal peripheral CORT synthesis. Behavior testing using the resident-intruder paradigm showed that dexamethasone-suppression of adrenal CORT synthesis increases the amount of time resident mice spent engaged in aggressive bouts, and CORT treatment 20 minutes prior to aggression testing abolished this effect. The findings presented here provide support for the importance of the interactions between classical and non-classical estrogen and glucocorticoid signaling pathways on hypothalamic spinogenesis and male territorial aggression. / 1 / Jennifer Rainville

Glucocorticoid (GC) signalling via the GC receptor (GR) regulates many aspects of lung development. To determine the need for epithelial GC-GR signalling, triple transgenic (TT) mice with doxycycline (dox) – inducible suppression of GR exclusively in the distal lung epithelium (DLE) were created. Following exposure to dox, E18.5 TT fetuses showed a reduction in GR mRNA levels and elimination of GR protein expression exclusively in the DLE. Newborn TT pups had decreased viability and TT fetal lungs had increased tissue to airspace ratios, decreased levels of proximal epithelial protein CC10, of all surfactant proteins (ATII cell proteins), of ion conductance channels β and γENaC, of water channel AQP5, and of ATI cell protein T1α. Thus DLE GC-GR signalling is important for neonatal viability and increased mortality of TT pups could be due to impaired epithelial differentiation, leading to decreased surfactant protein expression, delayed fluid clearance and/or increased lung cellularity.

Glucocorticoid (GC) signalling via the GC receptor (GR) regulates many aspects of lung development. To determine the need for epithelial GC-GR signalling, triple transgenic (TT) mice with doxycycline (dox) – inducible suppression of GR exclusively in the distal lung epithelium (DLE) were created. Following exposure to dox, E18.5 TT fetuses showed a reduction in GR mRNA levels and elimination of GR protein expression exclusively in the DLE. Newborn TT pups had decreased viability and TT fetal lungs had increased tissue to airspace ratios, decreased levels of proximal epithelial protein CC10, of all surfactant proteins (ATII cell proteins), of ion conductance channels β and γENaC, of water channel AQP5, and of ATI cell protein T1α. Thus DLE GC-GR signalling is important for neonatal viability and increased mortality of TT pups could be due to impaired epithelial differentiation, leading to decreased surfactant protein expression, delayed fluid clearance and/or increased lung cellularity.

Abstract The effects of dietary sucrose and systemic glucocorticoid treatment on the response of the pulpodentinal complex to dental caries were examined in an experimental rat model. The possible role of dentinal caries on dentin formation was also examined. After 5-6 weeks of a dietary and/or medication period, the areas of dentin formation and dentinal caries were quantified in the molars of growing animals. Also the number and severity of caries lesions were estimated. The 43% sucrose diet significantly reduced dentin formation and increased dentinal caries progression. Although glucocorticoid medication alone reduced dentin formation, without dietary sucrose it did not have an effect on caries. In combination of these two, glucocorticoids further increased the progression of dentinal caries, however without significant increase in the number of caries lesions. The cariogenic bacterial inoculation of rats fed a sucrose or control diet increased the progression of dentinal caries. The relationship between cariogenic bacteria and caries was not strong, but there was a stronger relationship between the total amount of dietary sucrose and dentinal caries. In addition to the overall reduction of dentin formation there was no difference in the amount of dentin formed between intact and carious fissures in the sucrose diet group. On the contrary, rats receiving the control diet positively responded to the dentinal caries by increasing dentin formation to prevent pulpal exposure. Whereas the high sucrose diet impaired both the deposition and mineralization of the dentin matrix, glucocorticoids affected matrix formation only. These results indicate that the functional alterations in the pulpo-dentinal complex might contribute to dentinal caries progression in a cariogenic environment, irrespective of the causative mechanism.

