O procedimento da Organização Mundial da Saúde (OMS) tem sido pouco estudado como instrumento para a detecção das alterações da tolerância à glicose na gravidez. Esta tese tem por objetivos descrever : as prevalências do diabete gestacional gestacional glicemia de (DMG) e da t olerância diminuída à glicose (TDGG); o desempenho da glicemia de jejum e da 1 hora no rastreamento dessas alterações. A amostra consiste de gestantes arroladas consecutivamente em serviços de pré-natal do Sistema Único de Saúde nas cidades de Porto Alegr e, São Paulo, Rio de Janeiro, Fortaleza, Salvador e Manaus. As 5010 gestantes 90% das gestantes arroladas - que realizaram um teste oral de tolerância à glicose (TTG-75g) padronizado de acordo com as orientações da OMS entre 24 e 28 semanas de gravidez constituem a amostra estudada. A prevalência do DMG é bastante baixa, 0,5); a da TDGG é de 7,2% (IC95%: Estratificando- se esses dados por faixa O, 3% (IC95% : O, 2 6,6 8,0). etária e por categorias de obesidade, observa-se que as prevalências são maiores nas gestantes de mais idade e nas mais obesas. Observa-se que a intolerância à glicose gestacional é representada fundamentalmente pela TDGG, sendo os casos de DMG raros . Se isso decorre da maior detecção do diabete prégestacional, reduzindo a prevalência do DMG e / ou de fatores que levam a um aumento da TDGG precisa ser melhor esclarecido. o emprego da glicemia como teste de rastreamento das alterações da tolerância à glicose gestacional foi estudado por meio de curvas ROC (receiver operator characteristic), primeiro definindo-se o ponto que otimiza igualmente sensibilidade e especificidade, e, a partir deste, priorizando-se o ponto de máxima sensibilidade sem perda expressiva da especificidade. O percentual de positividade de cada ponto da glicemia, que expressa o percentual de gestantes que devem realizar o teste diagnóstico, auxiliou na definição dos pontos que otimizam a sensibi lidade. Para a detecção do DMG, a glicemia de jejum que otimiza igualmente sensibilidade - 87,5% - e especificidade - 77,9% - é 89mg/dl, com percentual de positividade de 22,3% . Otimizando a sensibilidade - 94% - a glicemia é 85mg/dl, com especificidade de 65% e percentual de positividade de 35%. Para a detecção da TDGG, a glicemia de jejum que otimiza igualmente sensibilidade - 68,3% - e especificidade - 68,2% - corresponde ao valor de 85mg/dl (percentual de positividade de 34,5%) . Otimizando-se a sensibilidade - 81,3%, alcança-se um valor de 81mg/dl, que apresenta especificidade de 53,6% e percentual de positividade de 50% . A glicemia de jejum é bastante adequada para o rastreamento do DMG (89mg/dl), com desempenho apenas razoável na TDGG (85mg/dl) . Para o rastreamento do DMG, a glicemia de 1 hora que otimiza igualmente sensibi lidade - 93,8% - e especificidade - 97,2%- é 182mg/dl (percentual de positividade de 3 , 1%) . Para o rastreamento da TDGG, o ponto correspondente é o de 140mg/dl, com sensibilidade de 84 , 3%, especificidade de 80,2% e percentual de positividade de 24, 2%. Esse valor, aplicado ao rastreamento do DMG, alcança sensibilidade de 100% e especificidade de 75,6% . A glicemia de 1 hora tem um desempenho excelente no rastreamento das alterações da tolerância à glicose na gravi dez . A possível vantagem do emprego do ponto de corte de 182mg/dl para o diagnóstico do diabete gestacional precisa ser validada com desfechos obstétricos e neonatais da gravidez . Conclui-se que o procedimento da OMS pode ser aplicado na detecção das alterações da tolerância à glicose na gravidez, especialmente na detecção do DMG . A importância da TDGG aguarda validação por desfechos específicos da gravidez. / The World Health Organization's procedures for the detection of gestational glucose intolerance have been fairly studied in pregnancy . The aims of this work are to estimate the prevalences of gestational diabetes (GDM) and of gestational impaired glucose tolerance (GIGT) and to evaluate fasting and 1 hour plasma glucose as screening tools for gestational glucose intolerance. Pregnant women were consecutively enrolled at prenatal ambulatories of the Sistema Único de Saúde, in six cities in Brazil : Porto Alegre, São Paulo, Rio de Janeir o, Fortaleza , Salvador and Manaus . The 7 5g oral glucose tolerance test ( 7 5g-OGTT) was th th . performed between the 24 and the 28 weeks of gestat1on and full tests were available for 501 O women ( 90% o f the sample). The prevalence of GDM is very low, 0,3% 0,5) while that of GIGT is 7,2% (95%CI : (95%CI: O, 2 - 6,6 8,0). Prevalences were higher in older and heavier pregnant women. These data suggest that GDM is very rare in pregnancy and probably represents undiagnosed pre-gestational cases of noninsulin- dependent diabetes. Gestational glucose intolerance is mainly represented by cases of GIGT . The screening potentials of both fasting and 1 hour plasma glucose were evaluated by ROC (receiver operator characteristic) curves. Cut-off points that equally maximize sensitivity (Se) and specificity (Sp) were initially defined. Cutpoints that further maximize Se are described and the percent of positive tests - the percent of women requiring a full diagnostic test - is calculated in order to improve the screening potential of each cutpoint. For the screening of GDM, the fasting plasma glucose that equally maximizes Se - 87 , 5% - and Sp - 77,9% - is 89mg/dl (percent of positives: 22,3%). Maximizing sensitivity - 94% - leads to a cutpoint of 85mg/dl (Sp : 65% ; percent of positives : 35% ). For the detection of GIGT , the cutpoint that equally maximizes Se - 68,3% and Sp - 68 , 2% - is 85mg/dl (percent of positives: 34 , 5%) . A cutpoint of 81mg/dl maximizes Se- 81,3%- without undue loss of Sp 53 , 6% , but 50% of the sample require the diagnostic test with such cutpoint . Fasting plasma glucose is a good screening tes t for GDM and just fair for GIGT . The 1 hour plasma glucose that equally maximizes Se - 93,8% - and Sp- 97,2%- in the detection of GDM is 182mg/dl (percent of positives : 3,1%) . For the detection of GIGT the corresponding cutpoint is 140mg/dl: Se- 84 , 3% ; Sp- 80, 2%; percent o f posi ti ves 24, 2% . This cutpoint detects GDM with Se 100% and Sp 75, 6% . The 1 hour plasma glucose is an excellent screening test for both GDM and GIGT . The potential diagnostic value of the 1 hour cutpoint of 182mg/dl requires validation with obstet ric and neonatal outcomes. In conclusion, the WHO' s procedures can be used in pregnancy to evaluate gestational glucose intolerance. The importance of GIGT is yet to be validated with specific p r egnancy outcomes.

