Il est désormais clairement établi que la vélocité de croissance dans les premiers mois de la vie influence le risque d’apparition de maladies métaboliques, telles que l’obésité à l’âge adulte. L’un des mécanismes sous-jacents à cette programmation néonatale pourrait mettre en jeu des perturbations de l’environnement hormonal périnatal pendant des périodes critiques du développement. Cette hypothèse repose sur des travaux réalisés au laboratoire montrant que pendant le développement postnatal, des hormones telles que la ghréline et la leptine peuvent influencer le développement des circuits neuronaux impliqués dans la régulation de la prise alimentaire à l’âge adulte. Dans ce contexte, l’objectif général de ce travail de thèse a été de définir les médiateurs hormonaux par lesquels l’environnement nutritionnel périnatal peut influencer le développement de maladies métaboliques à l’âge adulte. Une attention particulière a été portée à la contribution de la ghréline et de la leptine.Pour induire une surnutrition postnatale, nous avons utilisé un modèle murin de surnutrition postnatale induit par la réduction de la taille de portées. Au troisième jour de vie post-natale, les portées ont été réduites à 3 petits (SL) pour induire une suralimentation et les portées contrôles ont conservé 7 petits. Comparées aux souris NL, les souris SL présentent un gain de poids rapide pendant la lactation, et présentent un surpoids à l'âge adulte même sous régime de nourriture standard. A l’âge adulte, les souris SL ont une adiposité et une glycémie à jeun plus élevée. Sous régime « obésogène » les souris SL présentent également une prise de poids et de graisse plus importante que les souris NL. La première partie de ces travaux a été dédiée à l’étude de la nutrition périnatale sur le développement du système ghrélinergique. Les souris SL présentent des taux réduits de ghréline total et active pedant la troisième semaine de vie post-natale. Ces diminutions des taux de ghréline sont associées à une diminution de l’expression de ghreline dans l'estomac. La normalisation de la ghrélinémie ne parvient pas à restaurer un phénotype métabolique normale chez les souris SL ce qui suggère que les souriceaux SL présenteraient une résistance a la ghréline. En accord avec cette hypothèse, les souriceaux SL présente une atténuation de la réponse centrale suite à l’injection périphérique de ghréline. Les mécanismes sous-jacents à cette résistance à la ghréline semblent mettre en jeu un défaut de transport de l’hormone via les tanycytes de l’éminence médiane. La seconde partie de mon travail de thèse a consisté à étudier l’importance de la leptine dans la programmation nutritionnelle. Etant connu que les souris SL présent des taux anormalement élevés de leptine pendant la deuxième semaine de vie postnatale, nous avons émis l’hypothèse que le blocage partiel de la leptine chez les souris SL pourrait avoir des effets bénéfiques sur le métabolisme de ces souris. Les animaux injectés néonatalement avec l’antagoniste de la leptine ne présentent pas de différences de poids par rapport aux animaux contrôles. En revanche, l’injection de l’antagoniste de la leptine chez les souriceaux SL induit une amélioration de leur masse grasse et une normalisation de leur glycémie. Cette amélioration du phénotype métabolique des souris SL est associée à un rétablissement de la sensibilité centrale à l’hormone leptine.Ainsi, les travaux réalisés au cours de ce doctorat confortent l’importance de l’environnement hormonal périnatal, en particulier de la ghéline et de la leptine, dans la programmation nutritionnelle. / Rapid weight gain after birth is associated with increased risk of metabolic disease in adulthood. This metabolic malprogramming may be mediated by endocrine disturbances during critical periods of development. Several hormones that act on hypothalamic feeding circuits in adulthood to regulate energy homeostasis, such as leptin and ghrelin, are also involved in the postnatal development of these circuits. Previous studies have shown that leptin promotes the growth of axons from the arcuate nucleus of the hypothalamus(ARH), whereas ghrelin inhibits arcuate axon growth. The aim of this thesis was to study whether accelerated growth during early postnatal life is associated with alterations in the ghrelin and leptin systems during postnatal development. Our general objective was also to examine the relative contribution of ghrelin and leptin in the metabolic malprogramming associated with neonatal overnutrition. To manipulate nutrient intake specifically during postnatal (pre-weaning) life, we manipulated litter size on postnatal day 3 by randomly distributing pups among mothers such that small litters (overfed) had 3 pups and normal litters (normal fed) had 7 pups. SL pups