Studies on Estradiol, growth hormone, and suppressors of cytokine signalling-2, and the influence in liver metabolism

Anexo II UNIVERSIDAD DE LAS PALMAS DE GRAN CANARIA Facultad de Veterinaria Programa de doctorado: Clínica Veterinaria e Investigación Terapéutica Título de la Tesis “Studies on Estradiol, Growth Hormone and Suppressor of Cytokine Signaling-2, and the influence in liver metabolism” Tesis Doctoral presentada por Dª. Mercedes de Mirecki Garrido Dirigida por el Dr. Leandro Fernández Pérez Codirigida por el Dr. Carlos Borja Guerra Hernández El Director, El Codirector, La Doctoranda, Las Palmas de Gran Canaria, a 07 de Mayo de 2014 Studies on Estradiol, Growth Hormone and Suppressor of Cytokine Signalling-2, and the influence in liver metabolism Mercedes de Mirecki Garrido Thesis for Doctoral degree (Ph.D.) University of Las Palmas de Gran Canaria Faculty of Veterinary Department of Clinical Sciences Pharmacology Unit Las Palmas de Gran Canaria 2014 Hasta la más larga caminata empieza por un primer paso (Confucio) To my Family LIST OF PUBLICATIONS This thesis is based on the following papers, which are referred to by their Roman numbers in the text: I. Fernández-Pérez L, Santana-Farré R, Mirecki-Garrido M, Garcia I, Guerra B, Mateos-Díaz C, Iglesias-Gato D, Díaz-Chico J.C, FloresMorales A, Díaz M 2014. Lipidic and transcriptomic analysis reveals a functional interplay between estradiol and Growth Hormone in Liver. PLOS ONE 9(5):e96305. II. Mirecki-Garrido M, Iglesias D, Borja Guerra, Mateos-Díaz C, FloresMorales A, Fernández-Pérez L 2014. Then Suppressor of Cytokine Signaling-2 influences the hepatic effects of 17β-Estradiol on liver transcriptome (manuscript in preparation). III. Zadjali F, Santana-Farré R, Vesterlund M, Carow B, Mirecki-Garrido M, Hernández-Hernández I, Flodström-Tullberg M, Parini P, Rottenberg M, Norstedt G, Fernández-Pérez L, Flores-Morales A 2012. SOCS2 inactivation protects against hepatic steatosis but worsens insulin resistance in high fat diet fed mice. FASEB J 26 (8):32823291. IV. Mirecki-Garrido M*, Alkharusi A*, Flores-Morales A, Fernández-Pérez L, Norstedt G 2014. Suppressor of Cytokine Signaling-2 gene deletion influences diabetes development induced by multiple low-dose streptozotocin (manuscript in preparation). *M. Mirecki-Garrido and A. Alkharusi have contributed equally to this work RELATED PUBLICATIONS I. Mirecki-Garrido M, Guerra B, Mateos-Díaz C, Jiménez-Monzón R, Díaz-Chico N, Díaz-Chico JC, Fernández-Pérez L 2012. Influences of estrogens on biological and therapeutic actions of Growth Hormone in liver. Pharmaceutical 5(7):758-778. II. Santana-Farré R*, Mirecki-Garrido M*, Bocos C, Henríquez-Hernández LA, Kahlon N, Herrera E, Norstedt G, Parini P, Flores-Morales A, Fernández-Pérez L 2012. Influence of neonatal hypothyroidism on hepatic gene expression and lipid metabolism in adulthood. PLoS ONE 7(5):7386. III. Zadjali F, Santana-Farre R, Mirecki-Garrido M, Ellis E, Norstedt G, Fernandez-Perez L, Flores-Morales A 2011. Liver X receptor agonist downregulates Growth Hormone signaling in the liver. Horm Mol Biol Clin Investig 8:471-478. * M. Mirecki-Garrido and R Santana-Farré have contributed equally to this work TABLE OF CONTENTS Page 1. INTRODUCTION 1 1.1.Growth Hormone 1 1.1.1. Growth Hormone Secretion - The sexually dimorphism of pituitary Growth Hormone secretion 1.1.2. Growth Hormone Signaling - Inactivation of Growth Hormone signaling 1.1.3. Growth Hormone regulates somatic growth, metabolism immunity and gender dimorphism of hepatic gene transcription 1.2. The Suppressor of Cytokine Signaling (SOCS) 12 1.2.1. SOCS Proteins Structure 1.2.2. Mechanism of action of SOCS proteins 1.2.3. Physiological actions of SOCS proteins 1.3. Estrogens 19 1.3.1 . Estrogens Signaling 1.3.2. Estrogens regulate somatic growth and lipid and glucose metabolism 1.4. Growth hormone and Estrogens interaction in liver 24 1.5. Relationship between the Metabolic Syndrome and GH, E2 and SOCS 27 1.6. Relevance of Pharmacogenomic to explore molecular mechanisms of GH, E2 and SOCS2 29 2. AIMS OF THE STUDY 31 3. MATERIALS AND METHODS 32 3.1. Animals Study design 3.2. Somatic growth and food intake analysis 3.3. Intraperitoneal Glucose Tolerance and Insulin Tolerance Tests. 3.4. Serum biochemistry 3.5. RNA isolation 3.6. cDNA microarray 3.7. Microarray data processing and analysis 3.8. Analysis of gene expression by real time quantitative-PCR (qPCR) 3.9. Histochemistry 3.10. Pancreas insulin analysis 3.11. Hepatic lipid analysis 3.12. General Stadistical Analisys 4. RESULTS AND DISCUSSION 42 4.1. Paper I: Lipidic and transcriptomic analysis reveals a functional interplay between E2 and GH in liver. 4.2. Paper II: Transcriptomic analysis reveals that SOCS2 influences the hepatic effects of E2 in vivo. 4.3. Paper III: SOCS2 inactivation protects against hepatic steatosis but worsens insulin resistance in high-fat-diet fed mice. 4.4. Paper IV: SOCS2 gene deletion influences diabetes development induced by multiple low-dose streptozotocin in mice. 5. CONCLUSSIONS AND GENERAL DISCUSSION 74 6. FUTURE PERSPECTIVES 77 ACKNOWLEDGEMENTS FUNDING RESUME OF THE MAIN ASPECT OF THE THESIS IN SPANISH REFERENCES MAIN PUBLICATIONS RELATED PUBLICATIONS LIST OF ABREVIATIONS ALT Alanine aminotransferase AR Androgen Receptor CD Control Diet CHO Cholesterol CIS Cytokine-inducible SH2 protein Cyp Cytochrome gene DM Diabetes Mellitus E2 17β-estradiol ER Estrogen Receptor FFA Free Fatty Acids GGT Gamma glutamil transpeptidasa GH Growth Hormone GHR Growth Hormone Receptor GR Glucocorticoid Receptor GHRH GH releasing hormone. HFD High-Fat Diet HOMA Homeostatic model assessment IGF-I Insulin-like Growth Factor I IGFBP Insulin Growth Factor Binding Protein ip intraperitoneally JAK Janus kinase KIR Kinase Inhibitory Region IL Interleukin LFD Low-Fat Diet LPS Lipopolysaccharide MLDSTZ Multiple Low Dose of Streptozotocin NAFLD Non-Alcoholic Fatty Liver Disease OX Orchidectomy PRL Prolactin sb subcutaneously SH2 Src homology 2 domain SHP SH2-Containing Protein Tyrosine Phosphatase SOCS Suppressor of Cytokine Signaling SOCS2-/- SOCS2 knockout mouse SS Somatostatin. SSI STAT-Induced STAT Inhibitor STAT Signal Transducers and Activator of Transcription TG Triacylglycerols TNFα Tumor Necrosis Factor α WT Wild type mice E2, GH, SOCS2, and Liver 1 1. INTRODUCTION 1.1. Growth Hormone Growth Hormone (GH) is the main regulator of somatic growth, metabolism, and gender dimorphism in liver [1,2,3,4]. GH is predominantly linked to linear growth during childhood, but continues to have important metabolic actions throughout life. Its primary somatic effect is the promotion of longitudinal growth [5,6] but also induce diverse effects on cell growth, differentiation and metabolism. GH promotes the proportional growth of several organs in the body including liver, muscle and bone and exerts effects on cellular differentiation, metabolism and nutrient uptake [7,8,9,10,11]. GH is also involved in the regulation of immune cells and hematopoiesis [12,13] and can act on the brain to influence emotion, behaviour and other cognitive responses [14,15]. GH actions are pleiotropic and influenced by factors such as age, gender (sex hormones) and pituitary secretory pattern [16,17,18]. 1.1.1. Growth Hormone Secretion. The somatotropic cells of the anterior pituitary gland are the primary producers of GH in the body. The regulation of pituitary GH secretion involves a complex neuroendocrine control system that includes the participation of several neurotransmitters and the feedback of hormonal and peripheral (metabolic) factors [19,20] (Fig.1). GH secretion from pituitary gland is regulated by two major hypothalamic peptides: the stimulatory GH releasing hormone (GHRH) and the inhibitory hormone somatostatin (SS). The balance of these stimulating and inhibiting peptides is in turn, indirectly, affected by many physiological stimulators (i.e, sex hormones, nutrients, sleep, and exercise,) and inhibitors (i.e., insulin growth factor I (IGF-I), and GH). In addition to hypothalamic (GHRS, SS) and endocrine (IGF-I, GH) factors, other peripheral (metabolic) factors influence pituitary GH release: free fatty acids (FFA), insulin, glucose, amino acids, leptin, neuropeptide Y, and ghrelin. These Mercedes de Mirecki Garrido 2 factors are primarily related to or derived from the metabolic status of the organism, which is consistent with the role of GH in regulating substrate metabolism, adiposity, and growth, and appear to coordinate the metabolic status of the organism with GH secretion [21,22,23,24]. One GH is released into the bloodstream it travels to target organs where it binds to the Growth Hormone Receptor (GHR). GHR is ubiquitously present with the highest levels found in the cells of the liver. GHR signaling triggers the expression and release of hepatic IGF-I, which is a major mediator of somatic GH action [25,26]. In addition to the pituitary, GH is produced in extra-pituitary tissues, which indicates that GH has local paracrine-autocrine effects, distinct from its classic endocrine somatotropic effects [27]. Figure 1. Regulation of pituitary GH secretion (Based on [28]). - The Sexually Dimorphism of Pituitary Growth Hormone Secretion. The secretion of GH is controlled by neuroendocrine factors, which regulate pituitary GH secretion in a sexually dimorphic manner in many species, including rats, mice and humans [29,30,31,32]. Gender dimorphism in GH secretory patterns can, at E2, GH, SOCS2, and Liver 3 least partly, explain differences in growth and liver physiology between males and females. The sexually dimorphic pattern of GH secretion is also seen in humans, but not as marked as in the rat. Sex steroids are physiological regulators of pituitary GH secretion and, indirectly, regulate sex-specific liver physiology [33]. From neonatal period of life, gonadal steroids play a critical role to maintain liver response to GH in adulthood. Sexual dimorphism in rodents seems to be regulated by estrogen secretion in adult females and by androgen secretion, neonatally and during adulthood, in males. Neonatal exposure to testosterone imprints the male program of neuroendocrine control of the pulsatile pituitary GH secretion that is first seen at puberty, when the adult pattern of GH secretion becomes evident, and continues throughout adulthood. The male characteristic metabolism in liver in adulthood is dependent on continuous androgen exposure. If such an androgen re-programming does not occur, the secretion pattern will remain as the feminine pattern (continuous GH secretion) [33]. In female rats, gonadectomy has little impact on hepatic steroid metabolism. However, estrogen treatment feminizes hepatic metabolism in male rats [34] (Fig.2). Figure 2. Plasma GH profile in adult male (A) and female (B) rats (Based on [35,36,37,38]) Mercedes de Mirecki Garrido 4 Figure 2A shows that the pulsatile plasma GH profile in adult male rats is characterized by peaks of plasma GH every 3-4 hours followed by interpulse interval where plasma GH levels are undetectable [33]. Consequently, intracellular activation of signal transducer and activator of transcription (STAT)-5b, main executor of GH signaling in liver, is also episodic and periods with low GH circulating levels are required to achieve maximal activation of STAT5. Adult female rats are characterized by more frequent pituitary GH release and a near continuous presence of GH in plasma (Fig. 2B). Female rats show reduced STAT5b activation compared with males [33]. These differences in STAT5b activation are responsible for several of the gender differences in hepatic gene expression. For example, the GH-free interval observed in the males (rats and mice) is required for the expression of male-specific liver genes, such as CYP2C11 [39], and most likely reflects the need to reset a GH-activated STAT5b signaling pathway. Genome-wide screens of gene expression have shown that GHand sexdependent regulation of hepatic gene expression is not confined to steroid or drug metabolism. Moreover, a number of other hepatic genes have been found to be upand/or down-regulated by the different patterns of GH or sex-steroid exposure. GHand sex-dependent hepatic transcripts encoding plasma proteins, enzymes, transcription factors and receptors involved in the metabolism of proteins, carbohydrates, lipids, or signaling regulation have been identified [40,41,42]. A consensus exists that the response to sex-different GH patterns is the major cause of gender dimorphism in liver; however, it is likely that factors other than the sexually dimorphic pattern of GH secretion are behind some sex differences in rat liver. Potential mechanisms that could contribute to “liver sexuality” are the pituitary-independent effects of estrogens through interaction with ERα or GHJAK2-STAT5 signaling pathway in liver. E2, GH, SOCS2, and Liver 11 protein synthesis without increasing proteolysis [81]. Muscle specific deletion of GHR mice are protect from High-Fat Diet (HFD) induced insulin resistance, this could be attributable to the decreased diabetogenic GH action on the muscle [82]. In adipose tissue GH promotes lipolysis, targeted deletion of GHR in fat resulted in a doubling of the fat mass, but has no effect on glucose homeostasis [83]. GH also seems to play a role in insulin secretion from the β cells in the pancreas. Mice with GHR deletion in the β cells exhibit decreased glucose stimulated insulin release and decreased β cell hyperplasia in response to HFD [84]. - Immunity. Besides its actions on postnatal growth and metabolism GH also plays a role in the regulation of immune system. Several different immune cells have been shown to express both GHR, which suggest that GH signalling directly affects immune function, and GH, which implies that paracrine and autocrine hormone secretion might play role in immunity [85]. GH is able to directly affect cytokine production by immune cells, but the mechanism and effects are unclear. GH have been shown to promote the production of inflammatory cytokines such as Interleukin (IL) -1α, IL-6 and Tumor Necrosis Factor α (TNFα) in immune cells both in vitro [86] and in vivo [87]. In a Finnish study, GH treatment of critically ill patients was found to increase both morbidity and mortality, likely through modulation of immune function [88]. While it is clear that GH modulates immune functions, the distinct effects and underlying mechanisms remain to be elucidated. - Gender Dimorphism of Hepatic Gene Transcription. Sex hormones imprint a sex-dependent pattern of pituitary GH hormone secretion which is a major player in establishing and maintaining the sexual dimorphism of hepatic gene transcription that emerges in rodents at puberty [33]. Genomic and bioinformatics analysis have contributed to solve molecular mechanisms involved in GHregulated hepatic gene transcription [40,41,46,47,89]. Sex-dependent expression and GH regulation characterizes several families of hepatic genes involved in Mercedes de Mirecki Garrido 12 endoand xenobiotic metabolism as well as relevant metabolic functions (e.g., lipid metabolism); 20-30% of all hepatic genes have a sex-specific expression pattern in rodents. Most of these hepatic sex differences are explained by the female-specific secretion pattern of GH, through the induction of femalepredominant transcripts and suppression of male-predominant. Estrogens cause opposite effects on lipid and glucose metabolism which represents a relevant point of regulatory interactions between estrogens and GH. 1.2. The Suppressors of Cytokine Signaling SOCS protein family was identified in 1997 by three different groups [90,91,92]. At present, the SOCS family represents eight different members: cytokineinducible SH2 protein (CIS) and SOCS-1 to SOCS-7. The SOCS proteins have been shown to modify cytokine actions through a classic negative feedback loop. The SOCS family utilizes a feedback loop to inhibit cytokine responses and block the activation of the JAK/STAT signaling pathway [93]. Although today their greatest importance is held to be in the regulation of the cytokine-induced JAK/STAT pathway, these proteins are both induced by and are negative regulators of more than just cytokines, suggesting a role as signal modulators much wider than implied by their names, evidence indicates that cytokines, growth factors and steroid hormones, and several xenobiotics can regulate SOCS expression [94]. As mentioned above, intracellular GH signaling is regulated by SOCS proteins that influence the action GH elicits. 