Osteoblasts are bone-forming cells and therefore they are responsible of the synthesis of type I collagen, the main component of bone matrix. However, there is an apparent disconnect between the regulation of osteoblast differentiation and bone formation since the synthesis of Type I collagen precedes the expression of Runx2, the earliest determinant of osteoblast differentiation. Recently, genetic experiments in the mouse have revealed the existence of an unexpected cross-regulation between bone and other organs. In particular this body of work has highlighted the importance of osteoblasts as endocrine cells to regulate whole-body glucose homeostasis by secretion of a hormone, osteocalcin. However, the fundamental question of why bone regulates glucose homeostasis remained to be answered. Therefore, in my thesis, considering that bone is a metabolically demanding organ that constantly renews itself, I hypothesized that characterizing the connection between the need of glucose as a main nutrient in osteoblasts and bone development will provide a key to deeper understanding of why bone regulates glucose homeostasis. My work shows here that glucose uptake through GLUT1 in osteoblasts is needed for osteoblast differentiation by suppressing the AMPK-dependent activation by phosphorylation at S148 of Smurf1 that targets Runx2 for degradation. I also uncovered the mechanism of action of Smurf1 in this setting. In a distinct but synergetic way, glucose uptake promotes bone formation by inhibiting a distinct function of AMPK. In turn, Runx2 favors Glut1 expression, and this feedforward regulation between Runx2 and Glut1 determines the onset of osteoblast differentiation during development and the extent of bone formation throughout life. Furthermore, I also identified that Smurf1 not only regulates osteoblast differentiation by targeting Runx2 for degradation but also contributes to whole-body glucose homeostasis by regulating the activation of osteocalcin by targeting the insulin receptor for degradation in vivo. These results identify Smurf1 as a determinant of osteoblast differentiation during development, of bone formation and glucose homeostasis post-natally. Most importantly, we show that these Smurf1 functions required AMPK-phosphorylation site S148 in vivo. Altogether, these results revealed the absolute necessity of glucose as a regulator of Runx2 accumulation during osteoblast differentiation and bone formation in vivo and highlight the fundamental importance of the intricate cross-talk between bone and whole-body glucose metabolism.

Wong Hei Yi. / Thesis (M.Phil.)--Chinese University of Hong Kong, 2004. / Includes bibliographical references (leaves 157-175). / Abstracts in English and Chinese. / Acknowledgements --- p.i / Abstract --- p.ii / Abstract 摘要 --- p.iv / List of Figures --- p.vi / List of Tables --- p.ix / List of Abbreviations --- p.x / Table of Contents --- p.xiii / Chapter Chapter 1: --- Introduction --- p.1 / Chapter 1.1 --- Importance of Glucose in Biological System --- p.1 / Chapter 1.2 --- Glucose Transporter Families --- p.2 / Chapter 1.2.1 --- Na+-Dependent Glucose Transporters --- p.2 / Chapter 1.2.2 --- Facilitative Glucose Transporters --- p.3 / Chapter 1.3 --- Glucose Transporter Type1 --- p.7 / Chapter 1.3.1 --- Primary Structure --- p.7 / Chapter 1.3.2 --- Secondary Structure --- p.8 / Chapter 1.3.3 --- Membrane Topology --- p.8 / Chapter 1.3.4 --- Tertiary Structure --- p.9 / Chapter 1.3.5 --- Kinetics Properties --- p.11 / Chapter 1.3.6 --- Affinity Reagents --- p.12 / Chapter 1.3.7 --- Tissue Distribution --- p.13 / Chapter 1.3.8 --- Multifunctional Property --- p.14 / Chapter 1.3.9 --- Characterization of GLUT1 Gene --- p.14 / Chapter 1.3.10 --- Regulation of GLUT1 Expression --- p.15 / Chapter 1.4 --- Glucose Transporter Type 1 and the Brain --- p.17 / Chapter 1.5 --- Glucose Transporter Type 1 Deficiency Syndrome --- p.20 / Chapter 1.5.1 --- Background of GlutlDS --- p.20 / Chapter 1.5.2 --- Clinical Features of GlutlDS --- p.23 / Chapter 1.5.3 --- Genotype-Phenotype Correlations --- p.24 / Chapter 1.5.4 --- Diagnosis --- p.26 / Chapter 1.5.4.1 --- Erythrocyte Glucose Transporter Activity --- p.26 / Chapter 1.5.4.2 --- Molecular Genetic Testing of GLUT1 Gene --- p.27 / Chapter 1.5.4.3 --- Glucose Concentration --- p.27 / Chapter 1.5.5 --- Management --- p.28 / Chapter 1.5.5.1 --- Ketogenic Diet --- p.28 / Chapter 1.5.5.2 --- Medication --- p.29 / Chapter 1.5.5.2.1 --- Glutl Activator --- p.29 / Chapter 1.5.5.2.2 --- Glutl Inhibitor --- p.29 / Chapter 1.6 --- Hypothesis and Objectives --- p.31 / Chapter Chapter 2: --- Identification of the First Two Asian GlutlDS Cases --- p.33 / Chapter 2.1 --- Materials --- p.34 / Chapter 2.1.1 --- Clinical History of Suspected GlutlDS Patients --- p.34 / Chapter 2.1.2 --- Blood Samples --- p.35 / Chapter 2.1.3 --- Reagents for Zero-trans Influx of 3-OMG Uptake in Erythrocytes --- p.35 / Chapter 2.1.4 --- Reagents for Zero-trans Efflux of 3-OMG Uptake in Erythrocytes --- p.37 / Chapter 2.1.5 --- Reagents for Glutl Gene Analysis --- p.37 / Chapter 2.1.6 --- Reagents and Buffers for Reverse Transcription --- p.38 / Chapter 2.1.7 --- Reagents and Buffers for Agarose Gel Electrophoresis --- p.39 / Chapter 2.1.8 --- Reagents for Erythrocytes Membrane Preparation and Detection --- p.41 / Chapter 2.2 --- Methods --- p.46 / Chapter 2.2.1 --- Zero-trans Influx of 3-OMG Uptake in Erythrocytes --- p.46 / Chapter 2.2.2 --- Zero-trans Efflux of 3-OMG out of Erythrocytes --- p.47 / Chapter 2.2.3 --- Glutl Protein Expression --- p.48 / Chapter 2.2.4 --- GLUT1 Gene Analyses --- p.51 / Chapter 2.2.5 --- Statistics --- p.58 / Chapter 2.3 --- Results --- p.59 / Chapter 2.4 --- Discussions and Conclusions --- p.69 / Chapter Chapter 3: --- Pathogenicity of GLUT1 Mutations --- p.78 / Chapter 3.1 --- Materials --- p.79 / Chapter 3.1.1 --- Construction of Glutl-Encoding Vectors --- p.79 / Chapter 3.1.2 --- Cell Lines --- p.80 / Chapter 3.1.3 --- "Cell Culture Media, Buffers and Other Reagents" --- p.81 / Chapter 3.1.4 --- Cell Culture Wares --- p.83 / Chapter 3.1.5 --- Reagents for Transfection --- p.83 / Chapter 3.1.6 --- Reagents for Protein Determination and Western Blot Analysis --- p.83 / Chapter 3.1.7 --- Reagents and Buffers for Flow Cytometry --- p.84 / Chapter 3.1.8 --- Reagents for 2-DOG Uptake in CHO-K1 Cells --- p.84 / Chapter 3.1.9 --- Reagents and Consumables for Confocal Microscopy --- p.85 / Chapter 3.2 --- Methods --- p.86 / Chapter 3.2.1 --- Cell Culture Methodology --- p.86 / Chapter 3.2.2 --- Construction of Glutl-Encoding Vectors --- p.87 / Chapter 3.2.3 --- Construction of Glutl Mutants --- p.91 / Chapter 3.2.4 --- Establishment of Wild Type and Mutant Glutl Expressing Cell Lines --- p.92 / Chapter 3.2.5 --- Glucose Influx Assays in CHO-K1 Cells --- p.96 / Chapter 3.2.6 --- Confocal Microscopy Studies on Glutl Cellular Localization --- p.97 / Chapter 3.2.7 --- Statistics --- p.98 / Chapter 3.3 --- Results --- p.99 / Chapter 3.4 --- Discussions and Conclusions --- p.112 / Chapter Chapter 4: --- Effects of Anticonvulsive Compounds on Cellular Glucose Transport --- p.117 / Chapter 4.1 --- Materials --- p.118 / Chapter 4.1.1 --- Cell Lines --- p.118 / Chapter 4.1.2 --- Cell Culture Media --- p.118 / Chapter 4.1.3 --- Blood Sample --- p.119 / Chapter 4.1.4 --- Anticonvulsive Compounds --- p.119 / Chapter 4.1.5 --- Reagents for Zero-trans Influx of 3-OMG Uptake in Fibroblasts --- p.120 / Chapter 4.1.6 --- Reagents for Zero-trans Influx of 2-DOG Uptake in Primary Astrocytes --- p.120 / Chapter 4.1.7 --- Reagents for Total RNA Isolation --- p.121 / Chapter 4.1.8 --- Reagents and Consumables for Real-Time PCR --- p.122 / Chapter 4.2 --- Methods --- p.123 / Chapter 4.2.1 --- Cell Culture --- p.123 / Chapter 4.2.2 --- Drug Concentrations --- p.123 / Chapter 4.2.3 --- Zero-trans Influx of 3-OMG Uptake in Erythrocytes --- p.123 / Chapter 4.2.4 --- Zero-trans Influx of 3-OMG Uptake in Fibroblasts --- p.124 / Chapter 4.2.5 --- Zero-trans Influx of 2-DOG Uptake in Primary Astrocytes --- p.125 / Chapter 4.2.6 --- Gene Expression Study --- p.127 / Chapter 4.2.7 --- Statistics --- p.130 / Chapter 4.3 --- Results --- p.131 / Chapter 4.4 --- Discussions and Conclusions --- p.148 / Chapter Chapter 5: --- General Conclusions and Future Perspectives --- p.154 / References --- p.157

Thesis. 1978. M.S.--Massachusetts Institute of Technology. Dept. of Chemical Engineering. / MICROFICHE COPY AVAILABLE IN ARCHIVES AND SCIENCE. / Bibliography: leaves 253-265. / M.S.