gained substantially more weight than NL pups during lactation, and remained overweight throughout adulthood on a standard chow diet. Adult SL mice displayed increased adiposity and and elevated fasting glucose levels. Adult SL mice also gained more weight and fat when exposed to an obesogenic diet. During postnatal development, SL pups displayed reduced total and active circulating ghrelin levels compared to NL pups, particularly during the third week postnatal week. This was associated with decreased stomach expression of ghrelin in the third postnatal week. Normalization of hypoghrelinemia in overnourished pups is relatively ineffective in ameliorating metabolic outcomes, suggesting that SL pups may present ghrelin resistance. Consistent with this idea, pups raised in small litters displayed an impaired central response to peripheral ghrelin. The mechanisms underlying this ghrelin resistance include diminished ghrelin transport into the brain.In addition to marked changes in ghrelin levels, neonatally overnourished pups also display a marked increase in circulating leptin levels during early postnatal life. To investigate whether these abnormally high levels in circulating leptin levels might be causally involved in the central leptin resistance and metabolic programming, we used a competitive leptin antagonist to partially block leptin action in SL and NL mice between P6 and 16, i.e. when when endogenous leptin levels are high in SL mice. Neonatal injection of the leptin antagonist had no effect on postnatal growth or adult body weight. However, SL pups treated with the leptin antagonist displayed a marked reduction in adult fat mass and normalized glucose tolerance and insulin sensitivity. Previous studies reported that SL exhibit early central leptin resistance. Remarkably, pretreatment with leptin antagonist significantly increased leptin-induced pSTAT3 levels in the ARH of SL pups, suggesting that neonatal leptin antagonism may normalize metabolic programming in neonatally overnourished mice by enhancing central leptin sensitivity. Together our data show that early postnatal nutrition influence the pattern of secretion of metabolic hormones and their ability to act on hypothalamic regions involved in appetite regulation. These hormonal alterations may contribute to the metabolic defects observed in subject exposed to overnutrition during early life.

A ghrelina, hormônio secretado principalmente por células gástricas, liga-se ao seu receptor, o receptor de secretagogo de GH (GHSR - Growth hormone secretagogue receptor), localizado no hipotálamo e na hipófise, estimulando a síntese e secreção do GH. Recentemente foram identificadas mutações no gene GHSR em crianças com baixa estatura idiopática (BEI) e com deficiência isolada de GH (DGH). No presente estudo investigamos a presença de mutações no gene GHSR em crianças com DGH isolada de causa não identificada e crianças com BEI, incluindo um subgrupo de crianças com atraso constitucional de crescimento e desenvolvimento (ACCD). Foram selecionados 14 pacientes com deficiência isolada de GH sem alterações anatômicas da região hipotálamo-hipofisária e 96 pacientes com BEI, destes 31 (32%) apresentavam ACCD. Também foram estudados 150 controles adultos e 197 crianças controle com crescimento e puberdade normais. A região codificadora do GHSR foi amplificada utilizando-se oligonucleotídeos iniciadores específicos, seguida de purificação enzimática e seqüenciamento automático. Encontramos 6 variantes alélicas em heterozigose no GHSR: nenhuma delas presente nos controles estudados, e quatro destas variantes estão localizadas em regiões conservadas do gene. Uma variante foi encontrada em uma paciente do grupo DGH (p.Val249Leu) e as outras cinco (c.-6 G>C, p.Ser84Ile, p.Val182Ala, p.Ala169Thr e p.Ala358Thr) foram encontradas em pacientes do subgrupo ACCD do grupo BEI. As variantes missense foram submetidas a estudo funcional que evidenciou que as mutações p.Ser84Ile e p.Val182Ala possuem diminuição na atividade basal associadas à diminuição da expressão do receptor na superfície celular. Adicionalmente, a mutação p.Ser84Ile também apresenta redução na atividade do GHSR induzida pelo ligante. A variante p.Val249Leu foi encontrada em uma paciente do sexo feminino com diagnóstico de DGH isolado. A falta de segregação familiar associada à ausência de déficit funcional da variante nos estudos in vitro sugere que, neste caso, a variante p.Val249Leu não é a causa do fenótipo de DGH nesta família e trata-se de uma variante alélica rara. As 5 variantes alélicas no GHSR (c.