A doença de Unverricht-Lundborg (DUL) é considerada uma doença rara, autossômica recessiva, sendo também denominada de Epilepsia Mioclônica Progressiva do tipo1 (emp1), causada por mutações no gene codificador (CSTB) da proteína cistatina B. A cistatina B é uma proteína essencial para a regulação dos processos fisiológicos do ser humano, e sua expressão reduzida parece ser a causa primária da EMP1. A doença em geral se inicia entre os seis e dezesseis anos, manifestando-se tanto como crises mioclônicas como por crises tônicoclônicas generalizadas. Trata-se de uma doença grave e limitante, cujo diagnóstico preciso é extremamente importante para as condutas apropriadas, incluindo aconselhamento genético. Este estudo tem como objetivo o estudo molecular e caracterização da expansão instável de repetição dodecamérica (CCCCGCCCCGCG) da região promotora 5\' não traduzida do gene CSTB entre pacientes com suspeita de EMP1 na população brasileira. No presente estudo, selecionamos 64 pacientes entre eles 54 casos índices do Hospital das Clínicas da Faculdade de Medicina de Ribeirão Preto (HCFMRP-USP) com suspeita de EMP1. Os restantes 10 casos eram parentes dos casos índices. Os 54 pacientes foram seguidos no setor de Epilepsia com diagnostico clinico e eletrofisiológico de EMP1, e foram encaminhados para o setor de Neurogenética para diagnostico molecular. Destes 54 casos índices, apenas 5 foram diagnosticados através da biologia molecular com expansão dodecamera acima de 30 repetições, sugestivo de DUL. Espera-se, com este estudo, identificar a população de pacientes com EPM1 que tenham mutação no gene CSTB e com os resultados de este projeto possibilitar assim um melhor entendimento da etiopatogenia e proporcionar um diagnóstico preciso dos casos de DUL. / Unverricht-Lundborg disease (ULD) is considered a rare autosomal recessive disease, also known as Progressive Myoclonic Epilepsy Type 1 (EMP1), caused by mutations in the CSTB gene, which provides instructions for making a protein called Cystatin B. Such protein is essential for regulating a person\'s physiological processes, and its reduced expression seems to be the first cause of EMP1. Affected individuals usually begin showing signs and symptoms of the disorder between the ages of 6 and 16, which manifests both as myoclonus or as generalized tonic-clonic crises. It is a grave and limiting condition whose precise diagnosis is extremely important for appropriate conducts, including genetic counseling. This study has as a goal the molecular evaluation and characterization of the unstable expansion of the decametric repetition (CCCCGCCCCGCG) from the non-translated CSTB\'s 5\' gene promoter region among Brazilian patients with suspected EMP1. In this study, 64 patients were selected, among them 54 key figures from Hospital das Clínicas of Faculdade de Medicina de Ribeirão Preto (HCFMRP-USP) with suspected EMP1. The other 10 figures were related to the key figures. The 54 patients were followed in the Epilepsy sector with EMP1 clinical and electrophysiological diagnostics and were forwarded to the neurogenetics sector for a molecular diagnostic. From such 54 key figures, only 5 were diagnosed through molecular biology with decametric expansion above 30 repetitions, suggestive of ULD. This study is aimed to identify the EPM1 patients with CSTB genetic mutation and hopefully the results of this identification will enable a better understanding of the etiopathogeny and provide with a exact diagnosis of ULD cases.

Spinal Muscular Atrophy with Progressive Myoclonic Epilepsy (SMA-PME), Farber Lipogranulomatosis (FL), and a rare variant form of Neonatal Progeroid Syndrome (NPS) are three monogenetic rare disorders caused by pathogenic variation in genes encoding lipid modifying proteins. FL and SMA-PME are caused by loss of function mutations in ASAH1, encoding the acid ceramidase (aCDase) enzyme. It is not, however, known how aCDase deficiency can produce either the isolated neurological symptoms of SMA-PME or the predominantly systemic symptoms of FL. Further, a recently identified variant form of NPS has been attributed to variants in ANO6, encoding a dual function calcium-activated chloride channel and glycerophosphoserine (GPS) scramblase. Here, it is not known how ANO6 mutation causes the premature aging phenotype that defines NPS. To address these questions, I sought to elucidate pathogenic changes in lipid metabolism that associate clinical phenotype. I show here that the different patient mutations in ANO6 cause a non-physiological gain of channel function and either a loss or gain of scramblase function depending on the variant expressed. Both variants, however, alter GPS metabolic homeostasis suggesting a common mechanism of action. To provide in vivo insight, I characterized a novel mouse model based on our NPS patient genetics, showing extremely low penetrance of disease symptoms in terms of live births yet confirming that affected animals show impaired GPS metabolism in affected organs. Next, I characterized the clinical presentation of six new patients with SMA-PME and identified distinct sphingolipid metabolic fingerprints in FL and SMA-PME cells. I show that FL is defined by a hypometabolic sphingolipid phenotype with cellular and molecular features of a classic lysosomal storage disorder. By contrast, SMA-PME has a hypermetabolic sphingolipid phenotype with features of non-classic lysosomal trafficking disorders. To provide clinical insight, I assessed the potential of enzyme replacement therapy, demonstrating a rescue of sphingolipid metabolism in SMA-PME patient cells. Together, this thesis identified changes in the cellular and tissue lipid profiles of patients with ANO6-NPS, SMA-PME, or FL, elucidating some of the lipid-centric pathomechanisms of these diseases.