Objetivo: O objetivo desse estudo foi identificar a associação entre fatores clínicos e laboratoriais com o uso de insulina em gestantes com DMG no momento do diagnóstico e analisar os possíveis fatores preditores do uso de insulina. Método: Foram estudadas, de forma retrospectiva, 294 pacientes com diabetes melito gestacional (DMG) diagnosticado por meio do teste de tolerância à glicose oral de 100 gramas (TTGO-100g) entre 24 e 33 semanas completas de gestação, cujo seguimento pré-natal foi realizado ambulatorialmente pelo setor de Endocrinopatias e Gestação da Clínica Obstétrica do Hospital das Clínicas da Faculdade de Medicina da Universidade de São Paulo, no período de 1 de julho de 2002 a 30 de junho de 2008. Os seguintes fatores clínicos e laboratoriais, que pudessem estar associados ao uso de insulina para controle glicêmico, foram analisados: idade materna, obesidade pré-gestacional - índice de massa corpórea (IMC) > 30 Kg/m2, antecedente familiar de diabetes melito (DM), tabagismo, hipertensão arterial, uso de corticosteróides sistêmicos, antecedente obstétrico de DMG e de macrossomia fetal, nuliparidade, multiparidade, antecedente obstétricos de natimortos e neomortos, idade gestacional no momento do diagnóstico, gemelidade, índice de líquido amniótico (ILA) aumentado ILA > 18 cm, polidrâmnio (ILA > 25 cm), número de valores anormais do TTGO-100g, glicemia de jejum anormal no TTGO- 100g glicemia de jejum > 95 mg/dL; média das quatro glicemias aferidas no TTGO-100g; valor da glicemia de jejum, de 1ª, 2ª e 3ª horas do TTGO-100g e hemoglobina glicada (HbA1c). A associação entre cada fator e a necessidade de insulinoterapia foi analisada individualmente (2 de Pearson / teste exato de Fisher e teste t de Student). O modelo de regressão logística para a análise multivariada foi usado para predizer a probabilidade desses fatores em relação ao uso de insulina. Resultados: Das 294 pacientes avaliadas, 39,8% (117/294) necessitaram de insulinoterapia para controle glicêmico. Observou-se correlação positiva entre o uso de insulina e obesidade pré-gestacional, antecedente familiar de DM, hipertensão arterial, antecedente obstétrico de DMG e de macrossomia fetal, número de valores anormais no TTGO-100g, glicemia de jejum > 95 mg/dL no TTGO-100g; média das quatro glicemias aferidas no TTGO-100g; valor da glicemia de jejum, de 1ª, 2ª e 3ª horas do TTGO-100g e HbA1c pela análise univariada (P<0,05). Na análise do modelo de regressão logística foram desenvolvidos dois modelos que incluíam os seguintes fatores preditores do uso de insulina: obesidade pré-gestacional, antecedente familiar de DM, número de valores anormais no TTGO-100g (só modelo 1) e valor da glicemia de jejum do TTGO-100g (só modelo 2). Os dois primeiros modelos foram novamente analisados, incluindo-se a variável HbA1c para verificação de sua contribuição na predição do uso de insulina. Curvas de probabilidade e escores foram construídos com base nas quatro combinações de fatores preditores. Conclusões: É possível estimar a probabilidade do uso de insulinoterapia para controle glicêmico em gestantes com DMG por meio de IMC pré-gestacional, antecedente familiar de DM, número de valores anormais do TTGO-100g, valor da glicemia de jejum no TTGO-100g e da HbA1c. / Objective: To determine the association between clinical and laboratory parameters and insulin requirement in pregnancies complicated by gestational diabetes mellitus (GDM), and to evaluate possible factors predicting the need for insulin therapy. Methods: A total of 294 patients with GDM diagnosed by the 100- g/3-h oral glucose tolerance test (OGTT) between 24 and 33 complete weeks of gestation were retrospectively studied. These patients were under prenatal follow-up at the Obstetric Clinic of the University of Sao Paulo School of Medicine (HCFMUSP) between July 1, 2002 and June 30, 2008. The clinical and laboratory factors which could be associated to the need for insulin therapy were analyzed: maternal age, prepregnancy obesity body mass index (BMI) > 30 Kg/m2, family history of diabetes mellitus (DM), smoking, hypertension, use of systemic corticosteroids, prior GDM, prior fetal macrosomia, nulliparity, multiparity, prior stillbirth, prior neonatal death, gestational age at diagnosis of GDM, multiple pregnancy, elevated amniotic fluid index (AFI) AFI > 18 cm, polyhydramnios (AFI > 25 cm), number of abnormal 100-g/3-h OGTT values, 100-g/3-h OGTT fasting plasma glucose > 95 mg/dL, mean of the four 