-6 G>C, p.Ser84Ile, p.Val182Ala, p.Ala169Thr e p.Ala358Thr) encontradas nos pacientes com BEI foram identificadas apenas naqueles com puberdade atrasada, ou seja, pertencentes ao subgrupo ACCD (3 do sexo masculino e 2 do sexo feminino). A freqüência de variantes neste grupo de pacientes foi de 16%, significativamente maior que nos outros grupos, e a ausência de variantes gênicas novas no grupo de crianças obesas com altura normal e mesmo no grupo de crianças com BEI sem ACCD sugere que nosso achado não foi casual e que as alterações descritas podem estar associadas ao fenótipo de ACCD. Os estudos in vitro mostraram prejuízos funcionais em 2 destas variantes (p.Ser84Ile e p.Val182Ala) porém, devido à limitação dos estudos funcionais (celulas heterólogas) não podemos afastar que as demais não tenham algum impacto funcional in vivo. Em conclusão, nossos resultados sugerem um envolvimento dos defeitos no GHSR na etiologia do atraso constitucional do crescimento e desenvolvimento em uma parcela de pacientes com esta condição / Ghrelin, hormone secreted by gastric cells, stimulates growth hormone secretion by acting on its receptor GHSR, located in the hypothalamus and pituitary. Recently, mutations in the GHSR gene were described in patients with growth hormone deficiency (GHD) and idiopathic short stature (ISS). In the present study we analyzed the GHSR gene in patients with isolated GHD and patients with ISS, including a subgroup of patients with constitutional delay of growth and puberty (CDGP). We studied 14 GHD patients with normal pituitary magnetic resonance imaging and 96 patients with ISS, 31 of them with CDGP. We also studied 150 adults and in 197 children with normal stature. The entire coding region as well as the exon-intron boundaries of GHSR were PCR amplified in all patients and control group and PCR products were bidirectionally sequenced. Six different heterozygous variants in GHSR were identified: none of them were found in the control group and four of these amino acid substitutions occurred at a conserved position within the GHSR. One variant (p.Val249Leu) was found in a GHD patient and the other five (c.-6 G>C, p.Ser84Ile, p.Val182Ala, p.Ala169Thr e p.Ala358Thr) were found in patients with CDGP. The missense variants were submitted to functional studies. Two of these variants (p.Ser84Ile and p.Val182Ala) result in a decrease in basal activity that was in part explained by a reduction in cell surface expression. The p.Ser84Ile mutation was also associated with a defect in ghrelin potency. The p.Val249Leu variant, found in a female patient with isolated GHD, did not segregate with the phenotype in the family and had no functional impairment in vitro. This suggests that p.Val249Leu is not the cause of the GHD in the family and may be a rare allelic variant. The other variants (c.-6 G>C, p.Ser84Ile, p.Val182Ala, p.Ala169Thr e p.Ala358Thr) were identified only in patients with CDGP (3 male and 2 female). The frequency of allelic variants observed in this group (16%) was higher than expected by chance in contrast with ISS and GHD children, and the absence of other GHSR mutations in the large group of control children suggests that the association between GHSR mutations and CDGP phenotype is unlikely to be fortuitous. Functional studies revealed that two of the identified missense variants (p.Ser84Ile and p.Val182Ala) are functionally significant. These functional studies were performed in heterologous cell expression systems; therefore it is not possible to completely rule out that the other identified variants might cause some unrevealed impairment on GHSR function or expression in vivo. In conclusion, our data raise the possibility that abnormalities in ghrelin receptor function may be implicated in the ethiology of CDGP in some patients

A ghrelina, hormônio secretado principalmente por células gástricas, liga-se ao seu receptor, o receptor de secretagogo de GH (GHSR - Growth hormone secretagogue receptor), localizado no hipotálamo e na hipófise, estimulando a síntese e secreção do GH. Recentemente foram identificadas mutações no gene GHSR em crianças com baixa estatura idiopática (BEI) e com deficiência isolada de GH (DGH). No presente estudo investigamos a presença de mutações no gene GHSR em crianças com DGH isolada de causa não identificada e crianças com BEI, incluindo um subgrupo de crianças com atraso constitucional de crescimento e desenvolvimento (ACCD). Foram selecionados 14 pacientes com deficiência isolada de GH sem alterações anatômicas da região hipotálamo-hipofisária e 96 pacientes com BEI, destes 31 (32%) apresentavam ACCD. Também foram estudados 150 controles adultos e 197 crianças controle com crescimento e puberdade normais. A região codificadora do GHSR foi amplificada utilizando-se oligonucleotídeos iniciadores específicos, seguida de purificação enzimática e seqüenciamento automático. Encontramos 6 variantes alélicas em heterozigose no GHSR: nenhuma delas presente nos controles estudados, e quatro destas variantes estão localizadas em regiões conservadas do gene. Uma variante foi encontrada em uma paciente do grupo DGH (p.Val249Leu) e as outras cinco (c.