1.2.1. SOCS Proteins Structure. All SOCS proteins display a three-part architecture [93]. A central Src homology 2 (SH2) domain is involved in substrate binding through recognition of cognate phosphotyrosine motifs. It is flanked by a variable N-terminal region and a conserved C-terminal domain known as the SOCS box. The N-terminal domain contains an extended SH2 subdomain (ESS) E2, GH, SOCS2, and Liver 13 that contributes to substrate interaction [95,96]. Additionally, SOCS1 and SOCS3 present an extra domain, which is named (KIR) and is located at the N-terminal region near the SH2 domain. Members of the SOCS family contain N-terminal regions of variable length. For example, CIS, SOCS1, SOCS2, and SOCS3 have relatively short (50-80 residue) N-terminal regions, whereas SOCS4, SOCS5, SOCS6, and SOCS7 have longer N-terminal regions of up to 2700 residues [97], suggesting these four proteins form a sub-group within the SOCS family. SOCS4 and SOCS5 share greater sequence homology with each other than with other members of the SOCS family [97], with conservation largely restricted to the SH2 domain (92% amino acid identity) and suggests that while the SH2 domains may have an overlapping binding specificity [97], the N-terminal regions will have unique protein targets (Fig.4). The SOCS box is a structural domain found at the C-terminus it is usually coupled to a protein interaction module such as an SH2 domain in case of SOCS proteins. The SOCS box participates in the formation of E3 ligase complexes, marking activated cytokine receptor complexes for proteasomal degradation. A similar mechanism was recently uncovered for controlling SOCS activity itself. The SOCS box can also add unique features to individual SOCS proteins [97,98]. Figure 4. Schematic representation of SOCS protein structure domains (Based on [93]). Mercedes de Mirecki Garrido 14 1.2.2. Mechanism of action of SOCS proteins. Since the discovery of the first member of the SOCS family, significant progress has been made in understanding the function and importance of these proteins. All SOCS proteins make use of their SH2 domain to bind to phosphorylated tyrosine residues. There are four major ways by which SOCS inhibits cytokine signaling: (1) blocking STAT recruitment to the cytokine receptor; (2) targeting the receptor for degradation by the proteasome; (3) binding to JAKs and directly inhibiting their kinase activity; (4) targeting JAKs for degradation by the proteasome (Fig. 5). First, SOCS proteins can suppress signaling by competing with downstream signal transducers for binding to share phosphorylated motifs of the activated receptor [99]. Therefore, the SOCS box can add unique features to individual SOCS proteins: it can function as an adaptor domain as was demonstrated for SOCS3 or as a modulator of substrate binding in case of CIS [100]. Second, SOCS proteins can regulate signal transduction by linking their substrates to the ubiquitination machinery via the SOCS box. Ubiquitination of a substrate via SOCS proteins can lead to its proteasomal degradation. In line with this, signal transduction of several cytokines is prolonged in the presence of proteasome inhibitors [101,102]. The SOCS box participates in the formation of E3 ligase complexes, marking activated cytokine receptor complexes for proteasomal degradation. A similar mechanism was recently uncovered for controlling SOCS activity itself, since SOCS2 was found to enhance the turnover of other SOCS proteins. Third, a small kinase inhibitory region (KIR) found in the N-terminal domain of SOCS1 and SOCS3 inhibits the activity of JAKs by acting as a pseudo-substrate [95]. SOCS1 can directly bind to the phosphorylated activation loop of JAK2, whereas SOCS3 shows only weak affinity for JAK2 and is thought to bind to the receptor in close proximity of the kinase [103,104]. E2, GH, SOCS2, and Liver 15 Figure 5. Schematic representation of SOCS-dependent negative regulation of GHRJAK-STAT signaling pathway (Based on [93]). Promiscuity and redundancy are notable features of the SOCS proteins system when tested in vitro. Any single SOCS can be induced by many cytokines in vitro and in turn act on several cytokine receptors, not necessarily the ones inducing their expression. GH induces the expression of CIS, SOCS1, -2, and -3; all of which have shown negative actions on the GHR when forcibly overexpressed in cell lines. Evidence also indicates that growth factors that do not belong to the 4helical bundle cytokine family (e.g. insulin, chemokines), and even steroid hormones, can induce SOCS expression. Consequently, regulation of SOCS protein expression provides the mean for cross-talk where multiple factors can regulate the activity of specific cytokines [55] being regulated the SOCS family at the transcriptional, translational and post-translational levels. Mercedes de Mirecki Garrido 16 Some emerging evidences suggest that SOCS molecules might also be degraded by others SOCS family members. SOCS2 appears to enhance the degradation of SOCS1 and SOCS3, and possibly CIS. SOCS6 and SOCS7 might also play a role in cross-modulation of other SOCS proteins. The SOCS box of SOCS2 appears to facilitate the targeting of others SOCS molecules for proteasomal degradation. [105]. 1.2.3. Physiological actions of SOCS proteins. During the last few years considerable progress has been achieved regarding the role of specific SOCS proteins by the generation of knockout or transgenic mice [106]. In vitro various SOCS proteins can inhibit different cytokine receptors and in turn individual SOCS proteins can be induced by a bewildering variety of different stimuli. Thus, in vitro redundancy and pleiotropy characterize the SOCS proteins. However, in vivo data have shown that individual SOCS proteins show at least some degree of specificity for certain cytokines [107]. - Cytokine Inducible SH2-containing protein (CIS) was the first SOCS family member to be described and it is the most closely related to SOCS2 [100]. CIS−/− mice did not present a severe phenotype; however, it was recently suggested that hematopoietic growth factors (e.g. IL-3, erythropoietin) might be affected [108]. CIS also plays a role in limiting GM-CSF, IL-2, prolactin and growth hormone signaling by inhibition of STAT5 activation. CIS-transgenic mice resemble Stat5b- /- mice, with defects in growth and lactation due to reduced growth hormone and prolactin signaling. CIS-transgenic mice additionally display enhanced TCR signaling and impaired responses to IL-2 [109]. - Suppressor of Cytokine Signaling (SOCS)-1, -3, are the most investigated members of the SOCS family. They have been implicated in the regulation of cytokine signaling related to immunological function but also affects hormone E2, GH, SOCS2, and Liver 17 signaling and metabolism. SOCS1−/− mice are normal at birth, they exhibit stunted growth and die within 3 weeks of birth, with a syndrome characterized by severe lymphopenia, activation of peripheral T cells, fatty degeneration and necrosis of the liver, and macrophage infiltration of major organs [110]. SOCS3−/− mice die during the embryonic stage of development due to placental function defects. In other words, deletion of SOCS3 causes embryonic lethality; these embryos can be saved, however, by a tetraploid rescue approach. These observations demonstrate SOCS3’s essential role in placental development and non-essential role in embryo development. Rescued SOCS3-/- mouse embryos exhibit prenatal lethality with cardiac hypertrophy, which suggests that SOCS3 is essential for regulating either LIF receptors or gp130 signaling [110]. SOCS3-deficient mice clearly show that although SOCS3 is essential for G-CSF, IL-6, LIF and leptin signaling, it is in fact, dispensable for regulation of IFNγ signaling [109]. - Suppressor of Cytokine Signaling (SOCS) -4, -5, -6 and -7. The latest additions to the SOCS family are also its largest member [97]. SOCS4−/− mice have not been reported and SOCS5 and SOCS6-deficient mice do not display an overt phenotype. Deletion of the SOCS7 gene highlights a critical role in regulation of insulin signaling [111,112,113]. SOCS7−/− mice are slightly smaller than the wild-tyoe littermates and they suffer from hydrocephalus and have enhanced insulin signaling, likely due to an increase in β-cell mass [111,114]. - Suppressor of Cytokine Signaling (SOCS)-2, The phenotype of SOCS2 null mice identifies SOCS2 as the key physiological player in the negative regulation of GH-dependent somatic growth [60,115,116]. As mentioned above, SOCS2 acts as an ubiquitin-ligase for the GHR, inhibiting GH signaling [60]. Consequently, SOCS2−/− mice are characterized by marked gigantism with increased bone length and proportional enlargement in skeletal muscles due to increased GH sensitivity [117,118]. SOCS2−/− male mice are characterized by a 40% increase in body Mercedes de Mirecki Garrido 18 weight, SOCS2−/− mice are giants but not obese [117]. SOCS2−/− female mice are a 20% heavier than wt littermates. A similar phenotype is also observed in the highgrowth (hg) mutation, which is caused by a deletion of the socs2 locus [115]. Dual-Knockout of SOCS2 and STAT5b and crossing of SOCS2−/− mice with GHRH signaling deficient mice abolished the original phenotype and confirmed that the increased growth observed in SOCS2-/- mice is due to increased sensitivity to GH but with normal systemic insulin-like growth factor-1 (IGF-I) levels [116,119]. Because SOCS2−/− mice do not have increased circulating IGF-1 levels, it is likely that the increased bone growth and observed structural differences within SOCS2−/− growth plates are a direct consequence of altered SOCS2mediated GH/IGF-1 signaling local to the growth plate [120,121,122,123]. Our own studies revealed that SOCS2−/− mice are protected from high-fat-diet (HFD)- induced steatosis. Paradoxically, the HFD-fed SOCS2−/− mice showed worsening of glucose tolerance and exacerbated inflammatory response that was manifested by enhanced production of inflammatory cytokines in liver and fat tissues [62]. Recently, we have identified SOCS2 as an important regulator of hepatic homeostasis (lipid and glucose metabolism and inflammation) under conditions of high-fat dietary stress [124]. In addition previous work has shown that SOCS2 deficiency induces changes in hepatic gene expression that only partially overlap with known GH induced effects, suggesting that not all the effects attributed to SOCS2 are GH dependent [118]. We have also identified SOCS2 as an important regulator of pancreas physiology (manuscript in preparation). Other studies have demonstrated that SOCS2 is essential for the regulation of GH actions not directly related to somatic growth. For example, SOCS2 can block GH-dependent inhibition of neural stem cell differentiation. Consequently SOCS2-/- mice have fewer neurons in the developing cortex, whereas SOCS2 over-expression results in increased neural differentiation [109,122]. It has also been demonstrated that SOCS2 inhibits intestinal epithelial proliferation [125]. A role of SOCS-2 in the differentiation process of mesenchymal precursors has also been shown in mice E2, GH, SOCS2, and Liver 19 myoblasts [126]. Stable transfection of SOCS2 into C2C12 cells resulted in the acceleration of proliferation and survival and inhibition of spontaneous myotube formation. In addition, SOCS2 potentiated bone morphogenic protein (BMP)- induced trans-differentiation of C2C12 cells into osteoblast phenotypes [126]. Further process affected by SOCS2 include mammary gland development [127], bone mineral density [128], and allergic response [129]. Besides its actions on GH signaling SOCS2 has been implemented in the regulation of other cytokines and immune cell regulation [94,130,131,132]. SOCS2-/- mice are more susceptible to infection and exhibit increased levels of pro-inflammatory cytokines in response to certain pathogens. Also, expression of SOCS2 is induced by lipoxins and has been demonstrated to mediate some of their anti-inflammatory actions [133]. 1.3. Estrogens Estrogens are a group of steroid hormones produced by enzymatic modification of cholesterol. Estardiol (E2) is the major natural estrogen in mammals and has physiological actions not only limited to reproductive organs both in females and males [134]. The role of E2 in the development and regulation of female reproduction is well established. However, recent studies demonstrate that E2 has a multiple effects on metabolism (e.g., anti-diabetic effects) [135]. Relevant to this Thesis, the liver is a direct target of estrogens because it expresses Estrogen Receptor alpha (ERα) which is connected, among others, with lipid and glucose homeostasis [136,137,138] and somatic growth [139]. 1.3.1. Estrogen signaling. Estrogen signaling can be mediated by multiple receptors [140]. Most of the known estrogenic effects are mediated via direct interaction of estrogen with the DNA-binding transcription factors, ERα and ERβ. Classical estrogen signaling occurs through a direct binding of ER dimers to estrogen responsive elements in the regulatory regions of estrogen target genes Mercedes de Mirecki Garrido 20 followed by activation of the transcriptional machinery at the transcription start site. In addition, estrogen can modulate gene expression by a second mechanism in which ERs interact with other transcription factors, like STAT5, through a process referred to as transcription factor cross-talk. Estrogen may also elicit effects through non-genomic mechanisms, which involve the activation of downstream kinase pathways like PKA, PKC, and MAPK via membrane-localized ERs. An orphan G protein-coupled receptor (GPR)-30 in the cell membrane mediates nongenomic and rapid estrogen signaling. Finally, E2 has a similar affinity for ERα and ERβ, these receptors are activated by a wide range of ligands including Selective Estrogen Receptor Modulators (e.g., raloxifen) as well as many other compounds. ERβ is expressed in the ovary, prostate, lung, gastrointestinal tract, bladder, and hematopoietic and the central nervous systems, while ERα is mainly expressed in reproductive tissues, kidney, bone, white adipose tissue, and liver. The liver expresses ERα but almost undetectable levels of ERβ which indicates that specific actions of estrogens in liver can be mimicked by using selective ERα agonists such as propyl-pyrazole-triol [141]. Collectively, the above mentioned data indicate that the mechanisms involved in ER signaling are influenced by cell phenotype, the target gene, and the activity or crosstalk with other signaling networks. The liver represents a site where physiologically and therapeutically relevant interactions between estrogens and GH can be developed. Particularly relevant is the interaction of estrogens with GHR-JAK2-STAT5 signaling pathway in the regulation of somatic growth, lipid and glucose metabolism, and “liver sexuality”. 1.3.2. Estrogens regulate somatic growth, metabolism and gender dimorphism of hepatic gene transcription. - Somatic growth. It is well known that sex steroids and GH interact closely to regulate pubertal growth [34]. Interestingly, loss of ERα (ERKO), but not ERβ, mediates important effects of estrogen in the skeleton of male mice during growth E2, GH, SOCS2, and Liver 27 Figure 7. Phyisiological effects of E2 and GH (female pattern) on lipid and glucose metabolism (Based on [28]). 1.5. Relationship between the Metabolic Syndrome and GH, E2 and SOCS. The Metabolic Syndrome is a complex disease that is increasing at epidemic rates in Western countries. Insulin resistance, obesity, diabetes and NAFLD (Non Alcoholic Fatty Liver Disease) are the components of the metabolic syndrome. Fatty liver development is a multifactorial process and certain metabolic disorders are associated with it, such as obesity, hyperlipidemia and insulin resistance. Fatty liver o steatosis (i.e, accumulation of triglycerides in the liver) is pathological. The prolonged lipid storage can result in an activation of inflammatory reactions and loss of metabolic competency. Insulin resistance is considered as the factor that most contributes to the development of NAFLD (as insulin controls energy Mercedes de Mirecki Garrido 28 balance by regulating the metabolism of carbohydrate and lipid; in relation to the lipid content of the liver, insulin modulates the synthesis and secretion and βoxidation of fatty acids). However, the molecular basis for liver steatosis formation is poorly understood. There are extensive clinical and experimental evidences suggesting that metabolic alterations associated with NAFLD are regulated, among others, by E2, GH, and SOCS2. E2 may interfere with endocrine, metabolic, and gender-differentiated functions in liver in both females and males. As mentioned above, estrogen, through its interaction with the ERα, exerts direct effects on liver [62,170]. Indirect mechanisms also play a crucial role because of the E2 influence on the pituitary GH secretion and the GHR-JAK2-STAT5 signaling pathway in the target tissues inducing the expression of SOCS2, which is a negative regulator of the GHRJAK2-STAT5 signaling pathway [59]. The GHR-JAK2-STAT5 signaling pathway is of particular importance in the regulation of endocrine, metabolic, and sexdifferentiated actions of GH in liver. The ability of GHR-JAK2-STAT5 signaling pathway to regulate hepatic lipid metabolism has been highlighted in recent mouse genetic studies showing that hepatic inactivation of GHR [80], it is associated kinase, JAK2 [170] or its downstream signaling intermediary, STAT5b [171] leads to fatty liver [1]. Hepatic steatosis is known to be associated with decreased insulin sensitivity and SOCS2 was shown to be a potent regulator of proinsulin processing and insulin secretion in β cells. We have now identified SOCS2 as an important regulator of hepatic homeostasis under conditions of High-Fat dietary stress [62]. An insufficient number of insulin-producing β cells is a hallmark of both type 1 and type 2 diabetes. SOCS proteins are powerful inhibitors of pathways involved in survival and function of pancreatic β cells, such as those induced by insulin and GH [116,172]. Lactogen signaling is critical for β cell proliferation and β cell E2, GH, SOCS2, and Liver 29 function during pregnancy. The most important mediator of lactogen signaling is the JAK2/STAT5 pathway [173] and CIS and SOCS2 are induced during pregnancy [174]. Constitutive production of SOCS2 in pancreatic β cells leads to hyperglycaemia and glucose intolerance [175]. In a Japanese cohort, several SNPs in the SOCS2 gene promoter region were found to be associated with increased risk of type 2 Diabetes Mellitus [176]. Elimination of SOCS1 increases interferon response and potentiates β cell death [177] while SOCS3 is a potent protector against type 1 Diabetes Mellitus through suppressing IL-β and TNFα in pancreatic β cells [178,179]. SOCS3 knock–down β cells also show resistance to apoptotic cell death [180]. Therefore, one therapeutic strategy to increase functional β cell mass in order to overcome insulin deficiency could be SOCS regulation. 