100-g/3-h OGTT values, 100-g/3-h OGTT fasting/one/two/three plasma glucose values, and glycated hemoglobin (HbA1c). The association between each factor and the need for insulin therapy was then analyzed individually (Pearsons chi-square/Fishers exact or Student t test). The performance of these factors to predict the probability of insulin therapy was estimated using a logistic regression model. Results: Among the 294 patients studied, 39.8% (117/294) required insulin for glycemic control. Univariate analysis showed a positive correlation between insulin therapy and prepregnancy obesity, family history of diabetes, hypertension, prior GDM, prior fetal macrosomia, number of abnormal 100-g/3-h OGTT values, 100-g/3-h OGTT fasting plasma glucose > 95 mg/dL, mean of the four 100-g/3-h OGTT values, 100-g/3-h OGTT fasting/one/two/three plasma glucose values, and HbA1c (P < 0.05). Two logistic regression models were developed and included the following parameters: prepregnancy obesity, family history of diabetes, number of abnormal 100-g/3-h OGTT values (just model 1) and 100-g/3-h OGTT fasting plasma glucose (just model 2). The two first models were analysed another time including the variable HbA1c to verify its contribution on prediction of the need for insulin therapy. Probability curves and scores were constructed based on the four combinations of predictive factors. Conclusions: The probability of insulin therapy can be estimated in pregnant women with GDM based on prepregnancy obesity, family history of diabetes, number of abnormal 100-g/3-h OGTT values, 100-g/3-h OGTT fasting plasma glucose, and HbA1c concentration.

Alzheimer’s disease (AD) is the most common cause of dementia, accounting for around 60-70% of cases. AD encompasses large-scale neuronal loss, resulting in progressive memory and other cognitive decline. Presently, there is no cure for dementia and in light of the ageing population demographic, this represents a clear unmet medical and socioeconomic challenge Worldwide. Much of the current AD research focuses on studying the brain once hallmark amyloid plaque and neurofibrillary tangle pathologies have presented. However their appearance is extremely end stage and to date, any therapeutic interventions aimed at alleviating them having failed to halt symptoms progression. It may therefore be beneficial to look for earlier changes, with metabolic and oxidative stress events as well as reduced cerebral metabolism thought to occur early on in disease progression. Evidence from rare, familial AD cases suggests a causative role for A in AD pathogenesis. For this reason, the enzyme beta-site amyloid precursor protein (APP) cleaving enzyme 1 (BACE1), the rate-limiting step in A production is currently of great therapeutic interest. With the prevailing view being that reducing BACE1 levels will be beneficial in AD, there remains a need to better understand the physiological roles of BACE1 to avoid potential side effects of BACE1 inhibition. Herein is presented data showing that, in agreement with the previous literature, BACE1 is fundamentally regulated by cell stress. Notably, both acute and prolonged bouts of oxidative and metabolic stress result in significant increases in BACE1 and APP protein expression. These changes also result in a shift in APP metabolism, with amyloidogenic processing of APP predominating during times of stress. It has also been shown that chronic elevation of BACE1 and/or manipulation of APP processing can alter cellular glucose uptake and use. These changes were determined through the use of radiolabelled substrate uptake and oxidation as well as extracellular flux assays. These data highlighted a fundamental shift in cellular metabolism, with aerobic glycolysis being utilised over oxidative metabolism of glucose. These changes were later shown to come as a result of metabolic lesions, which acted to impair substrate delivery to the electron transport chain of the mitochondria. Taken together, these data show that overexpression of the AD-associated protein BACE1 phenocopies a number of the earliest detectable changes observed in the brains of people who later develop AD. Finally, these data highlighted the potential importance of a number of novel pathways (Sirtuin, AMP-activated protein kinase, and peroxisome proliferator-activated receptor- coactivator signalling) that may underlie these changes and offer therapeutic avenues for earlier and more targeted treatment to halt AD progression.