-6 G>C, p.Ser84Ile, p.Val182Ala, p.Ala169Thr e p.Ala358Thr) foram encontradas em pacientes do subgrupo ACCD do grupo BEI. As variantes missense foram submetidas a estudo funcional que evidenciou que as mutações p.Ser84Ile e p.Val182Ala possuem diminuição na atividade basal associadas à diminuição da expressão do receptor na superfície celular. Adicionalmente, a mutação p.Ser84Ile também apresenta redução na atividade do GHSR induzida pelo ligante. A variante p.Val249Leu foi encontrada em uma paciente do sexo feminino com diagnóstico de DGH isolado. A falta de segregação familiar associada à ausência de déficit funcional da variante nos estudos in vitro sugere que, neste caso, a variante p.Val249Leu não é a causa do fenótipo de DGH nesta família e trata-se de uma variante alélica rara. As 5 variantes alélicas no GHSR (c.-6 G>C, p.Ser84Ile, p.Val182Ala, p.Ala169Thr e p.Ala358Thr) encontradas nos pacientes com BEI foram identificadas apenas naqueles com puberdade atrasada, ou seja, pertencentes ao subgrupo ACCD (3 do sexo masculino e 2 do sexo feminino). A freqüência de variantes neste grupo de pacientes foi de 16%, significativamente maior que nos outros grupos, e a ausência de variantes gênicas novas no grupo de crianças obesas com altura normal e mesmo no grupo de crianças com BEI sem ACCD sugere que nosso achado não foi casual e que as alterações descritas podem estar associadas ao fenótipo de ACCD. Os estudos in vitro mostraram prejuízos funcionais em 2 destas variantes (p.Ser84Ile e p.Val182Ala) porém, devido à limitação dos estudos funcionais (celulas heterólogas) não podemos afastar que as demais não tenham algum impacto funcional in vivo. Em conclusão, nossos resultados sugerem um envolvimento dos defeitos no GHSR na etiologia do atraso constitucional do crescimento e desenvolvimento em uma parcela de pacientes com esta condição / Ghrelin, hormone secreted by gastric cells, stimulates growth hormone secretion by acting on its receptor GHSR, located in the hypothalamus and pituitary. Recently, mutations in the GHSR gene were described in patients with growth hormone deficiency (GHD) and idiopathic short stature (ISS). In the present study we analyzed the GHSR gene in patients with isolated GHD and patients with ISS, including a subgroup of patients with constitutional delay of growth and puberty (CDGP). We studied 14 GHD patients with normal pituitary magnetic resonance imaging and 96 patients with ISS, 31 of them with CDGP. We also studied 150 adults and in 197 children with normal stature. The entire coding region as well as the exon-intron boundaries of GHSR were PCR amplified in all patients and control group and PCR products were bidirectionally sequenced. Six different heterozygous variants in GHSR were identified: none of them were found in the control group and four of these amino acid substitutions occurred at a conserved position within the GHSR. One variant (p.Val249Leu) was found in a GHD patient and the other five (c.-6 G>C, p.Ser84Ile, p.Val182Ala, p.Ala169Thr e p.Ala358Thr) were found in patients with CDGP. The missense variants were submitted to functional studies. Two of these variants (p.Ser84Ile and p.Val182Ala) result in a decrease in basal activity that was in part explained by a reduction in cell surface expression. The p.Ser84Ile mutation was also associated with a defect in ghrelin potency. The p.Val249Leu variant, found in a female patient with isolated GHD, did not segregate with the phenotype in the family and had no functional impairment in vitro. This suggests that p.Val249Leu is not the cause of the GHD in the family and may be a rare allelic variant. The other variants (c.-6 G>C, p.Ser84Ile, p.Val182Ala, p.Ala169Thr e p.Ala358Thr) were identified only in patients with CDGP (3 male and 2 female). The frequency of allelic variants observed in this group (16%) was higher than expected by chance in contrast with ISS and GHD children, and the absence of other GHSR mutations in the large group of control children suggests that the association between GHSR mutations and CDGP phenotype is unlikely to be fortuitous. Functional studies revealed that two of the identified missense variants (p.Ser84Ile and p.Val182Ala) are functionally significant. These functional studies were performed in heterologous cell expression systems; therefore it is not possible to completely rule out that the other identified variants might cause some unrevealed impairment on GHSR function or expression in vivo. In conclusion, our data raise the possibility that abnormalities in ghrelin receptor function may be implicated in the ethiology of CDGP in some patients