1.6. Relevance of Pharmacogenomic to explore molecular mechanisms of GH, E2 and SOCS2. In the field of Endocrinology different hormones affects the function of each other and their actions also depend largely on nutritional status and other parameters. The fact that most hormones have tissue specific effects, make such studies even more challenging. Nevertheless, finding out how different hormones affect the trasncriptome, proteome and metabolic content of cells is great importance for the understanding of hormone action and disturbances in endocrine system. The first papers describing microarray analysis in vivo gene expression in the endocrine field were published in 2001-2002 [40,181,182], since then the technology and the fields of investigation have extensively growth. The studies of gene expression used in this Thesis are particularly attractive to the Endocrine Pharmacology field, as regulation of gene expression is a key mechanism whereby hormones exert their actions. Some papers have been published in the field of the gene expression profiling in molecular studies of hormone actions in our group, [40,182,183,184]. In Rico-Bautista (2005) [118], were identify genes and metabolic parameters that Mercedes de Mirecki Garrido 30 might contribute to the SOCS2−⁄− phenotype by cDNA microarray, demonstrating that although SOCS2 deficiency induces significant changes in hepatic gene expression, only a fraction of these overlap with known GH-induced effects in the liver, suggesting that SOCS2 might be an important regulator of other growth factors and cytokines acting on the liver. Studying changes in gene expression after short-term hormonal treatment could help to elucidate direct mechanism of hormone action, while the effects of long-term treatment may help explain the physiological effects of hormones. E2 relationship in GH resistance models were study in some papers, Long-term administration of physiological doses of E2 to GHD male rats (hypothyroid) regulates several members of SOCS family by a complex interplay with GH and thyroid hormones [61] also a global expression analysis of GH actions in liver using microarrays clearly indicates that most of the known physiological effects of GH can be explained through its effects on the transcription of specific genes [40,47,48,184,185]. Pharmacogenomic has the potential to clarify tissue specific actions of hormones and explain observed physiological effects previously lacking molecular explanation. Pharmacogenomic is also a tool to personalize medicine combining bioinformatics with proteomic, metabolomics, and others new technologies, to explore pathophysiology and to characterize more precisely an individual´s risk for disease, as well as response to treatment. In this Thesis, gene expression profiling and system biology network studies have been used to construct hypotheses concerning the molecular mechanisms and metabolic networks underlying actions of E2, GH and SOCS2. By collecting and comparing transcription profiles concerning hormone actions in situations like varying hormonal status, diets, or gene deletions, it might be possible to elucidate the role of hormones in different states of disease. E2, GH, SOCS2, and Liver 31 2. AIMS The main aim of the work presented in this Thesis was to characterize the influence of E2 and GH interplay and SOCS2 on liver metabolism. Particularly, we focused on the following aims: - Characterize E2 and GH interplay on liver metabolism (Paper I). - Characterize the role of SOCS2 in E2-regulated somatic growth and liver metabolism (Paper II). - Characterize the role of SOCS2 in HFD-induced fatty liver (Paper III). - Characterize the role of SOCS2 in MLDSTZ-induced Diabetes (Paper IV). Mercedes de Mirecki Garrido 32 3. MATERIALS AND METHODS 3.1. Animals. All the studies were carried out in strict accordance with the recommendations in the Guide for the Care and Use of Laboratory Animals of the University of Las Palmas de Gran Canaria and conducted in accordance with European and Spanish laws and regulations. All Animals were kept under standard conditions, constant 12-h/12-h light/dark cycles, and in a controlled temperature (21-23 ºC) environment, and free access to autoclaved standard chow (A04 SAFE Panlab, Barcelona, Spain) and tap water throughout the experiments. SOCS2 deleted (SOCS2−/−) and wild-type (WT) littermates (C57BL/6J) male mice (Paper II and III) have been previously reported [124]  To investigate the molecular mechanisms of E2 and GH interplay in male liver (Paper I). Adult (2-3 months old) male Sprague-Dawley rats (n=6 per group) were used throughout these experiments. Generation of hypothyroid animals (TX) was performed as previously described [186,187] (see Paper I). Two weeks after starting MMI administration, male rats were orchidectomized (OX) or sham– operated to make TXOX rats or testis-intact controls (TX), respectively (Fig 1). Six rats were not treated with MMI and were subjected to sham-surgery to provide euthyroid testis-intact controls (INTACT). Four days after OX, we began HRT with E2 benzoate (Sigma) (50 µg/kg/day; sc; 5 days per week) (TXOXE2) or vehicle (0.2 ml corn oil; sc; 5 days per week) (TXOX) to TXOX rats for 20 days [188,189] before hormonal replacement for 7 days with E2 plus GH (TXOXE2GH) or vehicle plus GH (TXOXGH). GH (0.3 mg/kg/day) was administered as two daily sc injections at 12-h intervals (08:00h and 20:00h) to mimic the male-specific GH secretion [39,190]. TX and TXOX control animals received equivalent amounts of the vehicle alone. Hypothyroidism status was corroborated by monitoring the body weight gain at 7-day intervals and the serum levels of T4 and T3. Twenty-four hours (in the case of E2) or twelve hours (in the E2, GH, SOCS2, and Liver 33 case of GH) after the last injection, the animals were killed by exsanguinations. On PND94, blood samples were collected and serum stored at -80ºC until analysis. Portions of the liver were snap frozen in liquid nitrogen and stored at -80ºC until processed for mRNA analysis. Figure 1. Schematic representation for hormonal replacement in hypothyroidorchidectomized rats.  To investigate the role of SOCS2 on the effects of E2 on somatic growth and liver transcriptome (Paper II). We carried out two different studies in male mice. First, we investigated the influence of SOCS2 on E2-regulated somatic growth in intact (non-orchidectomized) male mice. For this purpose SOCS2−/− and WT littermates (C57BL/6) mice [124] (2 months old) (n=6 per group) were treated with E2 benzoate (Sigma, St. Louis, MO) (50µg/kg/48h; sc) or vehicle (corn oil:ethanol; 90:10; v/v; 50 µl/48h; sc) during 15 weeks [191] (see Fig.2). Second, we investigated the influence of SOCS2 on E2-regulate liver transcriptome in orchidectomized (OX) mice. For this purpose, SOCS2−/− and WT (n=6 per group) (4 months old) were OX or sham operated [192]. After one week of recovery period, they were injected with E2 benzoate (100 µg/kg; sc) or vehicle (corn oil:ethanol; 90:10; v/v) for 1 day, 2 days or 21 days (5 days/week) [193] (see Fig. 3). Mercedes de Mirecki Garrido 34 Figure 2. Schematic representation for long-term E2 treatment in SOCS2-/- and WT male mice (non-orchidectomized) that was used to investigate the role of SOCS2 on E2-regulated somatic growth. Figure 3. Schematic representation for E2 treatment in SOCS2-/- and WT male mice (orchidectomized) that was used to investigate the role of SOCS2 on E2-regulated liver transcriptome. The diet was removed from the cages, twelve hours before the mice were sacrificed, to minimize the effect of food. Serum samples were collected and stored at -80ºC until analysis. Portions of the liver were washed with cold PBS, snap frozen in liquid nitrogen, and stored at -80ºC until being processed for mRNA or biochemical analysis.  To investigate the role of SOCS2 in High-Fat Diet (HFD)-induced fatty liver and insulin resistance (Paper III). SOCS2−/− and WT littermates (8–10 weeks old) were fed with a standard control diet (CD) (SAFE-diet A04, Panlab SLU, Barcelona, Spain) or HFD (OpenSource Diets -D12492), New Brunswick, NJ, USA) (see Paper III). The mice were sacrificed after an ip injection of 0.75 U/kg human recombinant insulin (Actrapid, Novo Nordisk, Denmark) or saline as E2, GH, SOCS2, and Liver 35 a control. Tissues were stored at −80ºC until analysis. Pieces of pancreas and liver were excised for histological examination. Blood samples were collected through cardiac bleeding, and plasma was used to measure various analytes.  To investigate the role of SOCS2 on Diabetes Mellitus induced by multiple low-dose streptozotocin (MLDSTZ) (Paper IV). We used the MLDSTZ model to induce type 1 Diabetes [194] in SOCS2-/- and wild type littermates (Fig. 4). Six months-old male mice received 5 consecutive daily ip injections of STZ (50 mg/kg b.w.). Then, STZ administration was stopped and blood glucose, body weight, food and water intake, glucose tolerance, and insulin sensitivity were recorded. Mice were considered diabetics if the blood glucose levels were greater than 300 mg/dl. Sixty days after STZ treatment was stopped, mice were sacrificed and tissues were collected for further studies. Figure 4. Schematic representation for the multiple low dose streptozotocin (STZ) treatment in SOCS2-/- and WT male mice that was used as a model of autoimmune diabetes and β-cell destruction in vivo. 3.2. Somatic growth and food intake analysis. Body weights, tail lengths, and food intake were measured once a week for all animals. The measurement of tail length was used for monitoring growth [195,196,197]. The percentages of weight rate (WR) or tail rate (TR) were calculated by the following formulas: WR=[(W (g) (new) – W (g) (old)/W (g) (old)]*100 or TR = [(T (cm) (new) – T (cm) (old)/T (cm) (old)]*100, respectively. Food consumption was estimated by subtracting the amount of food left on the grid from initial food weight. Food spilled on the floor Mercedes de Mirecki Garrido 36 of the cage was not weighed, but spillage was minimal because the diet was supplied as pellets. The weekly caloric intake was calculated on the basis of food consumption x caloric value of the diet (2900 kcal/g). Feed efficiency (FE), which denotes the body weight increase per gram of food consumed or the ability to transform calories consumed into body weight [198], was calculated by the following formula: mean body weight gain (g)/total caloric intake. 3.3. Intraperitoneal glucose (ipGTT) and insulin (ipITT) tolerance tests. For ipGTT, mice were fasted 16h overnight followed by an intraperitoneally (ip) injection of D (+)-glucose (20% in 0.9% NaCl) at a dose of 2g/kg b.w. For ipITT, human insulin was ip injected into 4h fasted mice at a dose of 0.75 U/kg b.w. Blood glucose levels were measured using a glucometer (Roche Diagnostics, Switzerland). Insulin was measured in blood samples collected at fasting time point 0 during ipGTT. Homeostatic model assessment (HOMA)-IR (insulin resistance) index was calculated as follows: fasting insulin (ng/ml)×fasting glucose (mM). HOMA- (β-cell function) index was calculated as follows: 20×fasting insulin (μU/ml)/fasting glucose (mM) – 3.5 [199]. 3.4. Serum biochemistry. Cholesterol (CHO), triacylglycerols (TG), gamma glutamil transpeptidase (GGT), and creatinine were measured were measured by a CLIMA MC-15 clinical chemistry photometer (RAL, Spain). (Papers II and IV). CHO and TG were also assayed using colorimetric enzymatic kits (Roche/Hitachi Diagnostic GmbH, Mannheim, Germany) (see Paper III). Plasma alanine aminotrasferase (ALT) activity was measured using commercial assay kit (Cayman Chemical Co., MI, USA) (Paper III). Serum levels of insulin (Crystal Chem Inc.), leptin and IGF-I (Quantikine, R&D systems), and GH (Millipore, MA, USA) (Papers II, III and IV) were determined by using mouse/rat immunoassays (ELISA) according to manufacturer recommendations. The ELISA included quality controls provided by the manufacturers, and the standard curves of the E2, GH, SOCS2, and Liver 43 metabolism of amino acids and urea (i.e., OTC, ASS1, aminotransferases, and methyltransferases) are significantly down-regulated. This is in line with the positive effects of GH on nitrogen balance, which have been previously studied in hypophysectomized rats [8,40,213]. GH serves as an anabolic hormone that promotes lipolysis and prevents lipogenesis in adipose tissue, which increases the availability of fatty acids (FFA) for energy expenditure [8]. E2 is also able to interfere with this process by preventing the induction of some genes related to fat utilization, such as ApoC2, which activates the enzyme LPL that hydrolyzes TG. Therefore, E2 actions in liver can impact the peripheral metabolic actions of GH. Lipogenesis is often increased in situations of reduced energy expenditure such as hypothyroidism, GH deficiency, E2 deficiency, or aging [214]. Accordingly, our analysis of the hepatic lipid content revealed that TXOX rats contained significantly increased levels of total saturated fatty acids (SFA) compared to INTACT control rats. E2 replacement did not modify the mRNA expression levels of key regulators of hepatic lipogenesis [i.e., sterol regulatory element binding protein (SREBP)1c, acetyl-Co A carboxylase alpha (ACC), fatty-acid synthase (FAS)] [215], whereas it activated a PPARα transcriptional program that promotes fatty acid catabolism in liver [216,217]. This was evidenced by the E2 increased expression of the PPARα gene itself and the PPARα target genes involved in the β/-oxidation of fatty acids (i.e., CTE-I, CPT-2, Fasd6, Fasd1, Fasd2, Scd1, ACOX1, ECH1, BAAT, FGF21, CYP4A1, CYP4A3) (Table S2). Accordingly, E2 replacement caused a significant reduction in SFAs. Overall these findings are indicative of a positive crosstalk between E2 and PPARα that is supported by multiple independent studies [136,159,218,219]. Interestingly, despite the increased expression of genes involved in β-oxidation, we detected a significant increase in hepatic TG content in E2 treated TXOX rats, which is likely explained Mercedes de Mirecki Garrido 44 by effects on lipid transport. The first step of long chain fatty acids uptake is its translocation across the plasma membrane. Notably, E2 increased the transcription of several known PPARα activated genes encoding proteins that have been implicated in fatty acids uptake and activation such as CD36, ACSL4 and SLC27A5 (FATP5) [220,221]. We have previously demonstrated that the fatty acid transporter CD36 is predominantly expressed in female rat livers and proposed that this sexual dimorphism depends on the GH secretion pattern, which can be influenced by E2 treatment. E2 also increased transcripts of the SLC27A5 gene which encodes FATP5, an fatty acid transporter that is an acyl-CoA synthetase (bile acid ligase) that catalyzes the conjugation of bile acids with amino acids before excretion into bile canaliculi [222]. Following fatty acids uptake, the first step for the intracellular use of long chain fatty acids is its esterification with CoA. This reaction is catalyzed by acyl-CoA synthetases such as ACSL4 which was also induced by E2 in TXOX rat liver. The produced acyl-CoAs are substrates for ß-oxidation but also can prime the synthesis of TG, phospholipids, CE, and ceramides and therefore are also a primary source of signaling molecules [223]. The notion that E2 may regulate the formation of lipid signaling intermediaries is supported by the stimulation of fatty acids elongase-5 (Elovl5). Elovl functions with fatty acid desaturases to generate many of the long-chain PUFAs assimilated into cellular lipids (i.e., 20:4n-6 and 22:6n-3). However, it is worth mentioning that E2 administration did not alter VLCPUFA metabolism because the levels of 20:4n-6, 20:5n-3 and 22:6n-3 remained similar to values in the TXOX group. It has been reported that E2 might play a critical role in lipogenesis and Scd1 transcription [136], a gene that encodes a rate-limiting enzyme to generate MUFAs such as 18:1 n−9 and 16:1 n−7. Previous studies have reported that the absence of E2 or ERα in rats provoked a profound increase in lipogenesis and Scd1 transcription [137], which suggests that E2 inhibits Scd1 transcription. Interestingly, the antilipogenic effect of E2 therapy, while maintaining efficient TG export and reduced phospholipid transfer protein, has been reported to depend E2, GH, SOCS2, and Liver 45 on hepatic ERα [219,224]. Our study, however, shows that E2 increased the Scd1 gene expression and that this effect was paralleled by reduced hepatic content of 18:0 and increased of 18:1 n−9 (the main product of SCD reaction) contents, in total and, especially, in neutral lipids compared with TXOX animals, which indicates that E2 modulates SCD1 activity in TXOX liver. Surprisingly, E2 also downregulated Scd2 gene expression in TXOX rat livers. The significance of this opposed transcriptional regulation of Scd genes is unknown, but given that transcript levels of Scd1 are about 1800 times higher than that of Scd2 in the rat liver [225], changes in 18:1n-9 and 18:0 must be entirely attributed to variations in Scd1 gene expression. Overall, the changes in the lipid composition and gene expression profile seen in E2-treated TXOX rats support the finding that E2PPARα functional interactions play a physiological role in the regulation of hepatic lipid metabolism. E2 has the ability to reduce circulating CHO in women and in animal models fed on a high-fat diet [226]. However, E2 was unable to efficiently reverse hypercholesterolemia or hypotriglyceridemia in TXOX rats. This result may be due to the fact that E2 reduced expression levels of several transporters of CHO (and CE), including ApoB and ABCA1 in TXOX rats, which most likely contributed to maintaining an increased hepatic level CE. E2 may also induce intracellular CHO mobilization by modulating enzymes involved in CE and CHO synthesis and/or turnover [214,227]. Distinct enzymes can catalyze the CHO to CE conversion in liver: lecithin:cholesterol acyltransferase (LCAT), which uses phosphatidylcholine (PC) as a source of acyl changes and ACAT, which uses acylCoA. Because the levels of lysophosphatidylcholine (LPC) were undetectable in all groups, our initial conclusion was that E2 stimulated the ACAT2 reaction to increase CE. However, we did not detect changes in the expression level of the ACAT gene, which did not discard posttranslational modification of enzymes in the CE cycle in the liver from E2-treated TXOX rats. Mercedes de Mirecki Garrido 46 An increased level of hepatic CE, together with the increased TG and decreased FFA hepatic contents in GH-treated TXOX rats, resemble the effects of E2 on hepatic lipid composition and suggest that some effects of E2 might be GH mediated. A striking consequence of the combined replacement with E2 and GH is the complete restoration of MUFA levels from total and neutral lipids, an effect attributable