Objetivo: O objetivo desse estudo foi identificar a associação entre fatores clínicos e laboratoriais com o uso de insulina em gestantes com DMG no momento do diagnóstico e analisar os possíveis fatores preditores do uso de insulina. Método: Foram estudadas, de forma retrospectiva, 294 pacientes com diabetes melito gestacional (DMG) diagnosticado por meio do teste de tolerância à glicose oral de 100 gramas (TTGO-100g) entre 24 e 33 semanas completas de gestação, cujo seguimento pré-natal foi realizado ambulatorialmente pelo setor de Endocrinopatias e Gestação da Clínica Obstétrica do Hospital das Clínicas da Faculdade de Medicina da Universidade de São Paulo, no período de 1 de julho de 2002 a 30 de junho de 2008. Os seguintes fatores clínicos e laboratoriais, que pudessem estar associados ao uso de insulina para controle glicêmico, foram analisados: idade materna, obesidade pré-gestacional - índice de massa corpórea (IMC) > 30 Kg/m2, antecedente familiar de diabetes melito (DM), tabagismo, hipertensão arterial, uso de corticosteróides sistêmicos, antecedente obstétrico de DMG e de macrossomia fetal, nuliparidade, multiparidade, antecedente obstétricos de natimortos e neomortos, idade gestacional no momento do diagnóstico, gemelidade, índice de líquido amniótico (ILA) aumentado ILA > 18 cm, polidrâmnio (ILA > 25 cm), número de valores anormais do TTGO-100g, glicemia de jejum anormal no TTGO- 100g glicemia de jejum > 95 mg/dL; média das quatro glicemias aferidas no TTGO-100g; valor da glicemia de jejum, de 1ª, 2ª e 3ª horas do TTGO-100g e hemoglobina glicada (HbA1c). A associação entre cada fator e a necessidade de insulinoterapia foi analisada individualmente (2 de Pearson / teste exato de Fisher e teste t de Student). O modelo de regressão logística para a análise multivariada foi usado para predizer a probabilidade desses fatores em relação ao uso de insulina. Resultados: Das 294 pacientes avaliadas, 39,8% (117/294) necessitaram de insulinoterapia para controle glicêmico. Observou-se correlação positiva entre o uso de insulina e obesidade pré-gestacional, antecedente familiar de DM, hipertensão arterial, antecedente obstétrico de DMG e de macrossomia fetal, número de valores anormais no TTGO-100g, glicemia de jejum > 95 mg/dL no TTGO-100g; média das quatro glicemias aferidas no TTGO-100g; valor da glicemia de jejum, de 1ª, 2ª e 3ª horas do TTGO-100g e HbA1c pela análise univariada (P<0,05). Na análise do modelo de regressão logística foram desenvolvidos dois modelos que incluíam os seguintes fatores preditores do uso de insulina: obesidade pré-gestacional, antecedente familiar de DM, número de valores anormais no TTGO-100g (só modelo 1) e valor da glicemia de jejum do TTGO-100g (só modelo 2). Os dois primeiros modelos foram novamente analisados, incluindo-se a variável HbA1c para verificação de sua contribuição na predição do uso de insulina. Curvas de probabilidade e escores foram construídos com base nas quatro combinações de fatores preditores. Conclusões: É possível estimar a probabilidade do uso de insulinoterapia para controle glicêmico em gestantes com DMG por meio de IMC pré-gestacional, antecedente familiar de DM, número de valores anormais do TTGO-100g, valor da glicemia de jejum no TTGO-100g e da HbA1c. / Objective: To determine the association between clinical and laboratory parameters and insulin requirement in pregnancies complicated by gestational diabetes mellitus (GDM), and to evaluate possible factors predicting the need for insulin therapy. Methods: A total of 294 patients with GDM diagnosed by the 100- g/3-h oral glucose tolerance test (OGTT) between 24 and 33 complete weeks of gestation were retrospectively studied. These patients were under prenatal follow-up at the Obstetric Clinic of the University of Sao Paulo School of Medicine (HCFMUSP) between July 1, 2002 and June 30, 2008. The clinical and laboratory factors which could be associated to the need for insulin therapy were analyzed: maternal age, prepregnancy obesity body mass index (BMI) > 30 Kg/m2, family history of diabetes mellitus (DM), smoking, hypertension, use of systemic corticosteroids, prior GDM, prior fetal macrosomia, nulliparity, multiparity, prior stillbirth, prior neonatal death, gestational age at diagnosis of GDM, multiple pregnancy, elevated amniotic fluid index (AFI) AFI > 18 cm, polyhydramnios (AFI > 25 cm), number of abnormal 100-g/3-h OGTT values, 100-g/3-h OGTT fasting plasma glucose > 95 mg/dL, mean of the four 100-g/3-h OGTT values, 100-g/3-h OGTT fasting/one/two/three plasma glucose values, and glycated hemoglobin (HbA1c). The association between each factor and the need for insulin therapy was then analyzed individually (Pearsons chi-square/Fishers exact or Student t test). The performance of these factors to predict the probability of insulin therapy was estimated using a logistic regression model. Results: Among the 294 patients studied, 39.8% (117/294) required insulin for glycemic control. Univariate analysis showed a positive correlation between insulin therapy and prepregnancy obesity, family history of diabetes, hypertension, prior GDM, prior fetal macrosomia, number of abnormal 100-g/3-h OGTT values, 100-g/3-h OGTT fasting plasma glucose > 95 mg/dL, mean of the four 100-g/3-h OGTT values, 100-g/3-h OGTT fasting/one/two/three plasma glucose values, and HbA1c (P < 0.05). Two logistic regression models were developed and included the following parameters: prepregnancy obesity, family history of diabetes, number of abnormal 100-g/3-h OGTT values (just model 1) and 100-g/3-h OGTT fasting plasma glucose (just model 2). The two first models were analysed another time including the variable HbA1c to verify its contribution on prediction of the need for insulin therapy. Probability curves and scores were constructed based on the four combinations of predictive factors. Conclusions: The probability of insulin therapy can be estimated in pregnant women with GDM based on prepregnancy obesity, family history of diabetes, number of abnormal 100-g/3-h OGTT values, 100-g/3-h OGTT fasting plasma glucose, and HbA1c concentration.