Glucocorticoids (GCs) exert diverse effects on multiple cell types and tissues. The variability in GC sensitivity can give rise to disease states hence the importance of GC sensitivity modulators. GCs act through the glucocorticoid receptor (GR), a ligand-activated nuclear hormone receptor (NR), which interacts with the DNA to regulate gene transcription depending on the chromatin structure. GR itself modulates chromatin through epigenetic modification of histone residues. In the present study, novel modulators of GC sensitivity, altering GR-mediated gene expression through dynamic or epigenetic regulatory mechanisms, are identified and explored. Metastasis-related methyltransferase1 (Merm1/WBSCR22), is a histone methyltransferase, previously shown to methylate histone H3 Lysine 9 (H3K9), a repressive methyl mark, to inhibit target gene transcription. Our GR reporter transient transfections assays showed that Merm1 potentiated GR transactivation through its methyltransferase and SAM domains. Merm1 knockdown significantly impaired both GR transactivation, and transrepression of endogenous genes, including GILZ. The ChIP assay analysis confirmed that both GR and Merm1 bound the GILZ promoter and Merm1 regulated ligand-induced GR recruitment. Merm1 regulated tri-methylation of H3K4 (H3K4me3) and di-methylation of H3K79 (H3K79me2). At the GILZ locus, GR induced H3K4me3 and inhibited H3K79me2. Merm1 regulated both of these and also maintained basal H3K79me2. The GR-induced H3K4me3 followed by loss of H3K79me2 showed that these events were driven by H3K4 methylation. In conclusion, Merm1 regulates chromatin structure to affect GR recruitment, and mediates GR actions of transcription by histone methylation. In the second part of the thesis, the biological consequence of temporal dynamics of GC delivery to target cell gene expression and apoptosis has been investigated. For this purpose a flow-through culture system was designed and modified for pulsatile and continuous delivery of GC to HeLa cells and primary T cells. Pulsatile cortisol caused a significant reduction in cell survival compared to continuous exposure of the same cumulative dose in HeLa population. This was due to increased apoptosis. Transcription factor (TF) binding site analysis of the microarray data identified CCAAT-displacement protein (CDP) as a common TF binding site in the differentially regulated target genes. Mouse mammary tumour virus (MMTV) gene is regulated by CDP and is also GC responsive. MMTV-Luc was also differentially regulated between pulsatile and continuous cortisol. In primary T cells, GILZ and FKBP5 genes were more highly induced with continuous than with the same equivalent concentration given in pulses. In conclusion, cortisol oscillations exert important effects on target cell gene expression, and phenotype. In summary, GC sensitivity is modulated via different mechanisms. Our data illustrate a novel regulatory mechanism whereby GR activity is altered through histone modifications and chromatin remodelling. In addition, GC oscillations provide frequency modulation to GR-mediate gene expression with a resulting differential pattern of gene transcription and cellular response.

We compared the binding of [3H]-dexamethasone in cytosols prepared from atria and ventricles. The effects of steroid treatment and adrenalectomy on receptor concentration were measured in both male and female rats. In male rats the distribution of receptors was similar in atria and ventricles. In contrast, the atria of female rats in all treatment groups had twice the number of receptors as did the ventricles. Adrenalectomy in females resulted in receptor up-regulation, but dexamethasone treatment, which was effective in reducing binding in males and in the ventricles of females, failed to alter atrial binding in females. These results suggest that the atria of female rats may be more responsive than ventricles to the effects of circulating glucocorticoids.

Duchenne muscular dystrophy (DMD) is the most common type of muscular dystrophy caused by the loss of functional dystrophin. DMD is characterized by scoliosis, muscle wasting, loss of ambulation and a reduced life span. The first line of treatment for DMD is glucocorticoids (GCs). GCs are prescribed primarily for their anti-inflammatory and immunosuppressive effects; however, GC treatment is known to cause significant muscle atrophy. In DMD, GC treatment has been shown to improve muscle strength for the first 6 months and stabilization of the disease for up to 3 years. However, long term treatment reduces muscle function and accelerates disease progression. It is paradoxical that we use a medication that causes muscle wasting to treat a muscle wasting disease. The regeneration and function of muscle is dependent on the proper regulation and functioning of muscle satellite cells (MuSCs) to restore and repair muscle tissue. The impact GCs have on MuSCs from activation to proliferation and differentiation into muscle fibers is not well understood. GCs have many mechanisms of action by acting as a ligand to the glucocorticoid receptor (GR) to cause downstream effects by direct DNA binding or indirectly by regulating proteins. To study the role of GCs, we examined the effects of GC treatment on myoblast morphology, the cytoskeletal network, post-translational modifications (PTMs) of tubulin subunits, and the organization of microtubule organizing centers (MTOCs) in proliferating and differentiating myoblasts. This study shows that the GR is an essential regulator of myotube morphology and proper myonuclei placement. Furthermore, dexamethasone (DEX) treatment causes branching of the MT network, as well as an increase in the expression of the stabilizing MT markers, acetylated and detyrosinated tubulin during early differentiation. DEX treatment was also found to misposition the Golgi complex, a primary MTOC for the cytoskeletal network, from the periphery of the nucleus to the center of the nucleus during early differentiation. Finally, we found very few differentially expressed genes between WT and GRMuSC-/- myoblasts between early and late differentiation, indicating that these morphological defects we see are not due to GCs regulating gene expression. Thus, GCs act through the GR to modify the MT network during early differentiation, causing morphological changes in myoblasts that persist throughout differentiation.