to the increase in 18:1n9, likely through alteration of 9 desaturase expression. Moreover, GH and E2 increased hepatic CE and the combined effect of the two hormones were additive with regard to CE because its levels doubled those found in INTACT animals and were approximately 30% higher than in the E2 or GH groups which indicates a more efficient hepatic CHO metabolism. Accordingly, in the presence of E2, GH reduced hepatic CHO content compared not only to the TXOX group but also in relation to the E2or GH-treated TXOX groups. The hepatic content of TG was, however, significantly increased by GH in E2-pretreated TXOX rats, which suggests that combined treatment by E2 and GH dramatically enhances lipogenesis. It is known that in contrast with its lipolytic effects in adipose tissue, GH exerts lipogenic actions in liver through stimulation of SREBP1, which is usually accompanied by increased hepatic TG (VLDL) secretion [8]. Indeed, our lipid profiling analysis suggested that intermittent GH administration to TXOX rats increased lipogenesis in the liver. However, in contrast to the effects of a continuous infusion of GH in hypophisectomized rats [40], intermittent GH administration to TXOX rats did not increase SERBP1, whereas several genes involved in fatty acids transport (e.g., FABP) and the biosynthesis of unsaturated fatty acids from 18:2n-6 and 18:3n-3 (e.g., fatty acid desaturases 4, 5 and 6) were induced. Interestingly, intermittent GH administration to TXOX rats down-regulated the expression of the lipin gene, an SREBP1c target gene, which is critical in the regulation of cellular levels of DG and TG and a key regulator of fatty acid oxidation in adipose tissue, skeletal muscle, and liver tissue [228]. These findings support the hypothesis that the female pattern of GH E2, GH, SOCS2, and Liver 47 administration is a more efficient stimulus to induce lipogenic effects in the liver than the male pattern [184,229]. Another mechanism whereby GH might promote lipogenesis in the liver is through the down-regulation of lipid oxidation. We have previously shown that continuous GH administration to hypophysectomized [40] and to old-intact [46] male rats inhibited PPARα. Accordingly, our lipidomic and genomic analysis showed that intermittent GH administration to TXOX rats also leads to down-regulation of the PPARα signaling pathway. In particular, GH represses the expression of PPARα itself, ACOX-1, CPT-1, FGF21, and several members of the CYP4A family, which are involved in fatty acid oxidation. In summary, our study adds novel data that highlight the impact of subcutaneous E2 administration on liver physiology and its interplay with GH. These results highlight the role of E2 as a critical regulator of liver metabolism in mammals and add further weight to the hypothesis that E2 acts as an important regulator of GH actions in the liver. The E2-GH interplay in the liver is relevant because of the physiological roles that these hormones have in mammals and the widespread use of estrogen and estrogen-related compounds in human. Notably, this is the first study to demonstrate that hepatic lipid profiles are endowed with singular fingerprints that may be used to segregate different groups with altered hormone status. This includes different hormonal replacements (E2 or GH) that induced overlapping changes in gene expression. Therefore, liver lipid profiling can serve to identify cryptic hormone deficiencies or exposure to hormones or hormone-like substances. Mercedes de Mirecki Garrido 48 4.2. The Suppressor of Cytokine Signaling-2 influences the effects of E2 on somatotropic-liver axis and liver transcriptome (Paper II; manuscript in preparation). In Paper I, we added novel data that highlight the impact of subcutaneous E2 administration on liver physiology and its interplay with GH. The E2-GH interplay in the liver is relevant because of the physiological roles that these hormones have in mammals and the widespread use of estrogen and estrogen-related compounds in human. In this study (Paper II), we have carried out a transcriptomic analysis to obtain comprehensive information on the molecular mechanisms of E2 effects in vivo in the presence (WT) or in the absence (SOCS2−/−) of SOCS2. The influence of SOCS2 in E2-regulated somatic growth and liver transcriptome seen in this study could contribute to better understand the molecular mechanisms involved in the endocrine and metabolic consequences of exposition to estrogens or novel estrogen-related compounds and their influence on the GH-liver axis. As first approach to assess the influence of SOCS2 on the effects of E2 in liver we studied changes in somatotropic-liver axis in the absence or in the presence of SOCS2. After 30 days, E2 reduced body weight gain in the two genotypes. However, reduction of body weight by E2 was more prolonged and increased in the WT group (data not shown). As expected [118], when E2 treatment was started (time 0), SOCS2-/- mice had higher body weight than WT. Therefore, body weight gain measurements were normalized by body weight at time 0 this shows that E2, in comparison with vehicle-treated mice, caused higher reduction of body weight in WT (Fig.1A) than in SOCS2-/- (Fig.1B) mice. At end point, significant differences, in comparison with vehicle-treated mice, in body (Fig.1C) and femur (Fig.1D) lengths remained in E2-treated WT but not in E2-treated SOCS2-/- mice. Notably, E2 increased hepatic IGF-I mRNA levels (Fig. 1E) in both genotypes without changes in circulating IGF-I (Fig. 1F). E2, GH, SOCS2, and Liver 49 Figure 1. Influence of SOCS2 on E2-regulated somatotropic-liver axis. Wild-type (SOCS2+/+) and SOCS2−/− male mice were treated with E2 benzoate (E2B) (50µg/kg/48h) or vehicle (VEH) during 90 days as described in Material and Methods. Normalized body weights on postnatal day 50 (time 0 of E2 or vehicle treatment) in WT (A) and SOCS2-/- (B) mice were monitored at 7 days intervals in the absence (white square) or in the presence of E2 (black square). The lengths of body (C) and femur (D), the hepatic IGF-I mRNA levels (E), and circulating IGF-I (F) were measurements at end point. Results are expressed as mean ± SEM from six individual animals in each group. The mean mRNA expression level of IGF-I in the VEH SOCS2+/+ group is defined as 1, with all other expression values reported relative to this level. *P<0.05; **P<0.01; ***P<0.001 for comparison with vehicle treated groups; #P<0.05; ##P<0.01; ###P<0.001 for comparison between genotypes. E2 has been shown to induce SOCS2 in liver, which in turn negatively regulates STAT5-mediated transcriptional activity [61,62,193]. In contrast, the molecular consequence of SOCS2 gene deletion is a prolonged activation of STAT5 [230]. Therefore, we next carried out mRNA quantitative analysis of SOCS2, CIS, SOCS1 and SOCS3, which are negative regulators of GHR-STAT5 signaling [48,210]. E2 increased the hepatic levels of SOCS2 mRNA (Fig. 2A) in WT whereas it was undetectable in SOCS2-/- mice. In WT mice, E2 induced hepatic mRNA expression levels of CIS (Fig.2B) and SOCS1 (Fig.2C) whereas SOCS3 Mercedes de Mirecki Garrido 50 (Fig.2D) was unchanged. Notably, in vehicle-treated SOCS2-/- mice, the mRNA expression levels of CIS (Fig.2B) and SOCS3 (Fig.3D were significantly reduced whereas SOCS1 (Fig.2C) stayed unchanged, in comparison with WT mice. In contrast, E2 was not capable of increasing the mRNA expression levels of CIS, SOCS1 or SOCS3 in SOCS2-/- mice. These findings suggest that SOCS2 can modulate the effects of E2 on negative modulators of somatotropic-liver axis. Figure 2. Influence of SOCS2 on E2-regulated Suppressors of Cytokine Signaling genes in liver. Wild-type (SOCS2+/+) and SOCS2−/− male mice were treated with E2 benzoate (E2B) (50 µg/kg/48h) or vehicle (VEH) during 90 days as described in Material and Methods. The hepatic mRNA levels of SOCS2 (A), CIS (B), SOCS1 (C), and SOCS3 (D) were measured by qPCR after mice were sacrificed on postnatal 90 day as described in Material and Methods. Results are expressed as mean ± SEM (n=6). The mean mRNA expression level of each gene in the VEH SOCS2+/+ group is defined as 1, with all other expression values reported relative to this level. *P<0.05; **P<0.01; ***P<0.001 for E2, GH, SOCS2, and Liver 51 comparison with vehicle treated groups; #, P<0.05 ##, P<0.01, ###P<0.001 for comparison between genotypes. Most previous studies have been focused on the influence of estrogens on pituitary GH secretion [34] and the gender-specific GH secretion release from pituitary has been shown to have a great impact on hepatic transcriptional regulation [33]. Thus, we next measured the influence of E2 on GH-regulated gene markers of hepatic gender dimorphism [33] in WT and SOCS2-/- mice. In the absence of E2 treatment, mRNA levels of MUP, a male-specific gene, decreased (Fig.3A), whereas Cyp2B9 and A1bg (Fig.3B), two female-specific genes, were increased. These findings suggest that absence of SOCS2 is enough by itself to feminize male liver and influence the effects of E2 on gender dimorphism. In both genotypes, E2 decreased MUP (Fig.3A) whereas Cyp2b9 (Fig.3B) and A1bg (Fig.3C) were highly increased which show that E2 feminizes the adult liver. Overall, these findings support our hypothesis that SOCS2 can influence the effects of E2 on somatotropic-liver axis. Figure 3. SOCS2 influence on E2-regulated gender dimorphism in liver. Wild-type (SOCS2+/+) and SOCS2−/− male mice were treated with E2 benzoate (EB) (50 µg/kg/48h) or vehicle (VEH) during 90 days as described in Material and Methods. The hepatic mRNA levels of MUP (A), Cyp2b9 (B), and A1bg (C) were measured by qPCR after mice were sacrificed on postnatal 90 day as described in Material and Methods. Results are expressed as mean ± SEM (n=6). The mean mRNA expression level of each gene in the VEH SOCS2+/+ group is defined as 1, with all other expression values reported relative to this level*P<0.05; ** P<0.01; ***P<0.001 for comparison with vehicle treated groups; #, P<0.05; ##, P<0.01; ###, P<0.001 for comparison between genotypes. Mercedes de Mirecki Garrido 52 Several studies suggest that E2-mediated signaling can have an important role in the control of lipid and glucose metabolism [136,137]. Studies in both human and rodents suggest that altered levels of E2 or its receptors can lead to a metabolic syndrome-like phenotype (i.e, insulin resistance, adiposity, dyslipidemia). As mentioned in this Thesis (Introduction), the beneficial influence of E2 in relation to normalizing lipid and glucose homeostasis has been evidenced in ob/ob and high-fat diet fed mice, two models of obesity and type 2 diabetes, respectively [158,159]. Furthermore, treatment of ob/ob mice with the ERα selective agonist propyl-pyrazole-triol can improve glucose tolerance and insulin sensitivity which supports the critical role of ERα in the control of glucose homeostasis. In the absence of E2 treatment, SOCS2-/- mice had lower levels of TG (Fig.4B), leptin (Fig.4C), and perirenal fat (Fig.4D), in comparison with WT mice. In agreement with previously reported data [124], these findings show that SOCS2-/- male mice have reduced adiposity. Interestingly, E2 further reduced circulating TG (Fig.4b) and leptin (Fig.4C) in WT mice but not in SOCS2-/-. In contrast, circulating CHO (Fig. 4A) was reduced in SOCS2-/- but not in WT mice. E2, GH, SOCS2, and Liver 59 ID Probe Unigene/Refseq Symbol Gene description R q(%) mean ± SD A_55_P1995537 NM_010824 Mpo Myeloperoxidase, mitochondrial protein 3,41 0,85 0,00 A_52_P15388 NM_008522 Ltf Lactotransferrin 2,13 0,61 0,67 A_55_P1983921 NM_021352 Crybb3 Crystallin, beta B3, transcript variant 1 1,95 0,93 1,49 A_51_P167292 NM_009892 Chi3l3 Chitinase 3-like 3 1,93 0,55 0,67 A_55_P2163098 NM_134066 Akr1c18 Aldo-keto reductase family 1, member C18 1,90 0,36 0,00 A_51_P461067 ENSMUST000001 03420 G1m Immunoglobulin heavy constant gamma 1 1,85 0,64 0,93 A_55_P1953387 NM_001272097 Fabp5 Fatty acid binding protein 5, epidermal 1,74 1,31 4,03 A_51_P199168 NM_007702 Cidea Cell death-inducing DNA fragmentation factor, alpha subunitlike effector A 1,73 0,72 0,99 A_52_P213889 NM_172476 Tmc7 Transmembrane channel-like gene family 7 1,66 1,08 3,07 A_51_P488196 NM_028472 Bmper BMP-binding endothelial regulator 1,65 0,28 0,00 A_55_P2024155 NM_001033324 Zbtb16 Zinc finger and BTB domain containing 16 1,65 1,16 3,38 A_51_P520650 NM_177639 Dlgap1 Discs, large (Drosophila) homologassociated protein 1 1,62 1,26 4,03 A_55_P2002975 NM_205795 Mrgprb4 MAS-related GPR, member B4 1,55 0,96 2,80 A_55_P2054913 NM_001011863 Olfr406 Olfactory receptor 406 (Olfr406) 1,54 0,56 0,99 A_55_P2149921 TC1780716 Q80XJ7 Aldo-keto reductase family 1, member A4 1,52 0,91 2,80 A_55_P2185900 NM_032002 Nrg4 Neuregulin 4 1,49 0,87 2,80 A_51_P295896 NM_028934 4930452B06Rik RIKEN cDNA 4930452B06 gene 1,47 0,56 0,99 A_55_P2290388 NM_001043354 Rorb RAR-related orphan receptor beta (Rorb), 1,46 0,53 0,99 A_55_P2001553 NM_020043 Igdcc4 Immunoglobulin superfamily, DCC subclass, member 4 1,45 0,16 0,00 Table 1. Representative list of hepatic genes that were upregulated in SOCS2-/- mice instead of WT mice after 2 days of E2 administration. Two days of E2 administration to OXSOCS2-/- and OXWT mice was performed as described in Material and Methods. Then, differently expressed genes in the liver trancriptome were identified by DNA microarrays. The analysis is based on the SAM statistical technique and differentiallyexpressed genes were discovered using a FDR less than 5% and a mean ratio of log2 > |0.58|. The table shows probe ID, Unigene/Refseq, gene symbol, gene description, R (log2 E2-treated OXSOCS2-/- / vehicle-treated SOCS2-/-), and q (%). Mercedes de Mirecki Garrido 60 ID Probe Unigene/Refseq Symbol Gene description R q(%) mean ± SD A_55_P2408588 NM_007489 Arntl Aryl hydrocarbon receptor nuclear translocator-like -2,29 1,22 1,93 A_52_P480044 XR_105914 BC023105 PREDICTED: cDNA sequence BC023105 -2,21 0,72 0,94 A_55_P2213968 NR_045840 4933416M07Rik RIKEN cDNA 4933416M07 gene, non-coding RNA -2,14 0,88 1,17 A_66_P120728 NM_001011791 Olfr193 Olfactory receptor 193 -2,10 0,90 1,49 A_51_P107315 NM_145423 Slc5a8 Solute carrier family 5 (iodide transporter), member 8 -2,02 0,52 0,94 A_65_P11137 NM_019541 Cts8 Cathepsin 8 (Cts8) -1,96 0,22 0,00 A_55_P2056714 NM_001042612 Nlrp9c NLR family, pyrin domain containing 9C -1,95 0,90 1,57 A_55_P2151591 ENSMUST000000 32909 Pde3b Phosphodiesterase 3B, cGMPinhibited -1,94 0,73 1,09 A_51_P210031 NM_001252679 Smr2 Submaxillary gland androgen regulated protein 2 (Smr2) -1,90 0,94 1,93 A_55_P2121042 NM_001082531 Pla2g2a Phospholipase A2, group IIA (platelets, synovial fluid) -1,85 0,53 0,94 A_52_P122649 NM_175647 Dmrta1 Doublesex and mab-3 related transcription factor like family A1 -1,81 0,67 1,09 A_52_P669922 NM_032541 Hamp Hepcidin antimicrobial peptide -1,80 0,28 0,00 A_55_P2095859 NR_037604 Rdh18-ps Retinol dehydrogenase 18, pseudogene, non-coding RNA -1,78 0,49 0,94 A_52_P186751 NM_198677 BC061237 cDNA sequence -1,77 0,58 1,09 A_51_P267969 NM_011259 Reg3a Regenerating islet-derived 3 alpha -1,77 0,60 1,09 A_55_P2073024 NM_001034859 Gm4841 Predicted gene 4841 -1,76 0,46 0,94 A_51_P322473 NM_172417 2310042D19Rik RIKEN cDNA 2310042D19 gene -1,74 0,07 0,00 A_55_P2051082 XR_168616 Gm11634 PREDICTED: predicted gene 11634 -1,72 0,25 0,00 Table 2. Representative list of hepatic genes that were downregulated in SOCS2-/- mice instead of WT mice after 2 days of E2 administration. Two days of E2 administration to OXSOCS2-/- and OXWT mice was performed as described in Material and Methods. Then, differently expressed genes in the liver trancriptome were identified by DNA microarrays. The analysis is based on the SAM statistical technique and differentially-expressed genes were discovered using a FDR less than 5% and a mean ratio of log2 > |0.58|. The table shows probe ID, Unigene/Refseq, gene symbol, gene description, R (log2 E2-treated OXSOCS2-/- / vehicle-treated SOCS2-/-), and q(%). E2, GH, SOCS2, and Liver 61 4.3. Relationship between SOCS2, hepatic steatosis and insulin resistance in high fat diet fed mice (Paper II; FASEB J 2012, 26:32823291). In this study, we analysed the metabolic response of the SOCS2−/− mice to a hypercaloric, fatrich diet. The SOCS2−/− mice exhibited enhanced hepatic TG secretion and were protected from HFD-induced liver steatosis. However, they displayed severe systemic insulin resistance associated with hyperinsulinemia and worsened insulin sensitivity in the liver compared with the WT mice on a similar diet. The HFD-fed SOCS2−/− mice also exhibited enhanced expression of inflammatory cytokines in liver demonstrating a novel role of SOCS2 as a negative regulator of macrophage activation under conditions of high-fat dietary stress. Importantly, the SOCS2−/− mice phenotype is clearly different from the liver-specific SOCS3 and SOCS1 knockout mice phenotypes [234,235], which show enhanced liver steatosis, highlighting the functional differences between members of the SOCS family in the regulation of hepatic metabolism. In previous studies, we showed that SOCS2 is a negative regulator of hepatic GH receptor signaling [116]. The reduced steatosis observed in the HFD-fed SOCS2−/− mice compared with the HFD-fed WT mice is in agreement with the well-known actions of GH in promoting hepatic lipid mobilization [236] and strongly suggests that enhanced GH signaling in the liver is in part responsible for these effects. Indeed, we found mRNA levels of the known GH-regulated genes ApoB, FABP1 and DGAT2, which are involved in TG assembly and secretion to be increased in the SOCS2−/− mice. Detailed analysis of mice with a hepatic specific deletion of JAK2 has unveiled another mechanism whereby GH controls liver fat content [170]. These mice, denoted JAK2L, have elevated levels of circulation GH, which in turns leads to increased adipose tissue lipolysis. This increases FFA supply to the liver leading to steatosis [170]. The SOCS2-/- mice have reduced levels of circulating GH and in opposition to JAK2L exhibit increased fat mass, suggesting that reduced adipose tissue lipolysis in SOCS2-/- mice may also contribute to Mercedes de Mirecki Garrido 