Changes in cognitive status occur with aging and significant attention has been placed on developing interventions to possibly delay cognitive decline and identifying risk factors that exacerbate cognitive deficits. One issue that arises when studying interventions is that they do not always effectively translate from animal models to human subjects. When testing potentially modifiable risk factors related to cognitive impairment, more sensitive metrics could help in identifying targets for intervention at earlier time-points. Therefore, the aims of the current dissertation were twofold. The first study examined the ability to translate between species using a common behavioral paradigm, the Morris water maze (MWM). The second study evaluated human MWM performance and commonly used neuropsychological test performance in relation to a marker of glucose regulation, HbA1c. The first study tested translation between rats and humans using the MWM paradigm. Using secondary data sources from a study of nutrition, inflammation, and aging among rats and the Czech Brain Aging Study among humans, differences in average performance and across trial learning were examined between young (3 months; n=10) and aged (20 months; n=13) rats as well age young-old (age 53-70; n=47) and old-old (age 71-85; n=30) human subjects. The cumulative distance was measured in rats and distance error to the hidden goal was measured in human subjects. Results indicated that age-related deficits in performance are greater in magnitude for rats than human subjects. Further, the across trial learning data is more sensitive to change in performance than average performance metrics. Across learning trials indicated poorer performance for aged rats than young rats. Significant effects of age were also observed for human subjects using with the allocentric and egocentric subtests. The second study examined the influence of a measure of glucose regulation (HbA1c) on commonly used neuropsychological tests and a test of spatial navigation abilities among human subjects. Participants classified as cognitively normal, subjective cognitive decline, amnestic mild cognitive impairment (aMCI), and Alzheimer’s disease (AD) were evaluated on verbal memory, nonverbal memory, working memory, visuospatial skills, and executive function in addition to the virtual and real space versions of a human MWM paradigm. A total of 116 participants were included in the complete data sample and 133 participants were included in the multiple imputation sample. Results indicated that HbA1c influenced executive function but not any other measures of cognition. Higher HbA1c levels were associated with poorer performance. A significant interaction was observed between cognitive status and HbA1c. Those with cognitive impairment and higher HbA1c levels had poorer executive function performance. This effect, however, was not observed with the imputation sample. Results of the first study indicated that the MWM paradigm serves as a good tool to assess translation between rats and human subjects. This would be helpful in examining interventions designed to improve normal age-related changes in cognition. It is important to note, however, that the differences observed among animals tend to be greater than human subjects. Therefore, the margin of improvement may be greater following treatment with studies utilizing animals rather than human subjects. The second study indicated that glucose levels may have an impact on cognitive abilities, particularly those related to executive function. Targeting blood glucose levels may be one effective way to keep executive function abilities more intact with age. Taken together, these studies will better inform future work related to delaying cognitive decline among older adults.

Glukosetransporter spielen eine wichtige Rolle in der Versorgung des Gehirns mit Nährstoffen und somit für den Erhalt der physiologischen Zellintegrität. Glukose wird über die Blut-Hirn-Schranke (BHS) mittels spezifischen transmembranen Transportproteinen der SLC-Genfamilie (GLUT, SGLT) befördert. Dabei scheint während physiologischen Bedingungen hauptsächlich der Glukosetransporter GLUT1 (SLC2A1) für die Energieversorgung des Gehirns zuständig zu sein. Die Erforschung der SGLT-Expression ist in den letzten Jahren ein wichtiger Ansatzpunkt für neue Behandlungsstrategien vieler Erkrankungen, wie Diabetes Mellitus, maligne Neoplasien oder eines Herzinfarkts, geworden. Jedoch ist über deren Expression und Funktion im menschlichen Gehirn nur wenig bekannt. Besonders die Lokalisation entlang der BHS bleibt fraglich. Ein Großteil bisheriger Forschungsarbeiten beschäftigt sich hauptsächlich mit der Expressionsanalyse des Transporters SGLT1 im tierischen Gehirn in vivo (Poppe et al. 1997; Balen et al. 2008; Yu et al. 2013). Es konnte aufgezeigt werden, dass SGLT1 mRNA exklusiv in Neuronen und nicht an der BHS exprimiert wird. Dies wird durch in vitro Analysen einer humanen Hirnendothelzelllinie bestätigt. Demnach kann kein SGLT1 unter physiologischen Bedingungen nachgewiesen werden (Sajja et al. 2014). Im menschlichen Hirngewebe besitzen SGLTs somit keine zentrale Funktion für den Glukosetransport an der BHS. Im Gegensatz dazu konnte eine Expression von SGLT sowohl in vivo als auch in vitro während hypoglykämischen Bedingungen belegt werden (Vemula et al. 2009; Sajja et al. 2014). Die Expression der SGLT-Transporter während einer ischämischen Hypoglykämie führt zu der Annahme, dass diese Transporter für die Aufrechterhaltung der Energieversorgung des geschädigten Hirngewebes notwendig sind. Um die physiologischen Mechanismen nach einem Glukosemangel zu untersuchen, wurden SHT-Modelle etabliert (Salvador et al. 2013). In einem experimentellen Modell des Schädel-Hirn-Traumas im Rahmen eines DFG-gefördertes Projekts ist ein Expressionsverlauf von Glukosetransportern im Maushirn und in Hirnendothelzellen erarbeitet