Glucocorticoids are steroid hormones that affect a variety of physiological and pathological processes both throughout development and in adult life. During mammalian fetal growth, the late gestation rise in fetal glucocorticoid levels is essential for the maturation of tissues and organs in preparation for birth. In humans, glucocorticoids are routinely administered to women threatened by a preterm labour to accelerate fetal lung maturation and prevent neonatal respiratory distress and mice lacking glucocorticoid receptor (GR-/- mice) die neonatally as they are unable to inflate their lungs due to severe pulmonary immaturity. Apart from their importance for proper lung maturation, the physiological role of glucocorticoids in the development of other organs and tissues is not well known. However, prenatal exposure to excess glucocorticoids was shown to elicit detrimental “programming” effects, raising the susceptibility to adult diseases such as hypertension, obesity and metabolic disturbances in both humans and animal models. I therefore used global and conditional GR knock out mouse models to investigate the role and importance of adequate glucocorticoid signalling in fetal heart development and maturation. I further confirmed the direct effects of glucocorticoids on the cardiomyocyte structure and function in an in vitro setting. GR-/- fetuses are under-represented in late gestation (>50% of the number of GR+/+ littermates) but are present in the expected mendelian ratio at E14.5. At E17.5, GR-/- fetuses show edema (increased fluid accumulation and body sodium content). Excess extracellular fluid accumulation could be a result of a congenital heart failure. During development, corticosterone levels sharply increase within the fetal hearts at E15.5-E16.5, coincident with nuclear translocation of GR. Consistent with activation of GR only after this time, the phenotypic consequences of GR deficiency can be seen after E16.5 and not before. At E17.5, hearts of GR-/- fetuses are smaller than in GR+/+ but display no structural abnormalities. Cardiac function however is severely impaired, with left ventricular systolic and diastolic performance inferior in GR-/- fetuses compared to their wild-type littermates. Microscopically, at E17.5, the structure of the cardiac muscle and individual cardiomyocytes are affected by the lack of GR. The normal outer muscle layer, with characteristic rod-shaped, aligned cardiomyocytes is not discernable in the GR-/- heart. Within the cardiomyocytes, myofibrils are short, undefined and randomly scattered within the cell. Lack of the maturational progression in the GR-/- hearts at E17.5 is evident in the pattern of gene expression. GR-/- fetuses do not display the normal gestational changes between E14.5 and E17.5 that are seen in control mice, including in genes involved in the maturation of cardiac structure (eg myosin heavy chain-α, MyHC-α), function (atrial natriuretic peptide, ANP), energy metabolism (eg hexokinase-1, PPARγ coactivator-1α, PGC-1α) and calcium handling (ryanodine receptor, RyR; sarcoplasmic reticulum Ca2+-ATPase, SERCA2a). However, there are no genotype or gestational alterations in mRNA encoding the mineralocorticoid receptor, which is also a receptor for glucocorticoids in the heart. The normal gestational changes in the levels of modified histone H3 associated with the promoters of some of the genes (MyHC-α, ANP, PGC-1α) are not seen in hearts of GR-/- fetuses. This cardiac phenotype was not secondary to adrenal catecholamine insufficiency reported in other GR-/- models, as peripheral tissue levels of adrenaline were not different between genotypes. In order to test the hypothesis that the effects of glucocorticoids on the heart are mediated via GR in cardiomyocytes and to further elucidate the direct effects of GR deficiency specifically within the heart, mice with conditional deletion of GR selectively in cardiac and vascular smooth muscle cells were generated ("SMGRKO" mice). These show ~65% reduction in cardiac GR mRNA and protein levels. Circulating levels of corticosterone do not differ between genotypes at E17.5. SMGRKO fetuses at E17.5 display a phenotype strikingly similar to that of global GR-/-, namely edema, impaired cardiac function, impaired cellular architecture within the ventricle and alterations in the gene expression, implying that the GR-deficient phenotype is largely due to the direct actions of GR within the heart and not secondary to effects on other systems (eg kidney or liver). In order to investigate the pathways by which GR stimulates cardiomyocyte maturation, an in vitro model of murine primary fetal (E15.5-E16.5) cardiomyocytes was developed. Cultures contain >98% of troponin Tpositive cells which beat spontaneously. Treatment of cardiomyocytes with either synthetic (dexamethasone) or physiological (corticosterone) glucocorticoid induces time- and dose-dependent changes in gene expression, consistent with glucocorticoid-dependent changes seen in vivo in the late gestation heart. The effects of glucocorticoids on gene expression were abolished by either siRNA mediated knock-down of GR or RU486 antagonism of GR, but were unaffected by a mineralocorticoid receptor (MR) antagonist. Moreover, cycloheximide pretreatment (to block protein synthesis) suggested PGC-1α as a direct genomic target of GR. RNAseq transcriptome analysis performed on cardiomyocytes treated with dexamethasone and cycloheximide for 2h identified >600 genes as possible rapid and direct glucocorticoid response targets. Among them are genes involved in energy metabolism, calcium handling and sarcomere assembly. Glucocorticoid treatment of fetal cardiomyocytes also induces striking structural changes – formation of stress troponin T-associated actin fibers and sarcomere assembly. Spontaneous contractile activity is improved by glucocorticoid treatment, with a decrease in both contraction and relaxation time (without a change in frequency) and an improvement in the relaxation kinetics. In summary, glucocorticoid signalling in cardiomyocytes is required for the functional, structural and transcriptional maturation of the fetal heart in late gestation in vivo. Glucocorticoid treatment of primary murine fetal cardiomyocytes replicated the contractile, transcriptional and structural changes seen in vivo and was dependent on GR. Thus, GR is essential in cardiomyocytes for the structural and biochemical changes that underlie the maturation of heart function around the time of birth and an inadequate glucocorticoid environment could potentially lead to detrimental and permanent changes in postnatal cardiac function. Since prenatal glucocorticoids are routinely used clinically, it is important to consider any possible effects they might have on the heart development and its function later in life.