62 reduced hepatic TG accumulation. Noticeably, we did not detect significant changes in circulating FFA or hepatic expression of CD36 in SOCS2-/- mice compared to what has been reported in the JAK2L mice. Further experiments are needed to analyse the influence of GH and adipose tissue lipolysis in the SOCS2-/- mice liver. Surprisingly, the SOCS2−/− mice showed an exacerbated response to HFD feeding leading to worsened insulin sensitivity, whereas the SOCS2−/− and WT mice showed few differences in insulin signaling when maintained on a normal control diet. This suggests that hyperactivity of the hepatic GH receptor signaling alone, as it is observed in SOCS2−/− mice, is unlikely to account for the diet-dependent deterioration in glucose control observed in this study. A more likely explanation is that the anti-insulinic actions of GH are exacerbated by diet related mechanisms under SOCS2 control. We provide evidence for the existence of three of such mechanisms: hyperinsulinemia, peripheral adiposity and excessive production of inflammatory cytokines. Toll-like receptor driven macrophage activation in liver and adipose tissues by high-fat feeding leads to the production of inflammatory cytokines, a process that is required for diet-induced hepatic insulin resistance [237]. The loss of SOCS2 leads to an altered response to HFD in mice, resulting in This seems to be mediated by direct actions of SOCS2 on macrophage activation, as demonstrated in ex vivo experiments showing that BMDM from the SOCS2−/− mice exhibit increased phagocytic activity in vitro and are hyper-responsive to LPS stimulation leading to expression of IL-6, iNOS, IL-1  and INF-  .A previous study described an anti-inflammatory role of SOCS2, mediating the activity of aspirin-triggered lipoxins [133]. Our data suggest that the anti-inflammatory actions of SOCS2 may be mediated to some extent through inhibition of the LPS response in macrophages leading to NFκB activation. Transgenic mice with enhanced NFκB activity in hepatocytes exhibit insulin resistance driven by E2, GH, SOCS2, and Liver 63 inflammation in the absence of TG accumulation [238], which resembles the HFD fed SOCS2-/- mice. Measurements of hepatic cytokine production provide a strong indication that liver non-parenchimal cells participate in the hepatic response to HFD in SOCS2−/− mice. However, whether this altered reactivity is related to GH or other signals remains unknown. In relation to GH, we previously showed that GH treatment of SOCS2−/− mice results in enhanced hepatic expression of cytokine-regulated genes [116,118] and GH is known to exacerbate the inflammatory response in LPStreated rodents and worsen the conditions of critically ill patients [88]. Further work is needed to clarify the role of GH signaling in different tissues and cell types in the HF-fed SOCS2−/− mice. Studies in both human and rodents suggest that E2 protects from HFD-induced diabetes, steatosis, and adiposity [226,239,240,241]. Now, data from our lab also suggest a gender dimorphism in the HFD-fed SOCS2−/− model (manuscript in preparation). Interestingly, when SOCS2−/− mice were fed with standard diet (CD), there were no differences in glucose tolerance (ipGTT) between females and males. However, HFD diet caused more marked glucose intolerance in female than in male SOCS2−/− mice (Fig. 1). Mercedes de Mirecki Garrido 64 Figure 1. Dimorphism glucose tolerance in HFD-fed SOCS2-/- mice. WT (white columns) and SOCS2-/- (black columns ) mice were fed with chow diet (CD) or high-fat diet (HFD) as described in Material and Methods. Two weeks before mice were sacrificed, the glucose tolerance tests (ipGTT) were performed in males and females. Results are expressed as mean ± SEM (n=6). *P<0.05, **P<0.001, ***P<0.0001 The test for insulin sensitivity (ipITT) showed that, under CD treatment, both SOCS2−/− and wild type littermates, independently of gender, behave similarly. As described above, both HFD-fed SOCS2−/− groups (females and males) showed lower hypoglycemic response to exogenous insulin than wild type mice. Surprisingly, the HFD-fed SOCS2−/− female group showed better insulin sensitivity compared to HFD-fed SOCS2−/− males (Fig. 2). E2, GH, SOCS2, and Liver 65 Figure 2. Dimorphism insulin sensitivity in HFD-fed SOCS2-/- mice. WT (white columns) and SOCS2-/- (black columns ) mice were fed with chow diet (CD) or high-fat diet (HFD) as described in Material and Methods. A week before mice were sacrificed, the insulin tolerance tests (ipITT) were performed in males and females. AUC = area under curve of blood glucose levels. Results are expressed as mean ± SEM (n=6). *P<0.05, **P<0.001, ***P<0.0001 Furthermore, HFD diet caused less grade of steatosis in female than in male SOCS2−/− mice with a lower level of hepatic TG content. Finally, HFD-fed SOCS2−/− females, in comparison with HFD-fed SOCS2-/- male, showed 2,5-fold higher levels of circulating leptin (Fig. 3 A) and up to 9-fold lower plasma levels of inflammatory markers (i.e., IL6, INFγ, RANTES, TNFα, and IL1-β) (Fig. 3 BF). Our data suggest that E2 protects from HFD-induced steatosis and insulin resistance in the absence of SOCS2. Mercedes de Mirecki Garrido 66 Figure 3. Dimorphism in inflammatory response in HFD-fed SOCS2-/- mice. WT (white columns) and SOCS2-/- (black columns ) mice were fed with chow diet (CD) or high-fat diet (HFD) as described in Material and Methods. Then, the circulating levels of leptin (A), IL-6 (B), INFγ (C), RANTES (D),TNFα (E), and IL1β (F) were analysed in males and females. Results are expressed as mean ± SEM (n=6). *P<0.05, **P<0.001, ***P<0.0001 At physiological levels, the SOCS2−/− mice constitute a novel model system for the study of the metabolic syndrome with unique features that are relevant to the human disease. SOCS2−/− display no clear metabolic phenotype, these mice do not show obvious alteration in insulin sensitivity [118], in contrast GH transgenic mice develop marked insulin resistance [59,77,242]. HFD-fed SOCS2−/− mice displayed exacerbated insulin resistance compared to HFD fed WT. Fasting plasma insulin levels were increased in SOCS2−/− mice on HFD but we did not observe any differences in pancreatic levels of insulin or islet morphology, suggesting that the impaired insulin sensitivity is not cause by altered β-cell function. Moreover HFD induced a marked upregulation of the hepatic gene expression of proinflammatory cytokines in the SOCS2−/− mice. We conclude that SOCS2 has a protective role in the development of insulin resistance, probably by suppressing the release of proinflammatory cytokines. E2, GH, SOCS2, and Liver 67 In conclusion, our results demonstrated SOCS2 plays an important role in regulating the response to high-fat dietary stress. The SOCS2−/− mice provide a novel model to understand the complex relationship between inflammation, GH actions and nutrition in the control of hepatic glucose and lipid homeostasis. Future use of this model may help to outline the contribution of different mechanisms in the development of fat-induced hepatic insulin resistance and components related to lipid overload in comparison with components related to inflammatory stress. 4.4. Suppressor of Cytokine Signaling (SOCS)-2 gene deletion influences diabetes development induced by multiple low-dose streptozotocin (Paper IV; manuscript in preparation). It has been reported that experimental elimination of different SOCS has relevant physiological functions. For example, some studies have shown that SOCS are potent regulators of insulin signaling and beta cell functioning. Therefore, elimination of SOCS1 increases interferon response and potentiates islet cell death [177] while SOCS3 is a potent protector against type 1 Diabetes Mellitus (DM) through suppressing IL-B and TNF in pancreatic beta cells [178,179]. SOCS3−/− beta cells also show resistance to apoptotic cell death [180]. On the other hand, in the context of diabetes, SOCS2 has not been extensively studied, because it has mainly been implicated in GH and prolactin (PRL) signaling. However, it has been reported that SOCS2 is a potent regulator of pro-insulin processing and insulin secretion in β-cells [243]. Moreover, constitutive production of SOCS2 in β-cells leads to hyperglycaemia and glucose intolerance [175]. We have previously identified that SOCS2 negatively regulates GH signaling and SOCS2−/− mice grow larger than controls without elevation of GH [118,119]. The absence of SOCS2 makes mice highly sensitive to GH because SOCS2 deficiency leads to a reduced GH receptors (GHR) breakdown and, subsequently, more GHR. The elevated sensitivity of the SOCS2−/− mice to GH, and probably also to prolactin (PRL), is the rationale to investigate, in this study Mercedes de Mirecki Garrido 68 IV, how SOCS2 ablation may influence the development of diabetes in a mice model of autoimmune diabetes and β-cell destruction (multiple low-dose of streptozotocin (MLDSTZ)) [194], fed with standard diet. To analyze SOCS2 glucose homeostatsis and metabolism, the MLDSTZ was used as a model of autoimmune diabetes and B-cell destruction in vivo [194]. All diabetes development parameters were monitored, and both MLDSTZ-treated SOCS2-/- and WT mice developed sever diabetes after day 9 from injecting first dose of STZ, whereas vehicle-treated mice remained normoglycemic (Fig. 1). However, Figure 1A, shows that SOCS2−/− mice were more resistant to develop diabetes. E2, GH, SOCS2, and Liver 75 High-Fat-Diet (HFD)-induced diabetes, steatosis, and adiposity. These effects show a gender dimorphism being female more protected than male mice from HFD-induced damage. The results presented in this Thesis suggest that SOCS2, a physiological inhibitor of cytokine signaling, might play a physiological role in the regulation of hepatic steatosis in HFD-fed male mice. In addition, SOCS2 can mediate the inhibitory effects of E2 on GH-JAK2-STAT5 signaling pathway, which controls the hepatic lipid metabolism, the body size in mammals and it is an important regulator of glucose metabolism. In fact, GH has diabetogenic action inducing insulin resistance. SOCS2 Knockout mice show a GH dependent gigantism, but do not show alterations in insulin sensitivity. In contrast, GH transgenic mice are marked insulin resistant. Thus, SOCS2-/- mice provide a novel model to understand the complex relationship between inflammation, GH actions and nutrition in the control of hepatic glucose and lipid homeostasis. SOCS2−/− mice exhibit enhance liver TG secretion and are protected from HFD induced liver steatosis, being the females more protected than males. However, SOCS2−/− HFD mice display severe systemic insulin resistance, associated to hyper-insulinemia and worse insulin sensitivity in liver compared to WT mice fed similar diet. SOCS2−/− mice fed with a HFD also exhibit enhanced expression of inflammatory cytokines in liver and fat tissue, demonstrating a novel role of SOCS2 as negative regulator of macrofague activation in situations of dietary stress. Finally, experimental elimination of SOCS2 identifies this protein as an important regulator of insulin homeostasis in vivo. Our results demonstrate that MLDSTZinduced diabetic SOCS2-/- mice are more resistant to develop diabetes. Moreover, we also found that SOCS2 ablation is related to better conserved insulin sensitivity. We also observed that SOCS2-/- mice have higher beta cell , bigger beta islets size in pancreas, and higher serum insulin concentrations and HOMA-IR compared to Mercedes de Mirecki Garrido 76 WT. SOCS2-/- mice are more sensitive to GH (and probably to PRL) and exhibit higher pancreatic GHR and PRLR levels compared to WT. GH stimulates pancreatic β-cell growth, survival and insulin production through GHR-JAK2STAT5b signaling pathway. Thus, SOCS2 ablation may be related to improved insulin sensitivity and lower insulin resistance through a stronger and nonnegatively regulated action of GH on pancreas. These results suggest that SOCS2 inhibition may be used as therapeutic target to ameliorate diabetes development. E2, GH, SOCS2, and Liver 77 6. FUTURE PERSPECTIVES The findings in this thesis increase our knowledge about the SOCS2, E2 and GH interactions. Understanding E2, SOCS2 and GH interactions in physiological and pathological states may contribute to prevent health damage and improve clinical management of patients with growth, developmental and metabolic disorders. - In the general population, the endocrine and metabolic consequences of long-term exposition to estrogens or novel estrogen-related compounds and their influence on the GH axis are largely unknown; key importance knowledge in the understanding of the relationship of E2 and GH is SOCS2 and its regulation. Disturbances of SOCS2 would require further investigations of the hormonal and environmental factors in the control of its expression. - Future experiments in SOCS2−/− mice treated with E2 should elucidate the molecular mechanism that lead SOCS2 to effects E2 actions (metabolic, somatic growth and sexual-dimorphism in liver ) in vivo. - Further work is needed to clarify the role of GH signaling in different tissues and cell types in SOCS2−/− mice in response to HFD. - Future use of SOCS2−/−-HFD model may help to outline the contribution of different mechanisms in the development of fat-induced hepatic insulin resistance and components related to lipid overload in comparison with components related to inflammatory stress. - Further experiments should be done to understand the complex interplay between SOCS2 and DM. Mercedes de Mirecki Garrido 78 ACKNOWLEDGEMENTS I would like to thank my supervisor Leandro Fernandez, for give me the opportunity to work with you. It was an experience that I will not forget. You have been a constant throughout this long journey, and all this could not been without you. Borja Guerra, thank you for sharing his experience and knowledge. Lab people! Thanks a lot! I will not thanks enough for all you help, Dionisio Lorenzo, Ruyman Santana, Yeray Brito, Carlos Mateos, Sara Rubio, Patricia Martín, Roberto Jimenez, Cristina Bilbao, Laura Lopez, Julia Wiebe, Ana Wägner, Nico Martel, Raquel Ramirez, German Rodriguez, Arima Santana, Teresa Fuentes, the work in the laboratory and all the hours I've spent in the Culp could not have been the same without you. “El grupo de Castrillo”: Antonio Castrillo, Merci Diaz, José Guillén, Vladimir, Irene Hernández, Ana Ramón, Susana Beceiro, Noelia Alonso. Ignacio Robaina, Juan Carlos Diaz Chico, Ricardo Chirino, Nicolas Diaz Chico and Pilar, thank you for your support. Natalia Santana y Martin, “the animalario” would be a horror-movie without you. And I don´t want to forget to thank “the foreigners” Jorge Marrero, Fahad AlZadjali, Matias Vesterlund, Maria Claudia Sandoval, thank you for the time we share. And also to my colaborators Amira Alkharusi, Gunnar Norstedt, Amilcar Flores-Morales, Mario Díaz, and Diego Iglesias, thank you for everything. To Carmen Garrido for invited me to your laboratory, It was a pleasure for me stay with you in France, thank you to your research group for accepted me, INSERM Research Centre «Lipid, Nutrition and Oncology» Université de Bourgogne (Dijon). E2, GH, SOCS2, and Liver 79 To my family for being always by my side, giving me support at all times. And my couple thanks to prove that I can overcome anything. Mercedes de Mirecki Garrido 80 FUNDING This work was supported by grants from: 1. The Spanish Ministry of Science and Innovation with the funding of European Regional Development Fund-European Social Fund SAF200302117. 2. The Spanish Ministry of Science and Innovation with the funding of European Regional Development Fund-European Social Fund SAF200607824. 3. Agencia Canaria de Investigación, Innovación y Sociedad de la Información (ACIISI) PI2007/033. 4. Agencia Canaria de Investigación, Innovación y Sociedad de la Información (ACIISI) PI2010/0110. 5. Convenio de Investigación Alfredo Martin-Reyes Foundation (Arehucas)-Fundación del Instituto Canario de Investigación contra el Cáncer (FICIC). 6. Predoctoral fellowship from Universidad de Las Palmas de Gran CanariaMinisterio de Educación y Ciencias (ULPGC-MEC - AP2001-3499). 