Orientador: Gabriela Castellano / Tese (doutorado) - Universidade Estadual de Campinas, Instituto de Fisica Gleb Wataghin / Made available in DSpace on 2018-08-15T22:34:24Z (GMT). No. of bitstreams: 1 Oliveira_MarciaSilvade_D.pdf: 17533323 bytes, checksum: 32a83eb4815b68f061baa91a4b0ab2e4 (MD5) Previous issue date: 2010 / Resumo: Análise de textura em imagens digitais é um termo que se refere a um grupo de técnicas de processamento de imagens que objetivam a extração de descritores da imagem ou de regiões de interesse (ROIs) de forma a simplificar a caracterização das mesmas. A textura pode ser entendida como características intrínsecas da imagem (por exemplo: brilho, cor e distribuição de formas) que remetem à ideia de regularidade, rugosidade, suavidade, entre outras, por isso o nome 'textura'. Um tipo particular de análise de textura, bastante utilizado em imagens médicas, se baseia em medidas estatítsticas relativas à distribuição de níveis de cinza da imagem (matriz de coocorrência). Os descritores de textura, baseados na matriz de coocorrência, são conhecidos como descritores de Haralick. Este trabalho consistiu na aplicação de vídeo deste tipo de análise em imagens de Tomografia Computadorizada (TC) de vítimas de Acidente Vascular Cerebral Isquêmico e em imagens de Ressonância Magnetica (RM) de portadores de Epilepsia Mioclônica Juvenil, Doença de Machado-Joseph, Déficit Cognitivo Leve e Doença de Alzheimer, visando o desenvolvimento de uma ferramenta computacional que auxilie o neurologista na identificação de areas atingidas por estas doenças e que não sejam distinguíveis em uma análise visual. Neste trabalho foram selecionadas regiões de interesse (ROIs) e calculados os parâmetros de textura para cada grupo de imagens. Após o cálculo dos descritores de textura foi realizada uma análise estatística para verificar se havia diferenciação entre os vários tipos de tecidos. Os resultados obtidos mostraram que a análise de textura pode, de fato, ser utilizada para a extração de características discriminantes, tanto nas imagens de TC quanto nas imagens de RM para as cinco patologias analisadas / Abstract: Texture analysis in digital images is a term that refers to a group of image processing techniques that aim to extract descriptors of the image or regions of interest (ROIs) in order to simplify their characterization. Texture may be understood as intrinsic characteristics of the image (such as brightness, color and distribution of forms) that refer to the idea of regularity, roughness and smoothness, hence the name 'texture'. A particular type of texture analysis, widely used in medical imaging, is based on statistical measurements related to the image gray level distribution (coocurrence matrix). The texture descriptors based on the coocurrence matrix are known as Haralick descriptors. This work consisted on applying this type of analysis to computed tomography (CT) images of victims of Ischemic Stroke and magnetic resonance images (MRI) of patients with Juvenile Myoclonic Epilepsy, Machado-Joseph disease, mild cognitive impairment and Alzheimer's disease, in order to develop a computational tool to assist neurologists in the identification of areas affected by these diseases and which are not perceived in a visual analysis. In this work we selected regions of interest (ROIs) and calculated the texture parameters for each group of images. After the calculation of the texture descriptors, a statistical analysis was performed to determine whether there was differentiation between the various types of tissues. The results showed that texture analysis can indeed be used for the extraction of discriminant features in both the CT an the MR images for the five studied pathologies / Doutorado / Física / Doutora em Ciências