worden (Wais 2012; Salvador et al. 2015). Somit könnten SGLTs als Ansatzpunkt für den Nachweis der Überlebenszeit nach einem SHT fungieren. Die vorliegende Arbeit fokussiert sich auf die Expression der Natrium-abhängigen Glukosetransporter SGLT1 und SGLT2 im menschlichen Gehirn. Hierbei liegt das Hauptaugenmerk auf der Lokalisation dieser Transporter an der menschlichen BHS von post mortalem Hirngewebe. Weiterhin wird untersucht ob die Expressionsstärke von SGLT1 und SGLT2 eine Aussage über die Überlebenszeit von Verstorbenen nach einer traumatisch bedingten Hirnveränderung zulässt. Die Lokalisation von SGLT1 und SGLT2 an der menschlichen BHS konnte durch die Etablierung eines Protokolls zur Isolation von Hirnkapillaren erfolgen. Vorab wurden alle verwendeten Antikörper auf ihre Spezifität mittels siRNA Transfektion und Blockierung der Immunfluoreszenzsignale mittels immunisierten Peptids getestet. Somit ist die Spezifität der detektierten SGLT1- und SGLT2-Expression in menschlichen Hirnkapillaren gewährleistet. Anschließend wird untersucht, in welchen zeitlichem Verlauf nach einer traumatisch bedingten Hirnveränderung die verschiedenen Formen der Glukosetransporter exprimiert werden und ob ggf. der Umfang und die Verteilung von SGLT1, SGLT2 und GLUT1 sowie das Verhältnis zueinander Auskünfte über eine vitale bzw. postmortale Entstehung eines Traumas bzw. dessen Überlebenszeit zulässt. Hierfür wird ein Expressionsschema der Glukosetransporter generiert, abhängig von Todeszeitpunkt und Todesursache. Es konnte festgestellt werden, dass GLUT1 nicht als Target für die Ermittlung der Überlebenszeit nach einem Trauma geeignet ist. Dahingegen zeigen SGLT1 und SGLT2 eine signifikante Änderung der Expressionsstärke im contusionalen Gewebe in Abhängigkeit von der Überlebenszeit. Obwohl diese vorläufigen Daten einen neuen Ansatzpunkt für die forensische Fragestellung aufzeigen, müssen weitere Experimente mit einem erhöhten Umfang der Probenanzahl und kürzere Zeitspannen der Überlebenszeiträume durchgeführt werden. / The transport of glucose across the endothelial cells of the human blood-brain barrier (hBBB) plays a major role for energy supply of the brain and therefor for neuronal integrity. Glucose enters the brain cells through specific transmembrane transporter proteins of the SLC-gene family (GLUT, SGLT). Under physiological conditions glucose uptake across the BBB seems to be mediated primarily by facilitated diffusion through glucose transporter 1 (GLUT1). Although SGLTs are a known drug target for diabetes and furthermore play a role in other disease like cancer and cardiac ischemia, active glucose transport by SGLTs is hardly observed and very little is known about their expression or activity in human brain. Especially the function along the BBB remains uncertain. Up to now, expression analysis focused on SGLT1 and has been confirmed in vivo by analyzing brain tissue of animals (Poppe et al. 1997; Balen et al. 2008; Yu et al. 2013). Here detection mainly occurs in neurons, no SGLT1 mRNA in capillaries of the BBB could be found. Similarly in vitro experiments with a human brain microvascular endothelial cell line reveals no expression of SGLT1 under physiological conditions (Sajja et al. 2014). In human brain, SGLT1 is hardly expressed and so far could not be found along the BBB. In contrast to these findings, expression of SGLT1 could be detected in vivo as well as in vitro under hypoglycemic conditions (Vemula et al. 2009; Sajja et al. 2014). The occurrence of these transporters during ischemic hypoglycemia could lead to the conclusion that the secondary active glucose transport by SGLTs is necessary for additional glucose supply in injured brain. To investigate if SGLTs are required for the reconstruction of energy supply after glucose deficiency, traumatic brain injury (TBI) models were established to study secondary physiological mechanisms along the BBB (Salvador et al. 2013). In an experimental CCI (controlled cortical impact) mouse model within a DFG-funded project, an expression pattern of glucose transporters in the mouse brain and in brain endothelial cells has been developed (Wais 2012; Salvador et al. 2015). Thus it could lead as a Target for evidence of the time of survival after TBI. This study focuses on the sodium-dependent glucose transporters SGLT1 and SGLT2 expression in human brain. The main topic is to localize the sodium-dependent glucose transporters along the human BBB of post mortem brain tissue and to examine whether SGLT expression allow a conclusion to be drawn about the survival time of a patient after TBI. First of all the localization of SGLT1 and SGLT2 at the human BBB could be shown by establishment a capillary isolation protocol of human post mortem brain tissue. Therefore the antibody specificity was tested by a siRNA transfection protocol and blocking the immunofluorescence signal with an immunized peptide. Thus, specific SGLT1 and SGLT2 expression at the endothelial lining of the capillary lumen could be demonstrated. After attaching the value of SGLTs at the human BBB, the relationship of the glucose transporter expression in TBI tissue according to the survival time of the patient is presented. Hereby it should be clarified whether the expression and distribution of the transporters GLUT1, SGLT1 and SGLT2 as well as the relation to each other provide information on a vital or post mortal development of a trauma or its survival time. It could determine that GLUT1 is not suitable as a target for the representation of survival time after TBI. However, SGLT1 and SGLT2 show a significant change in the expression profile of traumatic brain regions. Here an increase according to the survival time after trauma can be shown. Although these preliminary data suggest a novel target for forensic questions, more experiments with an increased scope of survival time frames should be carried out.

No description available.