7. Fundación Universitaria de Las Palmas (FULP) programa INNOVA Canarias 2020, supported by Club de Leones de Gran Canaria y Haricana E2, GH, SOCS2, and Liver 81 RESUME OF THE MAIN ASPECT OF THE THESIS IN SPANISH Mercedes de Mirecki Garrido 82 E2, GH, SOCS2, and Liver 83 Estudios sobre el Estradiol, la Hormona de Crecimiento y el Supresor de la Señalización de Citoquínas-2, y la influencia en el metabolismo hepático Mercedes de Mirecki Garrido Tesis Doctoral Directores: Leandro Fernández Pérez Carlos Borja Guerra Hernández Faculta de Veterinaria Departamento de Ciencias Clínicas Farmacología Mercedes de Mirecki Garrido 84 1. OBJETIVOS El objetivo principal del trabajo presentado en esta Tesis es caracterizar la influencia del E2 y la GH y su relación con SOCS2 en el metabolismo hepático. En particular, nos hemos centrado en los siguientes objetivos específicos: - Caracterizar la relación que existe entre el E2 y la GH en el metabolismo hepático (Articulo I). - Caracterizar el rol de SOCS2 en el crecimiento somático y en el metabolismo hepático regulado por E2 (Articulo II). - Caracterizar el papel de SOCS2 en el desarrollo de hígado graso inducido por HFD (Articulo III). - Caracterizar el papel de SOCS2 en el desarrollo de la Diabetes Tipo I inducida por MLDSTZ (Articulo IV). E2, GH, SOCS2, and Liver 91 crecimiento se asocian en parte con un aumento de le catabolismo de aminoácidos y la síntesis de urea en el hígado [212]. Diversos análisis biológicos multidisciplinares muestran que una administración intermitente de GH a ratas TXOX produce una regulación positiva del catabolismo celular mientras los genes involucrados en el metabolismo de aminoácidos y urea (i.e., OTC, ASS1, aminotransferasas, and methyltransferasas) ven significativamente disminuida su expresión. Este efecto positivo de la GH sobre el balance de nitrógeno fue previamente estudiado en ratas hipofisectomizadas [8,40,213]. La GH actúa como una hormona anabolizante ya que promueve la lipolisis y evita la lipogénesis en el tejido adiposo, aumentado la disponibilidad de ácidos grados (FFA) para su uso como energía [8]. El E2 puede interferir en este proceso impidiendo la inducción de algunos genes relacionados con la utilización de lípidos, como es ApoC2. Por ello, las acciones del E2 en el hígado pueden afectar los efectos metabólicos periféricos de la GH. En situaciones de poco gasto energético, como el hipotiroidismo, la deficiencia en GH, E2 o el envejecimiento, se observa una estimulación de la lipogenesis [214]. Nuestros análisis del contenido lipídico del hígado revelan que las ratas TXOX presentan un aumento significativo de los niveles de ácidos grasos saturados totales (SFA) en comparación con el grupo intacto de ratas. El tratamiento con E2 no modifica los niveles de expresión de mRNA de los principales reguladores de la lipogenesis hepática [i.e., Sterol regulatory element binding protein (SREBP)1c, acetyl-Co A carboxylase alpha (ACC), fatty-acid synthase (FAS)] [215], mientras que sí activa el programa transcripcional de PPARα, que promueve el catabolismo de los ácidos grasos en el hígado [216,217]. Esto es evidente por el aumento de la expresión del gen de PPARα y sus genes diana involucrados en la β/-oxidación de ácidos grasos (i.e., CTE-I, CPT-2, Fasd6, Fasd1, Fasd2, Scd1, ACOX1, ECH1, BAAT, FGF21, CYP4A1, CYP4A3) mediante el E2. La terapia con E2 produce Mercedes de Mirecki Garrido 92 una disminución significativa de los SFAs. En conjunto nuestros hallazgos sugieren la existencia de una relación positiva entre E2 y PPARα, la cual esta respaldada por múltiples estudios independientes [136,159,218,219]. Sorprendentemente, a pesar del aumento de la expresión de genes involucrados en β-oxidación, detectamos un aumento significativo del contenido de TG hepáticos en las ratas TXOX tratadas con E2, lo cual probablemente pueda ser explicado por efectos en el transporte de lípidos. El primer paso en el en la absorción de ácidos grasos de cadena larga es su translocación a través de la membrana plasmática. Particularmente, el E2 aumenta la transcripción de varios genes activados por PPARα que codifican proteínas que están implicadas en la absorción y activación de los ácidos grasos como CD36, ACSL4 and SLC27A5 (FATP5) [220,221]. En estudios publicados previamente por nuestro laboratorio demostramos que el transportador de ácidos grasos CD36 se expresa predominantemente en los hígados de las ratas hembras y propusimos que este dimorfismo sexual depende de le patrón de secreción de la GH, el cual se puede ver influenciado por el tratamiento con E2. Además, el E2 aumenta la transcripción de genes que codifican para FATP5, como SLC27A5, y transportadores de los ácidos grasos como acil-CoA-sintetasa, que cataliza la conjugación de los ácidos biliares con aminoácidos antes de su excreción en los canalículos biliares [222]. Tras la absorción de los ácidos grasos, el primer paso para la utilización intracelular de los ácidos grasos de cadena larga es su esterificación con CoA. Esta reacción esta catalizada por las acil-CoA-sintetasas, como por ejemplo la ACSL4 la cual es inducida por E2 en los hígados de las ratas TXOX. Los acil-CoA producidos son sustratos para la ß-oxidación, pero también participan en la síntesis de triglicéridos (TG), fosfolípidos, CE y ceramidas y por lo tanto también son una fuente primaria de moléculas de señalización [223]. La idea de que E2 pueda regular la formación de intermediarios de la señalización lipídica se apoya en la estimulación de la expresión de la elongasa 5 de ácidos grasos (Elovl5). Elovl5 utiliza los ácidos grasos insaturados para general muchos de los PUFAs de cadena E2, GH, SOCS2, and Liver 93 larga que son asimilados por los lípidos celulares (i.e., 20:4n-6 and 22:6n-3). Sin embargo, merece la pena mencionar que la administración de E2 no altera el metabolismo de VLCPUFA, porque los niveles de 20:4n-6, 20:5n-3 and 22:6n-3 son similares a los observados para el grupo de TXOX. Se ha descrito que E2 puede desempeñar un papel crítico en la lipogénesis y la transcripción de SCD1 [136], gen que codifica para la enzima que limita la generación de MUFAs como 18:1 n−9 and 16:1 n−7. Estudios previos revelan que la ausencia de E2 o ERα en ratas da lugar un profundo incremento en la lipogénesis y en la transcripción de SCD1 [137], lo cual sugiere que el E2 inhibe la transcripción de SCD1. Se ha descrito que el efecto antilipogénico de la terapia con E2, manteniendo la eficiencia de la exportación de TG, depende del ERα hepático [219,224]. Nuestro estudio muestra que el E2 aumenta la expresión de SCD1 y que este efecto es paralelo a una reducción del contenido hepático 18:0 y a un aumento de 18:1 n−9 (producto principal de las reacciones de SCD), y lípidos neutrales en comparación con los TXOX, lo cual indica que E2 modula la actividad de SCD1 en los hígados de las ratas TXOX. Sorprendentemente, el E2 disminuye además la expresión de SCD2 en los hígados de las ratas TXOX. La importancia de esta regulación transcripcional opuesta de los genes SCD es desconocida, pero los niveles transcripcionales de SCD1 son aproximadamente 1800 veces más altos que los de SCD2 en el hígado de las ratas [225]. Esto último indica que los cambios en 18:1n9 y 18:0 deben atribuirse a las variaciones de la expresión de SCD1. En general, los cambios en la composición lipídica y en el perfil de expresión genético que se observan en las ratas TXOX tratadas con E2 sugieren que las interacciones funcionales E2 - PPARα juegan un papel fisiológico clave en la regulación del metabolismo hepático de los lípidos. El E2 es capaz de reducir colesterol (CHO) en mujeres y modelos animales alimentados con dieta alta en grasa [226]. Sin embargo, el E2 es incapaz de revertir eficazmente la hipercolesterolemia o la hipertriglicemia en las ratas TXOX. Mercedes de Mirecki Garrido 94 Esto se debe al hecho de que el E2 reduce la expresión de diversos transportadores de CHO, incluyendo Apob y ABCA1 en las ratas TXOX, los cuales contribuyen al mantenimiento y aumento de los niveles hepáticos de CE. Además, el E2 induce la movilización intracelular de CHO modulando las enzimas involucradas en las síntesis/movilización de CE y CHO [214,227]. Diferentes enzimas catalizan la conversión de CHO en CE en el hígado. Un claro ejemplo es la lecitina:colesterol aciltransferasa (LCAT), la cual usa fosfatidilcolina (PC) y ACAT, que es usada por acyl-CoA. Debido a que los niveles de lisofosfatidilcolina (LPC) son indetectables en todos los grupos, nuestra conclusión inicial es que el E2 estimula la reacción de ACAT2 para aumentar CE. Aunque no se detectaron en los niveles de expresión de ACAT, esto no descarta la posible modificación traduccional de enzimas del ciclo de CE en el hígado de las ratas TXOX tratadas con E2. El aumento de CE hepático junto con el aumento de TG y la disminución del contenido hepático de FFA en las ratas TXOX tratadas con GH, son similares a los efectos del E2 en la composición hepática lipídica lo cual sugiere que algunos efectos del E2 podrían estar mediados por la GH. Una consecuencia llamativa de la terapia combinada de E2 y GH es la completa recuperación de los niveles de MUFA de los lípidos totales y neutrales. Este efecto se le atribuye al aumento de 18:1n9, probablemente mediante la alteración de la expresión de la 9 desaturasa. Además GH y E2 aumentan los CE hepáticos y el efecto combinado de las dos hormonas es aditivo con respecto a CE, ya que sus niveles son el doble de los encontrados en los animales intactos, y son aproximadamente un 30% superiores a los encontrados en los grupos tratados con E2 o con GH por separado, lo cual es un indicativo de un metabolismo hepático de CHO más eficiente. Por consiguiente, en presencia de E2 y GH el nivel hepático de contenido en CHO se ve reducido no sólo respecto al grupo TXOX sino también respecto a los grupos tratados con E2 o GH. La GH aumenta significativamente el contenido de TG hepáticos en las ratas TXOX tratadas previamente con E2, lo cual sugiere que el tratamiento combinado de E2 y GH aumenta drásticamente la lipogénesis. Actualmente, es conocido que E2, GH, SOCS2, and Liver 95 en contra de este efecto lipolítico de la GH en los tejidos adiposos, la GH ejerce efectos lipogénicos en el hígado a través de la estimulación de SREBP1, la cual normalmente se ve acompañada por un aumento de la secreción de los TG hepáticos (VLDL) [8]. Nuestro análisis del perfil lipídico sugiere que la administración intermitente de GH a ratas TXOX aumenta la lipognesis en el hígado, al contrario que el tratamiento continuo con GH en ratas hipofisectomizadas [40]. La administración intermitente de GH a ratas TXOX no aumenta SERBP1, mientras que sí induce diferentes genes involucrados en el transporte de ácidos grasos (e.g., FABP) y en la biosíntesis de ácidos grasos insaturados de 18:2n-6 y18:3n-3 (por ejemplo las desaturasas de ácidos grasos 4, 5 y 6). La administración intermitente de GH a ratas TXOX disminuye la expresión del gen lipina, un gen diana de SREBP1c, el cual es un regulador clave de los niveles celulares de DG y TG, además de la oxidación de ácidos grasos en el tejido adiposo, en el musculo esquelético y en el hígado [228]. Estos hallazgos apoyan la hipótesis de que un patrón femenino de administración de GH es un estimulo más eficiente para inducir efectos lipogenicos en el hígado que un patrón masculino [184,229]. Otro mecanismo mediante el que GH fomenta la lipogénesis en el hígado es a través de la disminución de la oxidación lipídica. Estudios previos de nuestro laboratorio han demostrado que la administración continua de GH a ratas hipofisectomizadas [40] y a ratas adultas intactas [46] inhibe PPARα. De acuerdo con nuestro análisis lipídico y genómico la administración intermitente de GH a ratas TXOX da lugar a la regulación negativa de la vía de señalización de PPARα. En particular, la GH reprime la expresión de PPARα, ACOX-1, CPT-1, FGF21, además de diversos miembros de la familia CYP4A, involucrados en la oxidación de ácidos grasos. En resumen, nuestro estudio añade datos novedosos que resaltan el efecto de la administración subcutánea de E2 sobre la fisiología del hígado y su relación con la GH. Estos resultados ponen de manifiesto el papel de E2 como regulador clave del Mercedes de Mirecki Garrido 96 metabolismo en el hígado de los mamíferos y añaden más peso a la hipótesis de que E2 actúa como un importante regulador de las acciones de GH en el hígado. La interacción E2-GH en el hígado es relevante debido a las diferentes funciones fisiológicas que estas hormonas tienen en mamíferos y al amplio uso de los estrógenos y los compuestos derivados de los estrógenos en humanos. En particular, este es el primer estudio que demuestra que los perfiles de lípidos hepáticos tienen marcas singulares que se pueden utilizar para separar diferentes grupos con estados hormonales alterados. Esto incluye los reemplazos hormonales (E2 o GH) que inducen cambios que se superponen en la expresión génica. Por lo tanto, el perfil lipídico del hígado puede servir para identificar deficiencias hormonales crípticas o exposición a hormonas o a sustancias similares a las hormonas. 4.2. El Supresor de la Señalización de Citoquinas-2 influencia los efectos somáticos del E2 y el transcriptoma hepatico (Artículo II; manuscrito en preparación). En el artículo I, añadimos nuevos datos que ponen de manifiesto el impacto de la administración subcutánea E2 sobre la fisiología del hígado y su interacción con GH. La interacción E2-GH en el hígado es relevante debido a las funciones fisiológicas que estas hormonas tienen en los mamíferos y el amplio uso de los estrógenos y los compuestos derivados del estrógeno en humanos. En este trabajo (artículo II), hemos llevado a cabo un análisis del transcriptoma para obtener información completa sobre los mecanismos moleculares de los efectos de E2 in vivo en presencia (WT) o en ausencia (SOCS2-/-) de SOCS2. La influencia de SOCS2 en el crecimiento somático regulado por E2 y en transcriptoma hepático que observamos en el presente estudio podría contribuir a una mejor comprensión de los mecanismos moleculares implicados en las consecuencias endocrinas y metabólicas de la exposición a los estrógenos o compuestos derivados del estrógeno y su influencia sobre las acciones de la GH. Como primera aproximación para evaluar la influencia de SOCS2 sobre los efectos de E2 en el E2, GH, SOCS2, and Liver 97 hígado se estudiaron los cambios en las funciones somatotrópicas del hígado en la ausencia o en la presencia de SOCS2. Después de 30 días, E2 redujo la ganancia de peso corporal en los dos genotipos. Sin embargo, la reducción de peso corporal por E2 fue más prolongada y mayor en el grupo WT (datos no mostrados). Como era de esperar [118], cuando se inició el tratamiento E2 (tiempo 0), los ratones SOCS2-/- tenían un mayor peso corporal que los WT. De esta forma, las mediciones de ganancia de peso corporal se normalizaron por el peso corporal en el tiempo 0. Esta normalización demostró claramente que el tratamiento con E2, en comparación con los ratones tratados con el vehículo, causa una mayor reducción el peso corporal en los ratones WT que en los ratones SOCS2-/- (Fig. 1B). A punto final, observamos que el tratamiento con E2 produjo cambios significativos en la longitud del cuerpo (Fig. 1C) y el fémur (Fig. 1D) pero sólo en los ratones WT. Además, el E2 produjo un aumento de los niveles del mRNA hepático de IGF-I (Fig. 1E) en ambos genotipos sin cambios en los niveles circulantes de IGF-I (Fig. 1F). Mercedes de Mirecki Garrido 98 Figure 1. Influencia de SOCS2 sobre el crecimeinto regulado por E2 en el hígado. Ratones WT y SOCS2−/− se trataron con E2 benzoateo(E2B) (50µg/kg/48h) o vehiculo (VEH) durante 90 dias tal y como se describe ne el Material y Metodos. Peso corporal, normalizado a dia 50, de los ratones WT (A) and SOCS2-/- (B) se monitorizo cada 7 dias en ausencia de E2B (símbolo blanco) o en presencia de E2B (simbolo negro). Longitud corporal (C) y del femur (D), niveles hepaticos de expression de IGF-I mRNA levels (E), y nivel circulante de IGF-I (F) se midieron a teimpo fianl. Los resultados se expresan como media ± SEM de 6 individuos por grupo. La media de la expression de IGF-I en el VEH WT group se define como 1, y los demás valores de expression se relativizan respect a este valor. *P<0.05; **P<0.01, *** P<0.001; # P<0.05, ## P<0.01, ### P<0.001 para comparacion entre los dos genotipos Estudios previos han demostrado que el E2 induce la expresión de SOCS2 en el hígado, que a su vez regula negativamente la actividad transcripcional mediada por STAT5 [61,62,193]. Por lo tanto, la consecuencia molecular de la deleccion del gen de SOCS2 es una activación prolongada de STAT5 [230]. En este estudio realizamos además un análisis cuantitativo de la expresión de los mRNA de SOCS2, CIS, SOCS1 y SOCS3, todos reguladores negativos de GHR-STAT5 [48,210]. El tratamiento con E2 aumentó los niveles hepáticos del mRNA de SOCS2 (Fig. 2A) en los ratones WT, mientras que éste no pudo ser detectado en E2, GH, SOCS2, and Liver 99 ratones SOCS2-/-. En ratones WT, el E2 incrementó la expresión hepática de los mRNAs de CIS (Fig. 2B) y SOCS1 (Fig. 2C), mientras que no alteró la expresión del mRNA de SOCS3 (Fig.2D). Cabe destacar que en los ratones SOCS2-/- tratados con vehículo, los niveles de expresión del mARN de CIS (Fig. 2B) y SOCS3 (Fig.3D) se redujeron significativamente mientras que los de SOCS1 (Fig. 2C) se mantuvieron sin cambios. Sin embargo, el tratamiento con E2 fue capaz de incrementar los niveles de expresión del mRNA de CIS, SOCS1 o SOCS3 en los ratones SOCS2-/-.Estos resultados demuestran que la eliminación del gen de SOCS2 influye drásticamente en la actividad transcripcional activada por el E2 en el hígado. Figure 2. Influencia de SOCS2 sobre la regulacion ejercida por E2 en la expression hepatica de los Suprresores de la Señalización de Citoquinas. SOCS2 (A), CIS (B), SOCS1 (C), y SOCS3 (D) se midieron por qPCR tras el sacrificio. Los resultados se expresan como media ± SEM de 6 individuos por grupo. La media de la expression del VEH WT group se define como 1, y los demás valores de expression se relativizan respect a este valor. . *P<0.05; **P<0.01, *** P<0.001; # P<0.05, ## P<0.01, ### P<0.001 para comparacion entre los dos genotipos Mercedes de Mirecki Garrido 100 La mayoría de los estudios anteriores se han centrado en la influencia de los estrógenos sobre la secreción hipofisaria de GH [34] y sobre la influencia del género en esta secreción, demostrado que ésta tiene un gran impacto en la regulación transcripcional hepática [33]. Teniendo en cuenta lo anteriomente mencionado, en el presente estudio investigamos la influencia del E2 en los genes regulados por la GH y relacionados con el dimorfismo sexual hépatico [33] en ratones WT y SOCS2-/-. Al tratar con E2 se observó una reducción en la expresión del mRNA del gen MUP (un gen específicamente de machos) (Fig. 3A), mientras que los niveles del mRNA de Cyp2B9 y A1BG (dos genes específicos de las hembras) (Fig. 3B) se vieron incrementados. Estos hallazgos sugieren que la ausencia de SOCS2 es suficiente por sí misma para feminizar el hígado macho e influir en los efectos de E2 el dimorfismo sexual. En ambos genotipos, el tratamiento con E2 produjo un descenso de expresión del mRNA de MUP (Fig. 3A), mientras que los niveles de mRNA de Cyp2b9 (Fig. 3B) y A1BG (Fig. 3C) se vieron aumentados. Esto último sugiere que el tratamiento con E2 feminiza el hígado adulto. En conjunto, estos resultados apoyan nuestra hipótesis de que SOCS2 puede influir en los efectos del E2 sobre el eje somatotrópico-hígado. Figure 3. Influencia de SOCS2 sobre la regulacion de E2 en el dimorfismo del hígado. MUP (A), Cyp2b9 (B), and A1bg (C) se midieron por qPCR tras el sacrificio. Los resultados se expresan como media ± SEM de 6 individuos por grupo. La media de la expression del VEH WT group se define como 1, y los demás valores de expresion se relativizan respect a este valor. . *P<0.05; **P<0.01, *** P<0.001; # P<0.05, ## P<0.01, ### P<0.001 para comparacion entre los dos genotipos E2, GH, SOCS2, and Liver 107 Figure 9. Analisis de los efectos biologicos de 2 días de tratamiento E2 