This work presents the first evidence of a ligand-independent role for the glucocorticoid receptor (GR) as a positive regulator of gene expression in mammary cells. We have demonstrated that unliganded GR interacts directly with the promoter of the tumour suppressor gene BRCA1, and upregulates its expression. The presence of the stress hormone hydrocortisone (HC) abolished this interaction and resulted in repression of BRCA1. Since low levels of BRCA1 have been implicated in the development of sporadic breast cancer, this may represent a novel mechanism through which prolonged stress signaling increases breast cancer risk. We determined that the interaction between unliganded GR and BRCA1 is mediated through the beta subunit of the Ets transcription factor GABP at the RIBS promoter element. GR and GABPβ were shown to interact in both co-immunoprecipitation and mammalian two-hybrid assays, and this interaction involved the N-terminal to central regions of both proteins. To further characterize the role of unliganded GR in breast cells, we used shRNA to generate mouse mammary cell lines with depleted endogenous GR expression. Loss of GR resulted in an impaired capacity of cells to differentiate into acini, but this effect was rescued by the addition of glucocorticoids, implicating both the liganded and unliganded forms of GR as key regulators of differentiation. We performed expression microarray to identify targets of unliganded GR using the GR-depleted cell lines. This analysis revealed 260 genes negatively regulated and 343 genes positively regulated by unliganded GR. Many of the positively regulated genes were involved in pro-apoptotic networks, and appeared to oppose the activity of liganded GR targets. Validation and further analysis of five candidates of positive regulation by unliganded GR indicated that two of these, Hsd11b1 and Ch25h, were regulated by unliganded GR in a manner similar to Brca1. The Hsd11b1 enzyme regulates intracellular glucocorticoid levels by interconverting cortisol and its inactive metabolite, cortisone. Further investigation of Hsd11b1 expression and regulation indicated that Hsd11b1 activity appears to be unidirectional in breast cells, specifically inactivating cortisol. Overall, this work suggests that gene regulation by unliganded GR represents a mechanism for protecting the breast from tumourigenesis during stress. / Thesis (Ph.D, Biochemistry) -- Queen's University, 2012-11-29 11:18:14.596