en el transcriptoma del hígado en ratones SOCS2-/-. Mediante microarray de DNA analizamos las diferencias de expression de los genes hepaticos como se describe en Material y Métodos. Represnetamos los resulatdos del analisis functional y biologico (DAVID) usando Cytoscape. Nodo (círculo interno) el tamaño corresponde al número de genes regulados positivamente por E2; parte exterior de el Nodo (círculo exterior) el tamaño corresponde al número de genes regulados negativamente por E2; color del nodo y de la parte exterior corresponde a la importancia del conjunto de genes, sobreexpresados o inhibidos respectivamente, (rojo oscuro = enriquecido significativamente, rojo claro = enriquecido no significativamente, gris = ausente); El grosor de las lineas conectoras corresponde al número de genes que se superponen entre las dos situaciones analizadas. Lineas verdes corresponden a los genes sobreexpresados y lineas azules corresponden a los genes inhibidos. Cabe destacar que 625 y 678 genes fueron inducidos y reprimidos, respectivamente, por el E2, en los ratones SOCS2-/- pero no en animales WT (véase la Tabla 1-2 una pequeña representación). En general, estos resultados revelan que el E2 produce una profunda re-programación de la fisiología del hígado a corto plazo y que SOCS2 influencia notablemente los efectos del E2 in vivo. Mercedes de Mirecki Garrido 108 ID Probe Unigene/ Refseq Symbol Gene description R q(%) mean ± SD A_55_P1995537 NM_010824 Mpo Myeloperoxidase 3,41 0,85 0,00 A_52_P15388 NM_008522 Ltf Lactotransferrin 2,13 0,61 0,67 A_55_P1983921 NM_021352 Crybb3 Crystallin, beta B3 1,95 0,93 1,49 A_51_P167292 NM_009892 Chi3l3 Chitinase 3-like 3 1,93 0,55 0,67 A_55_P2163098 NM_134066 Akr1c18 Aldo-keto reductase family 1, member C18 1,90 0,36 0,00 A_51_P461067 ENSMUST00000 103420 G1m Immunoglobulin heavy constant gamma 1 1,85 0,64 0,93 A_55_P1953387 NM_001272097 Fabp5 Fatty acid binding protein 5, epidermal 1,74 1,31 4,03 A_51_P199168 NM_007702 Cidea Cell death-inducing DNA fragmentation factor, alpha subunitlike effector A 1,73 0,72 0,99 A_52_P213889 NM_172476 Tmc7 Transmembrane channel-like gene family 7 1,66 1,08 3,07 A_51_P488196 NM_028472 Bmper BMP-binding endothelial regulator 1,65 0,28 0,00 A_55_P2024155 NM_001033324 Zbtb16 Zinc finger and BTB domain containing 16 1,65 1,16 3,38 A_51_P520650 NM_177639 Dlgap1 Discs, large (Drosophila) homologassociated protein 1 1,62 1,26 4,03 A_55_P2002975 NM_205795 Mrgprb4 MAS-related GPR, member B4 1,55 0,96 2,80 A_55_P2054913 NM_001011863 Olfr406 Olfactory receptor 406 1,54 0,56 0,99 A_55_P2149921 TC1780716 Q80XJ7 Aldo-keto reductase family 1, member A4 1,52 0,91 2,80 A_55_P2185900 NM_032002 Nrg4 Neuregulin 4 1,49 0,87 2,80 A_51_P295896 NM_028934 4930452B06 Rik RIKEN cDNA 4930452B06 gene 1,47 0,56 0,99 A_55_P2290388 NM_001043354 Rorb RAR-related orphan receptor beta 1,46 0,53 0,99 A_55_P2001553 NM_020043 Igdcc4 Immunoglobulin superfamily, DCC subclass, member 4 1,45 0,16 0,00 Table 1. Lista representative de los genes hepaticos sobreexpresados en los ratones SOCS2-/- pero no en los WT tras 2 dos dias de tratamiento con E2. Mediante microarray de DNA analizamos las diferencias de expression de los genes hepaticos como se describe en Material y Métodos. Usamos el analisis estadistico SAM y discriminamos los genes que se expresan usando un FDR menor a 5% y log2 > |0.58|. La tabla se compone de probe ID, Unigene/Refseq, gene symbol, gene description, R (log2 E2-treated OXSOCS2-/- / vehicle-treated SOCS2-/-), and q (%). E2, GH, SOCS2, and Liver 109 ID Probe Unigene/Refseq Symbol Gene description R q(%) mean ± SD A_55_P2408588 NM_007489 Arntl Aryl hydrocarbon receptor nuclear translocator-like -2,29 1,22 1,93 A_52_P480044 XR_105914 BC023105 cDNA sequence BC023105 -2,21 0,72 0,94 A_55_P2213968 NR_045840 4933416M 07Rik RIKEN cDNA 4933416M07 gene -2,14 0,88 1,17 A_66_P120728 NM_001011791 Olfr193 Olfactory receptor 193 -2,10 0,90 1,49 A_51_P107315 NM_145423 Slc5a8 Solute carrier family 5 (iodide transporter), member 8 -2,02 0,52 0,94 A_65_P11137 NM_019541 Cts8 Cathepsin 8 -1,96 0,22 0,00 A_55_P2056714 NM_001042612 Nlrp9c NLR family, pyrin domain containing 9C -1,95 0,90 1,57 A_55_P2151591 ENSMUST0000 0032909 Pde3b Phosphodiesterase 3B, cGMPinhibited -1,94 0,73 1,09 A_51_P210031 NM_001252679 Smr2 Submaxillary gland androgen regulated protein 2 -1,90 0,94 1,93 A_55_P2121042 NM_001082531 Pla2g2a Phospholipase A2, group IIA (platelets, synovial fluid) -1,85 0,53 0,94 A_52_P122649 NM_175647 Dmrta1 Doublesex and mab-3 related transcription factor like family A1 -1,81 0,67 1,09 A_52_P669922 NM_032541 Hamp Hepcidin antimicrobial peptide -1,80 0,28 0,00 A_55_P2095859 NR_037604 Rdh18-ps Retinol dehydrogenase 18, pseudogene -1,78 0,49 0,94 A_52_P186751 NM_198677 BC061237 cDNA sequence BC061237 -1,77 0,58 1,09 A_51_P267969 NM_011259 Reg3a Regenerating islet-derived 3 alpha -1,77 0,60 1,09 A_55_P2073024 NM_001034859 Gm4841 Predicted gene 4841 -1,76 0,46 0,94 A_51_P322473 NM_172417 2310042D 19Rik RIKEN cDNA 2310042D19 gene -1,74 0,07 0,00 A_55_P2051082 XR_168616 Gm11634 Predicted gene 11634 -1,72 0,25 0,00 Table 2. Lista representative de los genes hepaticos inhibidos en los ratones SOCS2-/- pero no en los WT tras 2 dos dias de tratamiento con E2. Mediante microarray de DNA analizamos las diferencias de expression de los genes hepaticos como se describe en Material y Métodos. Usamos el analisis estadistico SAM y discriminamos los genes que se expresan usando un FDR menor a 5% y log2 > |0.58|. La tabla se compone de probe ID, Unigene/Refseq, gene symbol, gene description, R (log2 E2-treated OXSOCS2-/- / vehicle-treated SOCS2-/-), and q (%). Mercedes de Mirecki Garrido 110 4.3. Relación entre SOCS2, esteatosis hepatica y resistencia a la insulina en una Dieta alta en grasa (HFD) en ratones (Paper II; FASEB J 2012, 26:32823291). En este estudio analizamos al respuesta metabólica de los ratones SOCS2−/− a una dieta hipercalórica y rica en grasa (HFD). Los ratones SOCS2−/− muestraron un aumento de la expresión hepática de secreción de TG y están más protegidos de la esteatosis hepática inducida por HFD (Fig.1). Sin embargo, éstos presentan una severa resistencia sistémica a la insulina asociada a una hiperinsulinemia y a una peor sensibilidad a la insulina en el hígado en comparación con los WT alimentados con la misma dieta. Los ratones SOCS2−/− alimentados con HFD presentan además un aumento de la expresión hepática de citoquinas inflamatorias, sugiriendo así un novedoso papel de SOCS2 en la regulación negativa de la activación de macrófagos bajo condiciones de estrés, como es la HFD. Es importante señalar que el fenotipo de los ratones SOCS2−/− es claramente diferente a los fenotipos de los ratones deficientes en SOCS3 y SOCS1 en el hígado [234,235], los cuales muestran un aumento de la estatosis hepática. Esto último pone de manifiesto la existencia de diferencias funcionales entre los diferentes miembros de la familia SOCS en la regulación del metabolismo hepático. Figure 1. Analisis de el contenido lipidico en higado de ratones, WT y SOCS-/-, alimentados 4 meses con CD o HFD (A), tinción Oil Red de secciones de higado (imagenes magnificadas 40x) E2, GH, SOCS2, and Liver 111 En estudios previos realizados en nuestro laboratorio hemos demostrado que SOCS2 es regulador negativo de la señalización activada por la GH a través de su receptor en el hígado [116]. La reducida esteatosis observada en los ratones SOCS2−/− (HFD) en comparación con los WT (HFD), está en concordancia con la conocida acción de la GH fomentando la movilización de lípidos hepáticos [236] y sugiere, claramente, que el aumento de la señalización de GH en el hígado es en parte responsable de estos efectos. De hecho, en el presente estudio encontramos en los ratones SOCS2−/− un incremento de los genes involucrados en la formación y secreción de TG, regulados por la GH, como son ApoB, FABP1 y DGAT2. Un análisis detallado realizado en ratones con una delección hepática específica de JAK2 ha dado a conocer otro mecanismo mediante el cual la GH controla el contenido de grasa del hígado [170]. Estos ratones, llamados JAK2L, presentan elevados niveles de GH circulante, que a su vez da lugar a un aumento de la lipólisis en el tejido adiposo. Esto aumenta el suministro de FFA al hígado y se produce la esteatosis [170]. Al contrario que los JAK2L, los ratones SOCS2-/- tienen unos niveles de GH circulante reducidos y muestran un incremento en el tejidos graso, lo que sugiere que la reducida lipólisis en el tejido adiposo de los ratones SOCS2-/- podría contribuir a la reducida acumulación de TG en el hígado. En este estudio, no detectamos cambios significativos en los niveles circulantes de ácidos grasos libres (FFA) ni en la expresión hepática de CD36 en los ratones SOCS2-/- en comparación con lo previamente descrito en los ratones JAK2L. Teniendo en cuenta todo lo anteriormente comentado, consideramos que se necesitan experimentos adicionales para analizar la influencia de la GH y la lipólisis del tejido adiposo en los hígados de los ratones SOCS2-/-. Sorprendentemente, los ratones SOCS2−/− muestran una respuesta exagerada de empeoramiento de la sensibilidad a la insulina provocada por HFD, mientras que ante una dieta normal encontramos pocas diferencias entre los SOCS2−/− y los WT en la señalización activada por la insulina. Esto sugiere que la hiperactividad de la Mercedes de Mirecki Garrido 112 señalización hepática del receptor de GH por sí sola, tal como se observa en los ratones SOCS2−/−, no es suficiente para explicar el deterioro observado en este estudio en lo que se refiere a la homeostasis de la glucosa inducida por la HFD. Una explicación más plausible podría ser que los efectos anti-insulinicos de la GH se ven aumentados por mecanismos relacionados con la dieta bajo el control de SOCS2. En este estudio proporcionamos evidencias de tres de estos mecanismos: hiperinsulinemia, adiposidad periférica y excesiva producción de citoquinas inflamatorias. Los receptores tipo Toll fomentan la activación de macrófagos en el hígado y en el tejido adiposo mediante la producción de citoquinas inflamatorias inducidas por la HFD. Este proceso tiene que darse para que se produzca la resistencia hepática a la insulina inducida por dieta alta en grasa [237]. La perdida de SOCS2 da lugar a respuestas alteradas ante la HFD en ratones, produciendo un incremento en la expresión de citoquinas inflamatorias y un aumento de la activación de NFκB. Esto parece estar mediado por acciones directas de SOCS2 sobre la activación de los macrófagos, como demostramos en experimentos ex vivo, en los que BMDM de los ratones SOCS2−/− mostraban un incremento en la actividad fagocitica in vitro y además hiper-responden a la estimulación con lipopolisacárido (LPS) produciendo la expresión de IL-6, iNOS, IL-1  and INF-  . Algunos estudios previos describen el papel antiinflamatorio de SOCS2 mediando la actividad de lipoxinas activadas por la aspirina [133]. Nuestros datos sugieren que el efecto antiinflamatorio de SOCS2 podría estar mediado a través de la inhibición de la respuesta al LPS en los macrófagos mediante la activación de NFκB. Otros estudios han demostrado que ratones transgénicos que presentan un aumento de la actividad hepática de NFκB muestran una resistencia a la insulina fomentada por la inflamación en ausencia de acumulación de TG [238], lo cual nos recuerda al modelo de ratones SOCS2-/- alimentados con HFD. Las medidas de producción hepática de citoquinas proporciona evidencias de que las células no parenquimales del hígado participan en la respuesta a la HFD en los E2, GH, SOCS2, and Liver 113 ratones SOCS2−/−. Sin embargo, se desconoce si esta reactividad alterada es debida a la GH o a otros intermediarios. En relación con la GH, en estudios previos de nuestro laboratorio hemos demostrado que el tratamiento con GH en ratones SOCS2−/− da lugar a un aumento de la expresión hepática de genes regulados por citoquinas [116,118]. Además la GH aumenta la respuesta inflamatoria producida en respuesta al tratamiento con LPS en roedores y empeora las condiciones de pacientes en estado crítico [88]. Por tanto, se precisa de más investigación para clarificar el rol de la señalización activada por GH en diferentes tejidos y tipos celulares en ratones SOCS2−/− alimentados con dieta HFD. Diferentes estudios realizados en humanos y roedores sugieren que el E2 protege del desarrollo de la diabetes, la esteatosis y del incremento de grasa producido por la HFD [226,239,240,241]. Los datos de nuestro laboratorio sugieren un dimorfismo sexual en respuesta a la HFD en los ratones SOCS2−/− (manuscrito en preparación). Curiosamente, cuando se mantienen los ratones SOCS2−/− con dieta estándar (CD) no encontramos diferencias entre machos y hembras en la tolerancia a la glucosa (ipGTT). Sin embargo la HFD produce una intolerancia a la glucosa más marcada en las hembras que en los machos SOCS2−/− (Fig. 1). Mercedes de Mirecki Garrido 114 Figure 1. Dimorfismo en la tolerancia a la glucosa en ratones SOCS2-/- con HFD. WT (columnas blancas) y SOCS2-/- (comlumnas negras) se alimentaron con Dieta control (CD) o alta en grasa (HFD) como se describe en material y metodos. Dos semanas antes del sacrificio se realize un test de tolerancia a la glucose (ipGTT). Los resultados se expresan como media ± SEM (n=6). *P<0.05, **P<0.001, ***P<0.0001 El test de sensibilidad a la insulina (ipITT), muestra que ante la dieta CD ambos genotipos (SOCS2−/− y WT), independientemente del genero, se comportan de manera similar. Sin embargo, tal como se describió anteriormente, ambos grupos de ratones SOCS2−/− (machos y hembras) tratados con HFD muestran una menor repuesta hipoglucémica a la insulina exógena que los WT. Si bien es cierto que las hembras SOCS2−/− (HFD) presentan una mejor sensibilidad a la insulina si se las compara con los machos SOCS2−/− (HFD) (Fig. 1B). E2, GH, SOCS2, and Liver 115 Figure 2. Dimorfismo en la sensibilidad a la insulina en ratones SOCS2-/- con HFD. WT (columnas blancas) y SOCS2-/- (comlumnas negras) se alimentaron con Dieta control (CD) o alta en grasa (HFD) como se describe en material y metodos. Dos semanas antes del sacrificio se realize un test de tolerancia a la insulina (ipITT). Los resultados se expresan como media ± SEM (n=6). *P<0.05, **P<0.001, ***P<0.0001 Además, la HFD produce un menor grado de esteatosis en las hembras que en los machos SOCS2−/− con un menor contenido hepático de TG. Finalmente, las hembras SOCS2−/− (HFD) en comparación con los machos SOCS2−/− (HFD) presentan 2.5 veces más leptina circulante y unos niveles 9 veces inferiores de marcadores inflamatorios plasmáticos (i.e., IL6, INFγ, RANTES, TNFα, and IL1β) (Fig. 2B-C). Nuestros datos sugieren que el E2 protege de la esteatosis y la resistencia a la insulina inducidos por la HFD en ausencia de SOCS2. Mercedes de Mirecki Garrido 116 Figure 3. Dimorfismo en la respuesta inflamatori en ratones SOCS2-/- con HFD. WT (columnas blancas) y SOCS2-/- (comlumnas negras) se alimentaron con Dieta control (CD) o alta en grasa (HFD) como se describe en material y metodos. Se analizaron los niveles circulantes de leptina (A), IL-6 (B), INFγ (C), RANTES (D), TNFα (E), y IL1β (F). Los resultados se expresan como media ± SEM (n=6). *P<0.05, **P<0.001, ***P<0.0001 Desde el punto de vista fisiológico nuestros datos sugieren que los ratones deficientes en SOCS2 pueden representar un novedoso modelo para el estudio del síndrome metabólico con características únicas que son relevantes para las enfermedades humanas. Los ratones SOCS2−/− no muestran un fenotipo metabólico claro. De esta forma, los mencionados ratones no muestran alteraciones en la sensibilidad a la insulina [118], al contrario que los ratones transgénicos que carecen de GH en los que se desarrolla una marcada resistencia a la insulina [59,77,242]. Sin embargo, cuando los ratones SOCS2−/− son alimentados con HFD desarrollan una gran resistencia a la insulina comparándolos con los WT (HFD). Los niveles de insulina plasmática en ayunas se ven aumentados en los ratones SOCS2−/− HFD, pero no se observan diferencias en los niveles pancreáticos de insulina ni en la morfología de los islotes, lo cual sugiere que el deterioro de la sensibilidad a la insulina no es causada por E2, GH, SOCS2, and Liver 123 Figure 5. Analisis immunohistocquimico en pancreas de GHR enratones de cepa salvaje (WT) y deficientes en SOCS2 (SOCS2-/-). En conclusión, este estudio permite identificar a SOCS2 como un importante regulador de la homeostasis de la insulina in vivo y sugiere que la inhibición de SOCS2 podría usarse como una diana terapéutica para retrasar el desarrollo de la diabetes [251]. Mercedes de Mirecki Garrido 124 5. CONCLUSIONES Los estudios que componen esta tesis doctoral añaden nuevos datos que enfatizan el impacto de la administración subcutánea de E2 sobre la fisiología del hígado y su interacción con GH. Nuetros resultados ponen de manifiesto el papel de E2 como un regulador crítico del metabolismo en hígado de los mamíferos y añaden más peso a la hipótesis de que E2 actúa como un importante regulador de las acciones de GH en el hígado. La relación E2-GH en el hígado es relevante debido a las diferentes funciones fisiológicas que estas hormonas ejercen en los mamíferos y al amplio uso de los estrógenos y los compuestos relacionados con los estrógenos en humanos. Nuestros hallazgos muestran que el E2 induce una feminización del hígado de las ratas hipotiroideas castradas adultas (TXOX). En primer lugar, se observó que la administración de E2 a ratas TXOX incrementó la síntesis y secreción de GH desde la pituitaria, emulando un "patrón femenino", y dando lugar al aumento de la regulación de los genes femeninos. En segundo lugar, descubrimos que el efecto del E2 sobre las acciones de la GH en hígado de las ratas TXOX se puede atribuir a la interacción del E2 con la vía de señalización GH-JAK2-STAT5 a través de la inducción de los reguladores negativos de esta vía de señalización, como son SOCS2, la CIS y FGF21 (Articulo I). Nuestros resultados muestran claramente que el E2, directa o indirectamente, puede modificar los efectos de la GH sobre la fisiología del hígado. Curiosamente, los efectos del E2 producen una alteración de la expresión de los genes hepáticos relacionados con el crecimiento corporal y el metabolismo de los lípidos. Estos efectos dan lugar a una reducción en la longitud corporal en el grupo de los WT tratado con E2, mientras que estos efectos no resultaron significativos en los ratones SOCS2−/−. Por otra parte, el E2 tiene la capacidad de controlar el equilibrio energético, la ingesta de alimentos y la distribución de grasa corporal. La reducción de los niveles de E2 se asocia con E2, GH, SOCS2, and Liver 125 cambios en el peso corporal y en la distribución de la grasa en los seres humanos, encontrándose hallazgos paralelos en los animales. En nuestros resultados los ratones WT tratados con E2, sufrieron una disminución de la ganancia de peso corporal, mientras que en los ratones SOCS2−/− no se observó ninguna diferencia. Diversos estudios realizados tanto en humanos como en roedores sugieren que una reducción en los niveles E2 o sus receptores pueden dar lugar a un a un síndrome parecido al metabólico. El E2 protege de la diabetes, la esteatosis y la adiposidad inducidas por la dieta alta en grasa (HFD). Los efectos de la HFD presentan un dimorfismo de sexual, en el que las hembras están más protegidas que los machos de los daños inducidos por este tipo de dieta. Los resultados presentados en esta Tesis sugieren que SOCS2, un inhibidor fisiológico de la señalización de citoquinas, podría desempeñar un papel fisiológico en la regulación de la esteatosis hepática en ratones alimentados con HFD. Además, SOCS2 media los efectos inhibidores del E2 sobre la vía de señalización de GH-JAK2-STAT5, que controla el metabolismo hepático de los lípidos, el tamaño corporal de los mamíferos y además es una importante reguladora del metabolismo de la glucosa. De hecho, la GH tiene acción diabetogénica, ya que induce resistencia a la insulina. Los ratones SOCS2−/− muestran un gigantismo dependiente de GH, pero no presentan alteraciones en la sensibilidad a la insulina. A diferencia de los ratones transgénicos en GH que son altamente resistentes a la insulina. Por lo tanto, los ratones SOCS2−/− proporcionan un novedoso modelo que permitirá entender la compleja relación entre la inflamación, las acciones de GH y la nutrición en el control de la glucosa hepática y la homeostasis de los lípidos. Los ratones SOCS2−/− presentan una mayor secreción hepática de TG y están protegidos de la esteatosis hepática inducida por HFD, estando las hembras más protegidas que los machos. Sin embargo, los ratones SOCS2−/− alimentados con HFD muestran una severa resistencia sistémica a la insulina, asociada a una Mercedes de Mirecki Garrido 126 hiperinsulinemia y a una menor sensibilidad a la insulina en el hígado. Los ratones SOCS2−/− alimentados con HFD presentan también un aumento de la expresión de citoquinas inflamatorias en el hígado y la grasa, reflejando un nuevo papel de SOCS2 como regulador negativo de la activación de los macrófagos en situaciones de estrés debido a la dieta. Por último, la eliminación experimental de SOCS2 permite identificar esta proteína como un importante regulador de la homeostasis de la insulina in vivo. Nuestros resultados demuestran que los ratones SOCS2−/− son más resistentes al desarrollo de diabetes mediante MLDSTZ. Por otra parte, observamos que la supresión de SOCS2 se relaciona con una mejor conservación de la sensibilidad a la insulina. Además, los ratones SOCS2−/− tienen una mayor masa de células beta y mayor tamaño de los islotes beta en el páncreas que los ratones WT. Los ratones SOCS2−/− son más sensibles a la GH (y probablemente a la PRL) ya que muestran mayores niveles pancreáticos de receptores GH and PRL. La GH estimula el crecimiento, la supervivencia y la producción de insulina de las células β del páncreas, a través de la vía de señalización GHR-JAK2-STAT5b. Por lo tanto, la eliminación de SOCS2 podría estar relacionada con la mejora de la sensibilidad a la insulina y con una menor resistencia a la insulina a través de una acción no regulada negativamente de la GH sobre el páncreas. Estos resultados sugieren que la inhibición SOCS2 podría ser utilizada como diana terapéutica para retrasar el desarrollo de la diabetes. E2, GH, SOCS2, and Liver 127 6. PERSPECTIVAS DE FUTURO Las conclusiones que se extraen de esta tesis permiten aumentar nuestro conocimiento sobre la relación que existe entre SOCS2, E2 y GH. Entender esta relación en los diferentes estados fisiológicos y patológicos podría ayudar a prevenir enfermedades y mejorar el manejo clínico de los pacientes con trastornos del crecimiento, metabólicos y del desarrollo. - Las consecuencias endocrinas y metabólicas de la exposición prolongada a los estrógenos o a compuestos relacionados con ellos, así como su influencia sobre las acciones de la GH son ampliamente desconocidas por parte de la población general. Por tanto, es muy importante incrementar nuestro conocimiento sobre SOCS2 y su regulación para poder comprender las relaciones entre E2 y GH. De esta forma, se requieren futuras investigaciones para poder definir con exactitud las alteraciones hormonales y ambientales que producen variaciones en la expresión de SOCS2. - futuros experimentos en ratones SOCS2−/− ratones tratados con E2 permitirán elucidar con exactitud el mecanismo molecular por el que los efectos del E2,sobre el metabolismo, el crecimiento y el dimorfismo sexual, se ven afectados por SOCS2 in vivo. - Es necesario seguir trabajando para aclarar el papel de la señalización de GH en los diferentes tejidos y tipos celulares en los ratones SOCS2−/−. - La futura utilización del modelo de ratones SOCS2−/− alimentados con HFD ayudará a perfilar la contribución de los diferentes mecanismos implicados en el desarrollo de la resistencia a la insulina, la sobrecarga lipídica hepática y el estrés inflamatorio, inducido por dietas ricas en grasa. - Por último, consideramos que deben realizarse más experimentos para poder comprender la compleja interacción entre SOCS2 y Diabetes Mellitus. Mercedes de Mirecki Garrido 128 REFERENCES 1. Baik M, Yu JH, Hennighausen L (2011) Growth hormone-STAT5 regulation of growth, hepatocellular carcinoma, and liver metabolism. Ann N Y Acad Sci 1229: 29-37. 2. List EO, Sackmann-Sala L, Berryman DE, Funk K, Kelder B, et al. (2011) Endocrine parameters and phenotypes of the growth hormone receptor gene disrupted (GHR-/-) mouse. Endocr Rev 32: 356-386. 3. LeRoith D, Yakar S (2007) Mechanisms of disease: metabolic effects of growth hormone and insulin-like growth factor 1. Nat Clin Pract Endocrinol Metab 3: 302-310. 4. Lichanska AM, Waters MJ (2008) How growth hormone controls growth, obesity and sexual dimorphism. Trends Genet 24: 41-47. 5. Kopchick JJ, Andry JM (2000) Growth hormone (GH), GH receptor, and signal transduction. Mol Genet Metab 71: 293-314. 6. Isaksson OG, Jansson JO, Gause IA (1982) Growth hormone stimulates longitudinal bone growth directly. Science 216: 1237-1239. 7. Jorgensen JO, Rubeck KZ, Nielsen TS, Clasen BF, Vendelboe M, et al. (2010) Effects of GH in human muscle and fat. Pediatr Nephrol 25: 705-709. 8. Vijayakumar A, Novosyadlyy R, Wu Y, Yakar S, LeRoith D (2010) Biological effects of growth hormone on carbohydrate and lipid metabolism. Growth Horm IGF Res 20: 1-7. 9. Nam SY, Lobie PE (2000) The mechanism of effect of growth hormone on preadipocyte and adipocyte function. Obes Rev 1: 73-86. 10. Madsen K, Friberg U, Roos P, Eden S, Isaksson O (1983) Growth hormone stimulates the proliferation of cultured chondrocytes from rabbit ear and rat rib growth cartilage. Nature 304: 545-547. 11. Florini JR, Ewton DZ, Coolican SA (1996) Growth hormone and the insulinlike growth factor system in myogenesis. Endocr Rev 17: 481-517. 12. Smaniotto S, Martins-Neto AA, Dardenne M, Savino W (2011) Growth hormone is a modulator of lymphocyte migration. Neuroimmunomodulation 18: 309-313. 13. Kelley KW, Arkins S, Minshall C, Liu Q, Dantzer R (1996) Growth hormone, growth factors and hematopoiesis. Horm Res 45: 38-45. 14. Yoshizato H, Fujikawa T, Soya H, Tanaka M, Nakashima K (1998) The growth hormone (GH) gene is expressed in the lateral hypothalamus: enhancement by GH-releasing hormone and repression by restraint stress. Endocrinology 139: 2545-2551. 15. Creyghton WM, van Dam PS, Koppeschaar HP (2004) The role of the somatotropic system in cognition and other cerebral functions. Semin Vasc Med 4: 167-172. E2, GH, SOCS2, and Liver 129 16. Parkinson C, Renehan AG, Ryder WD, O'Dwyer ST, Shalet SM, et al. (2002) Gender and age influence the relationship between serum GH and IGF-I in patients with acromegaly. Clin Endocrinol (Oxf) 57: 59-64. 17. Schalch DS, Reichlin S (1966) Plasma growth hormone concentration in the rat determined by radioimmunoassay: influence of sex, pregnancy, lactation, anesthesia, hypophysectomy and extrasellar pituitary transplants. Endocrinology 79: 275-280. 18. Pilecka I, Whatmore A, Hooft van Huijsduijnen R, Destenaves B, Clayton P (2007) Growth hormone signalling: sprouting links between pathways, human genetics and therapeutic options. Trends Endocrinol Metab 18: 12-18. 19. Butler AA, Le Roith D (2001) Control of growth by the somatropic axis: growth hormone and the insulin-like growth factors have related and independent roles. Annu Rev Physiol 63: 141-164. 20. Kaplan SA, Cohen P (2007) The somatomedin hypothesis 2007: 50 years later. J Clin Endocrinol Metab 92: 4529-4535. 21. Carro E, Senaris R, Considine RV, Casanueva FF, Dieguez C (1997) Regulation of in vivo growth hormone secretion by leptin. Endocrinology 138: 2203-2206. 22. Howard AD, Feighner SD, Cully DF, Arena JP, Liberator PA, et al. (1996) A receptor in pituitary and hypothalamus that functions in growth hormone release. Science 273: 974-977. 23. Holst B, Schwartz TW (2006) Ghrelin receptor mutations--too little height and too much hunger. J Clin Invest 116: 637-641. 24. Svensson J, Lonn L, Jansson JO, Murphy G, Wyss D, et al. (1998) Twomonth treatment of obese subjects with the oral growth hormone (GH) secretagogue MK-677 increases GH secretion, fat-free mass, and energy expenditure. J Clin Endocrinol Metab 83: 362-369. 25. Salmon WD, Jr., Daughaday WH (1957) A hormonally controlled serum factor which stimulates sulfate incorporation by cartilage in vitro. J Lab Clin Med 49: 825-836. 26. Casanueva FF (1992) Physiology of growth hormone secretion and action. Endocrinol Metab Clin North Am 21: 483-517. 27. Waters MJ, Shang CA, Behncken SN, Tam SP, Li H, et al. (1999) Growth hormone as a cytokine. Clin Exp Pharmacol Physiol 26: 760-764. 28. Fernandez-Perez L, Guerra B, Diaz-Chico JC, Flores-Morales A (2013) Estrogens regulate the hepatic effects of growth hormone, a hormonal interplay with multiple fates. Front Endocrinol (Lausanne) 4: 66. 29. Jansson JO, Eden S, Isaksson O (1985) Sexual dimorphism in the control of growth hormone secretion. Endocr Rev 6: 128-150. Mercedes de Mirecki Garrido 130 30. MacLeod JN, Pampori NA, Shapiro BH (1991) Sex differences in the ultradian pattern of plasma growth hormone concentrations in mice. J Endocrinol 131: 395-399. 31. Veldhuis JD, Anderson SM, Kok P, Iranmanesh A, Frystyk J, et al. (2004) Estradiol supplementation modulates growth hormone (GH) secretory-burst waveform and recombinant human insulin-like growth factor-I-enforced suppression of endogenously driven GH release in postmenopausal women. J Clin Endocrinol Metab 89: 1312-1318. 32. Isaksson OG, Eden S, Jansson JO (1985) Mode of action of pituitary growth hormone on target cells. Annu Rev Physiol 47: 483-499. 33. Mode A, Gustafsson JA (2006) Sex and the liver - a journey through five decades. Drug Metab Rev 38: 197-207. 34. Kerrigan JR, Rogol AD (1992) The impact of gonadal steroid hormone action on growth hormone secretion during childhood and adolescence. Endocr Rev 13: 281-298. 35. Waxman DJ, Ram PA, Park SH, Choi HK (1995) Intermittent plasma growth hormone triggers tyrosine phosphorylation and nuclear translocation of a liver-expressed, Stat 5-related DNA binding protein. Proposed role as an intracellular regulator of male-specific liver gene transcription. J Biol Chem 270: 13262-13270. 36. Choi HK, Waxman DJ (1999) Growth hormone, but not prolactin, maintains, low-level activation of STAT5a and STAT5b in female rat liver. Endocrinology 140: 5126-5135. 37. Choi HK, Waxman DJ (2000) Plasma growth hormone pulse activation of hepatic JAK-STAT5 signaling: developmental regulation and role in male-specific liver gene expression. Endocrinology 141: 3245-3255. 38. Tannenbaum GS, Choi HK, Gurd W, Waxman DJ (2001) Temporal relationship between the sexually dimorphic spontaneous GH secretory profiles and hepatic STAT5 activity. Endocrinology 142: 4599-4606. 39. Waxman DJ, Pampori NA, Ram PA, Agrawal AK, Shapiro BH (1991) Interpulse interval in circulating growth hormone patterns regulates sexually dimorphic expression of hepatic cytochrome P450. Proc Natl Acad Sci U S A 88: 6868-6872. 40. Flores-Morales A, Stahlberg N, Tollet-Egnell P, Lundeberg J, Malek RL, et al. (2001) Microarray analysis of the in vivo effects of hypophysectomy and growth hormone treatment on gene expression in the rat. Endocrinology 142: 3163-3176. 41. Stahlberg N, Merino R, Hernandez LH, Fernandez-Perez L, Sandelin A, et al. (2005) Exploring hepatic hormone actions using a compilation of gene expression profiles. BMC Physiol 5: 8. E2, GH, SOCS2, and Liver 131 42. Waxman DJ, Holloway MG (2009) Sex differences in the expression of hepatic drug metabolizing enzymes. Mol Pharmacol 76: 215-228. 43. Brooks AJ, Wooh JW, Tunny KA, Waters MJ (2008) Growth hormone receptor; mechanism of action. Int J Biochem Cell Biol 40: 1984-1989. 44. Brown RJ, Adams JJ, Pelekanos RA, Wan Y, McKinstry WJ, et al. (2005) Model for growth hormone receptor activation based on subunit rotation within a receptor dimer. Nat Struct Mol Biol 12: 814-821. 45. Rico-Bautista E, Negrin-Martinez C, Novoa-Mogollon J, Fernandez-Perez L, Flores-Morales A (2004) Downregulation of the growth hormoneinduced Janus kinase 2/signal transducer and activator of transcription 5 signaling pathway requires an intact actin cytoskeleton. Exp Cell Res 294: 269-280. 46. Tollet-Egnell P, Parini P, Stahlberg N, Lonnstedt I, Lee NH, et al. (2004) Growth hormone-mediated alteration of fuel metabolism in the aged rat as determined from transcript profiles. Physiol Genomics 16: 261267. 47. Waxman DJ, O'Connor C (2006) Growth hormone regulation of sexdependent liver gene expression. Mol Endocrinol 20: 2613-2629. 48. Vidal OM, Merino R, Rico-Bautista E, Fernandez-Perez L, Chia DJ, et al. (2007) In vivo transcript profiling and phylogenetic analysis identifies suppressor of cytokine signaling 2 as a direct signal transducer and activator of transcription 5b target in liver. Mol Endocrinol 21: 293311. 49. Rowland JE, Lichanska AM, Kerr LM, White M, d'Aniello EM, et al. (2005) In vivo analysis of growth hormone receptor signaling domains and their associated transcripts. Mol Cell Biol 25: 66-77. 50. Gebert CA, Park SH, Waxman DJ (1999) Termination of growth hormone pulse-induced STAT5b signaling. Mol Endocrinol 13: 38-56. 51. Gebert CA, Park SH, Waxman DJ (1999) Down-regulation of liver JAK2STAT5b signaling by the female plasma pattern of continuous growth hormone stimulation. Mol Endocrinol 13: 213-227. 52. Udy GB, Towers RP, Snell RG, Wilkins RJ, Park SH, et al. (1997) Requirement of STAT5b for sexual dimorphism of body growth rates and liver gene expression. Proc Natl Acad Sci U S A 94: 7239-7244. 53. Davey HW, Park SH, Grattan DR, McLachlan MJ, Waxman DJ (1999) STAT5b-deficient mice are growth hormone pulse-resistant. Role of STAT5b in sex-specific liver p450 expression. J Biol Chem 274: 3533135336. 54. Kofoed EM, Hwa V, Little B, Woods KA, Buckway CK, et al. (2003) Growth hormone insensitivity associated with a STAT5b mutation. N Engl J Med 349: 1139-1147. Mercedes de Mirecki Garrido 132 55. Flores-Morales A, Greenhalgh CJ, Norstedt G, Rico-Bautista E (2006) Negative regulation of growth hormone receptor signaling. Mol Endocrinol 20: 241-253. 56. Fernandez L, Flores-Morales A, Lahuna O, Sliva D, Norstedt G, et al. (1998) Desensitization of the growth hormone-induced Janus kinase 2 (Jak 2)/signal transducer and activator of transcription 5 (Stat5)-signaling pathway requires protein synthesis and phospholipase C. Endocrinology 139: 1815-1824. 57. Leung KC, Waters MJ, Markus I, Baumbach WR, Ho KK (1997) Insulin and insulin-like growth factor-I acutely inhibit surface translocation of growth hormone receptors in osteoblasts: a novel mechanism of growth hormone receptor regulation. Proc Natl Acad Sci U S A 94: 11381-11386. 58. Stofega MR, Wang H, Ullrich A, Carter-Su C (1998) Growth hormone regulation of SIRP and SHP-2 tyrosyl phosphorylation and association. J Biol Chem 273: 7112-7117. 59. Rico-Bautista E, Flores-Morales A, Fernandez-Perez L (2006) Suppressor of cytokine signaling (SOCS) 2, a protein with multiple functions. Cytokine Growth Factor Rev 17: 431-439. 60. Vesterlund M, Zadjali F, Persson T, Nielsen ML, Kessler BM, et al. (2011) The SOCS2 ubiquitin ligase complex regulates growth hormone receptor levels. PLoS One 6: e25358. 61. Santana-Farre R, Flores-Morales A, Fernández-Pérez L (2008) Growth Hormone, Thyroid Hormones and Estradiol interplay in vivo to regulate gene expression of Suppressors of Cytokine Signalling (SOCS). In: A. Godoy-Matos JW, editor. International Proceedings 13th International Congress of Endocrinology. Rio de Janeiro (Brazil): Medimond, S.r.l. pp. 8-12. 62. Leung KC, Johannsson G, Leong GM, Ho KK (2004) Estrogen regulation of growth hormone action. Endocr Rev 25: 693-721. 63. Zhou Y, Xu BC, Maheshwari HG, He L, Reed M, et al. (1997) A mammalian model for Laron syndrome produced by targeted disruption of the mouse growth hormone receptor/binding protein gene (the Laron mouse). Proc Natl Acad Sci U S A 94: 13215-13220. 64. Klapper DG, Svoboda ME, Van Wyk JJ (1983) Sequence analysis of somatomedin-C: confirmation of identity with insulin-like growth factor I. Endocrinology 112: 2215-2217. 65. Powell-Braxton L, Hollingshead P, Warburton C, Dowd M, Pitts-Meek S, et al. (1993) IGF-I is required for normal embryonic growth in mice. Genes Dev 7: 2609-2617.